<%BANNER%>

Evaluating the Effects of the University of Florida Literacy Initiative (UFLI) on the Reading Skills of Spanish-Speaking...

Permanent Link: http://ufdc.ufl.edu/UFE0042039/00001

Material Information

Title: Evaluating the Effects of the University of Florida Literacy Initiative (UFLI) on the Reading Skills of Spanish-Speaking English Language Learners
Physical Description: 1 online resource (275 p.)
Language: english
Creator: Arriaza De Allen, Stephanie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: ell, intervention, reading, spanish
Special Education, School Psychology and Early Childhood Studies -- Dissertations, Academic -- UF
Genre: Special Education thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Many English language learners (ELLs) in schools experience reading difficulties, particularly Spanish-speaking students. This is a serious problem given the tendency for struggling readers to fall further behind as they advance in school. This issue becomes more pressing as the number of ELLs enrolled in schools rises and the availability of native language instruction diminishes. Fortunately, there is evidence that English reading interventions that have proven to be effective for struggling native English speakers are also effective for ELLs with different levels of English oral proficiency. Given that the literature is scarce, it is important to investigate the impact of other reading interventions on the reading outcomes of ELLs. The purpose of this study was to examine the effectiveness of a modified version of the University of Florida Literacy Initiative tutoring program in promoting the reading skills of second-grade Spanish-speaking ELLs who are struggling to read. The program was modified to include small-group instruction and practices that supported the language needs of ELLs. A multiple-baseline across groups design was used to examine students response to the intervention, as measured by the rate of correct pseudowords (CPPM) and sight words (CSPM) read per minute. Results showed that all groups improved from baseline to intervention in their rates of CPPM and CSPM and maintained appropriate rates two weeks after the intervention ceased. In addition, all groups showed a marked improvement in book reading accuracy of at least one half s year progress and up to a year and a half s progress in a limited number of sessions. Students improved reading skills were further corroborated by pre- and post-intervention measures of decoding, word recognition, fluency, and comprehension, as well as teachers ratings of students reading abilities and classroom behaviors. After intervention, students also demonstrated more positive attitudes towards reading. Most importantly, the intervention proved effective for all students regardless of their level of English oral proficiency. Social validity measures showed that the UFLI program was regarded as important, effective, and feasible among participants and teachers. These results are consistent with available literature that supports the implementation of early English interventions for ELLs.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Stephanie Arriaza De Allen.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Lane, Holly B.
Local: Co-adviser: Jones, Hazel A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042039:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042039/00001

Material Information

Title: Evaluating the Effects of the University of Florida Literacy Initiative (UFLI) on the Reading Skills of Spanish-Speaking English Language Learners
Physical Description: 1 online resource (275 p.)
Language: english
Creator: Arriaza De Allen, Stephanie
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: ell, intervention, reading, spanish
Special Education, School Psychology and Early Childhood Studies -- Dissertations, Academic -- UF
Genre: Special Education thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Many English language learners (ELLs) in schools experience reading difficulties, particularly Spanish-speaking students. This is a serious problem given the tendency for struggling readers to fall further behind as they advance in school. This issue becomes more pressing as the number of ELLs enrolled in schools rises and the availability of native language instruction diminishes. Fortunately, there is evidence that English reading interventions that have proven to be effective for struggling native English speakers are also effective for ELLs with different levels of English oral proficiency. Given that the literature is scarce, it is important to investigate the impact of other reading interventions on the reading outcomes of ELLs. The purpose of this study was to examine the effectiveness of a modified version of the University of Florida Literacy Initiative tutoring program in promoting the reading skills of second-grade Spanish-speaking ELLs who are struggling to read. The program was modified to include small-group instruction and practices that supported the language needs of ELLs. A multiple-baseline across groups design was used to examine students response to the intervention, as measured by the rate of correct pseudowords (CPPM) and sight words (CSPM) read per minute. Results showed that all groups improved from baseline to intervention in their rates of CPPM and CSPM and maintained appropriate rates two weeks after the intervention ceased. In addition, all groups showed a marked improvement in book reading accuracy of at least one half s year progress and up to a year and a half s progress in a limited number of sessions. Students improved reading skills were further corroborated by pre- and post-intervention measures of decoding, word recognition, fluency, and comprehension, as well as teachers ratings of students reading abilities and classroom behaviors. After intervention, students also demonstrated more positive attitudes towards reading. Most importantly, the intervention proved effective for all students regardless of their level of English oral proficiency. Social validity measures showed that the UFLI program was regarded as important, effective, and feasible among participants and teachers. These results are consistent with available literature that supports the implementation of early English interventions for ELLs.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Stephanie Arriaza De Allen.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Lane, Holly B.
Local: Co-adviser: Jones, Hazel A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0042039:00001


This item has the following downloads:


Full Text

PAGE 1

1 EVALUATING THE EFFECTS OF THE UNIVERSITY OF FLORIDA LITERACY INITIATIVE (UFLI) ON THE READING SKILLS OF SPANISH-SPEAKING ENGLISH LANGUAGE LEARNERS By STEPHANIE LISET A RRIAZA DE ALLEN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Stephanie Liset Arriaza de Allen

PAGE 3

3 To my loving husband Matt, dad Luis Francis co, mom Yolanda and loving family Thank you for all your love and support

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my gratitude to everyone who su pported me in the completion of my doctoral program. Most im portantly, I thank God and Jesus Christ for all the blessings I have received in my life. May this accomplishment be used for Their glory. I also want to thank our Holy Mo ther for Her uncondition al love and constant prayers. A special thanks goes to my husband, Matt, whose love and support encouraged me every step of the way. I am thankful for the many hours he stay ed by my side while I worked through the night, and for the long walks we took when I was stressed. I am also thankful for the words of enc ouragement he gave me when I felt exhausted, and for the countless times he held me close when I felt overwhelmed. I also want to thank my mom and dad for in stilling in me a passion for learning, for believing in me, and for always supporting my endeavors. They taught me the value of hard work and perseverance. Thanks go to my family and friends, as well, for their guidance and unceas ing encouragement. I thank Holly Lane, chair of my doctora l committee, for being a mentor and a dear friend. She was a source of inspiration to me because of her love of learning and her passion for helping children. I appreciate the sincere kindness and immeasurable generosity she showed me as I completed my doctoral program. I want to thank Hazel J ones, co-chair of my doctora l committee, for guiding me and helping me develop my knowledge of single-subject design. I am also grateful for my other committee members, Roxanne Hudson, Linda Lombardino, and Erica McCray. Their knowledge and expertise greatly contributed to my dissertation.

PAGE 5

5 Special thanks go to Penny Cox who was t he first person to suggest that I should pursue a doctoral degree and who guided me ever y step of the way. I also want to thank Shaira, Michell and Vicki for all the help they provided me throughout the years. They opened their hearts to me and became part of my life. I will always cherish their friendship. I also want to express my appreciati on to Donna Long, principal of Suwannee Elementary School and all t he school personnel, for welcoming me to the school and facilitating the completion of this study. T hanks go to all the wonderful students I had the privilege of tutoring. Their hard work and ent husiasm for learning were a source of inspiration. Finally, I want to thank all the graduat e students who helped me complete data collection, including Susan Helvenston, Maria Rodriguez, Yulia Aguilar, and Lindsay Bell. They always had a positive dispositi on and a great attitude toward the students. Their assistance was priceless.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDG MENTS .................................................................................................. 4LIST OF TABLES............................................................................................................ 9LIST OF FIGURES........................................................................................................ 10ABSTRACT................................................................................................................... 11CHAPTER 1 INTRODUCTION.................................................................................................... 13Statement of the Pr oblem....................................................................................... 14Ehris Phases of Word Reading Development........................................................ 17Pre-Alphabeti c Phase....................................................................................... 18Partial-Alphabet ic Phase.................................................................................. 18Full-Alphabeti c Phase...................................................................................... 19Consolidated-Alph abetic P hase....................................................................... 20Automatic Phase.............................................................................................. 21The Role of Instruction in Prom oting Word Readi ng Development......................... 21Purpose of the Study.............................................................................................. 242 REVIEW OF TH E LITERA TURE ............................................................................ 25Reading Re covery.................................................................................................. 26Direct Instruction..................................................................................................... 34Core Intervention Model................................................................................... 35Reading Mastery and Co rrective R eading........................................................ 38Proactive Reading............................................................................................ 43Read We ll......................................................................................................... 52Multiple Direct Instruction Inte rventions............................................................ 57Other Reading In tervent ions................................................................................... 59Word Reading In tervent ions............................................................................. 60Comprehensive Readi ng Interv entions............................................................ 62Discuss ion.............................................................................................................. 70Instruction in Wo rd Reading Skills.................................................................... 71Language and Reading Instruct ion................................................................... 72Strategies for E nglish Language Learners....................................................... 74Implications for Res earch................................................................................. 743 METHODS.............................................................................................................. 77Setti ng..................................................................................................................... 77Participants............................................................................................................. 79

PAGE 7

7 Assessment In strume nts......................................................................................... 94Screening In struments..................................................................................... 94Grouping Inst ruments....................................................................................... 99Supplemental Preand Post-I ntervention In struments................................... 100Intervent ion........................................................................................................... 106Early Literacy Components Addressed in UFLI.............................................. 107Effective Instructional Practices for ELLs....................................................... 109UFLI Se ssions................................................................................................ 110Materi als......................................................................................................... 115Dependent Variable.............................................................................................. 117Design................................................................................................................... 119Baseli ne.......................................................................................................... 120Intervent ion..................................................................................................... 120Maintenance................................................................................................... 121Tutor/Researcher........................................................................................... 122Data A nalys is........................................................................................................ 122Supplemental Data............................................................................................... 123Preand PostIn terventi on Data.................................................................... 123Book Levels duri ng Interv ention..................................................................... 124Interobserver Agr eement...................................................................................... 124Treatment Integrity................................................................................................ 125Social Va lidity....................................................................................................... 126Delimitations and Limita tions................................................................................ 128Delimita tions................................................................................................... 128Limita tions...................................................................................................... 129Description of One-onOne Tutoring Methodology............................................... 130Pilot Study............................................................................................................. 1314 RESULT S............................................................................................................. 133Dependent Variables............................................................................................ 133Baseline Phase.............................................................................................. 134Intervention Phase......................................................................................... 135Maintenance Phase........................................................................................ 137Summary of Depende nt Variables.................................................................. 138Total Number of Se ssions..................................................................................... 142Supplemental Data............................................................................................... 144Measures of Early Literacy Skills and Attitudes to wards Reading.................. 144Book Levels Read dur ing Intervention............................................................ 150Summary of Suppl emental Data..................................................................... 153Social Va lidity....................................................................................................... 153Student Inte rview............................................................................................ 154Teacher Ques tionnaire................................................................................... 154Summary of One-on-O ne Tutoring Data............................................................... 156Dependent Variables...................................................................................... 157Supplemental Data......................................................................................... 158Summary of Findi ngs............................................................................................ 160

PAGE 8

8 5 DISCUSSION....................................................................................................... 162Overview of the Study........................................................................................... 162Summary of Findi ngs............................................................................................ 164Interpretation of Fi ndings...................................................................................... 165Progression through Phases of Word Recognition Development................... 167Relation between CPPM and CSPM.............................................................. 174Interpretation of Supplemental Data..................................................................... 175Early Lite racy Sk ills........................................................................................ 175Attitudes towa rd Reading............................................................................... 178Teacher Reading Ability Rating Scal e (RARS)............................................... 180Social Va lidity....................................................................................................... 181Students Re sponses..................................................................................... 181Teachers Respons es..................................................................................... 183Effectiveness of the UF LI Tutori ng Progr am......................................................... 185Early Intervention and Language Profic iency................................................. 185Effective Instructional Practices for ELLs....................................................... 187Consistency with t he Response to Inte rvention A pproach.............................. 189Implications for Future Re search.......................................................................... 191Implications for Practice........................................................................................ 196Conclu sion............................................................................................................ 198APPENDIX A IRB DOCUME NTATION ....................................................................................... 200B HOME LANGUAGE SURVEY A ND REMINDER LETTERs ................................. 211C TEACHERS READING ABILITY RATING SC ALE (R ARS) ................................. 215D UFLI SMALL GROU P SESSION GUIDE .............................................................. 216E UFLI SMALL-GROU P SESSION NOTES ............................................................. 217F PSEUDOWORD PROBES ................................................................................... 219G SIGHT WORD PROBES ....................................................................................... 234H TREATMENT INTEGR ITY CHECK LIST ............................................................... 249I SOCIAL VALIDITY INSTRUM ENTS ..................................................................... 251J UFLI INDIVIDUAL TUTO RING SESSION GUIDE ................................................ 255LIST OF RE FERENCES............................................................................................. 256BIOGRAPHICAL SKETCH.......................................................................................... 274

PAGE 9

9 LIST OF TABLES Table page 3-1 Group com position ............................................................................................. 813-2 Summary of participants demographic in formation............................................ 823-3 Summary of participants instructional informa tion............................................. 843-4 Participants reading scores used for sele ction pur poses................................... 853-5 Students English and Spanish language levels based on the WMLS-R............ 853-6 Conners Abbreviated Teac her Rating Scal e (C-ATRS)..................................... 863-7 Running records result s used for groupi ng pur poses....................................... 1003-8 Treatment integrity............................................................................................ 1263-9 Running records re sults used fo r Miriam.......................................................... 1304-1 Individual rates of C PPM and CSPM dur ing basel ine....................................... 1354-2 Individual rate of CPPM and CSPM during intervent ion................................... 1374-3 Individual rate of CPPM and CSPM duri ng maintenance................................. 1394-4 Total number of tutoring se ssions received by each student............................ 1444-5 Preand post-intervention TOWR E raw scores and per centiles ranks............. 1454-6 Preand post-intervention KTEA-II raw scores and per centiles ranks............. 1464-7 Preand post-intervention DIBE LS Nonsense Word Fluency scores............... 1464-8 Pre-intervention DIBELS Oral Reading Fl uency scores................................... 1474-9 Post-intervention DIBELS Oral Reading Fl uency scores.................................. 1474-10 Pre/Post-intervention QR I-4 comprehensio n raw scores.................................. 1494-11 ERAS Pre-and post-intervention raw scores and perce ntile ranks................... 1504-12 Preand post-intervention mean rate scores for reading skills on the RARS... 1504-13 Preand post-intervention rate scores for classroom behaviors on the RARS. 151

PAGE 10

10 LIST OF FIGURES Figure page 4-1 Rate of correct pseudowords per minute (C PPM) across groups..................... 1404-2 Rate of correct sight words r ead per minute (CSPM) across groups................ 1414-3 Book levels read by eac h group during in tervention......................................... 1524-4 Teachers responses to the social validity questionna ire about the UFLI component s...................................................................................................... 1564-5 Teachers responses to the social validity questionna ire about the UFLI progr am............................................................................................................ 1564-6 Rate of correct pseudoword and sight word read per minute for Miriam.......... 1584-7 Book levels read by Mi riam during in tervention................................................ 160

PAGE 11

11 Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy EVALUATING THE EFFECTS OF THE UNIVERSITY OF FLORIDA LITERACY INITIATIVE (UFLI) ON THE READING SKILLS OF SPANISH-SPEAKING ENGLISH LANGUAGE LEARNERS By Stephanie Liset Arriaza de Allen August 2010 Chair: Holly B. Lane Cochair: Hazel A. Jones Major: Special Education Many English language learners (ELLs) in schools experience reading difficulties, particularly Spanish-speaking st udents. This is a serious problem given the tendency for struggling readers to fall further behind as they advance in school. This issue becomes more pressing as the number of ELLs enrolled in schools rises and the availability of native language instruction dimi nishes. Fortunately, there is evidence that English reading interventions that have proven to be effective for struggling native English speakers are also effective for ELLs with diffe rent levels of English oral proficiency. Given that the literature is scarce, it is important to investigate the impact of other reading interventions on the reading outcomes of ELLs. The purpose of this study was to examin e the effectiveness of a modified version of the University of Florida Literacy Initiati ve tutoring program in promoting the reading skills of second-grade Spanish-speaking ELLs who are struggling to read. The program was modified to include small-group instruction and practices that supported the

PAGE 12

12 language needs of ELLs. A multiple-base line across groups design was used to examine students response to the interventi on, as measured by the rate of correct pseudowords (CPPM) and sight words (CSPM) read per minute. Results showed that all groups improved from baseline to intervention in their rates of CPPM and CSPM and maintained appropriate rates tw o weeks after the intervention ceased. In addition, all groups showed a marked improvement in book reading accuracy of at least one halfs year progress and up to a year and a halfs progress in a limited number of sessions. St udents improved reading skills were further corroborated by preand post-intervention measures of decoding, word recognition, fluency, and comprehension, as well as teac hers ratings of students reading abilities and classroom behaviors. After intervention, students also demonstrated more positive attitudes towards reading. Most importantly, the intervention proved effective for all students regardless of their level of English oral proficiency. Social validity measures showed that the UFLI program was regarded as important, effective, and feasible among participants and teachers. These results are consistent with available literature that supports the implement ation of early English interventions for ELLs.

PAGE 13

13 CHAPTER 1 INTRODUCTION Learning to read is an important milestone in every ch ilds life. Due to its close relationship with academic achiev ement (Whi tehurst & Lonigan, 2002), proficient reading holds lifelong implications. Accordi ng to Adams (1990), reading is the key to education, and education is the key to succe ss for both individuals and a democracy (p. 13). The ability to read not only opens a wealth of professional opportunities for individuals, but also facilitates the completion of basic daily tasks, such as paying bills, reading the newspaper, reading medicine labels, reading inst ructions, and filling out forms (Anderson, Hiebert, Scott, & Wilk inson, 1985; Chhabra & McCardle, 2004). The recognition of the importanc e of reading has prompted m any initiatives throughout the years with the purpose of better understanding the processes involved in skilled reading, as well as identifying and setting in place effective ways to promote its development (Adams, 1990; Anderson et al., 1985; National Commission on Excellence on Education, 1983; National Reading Panel, 2000; No Child Left Behind, 2001; Snow et al., 1998). Unfortunately, there are many children in school who struggle with reading. In 2007, the National Center for E ducation Statistics (NCES) r eported that 33% of fourth grade students in public schools scored below basic levels on reading (Lee, Griff, & Donahue, 2007). The percentages across states showed great variance, from 19% in Massachusetts to 66% in the District of Co lumbia. The report also showed a 27-point gap between Blacks and Whites, and a 26point gap between Hispanics and Whites. There is also evidence that a gap already exists when children enter kindergarten. For example, the Early Childhood Longitudi nal Study examined a group of 19,000

PAGE 14

14 kindergarten students from a representative sample of 940 private and public schools in the nation and found that 18% of students were not familiar wit h the conventions of print (i.e., directionality of text, book structure), while 34% were unable to name the letters of the alphabet (NCES, 2001). Sc arborough (2003) stated that between 65% and 75% of children who experience reading difficulties in the early years continue to show poor reading skills during the course of their school years. These numbers are troublesome, given the tendency for struggling readers to fall further and further behind their higher performing peers (Whitehurst & Lonigan, 200 2), a phenomenon identified by Stanovich (1986) as the Matthew Effect in Reading (p. 380). That is, children who read well tend to read more, and as a result become bette r readers. On the ot her hand, students who struggle with reading tend to read less and, as a consequence, fail to become skilled readers. Statement of the Problem Children in the United States whose native language is not Englis h are at particular risk for experienci ng reading difficulties and falli ng behind in school (Snow et al., 1998). The growing number of language minority students in the U.S. (Klingner, Artiles, & Barletta, 2006) and the limited availability of bilingual programs (August, 2006; Gomez-Bellenge, Chen, & Schulz, 2008) make this problem even more pressing. The term English language le arner, or ELL, has been defined by the National Literacy Panel on Language-Minority Child ren and Youth (NLP; August, 2006) as language-minority students who are limited English profici ent (p. 44). Other names used in the literature include (a) limited Engl ish proficient (LEP) (IDEA, 2004), (b) English as a second language learners (ESL) (Fitzgerald, 1995), and (c) culturally and linguistically diverse (CLD) (Cartledge, &Gar dner, & Ford, 2009). According to Silliman,

PAGE 15

15 Wilkinson, and Brea-Spahn (2004 ), there are marked diff erences in achievement between English language learners and those w hose first language is English. This is significant given the changing demographics in schools today. According to the NCES (Planty et al., 2009), in 2007, 20% (10.8 millio n) of children between the ages of 5 and 17 spoke a language other than E nglish at home, marking a 2% increase from the year 2000. Among this group, 25% of children spoke English wit h difficulty. Furthermore, 75% of children who spoke English with difficu lty spoke Spanish as their first language (approximately 2.1 million). This is reflected in Florida, as well, where 78.8% of students who speak English with difficulty speak Spanish. As the number of ELLs increases in schools, educators face more challenges because many ELLs experience academic failure (Klingner, Artiles, & Barletta, 2006; McCardle, Mele-MacCarthy, Cutting, Leos, & DEmilio, 2005), performing below level in reading achievement tests (Bravo, Hiebert, & Pearson, 2006). According to the Survey of the States Limited Engl ish Proficient Students and Ava ilable Educati onal Programs and Services 2000-2001 Summary Report (Kin dler, 2002), only 18.7% of ELLs across 41 states scored above state norms on measur es of reading comprehension. According to Snow et al. (1998), Hispani c students are twice as likely as non-Hispanic Whites to perform below level on measur es of reading. Furthermore, these gaps are observed early on and tend to persist across the years. In order to help these student s catch up to their higher performing peers, schools need to implement effective reading interven tions at an early stage (Cartledge et al., 2009; Denton & Mathes, 2003). The existing lit erature in the field of reading has highlighted the importanc e of developing effective reading programs that include critical

PAGE 16

16 reading elements, like phonol ogical awareness, phonics, fl uency, vocabulary, and comprehension (National Read ing Panel, 2000; Snow et al., 1998). In particular, instruction in phonemic awareness and phonics becomes crucial because they play a key role in the early stages of reading dev elopment among native speakers of English (Adams, 1990; Ehri, 2004) and English l anguage learners (Ehri & Roberts, 2006; Helman & Burns, 2008). When students begin to read, they need to develop the ability to analyze and manipulate the sounds in speech, as well as to recognize the lettersound correspondences so that they can ef fectively decode unfamiliar words in print (Ehri, 2004). Furthermore, students must devel op the ability to read words accurately and automatically in order to achieve fl uent reading (Stahl, 2004). When students are able to read words automatically, they are ab le to focus their attention on higher order skills like comprehension (Adams, 1990; LaBerge & Samuels, 1974). This connection between word reading skills and reading comprehension has been well documented in monolingual readers (Gough, 1996; Stanovich, 1990; Vellutino, Scanlon, & Tanzman, 1994) and language minority student s (Hoover & Gough, 1990; Lesaux, Koda, Siegel, & Shanahan, 2006). Recently, the NLP (August & Shanahan, 2006) reported that word reading skills develop in similar ways between native spea kers of English and ELLs. Evidence of these similarities came from research studies that compared native speakers and ELLs on different measures of word reading skills, including pseudoword reading (Chiappe, Siegel, & Gottardo, 2002; Chiappe, Siegel & Wade-Wolley, 2002; Geva et al., 2000; Jackson & Lu, 1992; Limbos & Geva, 2001; Verhoeven, 1990, 2000; Wade-Woolley & Siegel, 1997). For example, Chiappe et al. (2002) found that ELLs and native speakers

PAGE 17

17 of English in kindergarten had similar perform ance on measures of letter identification and word reading and that by the end of fi rst grade no differences existed on measures of word and pseudoword reading, despite ELLs initial lower oral language proficiency. In another study, LeSaux and Siegel (2003) found that the va riables that accurately identified struggling be ginning readers who are native speak ers of English, also served to identify struggling ELLs. Since ELLs and na tive speakers of English learn to read in similar ways (Ehri & Roberts, 2006; Helman, 2009), Ehris explanation on how children learn to read can serve as a theoretic al foundation on which to build reading interventions. What follows is a description of Ehris theory of word reading development, which delineates how children learn to read words by sight (Ehri, 2005b). Ehris Phases of Word Reading Development According to Ehri (2005a), a skilled r eader is able to read isolated words and words in text accurately and quick ly. There are four ways in which words are read (Ehri & Snowling, 2004): (a) decoding (using grap ho-phonemic correspondence to identify unfamiliar words); (b) analogy (recognizing familia r words that have similar spellings to known words), (c) prediction (predicting a wo rd based on text cues like initial letters, surrounding words, pictures, etc.), and (d) si ght (automatically recognizing words that have been read before). Developi ng the ability to read words by sight is advantageous since it allows readers to focus their att ention on meaning (Ehri, 2005a). This ability to read words is acquired gradually. According to Ehri and McCormick (1998) readers go through five phases of word reading dev elopment, each one characterized by the learners understanding and use of the alphabetic system (p 140). These phases are: (a) pre-alphabetic, (b) partial-alphabetic, (c) fu ll-alphabetic, (d) cons olidated-alphabetic, and (e) automatic-alphabetic (Ehri & McCormick, 1998).

PAGE 18

18 Pre-Alphabetic Phase In this phase, learners read words based on nonalphabetic visual cues (Ehri & Snowling, 2004) that ar e unrelated to the sounds in the words. Arbitrary associations are made between visual aspects of the written word and the spoken form of a word, or between the written form and it s meaning (Ehri & McCormick, 1998). For example, the child who encounters a stop sign recognizes t he red, octagonal shape, and says stop, and the child who recognizes the familiar swirl of the Coca Cola logo says Coke. Yet, when the visual cues are taken away, the child is unable to read the words. This happens because in this phase children hav e limited knowledge of the alphabetic system. Therefore, they are unable to make connections between the letters and the sounds in words when they are trying to read (Ehri, 2005b). Since students rely on ineffective reading strategies, they are unabl e to read connected text (Ehri & Snowling, 2004). According to Ehri and Snowling (2004), this phase is char acteristic of children in preschool or kindergarten who have not received reading instru ction. Yet, older students with significant reading difficu lties might also be reading at a pre-alphabetic level (Ehri & McCormick, 1998). Partial-Alphabetic Phase The partial-alphabetic phase or phonetic cue reading Eh ri & Wilce (1985) is characterized by the partial knowledge of the alphabetic system. Students possess an incipient knowledge of letters and sounds, and use this knowledge to guess words. At this time, children are unable to use decoding or analogy as reading strategies (Ehri & Snowling, 2004). Furthermore, their phonemic s egmenting ability is still developing, so they are unable to identify all the s ounds in words (Ehri & Snowling, 2004). Consequently, when they encounter a word, only partial letter-sounds connections are

PAGE 19

19 made (Ehri & McCormick, 1998). For this reason, this phase has also been called rudimentary alphabetic (Ehri & Snowling, 2004). Some common reading behaviors characteristic of this phase include (a) writing words using only salient sounds and leaving out middle sounds (Ehri, 2005b), (b) misreading words as other words that have similar letters (i.e., tall for tell), (c) usi ng letter names to spell words (Ehri, 1989), (d) misspelling words that have sounds not included in the names of the letters (i.e., /g/ as in go), and (e) processing only partial grapheme-phoneme connections to learn sight words (i.e., remembering word book by t he /b/ and /k/ sounds) (Ehri & Snowling, 2004). Phonetic cue reading is characteristic of children in kindergarten and first grade, once letters are learned and reading instructi on begins (Ehri, 2005b). Yet, as with the pre-alphabetic phase, older st udents with learning disabilitie s might also be reading at this level (Cardoso-Martins, Rodrigues, & Ehri, 2003; Ehri & McCormick, 1998). While phonetic cue reading is more re liable than visual cue reading, it is not sufficient for skilled reading (Ehri, 1991, 1998). Full-Alphabetic Phase The full-alp habetic phase or cipher r eading (Gough & Hillinger, 1980) typifies students who have a full working knowledge of the alphabetic system, as well as phonemic segmentation ability. Learners are able to decode new words and learn sight words by using complete grapho-phonemic cues (Ehri & Snowling, 2004). With the ability to represent full sight words in memory, reading words by analogy becomes possible. Initially, word reading is slow and laborious because children are working to segment and blend all the sounds in the word. It seems that readers ar e glued to print (Chall, 1983). Yet, as students have opportunities to practice decoding and learning sight words, their reading becomes more fl uent (Ehri & McCormick, 1998). At this time,

PAGE 20

20 the ability to spell words correctly shows a marked improvement. Using their knowledge of the alphabetic system and their ability to identify sounds in word s, students are able to spell words more accurately (Ehri,, 1997, 2001). Thus, at this point children are acquiring the foundation for skilled reading and writ ing. However, for th is to take place, children need to receive systematic phonemic awareness and phonics instruction (Ehri & Snowling, 2004). Typically, this phase ta kes place in first grade, when children develop their alphabetic knowledge. A ccording to Ehri and McCormick (1998), once children master cipher reading, they are able to advance into the next phases of development. Consolidated-Alphabetic Phase The consolidated phas e or ort hographic phase (Ehri, 1991) initiates during the full alphabetic phase when children use their knowledge of grapheme-phoneme relations to read larger word units (Ehri & Snowling, 2004). As they enc ounter common letter patterns, readers recognize and store them as consolidated units. These larger units include rimes, syllables, morphemes, and root words (Ehri, 2005b). As children are able to read and store these larger units in memory, they gain accuracy and speed. They are able to remember multi-letter combinat ions, which helps decode and remember multisyllabic words (Ehri & Snowling, 2004). As children practice reading words in connected text, their word reading becomes automatic, facilitating comprehension (Ehri & McCormick, 1998). Readers in this phase of development are commonly found in second grade, but can also be found in hi gher grade levels among struggling readers (Ehri & Snowling, 2004).

PAGE 21

21 Automatic Phase Children who are considered proficient readers are typi cally in the automatic phase. According to Chall (1983) profic ient readers are able to read familiar and unfamiliar words with automatic ity and speed. At this point in time, readers have a copious sight vocabulary and multiple reading strategies that help them recognize most words in text and effectively decode uncommon words (Ehri & McCormick, 1998). Readers fluent word recognition allows them to allocate their attention on meaning when they are reading connected text. The Role of Instruction in Promoting Word Reading Development While Ehris phases describe the process in which norm ally developing students acquire the ability to read, it is important to recognize that not everyone moves through these phases with ease. According to Ehri and McCormick (1998) one reason why struggling readers fail to advance to the next phase is lack of adequate instruction. In general, instruction for childr en at risk for readi ng difficulties needs to be explicit, intensive (i.e., more time, smaller in structional groups), and supportive (e.g., scaffolding) (Torgesen, 2002). In addition, inte rvention needs to be provided in a timely manner to prevent reading di fficulties from getting worse (Lonigan, 2006; Stanovich, 1986; Torgesen, 1998). In relation to Spanish-speaking English language learners, there are two factors that researchers have considered when designing effective reading interventions: (a) language of instruction (Vaughn et al., 2006), and (b) level of oral English proficiency (Ehri & Roberts, 2006). Regarding language of instruct ion, many researchers believe that ELLs should receive instruction in their native language because it facilitates the acquisition of literacy skills by taking advantage of students oral langu age development (Francis,

PAGE 22

22 Lesaux, & August, 2006). In bilingual programs, students first learn to read in their native language before introduci ng literacy instruction in English (Tabors & Snow, 2002). According to the NLP (Francis, LeSau x, & August, 2006), supporters of nativelanguage instruction base their claims on evidence that (a) bilingual students can transfer procedural and declarative knowledg e from their first la nguage (L1) to their second language (L2), (b) bilingualism does not affect learning in either language, and (c) reading proficiency in ch ildrens native lan guage can predict reading proficiency in the second language. A synthesis of the literature on lang uage of reading instruction conducted by Slavin and Cheung (2005) and Gr eene (1997) reported posi tive results for bilingual programs compared to English-only reading program s. It is important to recognize, however, that many ELLs do not have access to native-language instruction due to geographic, political, and economic constr aints (Kelly et al., 2008; Goldenberg, 2008). Some reports show that approximately 60% of ELLs in the nation are receiving reading instruction in English (August, 2006; Goldenberg, 2008), while others report that up to 85% are in mainstream classroom s (Schirmer, Casbon, & Twiss, 1996). Furthermore, the percentage of students that have access to Spanish instruction has declined in recent years from 40.1% in 1993 to 20.4% in 2003 (August, 2006). Fortunately, the NLP (August & Shanahan, 2006) reported that ma ny of the instructional components that are effective fo r monolingual learner s appear to be effective for ELLs, and that those who have no access to first-language support can succeed when highquality instruction is provided in their second language (Snow, 2006). Regarding level of oral language profic iency, some researchers and educators believe that when students are learning in Eng lish, literacy instruction should be delayed

PAGE 23

23 until students develop sufficient English l anguage proficiency (E hri & Roberts, 2006; Tabors & Snow, 2002). One reason for this view is that students who have limited English proficiency lack the necessary vocabulary to understand the words they read (Helman, 2009), and consequently, are not able to make sense of texts (Peregoy & Boyle, 2005). The problem with this approach is that by the time interventions are implemented, children are already far behind and will need more intensive interventions to catch up to their peers. Fo rtunately, there is evidence that ELLs who participate in English reading interventions respond positivel y, despite having limited levels of English oral proficiency (Gunn et al., 2000, 2002, 2005; LeSaux & Si egel, 2003; Quiroga, Lemos-Britton, Mostafapour, Abbott, & Bern inger, 2002; Roberts, 2003; Stuart, 1999). Many of these studies implemented reading programs that had proven to be effective with monolingual str uggling readers (i.e., Proactive Reading, Reading Mastery, Corrective Reading, Reading Recovery). In some cases, researchers implemented the programs as they were designed (Gunn et al., 2000, 2002, 2005; Al Otaiba, 2005) and sometimes they modified the original vers ion to include activities and strategies designed to support the language needs of ELLs (Linan-Thompson et al., 2003; Vaughn, Linan-Thompson, Mathes et al., 2006; Vaughn, Mathes et al., 2006; Vaughn, Cirino et al, 2006). In all cases, results showed that students with differ ent levels of oral language proficiency made significant gains on measures of phonol ogical awareness, decoding, oral reading fluency, and passage comprehension. Theref ore, there is no compelling reason why literacy instructi on needs to be postponed until students develop sufficient oral English profic iency, and there is substantial support for providing early intervention (Ehri and Roberts, 2006; Gunn et al., 2006; Linan-Thompson et al., 2003).

PAGE 24

24 Purpose of the Study The findings from previous research prov ide guidance on how to effectively serve Spanish-speaking ELLs who are being instruct ed in English and w ho are struggling with reading. Unfortunately, the lit erature on effective intervent ions is still scarce (Gunn et al., 2000; Shanahan & Beck, 2006; Snow, 2006). There is a great need to identify other English int erventions that have proven to be effective with struggling monolingual students and evaluate how effective they ar e in promoting the reading skills of developing readers who are ELLs. Therefore, t he purpose of this study is to evaluate the effectiveness of a modified version of the University of Florida Literacy Initiative (UFLI) tutoring program with second-grade Spanish-speaking ELLs who are struggling with reading. The effectiveness of the pr ogram was examined through a multiple baseline across groups design. The findings fr om this study will not only contribute to the literature on effective r eading interventions, but also provide more options for teachers who are struggling to support ELLs as they teach them to read in English. In Chapter 2, a critical anal ysis of the literature on Engl ish reading interventions for struggling Spanish-speaking ELLs will be pr esented. Next, Chapter 3 will provide a detailed description of the met hodology used in this study. This is followed by Chapter 4, which provides a detailed description of the results obtained. Finally, a discussion of findings will be addressed in Chapter 5, along with an overview of limitations, implications for future research and implications for practice.

PAGE 25

25 CHAPTER 2 REVIEW OF THE LITERATURE Developin g the ability to read is vital for a child to succeed academically and in life, yet learning to read is not an easy task. M any young learners have di fficulties acquiring the skills necessary to read with fluency and to comprehend what they read (Gunn et al., 2000, 2005). English language learners ar e at particular risk for experiencing reading difficulties (Snow et al., 1998). In recent years, reports have showed that the majority of ELLs have received reading instruction in English (August, 2008; Goldenberg, 2008). Unfortunately, reports also show that this group of students is performing below average on measures of readi ng achievement (Kindler, 2002). As the number of ELLs in schools across the nati on continues to rise, particularly Spanishspeaking ELLs (Genesee, Lindhol m-Leary, Saunders, & Christian, 2005; Kindler, 2002), the need to identify effective English readi ng interventions becom es prominent (Vaughn, Linan-Thompson, et al., 2006). What follows is a critical synthesis of the literature on English reading interventions for Spanishspeaking ELLs who are struggling to read. The studies selected were identified th rough an extensive review of three databases: PsycInfo, EBSCO, and Google Scholar. The following key words were used in different combinations: English languag e learners, Spanish-s peaking, bilingual learners, second language learners, reading in terventions, supplemental interventions, and early interventions. In addition, a hand search of the latest issues in relevant journals in the field of educ ation was conducted (e.g., Reading Research Quarterly The Elementary School Journal Learning Disabilities Research and Practice ), as well as a ancestral search using key articles and reports (e.g., August & Shanahan, 2006; Vaughn, Linan-Thompson, et al., 2006). Of the available literat ure, studies that met the

PAGE 26

26 following criteria were selected: (a) at l east 40% of the partici pants were Hispanic or spoke Spanish as their first language; (b ) participants were considered English language learners or limited English proficient; (c) participant s were at risk for reading difficulties or performing below level of measures of reading ability; (d) included a supplemental reading intervent ion separate from the core reading program; (e) instruction was given in English; and (f) the intervention had a strong focus on foundational reading skills (e.g., phonological awareness, phonics, fluency). Only peer reviewed articles were included. A total of 18 studies met the inclusion criteria. The studies selected were grouped by type of inte rvention: (a) Reading Recovery, (b) Direct Instruction (Core Intervention Model, Proac tive Reading, Reading Mastery/Corrective Reading, and Read Well), and (c) other reading interventions (word reading interventions and comprehensiv e reading interventions). First, each study is described, addressi ng the purpose of the researchers, participants, intervention, dependent measures and results. Each section ends with a summary and analysis of findings from that particular type of interventions. The chapter ends with a general discussion of findings across studies, along with implications for research. Reading Recovery Reading Recovery (RR) is an intensive early intervention program designed to help first graders who are having difficultie s in reading and writ ing (Shanahan & Barr, 1995), and who have been identified by their teachers as the lowest reading achieving in their classrooms (N eal & Kelly, 1999; Swar tz & Klein, 1995). The main goal of this program is to accelerate the rate of progress of struggling reade rs in order to help them catch up to their cla ssmates (Clay, 1993b). The program lasts from 12 to 20

PAGE 27

27 weeks and provides daily, 30-minute individu alized lessons. Each lesson covers seven reading and writing activities that are individually des igned based on students daily progress: (a) rereading of two or more familiar books, (b) independent reading of previous days new book while the tuto r observes and records miscues, (c) letter identification tasks and/or word work breaki ng and making, (d) writing a story the child has created while listening and recording the sounds in words, (e) reassembling a cutup story created by the student, (f) intr oducing a new book, and (g) reading the new book using learned strategies (Center, Whendall, Freeman, Outhred, & McNaught, 1995; Clay, 1993b). Participants are selected based on two conditions: teachers recommendations and results from An Observation Survey of Early Literacy Achiev ement (lowest 20%; Clay, 1993a). The survey includes six measures related to early reading and writing: (a) letter identification (identify upper and lower case letters), (b) word test (read a list of 20 frequently used words), (c) concepts about pr int (tasks related to book reading, including directionality, concepts about letters and words), (d) writin g vocabulary (write as many words as possible in a ten-minute period), (e) hearing and recording sounds in words (writes a sentence read by the tutor) and (f) text reading level (highest leveled book read with a 90% or higher accuracy) (Clay, 1993a; Swartz & Klein, 1995). Students are discontinued form t he program when they show accelerated progress, when their scores on the Observation Survey are within the average range for first grade, and when they demonstr ate independent use of st rategies, becoming selfsustaining learners (Clay, 1993b; Neal & Kelly, 1999; Shanahan & Barr, 1995). In Reading Recovery, children who receive sixty or more lessons or who are discontinued

PAGE 28

28 with less than 60 lessons are referred to as p rogram children (Swartz & Klein, 1995, p. 3). Reading Recovery was developed by Ma ry Clay in the mid-70s and adopted nationwide in New Zealand in 1983. Various evaluations in New Zealand provided support for the effectiveness of the program in accelerating students progress (Clay,1985, 1987; Glynn, Croo ks, Bethune, Ballard, & Smit h, 1989). The program was brought to the United States by researchers at The Ohio State University in 1984, and since its inception, over 1.7 millio n students have been served around the nation (National Data Evaluation Center [NDE C], 2008). Research conducted by program developers in the United Stat es has found evidence of the effectiveness of the program in helping struggling readers (D eFord, 1995; Pinnell, Lyons, DeFord, Bryk, & Seltzer, 1994; Schwartz, 2005). A recent 3-year independent evaluatio n of the program conducted by the What Works Clearinghouse (2007) also reported positive effects on alphabetic skills and general reading achievement. This evaluation also reported potentially positive results on fluency and co mprehension. According to the NDEC (2008), in the year 2006-2007 almost 100,000 students were served. From these, 73% of the ones who successfully completed the program reached average levels in reading and writing. The rest were re ferred for further evaluation. In response to the growing number of E nglish language learners who do not have access to instruction in their native langu age, some researchers have evaluated the extent to which Reading Recovery produces positive gains when inst ruction is provided in English. Findings from a study conduc ted in England showed no significant differences between monolingual English s peakers and ELLs on a ll subtests of the

PAGE 29

29 Observation Survey after participating in RR (Hobsbaum, 1995). In the United States, two studies examined the impact of RR on English language learners (Ashdown & Simic, 2000; Neal & Kelly, 1999). Neal & Kelly (1999) reported data for students receiving RR in English and Descubriendo La Lectura (Spanish version of RR; Escamilla, 1994). Yet, in this review, onl y data for students participating in English interventions were addressed. What follo ws is a description of these studies. Ashdown and Simic (2000) evaluated the effectiveness of Reading Recovery on the reading achievement of struggling first grade native and non-native English speakers, from 37 Reading Recovery sites aff iliated with the New York University. Data from a six year period (1992-1998) were obtained from the Reading Recovery Data Sheet of the National Data Evaluation Center Three types of participants were selected based on their language status: (a) native E nglish speakers, (b) non-native speakers with limited English proficiency LEP, and (c) fluent non-native speakers ESL. English proficiency was determined by results from a language proficiency test, when available, or teacher judgment, as reported in t he RR national questionnaire (NDEC). These participants were then assigned to one of thre e groups based on their performance in the Observation Survey of Early Literacy Achievement : (a) RR group (natives = 20,863, LEP = 1814, ESL= 2924), representing the lowest 20% of first graders who received RR intervention; (b) comparison group (natives 8845, LEP = 995, ESL = 1427), representing at risk-students who did not re ceive intervention; and (c) random sample group (natives = 15,595, LEP = 731, ESL = 2037), representing the top 80% of first graders in Reading Recovery classrooms. A ll students received core reading instruction in English. The majority of non-native s peakers spoke Spanish as their first language

PAGE 30

30 (54% of LEP and 74% of ESL). Analysis sh owed that 63% of st udents in the RR group successfully exited the program. Chi-square analysis found significant differences in success rates among the language groups, point ing out that fluent ESL students had a higher success rate (66.3%) than native spea kers (62.2%) and LEP students (61.7%). Further analysis demonstrated that all gr oups, regardless of language background and English proficiency had equal op portunity to receive the fu ll Reading Recovery program (at least sixty lessons). Nevertheless, non-native speakers (ESL and LEP) were significantly underrepresented in the RR gr oup compared to native-English speakers and this may be a consequence of teacher bias in the selection process. The authors suggest that some educators might delay in tervention until non-native speakers develop their English skills or might select students who have higher probabilities of benefitting from the intervention, in this case, native speakers. To compare the reading achievement at the end of first grade for all groups, an analysis of variance was conducted with la nguage (native, ESL and LEP) and sample group (RR, comparison, or random sample) as fixed factor s, and Text Reading Level as the dependent variable. The analysis showed significant interaction between sample and language, with mean differences in the expected direction. The differences between the three language groups varied significant ly among the three samples, with nativespeakers outperforming the nonnative groups. Yet, the di fferences between language groups in the RR sample were much sma ller. The gap between LEP and ESL in the RR sample was narrower than those in the co mparison and random sample groups. ESL and LEP students in the RR sample signif icantly outperformed other ESL and LEP students at risk who did not receive intervent ion (comparison sample). Furthermore, the

PAGE 31

31 LEP students in RR outperformed the LEPs in the random sample, who were in the higher 80% of their classrooms (random sample ). The authors concluded that RR is an effective intervention for native and non-native speakers, independent of their level of English proficiency. Prov iding intensive, one-on-one in tervention in English helped promote the reading outcomes of non-native speakers who did not have access to bilingual programs. Additional support for the use of RR with ELLs comes from the work of Neal and Kelly (1999). They conducted a study with fi rst grade ELLs receiving Reading Recovery. The study compared data from the years 1993 to 1996 in the state of California. In this review, data are reported for three groups t hat received instruction in English. For analysis, the authors reported findings for three groups: (a) RR-ELL (n = 3992), comprising English language lear ners receiving instruction and intervention in English; (b) RR-English (n = 18,787), comprising all English-speaking chil dren participating in RR (including native speakers and ELLs); (c ) English random sample (typical first graders instructed in English, but not receiving RR), agai nst which the RR-ELL and the RR-English would be compared. The total number of participants for the random samples was not reported. Groups were compared based on three of the six subtests of the Observation Survey: writing vocabulary, hearing and recording sounds in words, and text reading levels. These three subtests were selected because they were considered valid indicators of childrens gr owth, with no ceiling effects. Results showed that RR groups made significant gains from preto postmeasures on the three subtests. Furthermore, students who were discontinued from the programs (reached average reading level compared to their classroom peers) outperformed the random

PAGE 32

32 sample (typical first graders) on all measures, while those who were not discontinued performed below their typically performing peers. It was al so estimated that 72% of children in the RR-ELL reached average perfo rmance; thus, they were discontinued from the program after an av erage of 67.69. These findings are comparable to the proportion of students in RR-En glish (75.2%) who were discontinued after 63.27 lessons on average. The author s concluded that RR was e ffective in helping at-risk children catch up to typical achieving first grader peers, regardless of language status. In regard to children who were not discontin ued, the authors stated that they performed at 8 to 10 text reading levels below childr en who were discontinued, an indication that they had not developed a system of literacy learning and t hat they would need more extensive interventions. An interesting finding related to not-discontinued children is that their progress in text reading was slower and that they stalled around a level five, equivalent to a pre-primer 2 level on basal series. This finding points out the need for more intensive interventions (Torgesen et al ., 2001). It is interesting to note that the studies that used RR with ELLs did not state making language accommodations, showing that the same intervention that worked for native speakers also worked for ELLs. In summary, these studies revealed that RR was effective in improving the reading outcomes of struggling first grade, Spanish-s peaking ELLs. In particular, students who were successfully discontinued from the program demonstrated average or above average performance compar ed to their first grade peers. These findings are comparable to the findings of studies that evaluated the effectiveness of RR with nativeEnglish speaking children ( Clay, 1985; Glynn et al., 1989; Pinnell et al., 1994). Of

PAGE 33

33 particular interest are the positive gains made by English language learners when instructed in English. While many researc hers highlight the benefits of using students native language during reading instruction (August & Hakuta, 1997; Greene, 1997; Krashen, 1991; Mortensen, 1984), it is certai n that many struggling readers do not have access to bilingual education programs (G oldenberg, 2008). These students are in need of effective early interventions that can s upport their literacy development in English. Reading Recovery can be considered a promis ing approach for this particular group of children. Notwithstanding, before any concluding stat ements are made about the effectiveness of RR, more research should be conducted. Particularly, it should address how effective is RR for different languag e groups and how students response to the intervention is impacted by leve l of English language proficiency. In the two studies that examined RR for ELLs, only Ashdown and Si mics (2000) reported data on the language composition of the sample. Yet, it did not disaggregate the data for each particular group. On the other hand, they reported that RR helped improve the reading outcomes of students with limited English proficiency as well as fluent English speakers. Despite the positive outcomes reported fo r RR, there are some areas of concern regarding the data collection proc edures, as well as the res earch designs. First, the two studies based their main analysis of student s performance on the Observation Survey (English and Spanish version), which was de signed by the develo per of the Reading Recovery program. According to Center and Wheldall (1992), the survey may be biased because it focuses on tasks that are explicit ly taught in the program, increasing the likelihood that students are identified as successful when t hey might not have acquired the necessary skills to thrive in the general classroom. Second, st udies did not report

PAGE 34

34 data on the long lasting effects of the inte rvention. There is a need for longitudinal studies that evaluate students ability to th rive once they leave the program. Third, none of the studies compared the effectiveness of RR to other individualized tutoring programs that address early reading skills. T hus, we cannot assure that RR is more effective for English language le arners than other available pr ograms. Fourth, no study included a control group with r andom assignment. This is a common criticism against Clays research design (Center & Wheldall, 1992). According to Nicholson (1989), by not having random assignment, progress on part of the lowest performing students may not be explained with certainty by participatio n in RR. It is possible that students who are performing at low levels at pre-test make progress even in the absence of an intervention. Fifth, according to Shanahan and Barr (1995), the empirical evaluations of this program have been limited to a set of unpublished technical reports. While this review did not include technical reports, it is important to note that both studies were published in Literacy Teaching and Learning a journal published by the Reading Recovery Council of North America. This calls attention to the need for independent evaluations of RR for Spanish-speaking English language learners, given its widespread implementation. Direct Instruction Direct Instruction (DI) i s an approach desig ned to accelerate student learning by the careful control of curriculum and inst ruction delivery (Engelmann & Engelmann, 2004; Marchand-Martella, Slocum, & Marte lla, 2004). There are three critical components of this approach: (a) organization of instruction (scheduling and material organization that promotes reading-engaged ti me); (b) program design (specifying clear objectives, teaching learning st rategies, clear instructional formats that specify what

PAGE 35

35 teachers and students will do, prov iding an optimal sequence of skills, provision of appropriate examples, and opportunities to pr actice the new skills); and (c) teacher presentation techniques (small group inst ruction, appropriate pacing, progress monitoring, teaching to mastery, motivation through high levels of student success (Carnine, Silbert, & Kameenui, 1990; Ma rchand-Martella et al., 2004). Various programs have been developed following the principles of Dir ect Instruction: Corrective Reading (Engelmann, Carni ne, & Johnson, 1988), Reading Mastery (Engelmann & Bruner, 1988), Proactive Reading or Earl y Interventions in Reading (Mathes & Torgesen, 2005), Core Intervention Model (Gerber & English, 2003), and Read Well (Sprick, Howard, & Fidanque, 1998). Recently researchers have evaluated the impact of these programs on the reading outcomes of Spanish-speaking ELLs. Core Intervention Model The Core Intervention Model (Gerber & E nglish, 2003) is based on principles of Direct Instruction (Carnine, Silber t, & Kame enui, 1990) and it focuses on the systematic reteaching of phonological skills through an exp licit correction process called correction staircase (Gerber, Jimenez Leafstedt, Villaruz, Richards & English, 2004, p.241). Correction staircase differentiates instruct ion by breaking down complex tasks into simpler steps and providing a scaffolding se quence of instruction to ensure that students provide the correct answer. Each s equence of instruction st arts with a supply question (e.g., What word rh ymes with cat?), where students construct and formulate an answer. If a student is unable to provi de an answer or makes a mistake, the instructor takes the student down the stair case to a less demanding question. In this case, a binary choice question is introduc ed, where students sele ct from the two options provided (e.g., What rhymes with cat, calf or bat?). If the student provides the

PAGE 36

36 right answer, the instructor moves up the staircase and ask the original question, allowing the student to answer it independently. If the student fails to answer the binary question, the next step down is to model-lead, where the instructor models and then elicits the right answer (e.g., Bat rhymes with cat. What word rhymes with cat?). If the student fails again, instructor moves down to the model-imitation step, where students are asked to imitate the correct answer (e .g., Say bat.). When students provide the correct answers, the instructor guides them back up the staircase until they answer the supply question again. Once students ans wer the supply question correctly, the instructor either asks a similar supply questi on to ensure that learning has taken place or moves to a different task (Gerber et al., 2004; Leafstedt, Richards, & Gerber, 2004; Richards, Leafstedt, & Gerber, 2006). Leafstedt et al. (2004) init ially tried the CIM model with a class of Spanishspeaking kindergarten students rece iving instruction in English (n=16) and found that students made significant grow th in phonological awarene ss and word reading after 10 weeks of intervention (15-minute sessions 2 times a week). The instruction incorporated a phonological awareness curriculu m, called Early Reading Project, which included developmentally sequenced skills (onset-ri me before phonemes) and tasks (identification before manipulation, and manipulation before production). Leafstedt and colleagues found that students ac ross ability levels responded to the CIM model, but that the lowest performing gr oup would have benefited from a longer intervention. While this model was used as a whole class intervention, Richards and colleagues (2006) decided to try the model as a supplement al intervention for Spanish-speaking students identified as struggling readers.

PAGE 37

37 The study conducted by Richards et al. (2006) examined the impact of the CIM model, following the developmental sequence of skills and tasks from the Early Reading Project. They examined the response of four at-risk Spanish-s peaking students in kindergarten to a 10-week intervention (15minute sessions, 2 times a week). Students were identified based on their low scores on the word identification and word attack subtests of the Woodcock Johnson-Achie vement (WJ-ACH). Based on a microgenetic methodology, used to analyze the process of change during a specific period of time (Siegler & Crowley, 1991), the authors collected weekly measures for fluency and strategy use. Fluency was examined with the nonsense word fluency and segmentation fluency measures of the Dynamic Indicators of Basic Early Literacy Skills (DIBELS). Strategy use was examined with three phonolog ical awareness tasks (e.g., onset, rime, and segmentation) from La Patera Phonologi cal Awareness Test (LP-PAT; Richards, Leafstedt, Gerber, Jimenez, & Filippini, 2003) Specifically, the instructors asked students a series of questions t hat allowed them to explain their responses (i.e., How do you know that? What letter do they st art with? How many sounds do you hear?). Strategies were coded based on a hypothes ized model of strategy development for kindergarten. Strategies were either coded as lower level (e.g., student says words sound the same, they repeat a word, or se gment one phoneme) or higher level (e.g., student says individual phonemes sound the sa me, segments two, three, or four phonemes). Results showed that, in fluency, 3 out of 4 students (responders) met midyear kindergarten benchmarks on nonword fluenc y and segmentation fluency. These students seemed to use more strategies from the higher level for all three phonological

PAGE 38

38 tasks. The fourth student progressed at a slower pace, failing to use higher level strategies (i.e., non-responder). Notwithstandi ng, the difference between the responders and the non responder was in timing, not in st rategy development. This means that with more time, the non responder would probably start using higher level strategies. Variation in response to explicit and system atic interventions has been reported by other researchers (Torgesen, 2000; Wanzek & Vaughn, 2008). This study not only provides additional support for the use of CIM as a promising phonological awareness intervention, but also highlights the importance of studying strategy use as an effective measure of response to intervention. The authors suggest replicating this study for generalization purposes. In replic ating, it is important to consider providing longer interventions to allow all students to develop reading strategies. Reading Mastery and Corrective Reading Reading Mastery is a reading progra m designed for beginning readers that provides explic it instruction in phonemic awareness, sound-letter correspondence, and decoding (Gunn, Biglan, Smolkowski, & Ary, 2000). Corrective Reading was designed for older students reading below grade level and struggling with basic decoding skills. It addresses reading accuracy (decoding), reading fluency, and reading comprehension (Smith, 2004). Following the principles of DI, both programs group students according to instructional needs, provide am ple opportunities to practice teach students to mastery, and provide frequent teacher modeling and feedback (Gunn et al., 2000). Research has found both programs to be effective in improv ing the reading outcomes of beginning and struggling readers (Campbell, 1984; Gregory, Hackney, & Gregory, 1982). Three studies conducted by Gunn and colle agues evaluated the effectiveness of Corrective Reading and Reading Mastery on the reading outcomes of Hispanic

PAGE 39

39 (Spanish-speaking) and non-Hispanic (European Am erican, English-speaking) students at risk for reading difficulties (Gunn, Biglan, Smolkowski, & Ary, 2000; Gunn, Smolkowski, Biglan, & Black, 2002; Gunn, Smolkowski, Bi glan, Black, & Blair, 2005). The first study by Gunn and colleagues (200 0) included a total of 256 students (158 Hispanics and 98 non-Hispanics) from K to 3rd grade. Students were screened in the spring of the year prior to intervention and were selected based on teachers rating of aggressive behavior and/or based on below grade level reading performance (158 selected through reading criter ia; 98 selected through aggre ssive/reading criteria). Students were matched and randomly assigned to one of two conditions: (a) Intervention including st ruggling beginning readers in first and second grade who received Reading Mastery and struggling thir d and fourth grade students who received Corrective Reading (73 Hispanics and 50 non-Hispanics); and (b) comparison including first to f ourth grade struggling r eaders who did not receive intervention (79 Hispanics and 45 Non-Hispan ics). The intervention students were tested three times: Time 1, in the fall of the fi rst year (year after screening), bef ore intervention began; Time 2, in the fall of the first year, after 6 to 7 months of intervention; and Time 3, in the fall of the second year, after 9 months of additional intervention. Students in the intervention condition partici pated in daily, small group instruction (1-3 members) for 25 to 30 minutes per lesson fo r 2 years. At the end of the first year of intervention (Time 2), analysis of variance showed that students in the intervention group significantly outperformed the compar ison group on measures of word attack (WJ-ACH), but not on word i dentification (WJ-ACH) or Or al Reading Fluency (Martson, 1989). No significant interaction between et hnicity and instructional condition were

PAGE 40

40 found. Effects due to ethnicity showed that Hispanics perform ed significantly lower than non-Hispanics on measures of oral reading fl uency. At the end of the second year of instruction (Time 3), students in the intervention group significantly outperformed the comparison group on four measur es of the WJ-ACH: word a ttack, word identification, reading vocabulary and passage comprehensi on. In addition, oral reading fluency approached significance (p<.056). Further anal ysis showed no significant differences between Hispanics and non-Hispanics on word ident ification, word attack, fluency, or passage comprehension. Non-Hispanics m ade greater gains in vocabulary than Hispanics. Additional analysis showed that Hispanics students who started the program with limited English proficien cy (n=19) benefitted as much as the rest of Hispanics. Gunn and colleagues analyzed the relationship between decoding, reading fluency, and comprehension. They found that improvement in decoding from Time 1 to Time 3 was associated to improvement in oral r eading fluency and passage comprehension. The strongest predictor of comp rehension in this study was oral reading fluency. The authors concluded that Corrective Reading and Reading Mastery were effective in improving the reading skills of Hispanic and non-Hispanic struggling readers, and that supplemental instruction is e ffective for ELLs regardless of language proficiency. Providing explic it and systematic instruction in alphabetic reading skills improved students decoding ski lls, which in turn improved fluent reading, and comprehension. Based on the greater gains made by students at the end of the second year, the authors state that providing longer interventions is necessary to make an impact on the reading skills of struggling readers.

PAGE 41

41 In a follow-up study, Gunn and colleagues (2002) examined the sustaining effects of Corrective Reading and Read ing Mastery one year afte r the intervention had ended (Time 4). From the original 256 students incl uded in the first study, the researchers collected data on 195 students. Analysis of vari ance showed that students that received the supplemental intervention conti nued to outperform the comparison group on measures of word attack and oral reading fl uency. Based on a significant interaction between condition and ethnicity, further analysis showed that Hispanics made significant gains in word attack, com pared to non-Hispanics. While no significant differences were found between conditi ons in relation to vocabulary and comprehension, the authors st ated that the effects approached significance for both. Analysis of the performance of Hispanics with different levels of English language proficiency showed no significant differenc e between students who spoke English at the onset of the intervention and students w ho did not. Yet, these results should be approached with caution because of the limited number of student data available related to language proficiency (n=16). The analysis also showed that Hispanics that received intervention made significant gains not only in word attack and oral reading fluency, but also in comprehension. These results provided additional support fo r the effectiveness of Reading Mastery and Corrective Reading as effective intervent ions for struggling readers, Hispanic and non-Hispanic. According to Gunn and colleagues, these findings go against a common belief that providing English language learners with English instruction is harmful. Quite the opposite, interventions that offer direct instruction, modeling, immediate feedback, and practice opportunities are very helpful in supporting their reading skills. In their

PAGE 42

42 concluding remarks, Gunn and colleagues pointed out that while students made significant gains after 2 years of intervent ion, they continued to perform below grade level. They hypothesize that reducing the duration of each lesson to 30 minutes instead of the recommended 40 to 50 minutes could have had a negative impact on the outcomes. In 2005, following the same format as the first study, Gunn and colleagues compared once again the effects of the two reading programs on the reading skills of kindergarten through third grade Hispanic and non-Hispanic stude nts at-risk for reading difficulties. This time, they included a larger sample (n= 299; 159 Hispanics and 140 Non-Hispanics) and an extended period of analysis (2 years of intervention and a 2 year follow-up). Students were matched and randomly assigned to one of two groups: (a) intervention (Reading Mastery for first and second graders and Corrective Reading for third and fourth graders) and (b) comparison group (struggling first to fourth graders with no intervention). In this st udy, the authors conducted an addi tional assessment, Time 5, which was conducted two years a fter the intervention had ended. Results demonstrate that at the end of the intervention, students who received the two reading programs made faster gains in letter-word identification (WJ-ACH) and Oral Reading Fluency than the comparison group Significant differences between intervention and comparison groups were also found for word attack, reading comprehension, and vocabulary. Two years after the intervent ion was over, significant differences between conditions were found for letter-word identification, oral reading fluency, and reading comprehension, with vocabul ary falling just below the chosen .05 alpha level. The authors found once again that Hispanics benefited from the intervention

PAGE 43

43 as much as non-Hispanics, and that language pr oficiency did not affect their response to the intervention. Furthermore, the progr ams continued to produce sustaining effects two years after the intervention had ended. In contrast to the prev ious study, Gunn and colleagues found that those w ho participated in the intervention approached national average levels on word identification (42nd percentile) and exceeded national levels on word attack (53rd percentile). Nevertheless, their performance in vocabulary and comprehension continued to lag behind the national standar ds (18th and 25th percentiles, respectively). To make an impact, the authors suggest that early interventions should emphasize t hese areas to a greater degree. All three studies provide support for the use of Reading Mastery and Corrective Reading as supplemental reading intervent ions for struggling readers that speak English as their second language. The tw o programs not only improved students outcomes at the end of the intervention, but produced sustained effects one and two years after the intervention was over. Gunn and colleagues also made evident the benefit of providing longer interventions to improve the outcomes of students at risk for reading difficulties. In addition, these studies highlight the import ance of not delaying reading interventions until second language learners develop their oral English language skills. English language learners can benefit from instruction provided in English that explicitly addresses critical early reading skills, like decoding, reading fluency, and comprehension. Proactive Reading Proactive Reading or SRA Early Interv entions in Reading is a small-group intervention program derived fr om DI (Carnine, Silbert, & Kameenui, 1990), designed to help struggling firstand second-grade st udents become competent readers (Mathes,

PAGE 44

44 Denton, Fletcher, Anthony, Francis, & Sc hatschneider, 2005). The program has 120 40minute lessons devised to promote phonemic awareness, phonetic decoding, reading fluency, and comprehension (Jordan, 2006). Participants complete daily lessons (6-10 short activities) in small homogeneous groups. For every activity included, instructors model the new content (very little new c ontent addressed in each lesson), facilitate guided and independent practi ce, monitor students progre ss, and provide immediate positive and corrective feedback (Mathes et al., 2005). The program has a specific scope and sequence of tasks to pr omote the gradual acquisition of knowledge and skills (e.g., phonetic elements in isolation before applying them in words, decoding words in isolation before applying them to connected text). Movement from one lesson to the other is dependent on mastery of content (1 00% accuracy on an activity) by every student in the group (Mathes et al., 2005; Jor dan, 2006). A typical lesson addresses five reading components: (a) phonemic awareness: phoneme segment ation, blending, and discrimination (from initial s ounds, to final and medial so unds, to consonant blends); (b) letter knowledge: saying and writing le tter-sound correspondences (a new letter-sound correspondence introduced every 2 to 3 days) ; (c) word recognition: sound out regular and irregular, high frequency words (moving from CVC words to more complex, multisyllabic words); (d) connected text fl uency: repeated reading of decodable text to improve rate and accuracy (2-3 times; c horal, paired, and individual reading); (e) comprehension: before reading activities (prediction, activation of background knowledge, establishing purpose for reading) and after reading activiti es (story-retell, sequencing, summarizing, story grammar elements; Jordan, 2006; Vaughn, Cirino et al., 2006; Vaughn, Mathes et al., 2006).

PAGE 45

45 Proactive Reading was first examined wit h monolingual Engl ish-speaking students with reading difficulties and was found to im prove many critical reading elements. Mathes et al. (2005), for exampl e, found that after two years of intervention, first grade students who received Proactive Reading si gnificantly improved in measures of phonological awareness, word reading accuracy, word reading fluency, word attack, word identification, spelling, and reading flue ncy. Yet, no significant differences were found in measures of reading comprehension. Four studies have examined the effect s of Proactive Reading on the reading outcomes of Spanish-speaking ELLs. Three studies implemented a modified version of Proactive Reading (Linan-Thompson, Vaughn, Prater, & Cirino, 2006; Vaughn, Mathes et al., 2006; Vaughn, Cirino et al., 2006), and one study implemented Proactive Reading as it was originally designed (Al Otaiba, 2005). Linan-Thompson et al. (2006) and Vaughn, Cirino et al. (2006) also reported data for students receiving Lectura Proactiva (Spanish version of Proactive Reading). Fo r the purpose of this review, only data addressing English interv entions will be presented. The modified version of Proactive Reading had an additional 10-minute component addressing vocabulary and oracy development. According to PollardDurodola et al. (2006), the parti cipants in these studies woul d greatly benefit from this added component due to their limited vocabulary and listening comprehension skills. The vocabulary and oracy component included a five-day cycle of shared book reading and vocabulary review (Hickman, Pollard -Durodola, & Vaughn, 2004). On day 1, students participated in a five-step routine: (1 ) introduction of a story and three to four Tier 2 vocabulary words; (2) read-aloud of a passage from an informational text or

PAGE 46

46 storybook; (3) rereading of text focusi ng on vocabulary words; (4) focus on deep understanding of vocabulary kn owledge; and (5) summarization. On days 2-4, the instructor reviewed the words and passage from the day before, and introduced a new passage with new target words, following t he same format as day one. On day 5, instructors reviewed five words students had difficulty with, reread and processed the whole text (Pollard-Durodola et al., 2006). In one study, Vaughn, Mathes et al. ( 2006) examined how effective Proactive Reading (modified version) was in impr oving the reading outcomes of first-grade Spanish-speaking stud ents experiencing reading difficu lties. Proactive Reading was selected to match the students language of instruction used in the schools. Participants who scored below the 25th percentile on the letter-word identificat ion subtest of the Woodcock Language Proficiency Battery-Revised (WLPB-R) and failed to read more than one word from a five-word experimental reading list in English and Spanish were selected to participate. Based on their pre-test scores, students were matched and randomly assigned to either a treatment group that received the intervention (n= 22) or a comparison group that did not received the intervention (n = 19). The lessons had an additional piece: embedded language support activities (3-8) designed to facilitate comprehension of ta sks and concepts. These activities, supported by available literat ure on effective English language learning strategies, included instructional scripts with pi ctures, use of gestures and fa cial expressions, explicit instruction of English language use, among others (Gerst en & Baker, 2000). Results showed that after 7 months of intervention (average 96.55 hours of instruction), students in the treatment group significantly outperformed the comparison

PAGE 47

47 group on various English measures, including rapid letter-naming (CTOPP, effect size 0.88), phonological awareness composite (CTO PP, effect size 1.24), letter sound identification (untimed, effect size 1.10), word attack (W LPB-R, effect size 1.09), passage comprehension with cloze procedure (WL PB-R; effect size 1.08), as well as verbal analogies (WLPB-R; effect size 0.77) No significant differences were found in oral reading fluency (DIBELS), picture vo cabulary and listening co mprehension (WLPBR). Results for Spanish measures showed significant difference for phonological awareness composite (CTOPP; effect size 0.76), Spanish oral language (WLPB-R, effect size 0.01), word atta ck (WLPB-R; effect size 0. 87), and comprehension (WLPB-R; effect size 0.81), but not for oral reading fluency (DIBELS). The intervention was effective for Spanis h-speaking students l earning to read in English. Furthermore, transfer of skills to Spanish was obs erved as a result of the intervention. It is possible that transfer from English to Spanish is easier because Spanish has a more transpar ent orthography and because students already posses oral language proficiency. Of particular interest is the fact that student s responded to the program despite initial low reading and language pr oficiency in English and Spanish. This supports the conclusion reached by Gunn et al. (2005) to not delay reading interventions until students develop their oral language skills. Another important finding from this study is the significant gains made in passage comprehension. Many early intervention programs tend to improve foundational skills (word attack and word identification) but tend to have limited impact on comp rehension. Comprehension strategies like story retell and the inclusion of the vocabulary and oracy component may

PAGE 48

48 be responsible for the outcome s. Vaughn and colle agues affirmed that more gains are attained when English as second language strategies are used. In a follow-up study, Linan-Thompson, Vaughn, Prater, and Cirino (2006) examined the number of student s from the previous stud y (Vaughn, Mathes, et al., 2006) who benefited from the in tervention provided at the end of 1st grade and who continued to benefit at the end of 2nd grade. Follo wing the Response to Intervention (RTI) model, students who received resear ch-based interventions and made expected gains based on predetermined criteria were identified as responders and were expected to continue to thrive when the suppl emental intervention was removed. On the other hand, students who at the end of an inte rvention failed to meet the established criteria or made minimal gains were identified as non-responders and were considered to be at-risk for reading difficulties. Responders and nonresponders were identified based on their end-of-year scores on two subtests of theWLPB-R: word attack and passage comprehension. Students who scor ed above 85 on either measure were considered responders, and those who scored below 85 were considered nonresponders. Data were available for 39 par ticipants at the end of first grade and 29 at the end of second grade. Results showed that 91% of students met criteria at t he end of first grade and 94% met criteria at the end of se cond grade. No data were available related to the specific subtest in which students met criteria (wor d attack or passage comprehension, or both). This shows that the interventions were effect ive for the majority of students and that the gains were sustained one year after t he intervention had ended. Linan-Thompson and colleagues also found that m any students in the comparis on group reached criteria at

PAGE 49

49 the end of first and second grade, even though they did not receive supplemental intervention. Yet, the percentages were not as high as wit h the treatment groups. For example, only 42% of students met criteria at the end of fi rst grade and 44% at the end of second grade. The authors stated that the par ticipating schools were considered exemplary or recognized because of the high percentage of students passing the third grade state-level reading exams. Their core reading programs were effectively serving the majority of student s. Therefore, it was expected that m any students would benefit from core reading in struction itself. The low percentage of student in the comparison group that reached criteria at the end of se cond grade suggests that students learning in English may need early, intense and sustained interventions. A word of caution: one of the reasons why this study possibly showed higher percentage of responders is t hat the measures utilized we re untimed. According to Linan-Thompson and colleagues, fluency measures (timed) are some of the hardest to influence through intervention (Torgesen et al., 2001; Vellutino et al., 1996). It is recommended that future research includes timed and untimed measures to see if different results are obtained. Vaughn, Cirino, et al. (2006) wanted to replicate the findings obtained with the modified versions of Proactive Reading du e to the limited research available on effective interventions for English language learners. Following the same methodology and format as the first study (Vaughn, Mat hes, et al., 2006), Vaughn and colleagues evaluated the program with a different sample of Spanish-speaking first-grade students. As before, students were selected, matc hed, and randomly assigned to one of two groups: intervention (n=43) or comparison (n = 48). Students in the intervention group

PAGE 50

50 participated in 50-minute lessons, five da ys a week, for 7 months (average of 115 sessions). Results showed that the interventi on group significantly outperformed the comparison group on letter-sound identification ( untimed; effect size 0.36), phonological awareness composite (CTOPP, effect size 0. 38), word attack (WLPB-R, effect size 0.42), word reading efficiency (TOWRE, effect size 0.41), oral reading fluency (DIBELS), and spelling (words spelled corre ctly from a list of 25 words in an experimental test, effe ct size 0.35). No significant di fferences were found for English language related measures, including listening comprehension, passage comprehension, picture vocabulary, and verbal analogies. This might be related to the extremely low English oral language sco res that this group had at pre-test. In sum, the study replicated the positiv e findings from the first study on many beginning reading skills, particularly in phonological awareness composite. Many significant gains were obtained, even though their level of performance was lower than those in the first study. It is important to note that the modified ve rsion of the programs had the vocabulary and oracy component to hel p students develop their vocabulary and oral language proficiency; yet, none of these were significant ly affected at the end of the intervention. Compared to the previous studies, passage comprehension was not significantly different from the comparison group. These mixed results suggest that further research is necessary before maki ng concluding claims about the need for the added component. While the previous studies focused on the modified version of Proactive Reading, Al Otaiba (2005) evaluated the effectiveness of Proactive Reading as it was designed,

PAGE 51

51 with no added vocabulary or oracy component. Using a case study design, the author studied the language and reading outcomes of eight ELLs (6 Spanish-speaking and 2 Arabic speaking). Of these, three students we re in kindergarten (2 Spanish and 1 Arabic), three in first grade (all Spanish), and two in third grade (1 Spanish and 1 Arabic). Students were identif ied as at high risk based on their limited progress in reading. Students participated in 15 sessi ons over a period of 10 weeks. Al Otaiba conducted t-tests on vocabular y (PPVT-R), sound matching (CTOPP), as well as word attack, word identific ation, and passage comprehension (Woodcock Reading Mastery R). For the comparisons, raw scores and standard scores were used. Results showed significant gains on ra w scores for sound matc hing, word attack, and passage comprehension, as well as gains in standard scores for word attack. The author reports a standard score gain of .38 on word attack, .18 on wo rd identification, and 0.30 on passage comprehension for each session. The response to intervention was not affected by the students English language proficiency. The author observed variation in the responses, where some made sharp gains, while others made slower progress. All of this was obtained in just 15 lessons, compared to the previous studies that lasted 7 months and had an average of 115 lessons. This investigation provided initial support for the use of Proactive Reading in its or iginal version as a supplemental program for struggling English language learners. In summary, the studies described support for Proactive Reading as a promising intervention for struggling Spanish-speakin g English language learners. Many basic reading skills were positively impacted by the intervention, particularly phonological awareness, word attack, word reading effici ency, letter-sound identification. Initially,

PAGE 52

52 Vaughn, Mathes, et al. (2006) attributed t he positive comprehension outcomes to the added vocabulary and oracy component. These findi ngs were contrary to what Mathes et al. (2005) found for monolin gual English-speakers. Not withstanding, the studies with English language learners did not include a condition where the original and the modified version of the program were co mpared. Therefore, the improvement in comprehension cannot be solely attribut ed to the added component. This is of relevance, since the replication study by Vaughn, Cirino et al. (2006) did not find significant differences in passage compr ehension for either program. Moreover, the added component, with its emphasis on vocabulary development, failed to produce significant gains in picture vocabulary. These results suggest the need for further research, including various comparisons to determine what elemen ts of the program have a greater impact on oral and reading outcomes, and what elements need to be intensified. For example, t he authors suggested focusing on fluency, which is critical for future academic achievement. Read Well Read Well (Sprick, Howard, & Fidanque, 1998) is a reading program for beginning readers in kindergarten and in first grade, as well as for struggling readers in need of remediation (Wahl, 2007). The program provides small group instruction focusing on the explic it and systematic teachi ng of English decoding, as well as sustained practice in reading of decodable texts, di scussion of vocabulary, and concepts in the text (Denton, Anthony, Parker, & Hasbrouck, 2004). Each lesson typically lasts 30 minutes (15 minutes on decoding and 15 minutes on story r eading). Lessons are organized by units that are thematically based. Each lesson starts with a new letter sound, which is related to the words in decoding activities, which, in turn, are related to the stories read. In each

PAGE 53

53 lesson, students read connected text, where sounds that have been previously learned are practiced. One particular characteristic of this program is the ty pe of texts that are read: duet stories and solo st ories. In duet stories, te xts have decodable segments for the students, as well as more advanced segm ents for the teacher. The solo stories include fully decodable texts that are r ead by the students (Wahl, 2007). A study conducted by Jitendra et al. (2004) evaluated the effectiveness of Read Well with diverse learners (e.g., learning disabilities, attenti on-deficit disorder, second-language learner) and found positive effects on reading, spelling, and comprehension for most students. Two studies have evaluated its effectiveness as a supplemental intervention program for struggling ELLs (Denton et al., 2004; Santoro, Jitendra, Starosta, & Sacks, 2006). The first study conducted by Denton et al. (2004) examined the effectiveness of two tutoring programs, Read Well and a revis ed version of Read Naturally (Hasbrouck, Inhot, & Rogers, 1999), on the reading achiev ement of Spanish-speaking bilingual students. A total of 93 students from sec ond to fifth grade with similar reading performance were selected based on the schools standardized tests and teachers recommendations. To be included, students had to show adequate oral English proficiency and basic reading proficiency in their native language as measured by the Language Assessment Scales Oral in Englis h and Spanish, as well as the Language Assessment Scales Reading and Writing. Students were then assigned to one of two programs based on their performance on the word attack subtest of the Woodcock Reading MasteryR. Those who scored below grade one (e.g. students in need of decoding instruction) were assigned to Read Well, and those who scored equal or

PAGE 54

54 above grade one (e.g. students with established decoding but needing practice in fluency and comprehension) were assigned to Read Naturally. Within each program, students were matched and randomly assig ned to one of two conditions: treatment (Read Well = 19; Read Naturally = 32) or co mparison group (not receiving intervention; Read Well = 14; Read Naturally = 28). Read Well and Read Naturally were not compared with each other, just with their comparison groups. Student s in the treatment conditions participated in small group (1-4 members), 40-minute sessions, 3 times a week, for 10 weeks (average of 22 sessions). To address the specific needs of Spanishspeaking students, the Read Well program focused on the differences between Spanish and English language in their activities. The Read Naturally program modified the lesson to include vocabulary instruction (di scussion of meaning for two words in each passage), decoding (teaching one high fr equency word per passage and teaching of unknown words), and compr ehension (through discussion). At the end of 10 weeks, Denton and colle agues compared students outcomes on three subtests of the Woodcock Readi ng Mastery Test-Revised (WRMT-R): word attack, word identification, and passage comprehension. They found significant differences between Read Well treatment and comparison groups only on word identification (WRMT-R), whic h showed the impact that a small amount of explicit English phonics instruction can have on struggling English language learners, particularly when attention is called to the differences between their first and second language. On the Read Naturally condition, no significant differences were found between treatment and comparison groups on any of the subtests. The authors speculated that in the Read Well condition, students did not have sufficient time to

PAGE 55

55 develop the reading automaticity needed to a ffect comprehension. They also suggested that the lack of systematic vocabulary t eaching might have hindered their ability to comprehend what they were reading. T hus, they recommended providing longer interventions and including a systematic vocabulary component when working with English language learners. In relation to the comprehens ion outcomes of the Read Naturally condition, Denton and colleagues also suggested the need for systematic vocabulary instruction. In addition, they st ated that increased fluency could have had an adverse effect on English l anguage learners, who might have needed additional time to process what they were read in English. Notwithstanding, they did not provide any evidence to support this claim. Little research has been conducted on the fluency skills of English language learners (Antunez, 2002), but available data on the fluency skills of struggling readers have demonstr ated that improved fluency is related to improved comprehension (Samuels, 1979). Most of th e studies dealing with the comprehension skills of English language learners point to the critical role of vocabulary (August, Carlo, Dressler, & Snow, 2005; Carlo et al., 2004; Gersten & Baker, 2000). Therefore, investigating the effect of word reading ski lls on fluency outcomes should be addressed in future studies. The second study, conducted by Santoro, Jitendra, Starosta, and Sacks (2006), examined the impact of Read Well on the reading skills of second-grade English language learners, two of which were Hispa nic and spoke Spanish as their native language. Using a multiple probe across par ticipants design, Santoro and colleagues were able to make evident the functional relationship between the intervention and the reading outcomes. The length of intervention varied for each participant. As specified in

PAGE 56

56 the program, each lesson lasted an average of 30 minutes and was provided individually by a trained tutor. Results demonstrated that the positive effect of Read Well was replicated across participants. The Test of Oral Reading Flue ncy (Martson, 1989) was used to report the number of correct words per minute read in a grade-level passage. All students reading rate increased from baseline to interventio n to follow up (2 weeks later). Lourdes, a Spanish-speaking student, comp leted 14 units in 14 weeks. She went from 36 cwpm at baseline to 61 at intervention, to 79 at follow up. Juan, the other Spanish-speaking student, completed 5 units in 7 weeks. He w ent from 2 cwpm at baseline, to 10 at intervention, to 21 at follow up. Despite the fact that both started at different levels, Read Well supported the development of fluency skills in both students. Similar patterns of improvement were reported for phoneme segmentation fluency, letter naming fluency, letter sound fluency, and non word fl uency (DIBELS). Preto posttest measures of word attack (WRMT-R) show ed gains for Lourdes (from 71 to 84) and Juan (39 to 53). Yet, measures of Word Identification (WRMT-R) did not show improvement from either student. This showed that wh ile Read Well was effective in supporting students ability to decode (Word Attack), it was not sufficient to impact students ability to read real words. In the passage co mprehension subtest of the WRMT-R, Lourdes went from 95 to 103, while J uan went down from 53 to 48. This can be attributed to Juans difficulties with reading fluency. T he passage comprehension requires the child to read a short passage and identify a key word that is missing from the passage. Thus, Juan might not have reached the level of aut omaticity necessary to understand the passages and select an appropriate answer. The authors suggested t hat the vocabulary

PAGE 57

57 and comprehension components of the program did not provide connections to students background and personal interests, and that this might have negatively impacted some of the outcomes. The studies presented showed mixed resu lts about the effectiveness of Read Well on the reading performance of Spanish-spea king English language learners. While Denton and colleagues (2004) found that only t he word identification skills of students receiving Read Well were significantly hi gher than those in a comparison group, Santoro and colleagues (2006) found that the two Spanis h-speaking students improved in oral reading fluency, non word fluency, phoneme segmentation fluency, letter naming fluency, letter-sound fluency and word attack, exc ept for word identification. Denton and colleagues stated that the short duration of their intervention may have limited the impact of the program. Probably, with longer interventio ns, researchers can more effectively measure the effectiveness of Read Well. In regard to comprehension, Denton and colleagues, as well as Santoro and colleagues agreed that including systematic teaching of vocabulary can improve the co mprehension outcomes of English language learners. More research is needed to evaluate the value of this program. Multiple Direct Instruction Interventions A study by Kamps, et al (2007) compared the effect of three Direct Instruction interventions (Read Well, R eading Mastery and Early Interventions in Reading or Proactive Reading) with an English as a second langua ge/balanced literacy intervention. The curriculums targeted phonem ic awareness, letter sound recognition, decoding, fluency and comprehension through ex plicit instruction, repeated practice, and small group instruction (3-6 students). The ESL/balanced intervention focused on word study, group and individual reading, and wr iting activities through small group

PAGE 58

58 instruction (5-12 students). A total of 230 first and second grade students receiving supplemental interventions participated in the study. Two groups were formed for comparison purposes: a) experimental group (n=117, 84 ELLs and 33 English speakers), representing at -risk students receiving one of the three supplemental programs mentioned above; and b) comparison group (n=113, 60 ELLs and 53 English speakers), representing at-ri sk students participating in the English as a Second LanguageESL/balanced literacy appr oach. The majority of ELLs in the sample were Spanish speaking (58%), with the rest speaking Somalian, Sudanese, or Vietnamese. The authors did not report t he total number of sessions that students participated in each of the programs, but reported pre-test data from the fall semester and post-test data from the spring semester. Analysis of students performance showed greater gains for students participating in the direct instru ction interventions. Specifically, an analysis of variance repeated-measures test showed that on nonword fluency (NWF; DIBELS), there were significant differences between experimental and comparison groups (in favor of experimental), but not between En glish-speakers and ELLs. This means that ELLs and English-speakers responded similarly to direct instruction interventions. In Oral Reading Fluency (DIBELS), there were significant differences between experimental and comparison group, as well as between English-speakers and ELLs. Pairwise comparisons showed that the diff erences were between English-speakers and ELLs in the comparison group, and not in t he experimental group. This finding further supports the positive effect of direct inte rventions on both language groups. Additional analysis (ANOVA) focusing solely on ELLs found significant differences between experimental and comparison gr oups. Specifically, those in the direct instruction

PAGE 59

59 condition showed larger gains from preto post-interventions in first grade on NWF (effect size .879) and second grade on ORF (e ffect size .947). Simi lar findings were reported for three subtests of the Woodcock Reading Master y Test (word attack, word identification, and passage comp rehension). At the end of firs t grade, students in direct instruction interventions showed larger gains in word attack (effect size 1.78), word identification (effect size 1.54), and pass age comprehension (effect size 1.04). At the end of second grade, they obtained larger gains in word identification (effect size 1.39) and passage comprehension (effect size 1.35) The authors also report that ELLs participating in direct instruction interventions demonstrated a faster rate of growth than those in ESL/Balanced approach in the firs t grade. This difference was no longer significant in second grade, but the authors state that 16 out of 19 students in the ESL/balanced approach had received direct instru ction in first grade, thus improving their performance in second grade. Kamps and colleagues concluded that providing supplemental, small group instruction is important for at-risk Engl ish language learners; and that programs like Read Well, Reading Mastery, and Early Interventions in Reading are effective first grade interventions. The authors suggested that students who participated in the ESL/balanced intervention did not make si gnificant progress due to the lack of systematic phonemic and phon ics instruction. They also speculate that the large group size had a negative impact on their performance. Other Reading Interventions This section addresses studies that focu sed on specific reading int erventions, but that were not identified as following the Direct Instruction approach. The section is

PAGE 60

60 divided in two sections. The first section addresses word reading interventions, and the second section addresses comprehensive reading interventions. Word Reading Interventions Word reading skills p lay a key role in beginning reading. According to Ehri and Snowling (2004), the ability to read words with accuracy in isolation or in context is crucial for skillful reading. Struggling readers take longer time to read words and make more mistakes because they have diffi culty making phonemegrapheme connections (Ehri, 2004).To be able to read words accurately and fluently, readers need to develop phonological awareness and decoding skills (G unn et al., 2005). The following study evaluates a reading intervention that focuses on word reading skills at the word and text level. Quiroga, Lemos-Britton, Mostafapour, A bbott, and Berninger (2002) examined whether an English intervention addr essing phonological aw areness, phonics, and fluency improved the reading outcomes of Spanish-speaking first-grade students learning to read in English. Participants we re identified based on their low scores (2 standard deviations below the mean) on the word attack and word identification subtests of the Woodcock Reading Mastery Test-R (WRMT-R). Each student received a total of 12 individual sessions over a 6-week period (30-minute sessions, 2 times a week). Each lesson included phonological awar eness activities in English and Spanish, as well as reading instruction at the subword, word, and text level. At the beginning of each session, students completed a short phonological awareness lesson from the Spanish Phonological Awareness Training Program (Lemos-Britton & Mostafapour, 1997). In the first six sessions, instruction focused on syllable segmentation, and on the last six sessions, on phoneme segmentation. Next, they participated in English sound

PAGE 61

61 games where students counted syllables an d phonemes in words selected from the text to be read later on the session. A fter that, students learned English phonemegrapheme correspondences from the Talking Letters Program. Using a set of 92 correspondences cards (summary of the spel ling unit together with pictured words that contain the spelling units), st udents were exposed to the spe lling units four times in the course of the intervention. Next, student s practiced the learned phoneme-grapheme correspondences in single words and c onnected text through repeated readings. Throughout the lessons, instructors monitore d for comprehension by asking students what they liked about the text or by predict ing what would happen next. However, this was not an area of focus for the intervention. At the end of the interv ention, the group made significant gains in word identification, but not in word attack. I ndividual scores showed that all participants improved their real word reading skills, while only six improved pseudoword reading skills. Furthermore, their performa nce in word reading at the end of the intervention was considered low average, which is more t han expected based on their low levels of oral language proficiency. Thus, the provision of an empirically-sound intervention that addressed phonological awareness in both languages, phoneme-grapheme correspondences in English, and repeated readi ng of connected text was effective in improving the reading outco mes of Spanish-speaking st udents who performed way below their peers. The authors stated that reading interventio ns should not be delayed because of limited language profic iency. On the contrary, st udents should learn to read while they continue to develop oral proficiency. Additional support for this intervention is

PAGE 62

62 needed, particularly studies that include fluen cy measures to see if the skills targeted also promote fluent reading. Comprehensive Reading Interventions Following recommendations from t he NRP (2000) and the NLP (August & Shanahan, 2006), the next four st udies inc luded reading interv entions that incorporated comprehension instruction in addition to word reading skills. The studies also examined the effect of increasing the in tensity of the intervention. So me studies reduced the group size, while others increased t he amount of intervention. Vaughn, Linan-Thompson, Kouzekanani et al. (2003) evaluated the effectiveness of an English intervention on the reading out comes of struggling second-grade English language learners. A total of 77students (39 monolingual English speakers, MES; 38 English language learners, ELL) were assigned to one of three groups: (a) 1:1 (15 ELLs and 12 MES), who received one-on-one intervent ion; (b) 1:3 (15 ELLs and 14 MES), who received instruction in groups of 3; and 1:10 (8 ELLS and 13 MES), who received intervention in groups of 10. Participants we re originally select ed based on teachers observations of failed reading or students low performance on the second-grade statelevel Texas Primary Reading Inventory (T PRI). The ELL group included all Hispanic students, while the MES group included 17 African Americans, 3 Whites, and 19 Hispanics who spoke English as the primary language. The students participated in a total of 58 sessions over a 13-week period (30 minutes, 5 times a week). Each session incl uded five components: (a) fluent reading (6 minutes), through repeated reading of familiar text in pairs and individually; (b) phonological awareness (6 minutes), oral activities focusing mainly on phonemic awareness selected from Ladders to Literacy (OConnor, Notary-Syverson, & Vadasy,

PAGE 63

63 1998) and Phonemic Awareness in Young Chil dren: A Classroom Curriculum (Adams, Foorman, Lundberg, & Beeler, 1998); (c) inst ructional-level reading (10 minutes), including reading of a new book while t eacher provided support in decoding and modeled comprehension strategies through think-alouds, while students provided the main idea or a summary of the text at the end of the reading; (d) word study (6 minutes), including explicit instruction in phonem e-grapheme correspondences and word patterns, as well as a 1-minute writing activity where students had to write as many words as they could with an opportunity to correct their ap proximations afterwar ds; and (e) progress monitoring (20 minutes weekly) through letter naming, phoneme segmentation and nonsense word probes from DIBELS, as well as connected text reading from Read Naturally. Results showed that all gr oups made significant gains in passage comprehension (WRMT-R), phoneme segmentati on (DIBELS), and fluency (Test of Oral Reading Fluency) from post-test to follow-up (4 wee ks later). Furthermore, in comprehension, groups of 1:1 and groups of 1:3 outperformed gr oups of 1:10. No significant differences were found between groups of 1 and groups of 3. On phonem e segmentation (DIBELS) and reading fluency (Test of Reading Fluency), results were somewhat different. For example, students who were in groups of 1:1 significantly outperform groups of 1:10. No significant differences were found betw een groups of 1:1 and groups of 1:3, and between groups of 1: 3 and groups of 1:10. Vaughn and colle agues observed that 1:1 grouping was superior to 1:10 grouping in phoneme segmentation and comprehension for MES and ELLs. In fluency, MES performed better in groups of 1:1 than 1:10, while ELLs performed better on groups of 1:3 than groups of 1:10. Ov erall, groups of 1:1 did

PAGE 64

64 not outperform the groups of 1:3 on any measure, showi ng that both sizes were effective for this particular intervention. This finding is relevant because many schools have limited resources, so small group (group of 3) can be chosen over one-one-one instruction as a way to effectively se rve larger number of students. Vaughn and colleagues also stated that E nglish language learners benefited from these explicit and intensive intervention as much as monolingual English speakers, and that this finding is noteworthy due to the high number of ELLs who are struggling to read. Yet, one limitation identified by the authors was the lack of a com parison group that would have included struggling readers who did not receive the intervention. Without this comparison, the association between student gains and the program it self is somewhat limited. Further research is recommended in which the different components of the program are evaluated to see which ones make a bigger impact. A second analysis based on the same cohort included in the previous study was conducted by Linan-Thompson, Vaughn, Hickm an-Davis, and Kouzekanani (2003). This analysis included data for 26 English language learners. These data were disaggregated and further examin ed, including two follow-ups (4 weeks and 4 months). The purpose was to determine the effectiv eness of small group intervention (1-3 students) over time. As mentioned previous ly, students participated in a total of 58 sessions over a 13-week period (30 minutes, 5 days a week). The authors affirmed that many accommodations were made for English language learners, including the use of picture cards to exemplify the meaning of un known words, identification of real and nonsense words, provision of qu ick definitions of unfamiliar words, activation of prior knowledge, opportunities to practice oral skills during instructional level reading,

PAGE 65

65 compare/contrast of Spanish and English phonemes and word patterns, and ample opportunities for practice. Based on a series of analyses of varianc e (ANOVA) results showed that students made significant gains from preto post-test on word attack (WRMT-R), passage comprehension (WRMT-R), phoneme segment ation (DIBELS), and fluency (TORF), with fluency showing the greatest gains. Gains were sustained four weeks later on passage comprehension and phoneme segmentation, but not on word attack or fluency. From follow-up 1 (4 weeks) to follow-up 2 (4 months ov er the summer) students maintained their gains on word attack passage comprehension, but only fluency increased significantly. In addition, it wa s observed that student s responded to the intervention regardless of their level of English oral proficiency. These findings provide support for small group interventions that focus on the critical components of reading and that make accommodations for English l anguage learners. Of particular interest are students significant improvem ent in fluency and comprehens ion. Research has shown that comprehension is closely related to fluent reading (LaBerge & Samuels, 1974). Thus, it can be implied that students improved fluency at the end of the intervention facilitated their performance in passage comprehension. Another study, conducted by Vaughn, Linan-Thompson, and Hickman (2003), examined the same reading program descri bed above with a new sample of secondgrade students. In addition, they altered the duration of the intervention to accommodate the needs of students who failed to respond (increments of 10 weeks). A total of 45 students (35 Hispan ics, 6 Whites, and 4 African Americans) participated in the intervention. They were identified bas ed on their low performance in the screening

PAGE 66

66 portion of the TPRI (five or less words r ead correctly out of 8 possible words). All students started the intervention in groups of three. After 10 weeks of intervention, students were assessed to determine response to intervention based on preestablished exit criteria: (a) five or more words read corr ectly from the screening portion of the TPRI, (b) above 55 correct words per minute (cwpm) on second-grade passage from the Test of Reading Fluency (TORF) and (c) 50 cwpm on three consecutive measures of second-grade fluency from Read Naturally. At the end of 10 weeks, 10 students met criteria (early exit) and exited the program Students who did not meet criteria (n=35) were regrouped and received 10 more weeks of intervention. At the end of the next 10 weeks, 14 students exited the program (mid exit) and 21 regrouped and received 10 additional weeks with a modifi ed version of the program. The modified version was designed to meet the individual needs of t he students, including more assessment of basic skills, more time on word study and fluency, and less work on phonological awareness. The researc hers based these changes on observed improvement in phonological awareness and increased need for word attack skills. At the end of the next 10 weeks, 10 students exit ed the program (late exit), while the 11 who failed to meet criteria (no exit) were referred for further evaluation. Assessment took place in two ways: (1) preand post-test (after 30 weeks) measures using the WRMT-R (word attack and passage comprehension) and the CTOPP (composite score for phonological awareness and rapid naming), and (2) four measures of the TORF (before intervention and after each of the three 10-week periods). Results for the TORF showed that all students made significant gains from preto post-test (30 weeks) Effect sizes reported for all groups were large: 3.18 for

PAGE 67

67 early exit group, 2.84 for mid exit group, and 6.06 for late exit. The no-exit group made significant gains (effect size 2.66) despite not meeting exit criteria A closer look at student data showed that the no exit group started at a much lower place in fluency compared to the early-exit group (10.55 and 32.50 mean score, respectively). This meant that progress was obs erved even in the no-exit group, but based on the exit criteria established, students were identified as unresponsive. Fuchs and Fuchs (2007) recommend that a child be identified as unr esponsive when they fail to make adequate progress and when they score below the norm at the end of the intervention or fail to meet a stipulated benchmark. Results also showed that all groups made significant gains at the end of 30 weeks in word attack, passage comprehension, phonological awareness, and rapid naming with effect sizes ranging from 0.47 to 2.22. This showed t hat the intervention was as effective for English language learners (His panics) as for English language speakers (Whites and African Americans). In fact, Vaughn and colleagues report that all ELLs exited the program (6 in early exit, 6 in mid exit, and 3 in late exit). A closer look at language proficiency showed t hat students from the early -exit group were more proficient in English than those from mid and late exit and that might have facilitated their learning. Notwithstanding, these findings show that even students with low oral language proficiency can benefit from the program; they might just need more time. The findings from this study are of vital importanc e to the field of readi ng, particularly with the amount of students who continue to fail despi te intensive interventions (Torgesen et al., 2001). It seems that increas ing the intensity by providin g longer interventions is an effective way of supporting struggling readers.

PAGE 68

68 Wanzek and Vaughn (2008) evaluated the e ffects of a similar reading program, focusing on word reading skills, as well as comprehension. They also examined whether increasing the amount of interv ention had an impact on students who had been identified as low responders. This inve stigation was conducted over two years and focused on first grade struggling readers who had already failed to meet exit criteria at the end of a 13-week program. The exit criteria established for the original intervention was set at more than 30 correct sounds per minute on the nonsense word fluency and more than 20 words per minute on oral reading fluency subtests form DIBELS. The first year (Study 1) had a total of 50 students (21 from the interv ention group and 29 from the comparison group no intervent ion) and the second year (Study 2) had a total of 30 students (14 from the intervention group and 22 from the com parison group). The majority of students in the samples were Hispanics (36 in year 1 and 23 in year 2). In each study, students who were selected from the intervention group remained in the intervention group, and students selected from the comparison group remained in the comparison group. In study 1, the intervention group rece ived an additional 13 weeks of instruction with one 30-minute lessons a day (single dose) five days a week. In study 2, the intervention group received 13 weeks of interv ention with two 30-minute lessons a day, five days a week (double dose). Each lesso n included the following components: (a) phonics and word recognition (15 minutes), where students practiced letter names, letter sounds, spelling of regular and irregular words, word family patterns, and word building activities; (b) fluency (5 minut es), where students participated in fluency activities at the subword, word, and te xt level; and (c) passage reading and reading

PAGE 69

69 comprehension (10 minutes), where students read short passages that included the patters learned previously and responded comprehension questions. Results for study 1 showed no significant differences between intervention and comparison group on all pretest measures (oral readi ng fluency from DIBELS, and word attack, word identification, and passa ge comprehension from WRMT-R), except for nonword fluency from DIBELS (in favor of treatment group). At post test, no significant differences were found on any measure between intervention and comparison group. Furthermore, all student s scored below first grade end-of-year benchmark for oral reading fluency set at 40 correct words per minute. This finding demonstrated that the additional dose of the program was not sufficient for this group of low responding students. Results for study 2 showed no significant differences between groups on all pretest measures. At post-test, significant differences were found only for word attack, but data revealed that more students in the tr eatment group made gains on word attack, word identification and comprehension. Ye t, all students in the treatment group continued to score below grade level on endof-year oral reading fluency (40 cwpm), even though 50% of the students increased their fluency by 10 cwpm. Wanzek and Vaughn (2008) concluded that the additional dos es of intervention did not seem to effectively support the reading skills of low responders, and suggested that low responders might need a smaller group size, more intense instructional routines, or a different intervention. The last suggesti on is a very important point, since the intervention was not compared against ot her programs that might benefit low responders when given in varying amounts of time.

PAGE 70

70 In summary, comprehensive interventions that include word r eading instruction as well as comprehension instruction proved to be effective for many struggling English language learners, except for the program implemented by Wanzek and Vaughn (2008). Notwithstanding, the limited re sponse might be associated to the sample itself. Wanzek and Vaughn stated that in a response to intervention model, these students would be identified and considered for special educat ion. In addition, the studies reported provided mixed results about the use of mo re intensive interv entions with Spanishspeaking students through smaller group size or increased intervention time. For example, Vaughn et al. (2003) found that small group interventions (1:1 and 1:3) were more effective than large group interventions (1:10), and Linan-Thompson et al. (2003) found positive effects of small group interv ention when data were disaggregated for English language learners. On the other hand, provision of extended intervention proved very effective for students in the study conducted by V aughn, Linan-Thompson and Hickman (2003), but not for students in Wanzek and Vaughn s (2008) study. This variation in results is consis tent with previous findings (B erninger et al., 2002; McMaster et al., 2005; Vellutino et al., 1996) and only hi ghlights the need for further research. In particular, these studies should be replic ated using a comparison group receiving a different intervention program in order to determine whether or not the lack of response of some students is due to the intervention itself. Discussion The review of the literat ure included 18 studies of reading interv entions for Spanish-speaking ELLs learning to read in English. Overall, these studies showed that Spanish-speaking ELLs who are struggling to read in English can benefit from English reading interventions that address word reading skills. The following discussion

PAGE 71

71 addresses (a) instruction in word reading sk ills, (b) language and reading instruction, (c) strategies for English l anguage learners, and (d) implications for research. Instruction in Word Reading Skills While the ultimate goal of reading is co mprehension, research has highlighted the important role of word reading skills in beginning readin g, specif ically phonological awareness, decoding, and word reading fluency (Ehri & Snowling, 2004; Stanovich, 1990; Wagner, Torgesen, & Rashotte, 1994). Fluent word recognition is closely related to reading comprehension (Fuchs & Deno, 1992; Stanovich, 1990) because it allows readers to allocate resources to higher order skills (LaBerge & Samuels, 1974). Conversely, when word reading is slow and laborious, readers expend cognitive resources on decoding instead of deriving meaning from text (Ehri & Snowling, 2004; Gunn et al., 2000; LaBerge & Samuels, 1974). This is true for native English speakers as well as ELLs, both of whic h learn to read in similar wa ys (Lindsey, Manis, & Bailey, 2003). In addition to word reading skills, the NRP (2000) also suggests that any comprehensive reading program needs to address vocabulary development and reading comprehension. In this review, most intervention programs focused to a greater degree on word reading skills (e.g., phonologic al awareness, phonics, and fluency). The emphasis on vocabulary and comprehension in struction varied across studies, with some briefly addressing them in each lesson (Denton et al., 2004; Santoro et al., 2006), while others dedicating a significant amount of time to them in each lesson (Vaguhn, Linan-Thompson, & Hickman, 2003; Vaughn, Linan-Thompson, et al., 2006; Vaughn, Linan-Thompson, Kouzekanani, et al., 2003; Vaughn, Mathes, et al., 2006; Vaughn, Cirino, et al., 2006; ). Findings from these studies showed that an emphasis on word

PAGE 72

72 reading skills produced a positive effect in the reading outcomes of Spanish-speaking struggling readers. This was demonstrated by significant gains made on measures of phonological awareness, word attack, word identification, and fluency. Furthermore, improvements in word reading skills were also associated with improvements in reading comprehension (Gunn et al., 2000, 2005). In relation to comprehension, most studies that added a comprehension component reported a positive e ffect on that skill, including the modified version of Proactive Reading and the comprehensive r eading interventions described last (LinanThompson, Vaughn et al., 2003; Vaughn, Linan-Thompson, & Hickman, 2003; Vaughn, Linan-Thompson, Kouzekanani et al., 2003; Vaughn, Linan-Thompson et al., 2006; Vaughn, Mathes et al., 2006) Therefore, it can be c oncluded that addressing word reading skills should be a major component of any early reading interventions. Also, adding comprehension instruction to r eading intervention enhances the reading outcomes of ELLs. Language and Reading Instruction Two important issues constantly addressed in the literature concerning effective literacy instruction for Span ish-speaking ELLs are (a) language of instruction (Vaughn et al., 2006), and (b) level of Englis h oral proficiency (Ehri & Roberts, 2006). Many researchers state that reading development is facilitated when instruction is given in the students native language (Slaving & Cheung, 2005; Greene, 1997). Yet, national data show that approximately 60% of all ELLs are being instructed in English (August, 2006; Goldenberg, 2008). Since Spanish-speaking ELLs are performing below level on measures of reading achievement (Klingner et al., 2006), it is important to identify interventions that match the students l anguage of instruction and that effectively

PAGE 73

73 support students reading development in English. This review of the literature provides ample support for the use of English interventions. Across studies, ELLs improved on different measures of readi ng, including word reading skills (i.e., phonemic awareness, decoding), fluency, and comprehension. Interest ingly, students responded positively to interventions designed origin ally for monolingual Englishspeaking students, some of which included language accommodations (i.e., Linan-Thompson et al., 2006; Vaughn, Cirino et al., 2006; Vaughn, Mathes et al., 2006) and some of which did not (i.e., Al Otaiba, 2005; Kamps et al., 2007; Neal & Kelly, 1999). Regarding level of language prof iciency, there is a common view that in order to benefit from reading instructi on students first need to acquire oral English proficiency (Ehri & Roberts, 2006). In fa ct, many educational program s choose to delay reading interventions for ELLs until they develop suffici ent English oral skills (Gunn et al., 2005). This practice is extremely harmful for young ch ildren given that it takes at least 5 to 7 years for a student to becom e a fluent speaker of English (Collier, 1989; Cummins, 1981). According to Torgesen (1998), the cons equences of a slow start in reading become monumental as they accumulate exponent ially over time (p. 1). Statistics show that a gap already exists between ELLs and English native speakers when they enter school (Lee, Griff, & Donahue, 2007). Therefor e, delaying instruction only places ELLs at a higher risk for reading difficu lties. Fortunately, the stud ies reviewed indicated that delaying instruction is not necessary, gi ven that even students with limited English proficiency responded positively to reading instruction (i.e., Gunn et al., 2000; 2002; 2005; Linan-Thompson et al., 2006; Vaughn, Ma thes et al., 2006). According to Gunn and colleagues (2000), since limited English proficien cy does not limit students

PAGE 74

74 response to intervention schools can help ELLs succeed by intervening as early as first grade. Strategies for English Language Learners Despite the fact that many ELLs improv ed their reading skills without receiving any type of language accommodations, researchers recommend including strategies that support the language needs of Spanish-speaki ng ELLs (Vaughn, Mathes, et al., 2006). For example, Linan-Thompson et al. (2003) used picture card s and quick definitions to explain unfamiliar words, distinguished betw een real and nonsense words during word work activities, and contrasted Spanish and English phonemes and word patterns to help students distinguish betwe en the two. The interventions reviewed also provided other important elements of effective practice for ELLs: (a) explicit and systematic instruction (Helman, 2009), (b) ample opportuni ties for practice (Gersten & Geva, 2003), (c) small group instruction (Gersten et al. 2007), and (d) scaffoldi ng (Watts-Taffe & Truscott, 2000). According to Goldenberg (2008) other strategies for ELLs who are being instructed in English include (a) visual cues and physical gestur es, (b) clarification of words or passages to support comprehension, (c) providing opportunities to interact with the instructor and with peers, and (d) adjusting rate and complexity of speech to accommodate the language abilit y of the students. Implications for Research The findings from this l ite rature review provide an empirical foundation for the present study. Results showed that Spanish-speaking ELLs who are instructed in English and who are struggling to read can benefit from English reading interventions that target word reading skills. Furthermore interventions originally developed for monolingual struggling readers offer the kind of instruct ion that supports the reading

PAGE 75

75 development of ELLs with different levels of oral language proficiency. Due to the high percentage of Spanish-speaking ELLs receiving reading instruction in English, it is important to identify other English interventions that have already proven to be effective for monolingual struggling readers and evaluate their effectiveness with this population. In particular, it is important to cl osely examine how students respond to intervention. While many studies have shown significant gains for students who participate in intensive reading programs, almost every study reported that some students fail to respond and continue to struggl e in reading (Wanzek & Vaughn, 2008). According to Torgesen (2000), between 2% and 6% of students who participate in reading interventions continue to struggle afterwards. Torgesen suggested that these students may benefit from more intensive inte rventions. To make interventions more intensive, some researchers have decr eased the group size and/or increased the amount of instruction provi ded (Berninger et al., 2002; McMa ster et al., 2005; Torgesen et al., 2001), finding some positive resu lts. Vaughn, Linan-Thompson and Hickman (2003) found that when more intervention wa s provided and instruction was modified to meet the specific needs of each individual, students responded positively. In contrast, Wanzek and Vaughn (2008) found that double doses of intervention were not effective for low responders. Therefore, it is important to not only exam ine what interventions are effective, but also what interventions are effective for tre atment resisters and under what conditions (Torgesen, 2000). For this reason, single su bject design becomes a valuable research method because it allows researchers to identify the specific factors that either inhibit or enhance the efficacy of a particular intervention (Jitendra et al., 2004). Therefore, in the

PAGE 76

76 study presented here, a multiple-baseline across groups design was used to evaluate the effectiveness of a modified version of the University of Florida Literacy Initiative (UFLI) tutoring program in developing the word reading skills of Spanish-speaking ELLs in second grade who are struggling to read.

PAGE 77

77 CHAPTER 3 METHODS This study implemented a multiple base line across groups design to examine the effects of a modified version of the Universit y of Florida Literacy Initiative (UFLI) smallgroup tutoring program on the reading ski lls of Spanish-speaking English language learners in second grade, who are experiencing reading difficulties. The purpose of this chapter is to provide a detailed description of the methods used to complete the study. The chapter addresses the following components: (a) setting, (b) participants, including selection criteria, (c) assessment instruments (d) interv entions, (e) dependent variable, study design, and data analysis, (f) supplemental data, (g) interobserver agreement, treatment integrity, and social validity, (h) delimitations and limitations, (i) one-on-one tutoring methodology, and (j) pilot study data. Setting This study was conducted at Suwannee Elementary School in Suwanne e County School District in Florida. Suwannee County Sc hool District is a rural district located in North Florida and is composed of seven school s and one technical center. For the year 2008-2009, the racial/ethnic composition of t he district was approxim ately 73.9% White, 13.9% Black, 9.5% Hispanic, 0.5% Asian, 0. 3% Indian, and 1.8% mu ltiracial. Of these students, approximately 4% we re identified as English language learners and 1.8% as migrant students. The district had approximately 60.3% of students participating in the free/reduced price lunch program (FLDOE, 2010b). Suwannee Elementary is a Title I school that serves only second and third grade students. Approximately 73% of the schools students participated in the free/reduced price lunch program in the year 2008-2009. That same year the school had a total of

PAGE 78

78 778 students enrolled. Of t hese, 392 students were in second grade and 385 students in third grade (FLDOE, 2010c). The racial co mposition of the school was 64.9% White, 17.5% Black, 13.6% Hispanic, 0.8% Asian, 0.8% Indian, and 2.4% Multiracial. The school identified 7.6% of the students as English language learners and 2.3% as migrant (FLDOE, 2010a). According to the Fl orida Department of Education (FLDOE, 2009), the percentage of third grade students in each achi evement level of the FCAT reading for the school in the year 2008-2009 was as follows: 16% at Level 1, 11% at Level 2, 35% at Level 3, 31% at Level 4, and 6% at Level 5. FCAT reading data for ELLs identified 53% at Level 1, 24% at Level 2, 18% at Level 3, 6% at Level 4, and 0% at Level 5. Because Suwannee Elementary School ha d only second and third grades, there were more classrooms at t hese two grade levels than at a typical elementary school. According to school records, on the year 2009-2010, there were 21 second-grade classrooms with a total of 374 st udents. Six of these classrooms (12 to 16 students per class) had been designated for lower performi ng students, and all participants in the study were from these classrooms. The lower performing classrooms used a double dose approach to reading in an effort to hel p students catch up to grade level. This meant that these classroom s had 180 minutes of reading instruction daily. The school used the Harcourt StoryTown Core Reading Program (Beck et al., 2007), which is one of the approved programs for adopt ion in Florida because it meets the criteria outlined by the State. Within the180 minutes of reading instruction, teachers also provided 45 minutes of small group inte rvention using one of two programs: Harcourt Reading Interventions or SRA Early Interventions in Reading (SRA/EIR; Mathes & Torgesen,

PAGE 79

79 2005). Teachers had the option of selecting whic h program to use in their classrooms. All of the participants in the study received at least one of these interventions. All the teachers of the participating students had Flor ida ESOL endorsement on their teaching certificates. Participants The purpos e of this section is to provide a description of the participants, including selection criteria and selection process. As required by the University of Florida Institutional Review Board (IRB), the inve stigator obtained inform ed consent from the parents of children partic ipating in the study, as well as informed assent from each child. For copies of IRB documentation, including the study protocol, consent letters, and assent form, see Appendix A. The following criteria were used to select participants for this study: 1. The child was identified by the school as being an English language learner whose native language was Spanish. This criterion was established because the focus of the study is on the effects of the UF LI intervention with this population. 2. The child was in second grade. This criterion was established because the UFLI program recommends that intervention beginning during the fall should target second graders. First gr aders tend to need a semester of reading instruction before they are ready to benefit from UFLI tutoring. 3. The child was performing below level based on three measures of reading ability. Although teacher judgment wa s the initial identifying fa ctor, this criterion was established to ensure that the judgment of the teacher c ould be corroborated with assessment data. 4. The childs record for the past year showed regular school attendance and low percentage of tardiness. This criterion was established to ensure that the intervention could proceed without excessive interruption and to minimize participant attrition. The selection process proceeded accordi ng to several prescribed steps. School personnel (i.e., principal, reading coach, and teachers) were asked to identify second

PAGE 80

80 grade English language learners, whose nat ive language was Spanish, and who were performing below grade level based on their la test end-of-year Stanford Achievement Test-10 (SAT-10) and the Dynamic Indicators of Basic Early Literacy Skills (DIBELS) scores. A total of 24 students were identified. To corroborate that these students were still performing below grade level, teachers were asked to administer the Invented Spelling Assessment (Lane & Pull en, 2004). The spelling sheets were given back to the researcher to be graded. In order to prot ect the identity of the students, teachers assigned a number to each students spelling sheet. Based on these results, a total of 16 students who scored below the 40th percentile on Invented Spelling were selected to participate in the study. Consent forms were s ent by the school principal to the parents of these students asking their permission to conduct further assessment and to participate in the study. Up until this poin t, the researcher did not have access to students names. A total of 13 students received parental consent, but one of them was dropped from the study because his mot her stated that the child did not speak Spanish. The new pool of participants consisted of 12 students. Next, the researcher took running records from each child using Reading Recovery books to identify their instructional reading level at which students were reading with 90 % to 95% of accuracy. Out of the 12 students, one was dropped because he read a level 19 book with 96% reading accuracy, when the goal for the beginning of second grade is level 17. Thus, based on this criterion, the student was not in need of reading intervention and would not have benefitted from the study. Later, when groups were fo rmed, a second student was dropped from the group intervention because she could not be placed in any of the

PAGE 81

81 three reading groups. This student was readi ng at a book level that was too high for Group 2 and too low for Group 1. Still, she received the UF LI tutoring program on a oneon-one basis, but not as part of the multip le-baseline design. A description of the methods used with this student is provided at the end of this chapter, and results obtained are presented at the end of chapter 4. The final pool of students who received group instruction included 10 participants. Students data on group composition and initial book levels can be f ound in Table 3-1. A detailed description of the procedure used to identify initial book levels for eac h participant and the percentages of accuracy can be found in the assessment section, under Grouping Inst ruments. A summary of this procedure for each participant is also provided in Table 3-7. Table 3-1. Group composition Group number Number of members Males Females Beginning book level 1 3 1 2 9 2 4 1 3 2 3 3 2 1 1 What follows is a description of each par ticipants (a) demographic information, (b) instructional information, (c) reading scores used for selection purposes, (d) English and Spanish oral language pr oficiency, and (e) classroom attitudes and behaviors. Throughout this report of the study, pseudonyms are used to protect the identity of the participants. Demographic information was obtained from parents through a home language survey (Appendix B), teachers, and students when needed. This information includes age, gender, parent nationality, languages spoken at home, language used to communicate with child, year s of schooling in US, and parent al level of education. Demographic data can be found in Table 3-2. Instructional information was gathered

PAGE 82

82 Table 3-2. Summary of participants demographic information Name Age Gender Parents nationality Language(s) spoken at home Language used to communicate with child Years of schooling in US Parents level of education Amelia 78 F Mexico Spanish/ English Spanish/ English 3 years Not Available Pedro 8-8 M Cuba Spanish/ English Spanish 4 years Community College Jessica 8-3 F Mexico Spanish/ English Spanish 6 years No schooling Viviana 8-5 F Guatemala Spanish/ Conjoval Spanish 5 years No schooling Ernesto 8-10 M El Salvador Spanish Spanish 5 years High school Maggie 8-8 F Mexico Spanish Spanish 5 years No schooling Maria 8-8 F Mexico Spanish Spanish 5 years High school Jennifer 10-1 F Guatemala Spanis h Spanish 1 years No schooling Aldo 9-3 M Mexico Spanish/ English Spanish 4 years No schooling Jorge 9-5 M Mexico Spanish Spanish 5 years No schooling

PAGE 83

83 from teachers and school personnel (e.g., readi ng coach) about the grades in which students have been retained, the amount of readi ng instruction students receive daily, the core reading program covered in the classroom, additional reading interventions students received, language status (i.e., LEP /limited English profic iency, ELL/English language learner), LEP classes provided, migrant status, ESE classification, and ESE services rendered. A summary of this information can be found in Table 3-3. Reading scores are reported based on the latest scores from the end-of-year SAT10 and the DIBELS Phoneme Se gmentation Fluency (PSF), Nonsense Word Fluency (NWF) and Oral Reading Fluency (ORF) subtests. For students that were retained in second grade, only NWF and ORF scores are reported. Reading scores also included beginning-of-year Invented Spelling Assessm ent scores for all students (Lane & Pullen, 2004). These scores can be found in Table 3-4. English and Spanish oral language proficiency levels were obtained using t he Woodcock-Munoz Language Survey-Revised (WMLS-R; Woodcock, Munoz-Sandoval, Ru ef, & Alvarado, 2005). Language levels are reported in Table 3-5. Finally classroom attitudes and behaviors were rated by teachers using the Conners Abbreviated Teacher Rating Scale (C-ATRS). A summary of results can be found in Table 3-6. Refer to the asse ssment section later in this chapter for a detailed description on each of these measures. Amelia Amelia was a Hispanic female, aged 7 years-8 months, enrolled in second grade for the first time. Given that Amelias parents did not return the home language survey, the demographic information reported is based on data gathered from Amelia and her teacher. Amelia stated that her parents are from Mexico and they speak

PAGE 84

84 Table 3-3. Summary of partici pants instructional information Name Grades Retained Other reading interventions Migrant status Language status Received LEP classes ESE classification Received ESE services Amelia None HR No ELL/LEP No None N/A Pedro 2nd SRA/EIR No ELL No None N/A Jessica K SRA/EIR No ELL/LEP No None N/A Viviana K HR Yes ELL/LEP Yes (30 min/3 times week) Language disability None Ernesto K SRA/EIR HR No ELL/LEP No Language disability Speech therapy Maggie k, 2nd SRA/EIR No ELL/LEP As needed None N/A Maria 2nd SRA/EIR No ELL/LEP No None N/A Jennifer 2nd HR Yes ELL/LEP Yes (30 min/3 times a week) None N/A Aldo k, 2nd SRA/EIR Yes ELL/LEP Yes (30 min/3 times a week) Language disability Speech therapy Jorge k, 2nd SRA/EIR No ELL Yes (30 min/3 times a week) None N/A HR = Harcourt Reading Interventions, SRA/EIR = SRA Early Interv entions in Reading, ELL = English language learners, and LEP = l imited English proficient. For ESE services, N/A stands for not applicable. This was used fo r students that did not have an ESE classi fication. None was used when a student had an ESE classifica tion, but did not receive ESE services.

PAGE 85

85 Table 3-4. Participants reading scores used for selection purposes Name SAT-10 (percentiles) DIBELS/PS F (number of phonemes per minute) DIBELS/NWF (number of correct lettersounds read per minute) DIBELS/ORF (number of correct words per minute) Invented Spelling (percentiles) Amelia 42 56 (AA)67 (LR)48 (LR) 20-30 Pedro 35 89 (AA)68 (HR) 20 Jessica 46 36 (LR)52 (LR)57 (LR) <10 Viviana 9 49 (LR)49 (MR)26 (MR) <10 Ernesto 13 43 (LR)46 (MR)26 (MR) 10-20 Maggie 13 58 (LR)39 (HR) <10 Maria 15 58 (LR)23 (HR) 20-30 Jennifer 49 (MR)2 (HR) <10 Aldo 7 4 (HR)4 (HR) <10 Jorge 11 17 (HR)5 (HR) <10 Scores for SAT-10 and DIBELS correspond to the end of the previous academic year. For Amelia, Jessica, Viviana, and Ernesto, those scores correspond to the end of first grade. For the rest of the students, the scores correspond to the end of second grade, since they were retained the previous year. Those students do not have PSF scores because that test is not implemented at the end of second grade. No SAT-10 score was available for Jennifer. Invented spelling scores correspond to the beginning of the current year. Table 3-5. Students English and Span ish language levels based on the WMLS-R Name English CALP English language ability Spanish CALP Spanish language ability Amelia Level 3 Limited Level 3 Limited Pedro Level 3 Limited Level 4 Fluent Jessica Level 3 Limited Level 3 Limited Viviana Level2 Very limited Level 3 Limited Ernesto Level 3 Limited Level 3 Limited Maggie Level 3 Limited Level 3 Limited Maria Level 4 Fluent Level 3 Limited Jennifer Level 1 Negligible Level 3 Limited Aldo Level 3 Limited Level 2 Very limited Jorge Level. 3.5 Limited to fluentLevel 3.5 Limited to fluent CALP scores correspond to the level of cognitive-ac ademic language proficiency. CALP scores go from Level 1 (negligible English ability) to Level 6 (very advanced English ability). Spanish and English at home. According to Amelia, she uses both languages to communicate with her parents. Information about her parents level of education is unknown. School records showed that she had attended school in the US at least since kindergarten. Her teacher r eported that Amelia receiv ed the Harcourt Reading

PAGE 86

86 Table 3-6. Conners Abbreviat ed Teacher Rating Scale (C-ATRS) Name Restles or Overactive Excitable/ Impulsive Disturbs others Fails to finish things Constant Fidgeting Inattentive Easily frustrated Cries often Quick mood changes Temper outburst/ unpredictable behavior Amelia NA NA JL JL JL JL PM JL JL JL Pedro PM JL PM NA JL JL NA NA NA NA Jessica NA NA NA NA NA NA NA NA NA NA Viviana NA NA NA NA NA NA NA NA NA NA Ernesto VM PM NA JL VM VM JL NA NA NA Maggie NA NA NA NA NA JL NA JL NA NA Maria NA NA NA NA NA NA NA NA NA NA Jennifer NA NA NA NA NA NA NA NA NA NA Aldo PM PM JL PM JL VM NA NA NA NA Jorge PM JL PM VM VM VM JL NA NA NA For the C-ATRS, NA stands for not at all, JL stands for just a little, PM stands for pretty much, and VM stands for very much. Intervention program 45-minutes a day, five days a week. Amelia was identified by the school as being LEP (limited English profic ient) and enrolled in LEP classes, but her teacher stated that Amelia di d not receive these services. At the end of first grade, Amelia scored in the 42nd percentile on the SAT-10. Her DIBELS scores for the end of first grade were 56 on PSF (above average), 67 on NWF (low risk), and 48 on ORF (low risk). At the beginning of second grade, she scored in the 20th to 30th percentile on the Invented Spelling Assessment. Based on the WMLS-R, Amelia demonstrated limited English and Spanish oral language proficiency (Level 3). On the C-ATRS, her teacher reported that Amelia would get easily frustrated and tended to be a little distracted in class. She st ated that Amelia does not like to not know what to do. She also affirmed that Amelia withdraws if s he is unsure of something. One area of concern reported by her teacher on different occasions was that Amelia had no support from home. Her teacher also reported that Amelias mother had spent

PAGE 87

87 extensive time in jail, and attempts for wr itten or oral communica tion with her guardian tended to be unsuccessful. Pedro. Pedro was a Hispanic male, aged 8 years-8months, enrolled in second grade for the second time. His parents were from Cuba and they spoke Spanish and English, but only used Spanish to communica te with Pedro. Pedro s parents stated that their highest level of educati on corresponded to community college. They also reported that Pedro had had four years of schooling in the US. His current teacher stated that Pedro received SRA Early Reading Interventi ons (SRA/EIR) Level 1 (first grade) for 45 minutes a day, five days a week. The school considered Pedro to be an English language learner, but he no longer had an LEP status. He did not receive LEP classes. At the end of second grade from the previous academic year, Pedro scored in the 35th percentile on the SAT-10. His DIBELS scor es for the end of second grade were 89 on NWF (above average) and 68 on ORF (high risk). At the beginni ng of second grade of the current year, he scored in the 20th percentile on the Invented Spelling Assessment. Language scores based on the WMLS -R, showed that Pedro had limited English oral language ability (Level 3) and fl uent Spanish oral language ability (Level 4). According to teachers report on the C-ATRS Pedro is a good student but likes to talk to other students. The teacher also reported that he was a little bit impulsive and easily distracted. Jessica Jessica was a Hispanic female, aged 8 years-3months, enrolled in second grade for the first ti me. Her parents were from Mexico and they spoke English and Spanish. At home, they used Spanish to communica te with Jessica. Her parents reported not having any schooling experience. Jessicas parents stated that she had

PAGE 88

88 received six years of school ing in the US. School records showed that she was previously retained in kindergarten. In second grade, Jessica participated in the SRA/EIR program Level 1 (fir st grade) for 45 minutes a day five days a week. Jessica had an LEP status and the school reported that she was enrolled in LEP classes; yet, her teacher affirmed that she di d not receive those services. Reading scores for the end of first gr ade showed that Jessica scored in the 46th percentile on the SAT-10. Her DIBELS at the time were 36 on PSF (low risk), 52 on NWF (low risk), and 57 on ORF (low risk). At the beginning of second grade, she scored below the 10th percentile on the Invented Spelling Assessment. Language scores on the WMLS-R showed that Jessica had limited Eng lish and Spanish oral language ability (Level 3). On the C-ATRS, her teacher reported that Jessica did not have any attitude or behavior problems in class. Viviana. Viviana was a Hispanic female, aged 8 years-5 months, enrolled in second grade for the first ti me. Her mother was from Guatemala and they spoke Spanish and Conjoval at hom e. Conjoval is a Mayan or al language, with no written form. To communicate at home, they used Spanish. Her mother r eported not finishing high school. Her mother stated that Viviana had had five years of instruction in the US. School records showed that she was retained in kindergarten. In second grade, she received Harcourt Reading Interventions in sm all groups, five times a week, for 45 minutes. Viviana had both migrant and LEP status. She received LEP classes 3 times a week for 30 minutes. She was also identif ied as having a language disability. Her teacher reported that Viviana did not receive ESE services.

PAGE 89

89 Reading scores showed that at the end of first grade, Viviana scored in the 9th percentile on the SAT-10. Her DIBELS scores for the end of first grade were 49 on PSF (low risk), 49 on NWF (moderate risk), and 20 on ORF (moderate risk). At the beginning of second grade, she scored below the 10th percentile on the Invented Spelling Assessment. Language scores on the WMLS-R demonstrated that she had very limited English oral language ability (Level 2) and li mited Spanish oral language ability (Level 3). The teachers report on the C-ATRS sh owed that Viviana had no attitude or behavior problems in the class. Her teacher stated on many occasions that Viviana was a very sweet girl, very respectful and that she always smiled. Ernesto. Ernesto was a Hispanic male, aged 8 years-10months, enrolled in second grade for the first time. He was previously retained in second grade. His parents were from El Salvador and they only spok e Spanish at home. The highest level of education that parents reported having was high school. Ernesto had had five years of schooling in the US. His t eacher reported that he rece ived the SRA/EIR program Level 1 in small groups, for 30 minutes a day, five days a week. He also received the Harcourt Interventions small group program fi ve times a week for 45 minutes. Ernesto had an LEP status but did not receive LEP classes, despite records showing that he was enrolled in them. He was identified as having a language disability and he received speech therapy two times a week for 20 mi nutes. Ernesto was also referred to the school psychologist to be tested for autism, but result s are still pending. At the end of first grade, Ernesto scored in the 13th percentile on the SAT-10. His DIBELS scores for the end of first grade were 43 on PSF (low risk), 46 on NWF (moderate risk) and 26 on ORF (m oderate risk). At the beginnin g of second grade, he

PAGE 90

90 scored between the 10th and 20th percentile on the Invented Spelling Assessment. Language results on the WMLS-R showed that Ernesto had limited E nglish and Spanish oral language ability (Level 3). On the CATRS, his teacher reported that he is sweet and loves to learn, but tends to be restless, impulsive and inattentiv e. She stated that Ernesto lacked focus in a whole group but did much better on one-on-one or small group. Maggie. Maggie was a Hispanic female, aged 8 years-8months, enrolled in second grade for the second time. She was re tained once in kindergarten as well. Her parents were from Mexico and they only sp oke Spanish at home. They reported not having any schooling experience. They also stated that Maggie ha d had five years of schooling in the US. Her t eacher reported that Maggie participated in the SRA/EIR intervention for 45 minutes, five days a we ek. She was identified as being LEP and receiving LEP classes, but her teacher stated that those services were provided only as needed. Reading scores showed that at the end of second grade from the previous year, she scored in the 13th percentile on the SAT-10. Her DIBELS scores for the end of second grade were 58 on NWF (low risk), an d 39 on ORF (high risk). At the beginning of second grade of the current year, she scored below the 10th percentile on the Invented Spelling Assessment. Based on t he WMLS-R, Maggie demonstrated limited English and Spanish oral language ability (Level 3). Her teacher reported on the CATRS that Maggie would get easily frustr ated and tended to cry often and easily. She also affirmed on different occasions that sh e was insecure and constantly looked for her approval.

PAGE 91

91 Maria. Maria was Hispanic female, aged 8 year s-8 months, enrolled in second grade for the second time. Her parents were from Mexico and they only spoke Spanish at home. Their highest level of education reported was high school. Maria had had five years of schooling in the US. In second gr ade, she received 45 minutes of SRA/EIR small group intervention, five times a week School records showed that Maria had an LEP status and was enrolled in LEP classes; still, her teacher repor ted that she did not receive those services. Maria was the only st udy participant who did not participate in the free or reduced-pr ice lunch program. Marias end-of-second-grade scores from the previous year showed that she scored in the 15th percentile on the SAT-10. Her DIBELS scores were 58 on NWF (low risk) and 23 on ORF (high risk). At the beginni ng of second grade of the current year, she scored between the 20th and 30th percentile on the Invented Spelling Assessment. Language scores on the WMLS-R demonstrated that Maria had fluent English oral language ability (Level 4) and limited Spanish oral language abi lity (Level 3). On the CATRS, her teacher stated that Maria did not have any attitude or behavioral problems in class. Jennifer. Jennifer was a Hispanic female, aged 10 years-1 month, enrolled in second grade for the second time. Her mot her was from Guatemala and they only spoke Spanish at home. Her mother report ed not having any schooling experience. She also stated that Jennifer had had one year of schooling in the US. Her teacher reported that she received 45 minutes of Harcourt In tervention five times a week. She had both migrant and LEP status. School records showed that she received LEP classes three times a week for 30 minutes.

PAGE 92

92 Reading scores at the end of second grade from the previous year showed that Jennifer scored 0 on the SAT-10. Her DIBELS scores for the end of second grade were 49 on NWF (moderate risk), and 2 on ORF (high risk). At the beginning of second grade of the current year, she scored below the 10th percentile on the Invented Spelling Assessment. The WMLS-R showed that Jenni fer had negligible Engl ish oral language ability (Level 1) and limited Spanish oral language ability (Level 3). On the C-ATRS, her teacher reported that Jennifer had no attit ude or behavioral problems in class. On one occasion, her teacher affirmed that one of Jennifers biggest challenges was her limited English oral skills, but t hat she was eager to learn. Aldo. Aldo was a Hispanic male, aged 9 years-3months, enrolled in second grade for the second time. He was also previously retained in kindergarten. His parents were from Mexico and were able to speak Engl ish and Spanish. At home, they only used Spanish to communicate with Aldo. They r eported not having any schooling experience. Aldos parents reported that he had had 4 years of schooling in the US. Aldo participated in the SRA/EIR program Level 1, 45 minutes a day, five days a week. School records showed that Aldo had migrant and LEP status. He participated in LEP classes three times a week for 30 minutes Aldo was identified as having a language disability and received speech therapy one time a week for 30 minutes. At the end of second gr ade the previous year, he scored in the 7th percentile on the SAT-10. His DIBELS scores for the end of second grade were 4 on NWF (high risk) and 4 on ORF (high risk). At the beginning of second grade of the current year, he scored below the 10th percentile on the Invented Spelling Assessment. Based on the WMLS-R, Aldo demonstrated limited English oral language ability (Level 3) and very

PAGE 93

93 limited Spanish oral language abi lity (Level 2). Aldos teac her reported on the C-ATRS that he had a short attention span, and t hat he was restless, impulsive, and easily distracted. She also stated t hat Aldo did not make much pr ogress the previous year and that he was the lowest perfo rming student in her class. Jorge. Jorge was a Hispanic male, aged 9 year s-5 months, enrolled in second grade for the second time. He was also prev iously retained in kindergarten. Jorges parents were from Mexico and they only spoke Spanish. They reported not having any schooling experience. Jorges parents stated that he had had 5 years of schooling in the US. His teacher affirmed that Jorge participa ted in the SRA/EIR program for 45 minutes a day, five days a week. School records showed that Jorge no longer had an LEP status, but his teacher affirmed that LEP classes were provided as needed. At the end of second grade of the pr evious year, he scored in the 11th percentile on the SAT-10. His DIBELS scores for the end of second grade were 17 on NWF (high risk) and 5 on ORF (high risk). At the beginni ng of second grade of the current year, he scored below the 10th percentile on the Invented Spelling Assessment. Based on the WMLS-R, Jorge demonstrated limited to fl uent English and Spanish oral language ability (Level 3.5). On t he C-ATRS, his teacher reported that he was impulsive and restless in class, and that he tended to avoi d work and to disturb his classmates. She also affirmed that Jorge had a short attenti on span and was very inattentive in class. According to her, one of Jorges biggest ch allenges was his behavior. He had been referred and suspended for stealing, for being disrespectful, and for making threatening statements toward his peers.

PAGE 94

94 Assessment Instruments There were four types of assessments instruments given to the participants: (a) screening instruments, (b) grouping instrument s, (c) supplemental instruments, and (d) probes of student progress on the dependent variabl e. The following sections contain a description of each of the first three types of assessments. A description of the probes is presented in the Dependent Variables section. Screening Instruments As part of the selection process, students scores on three measures of reading performance were analyzed to help determine e lig ibility for this study: (a) SAT-10, (b) DIBELS, and (c) Invented Spel ling Assessment. Additional measures were given to selected participants to establish their level oral language proficiency (WMLS-R), and their classroom attitudes and behaviors (C-ATR S). What follows is a brief description of each measure. Stanford Achievement Test-10 (SAT-10). The SAT-10 is a group-administered, standardized measure of reading achievement for grades K to 12, currently used in Florida Reading First schools. The test is untimed and uses a multiple-choice format. The reading portion assesses phonemic aw areness, decoding, phonics, vocabulary, and comprehension. Scoring is conducted by the test publisher. Student scores are reported in percentiles. Scores for SAT -10 were obtained from the school. Dynamic Indicators of Basic Early Literacy Skills (DIBELS). DIBELS (Good & Kaminski, 2002) is a set of tests designed to screen and monitor students reading skills from K to 3rd grade. In this study, the latest end-of-year scores were obtained from the school for each of the 10 participants. For Am elia, Jessica, Viviana, and Ernesto the DIBELS scores reported corresponded to the end of first grade. For all other

PAGE 95

95 participants, end-of-year scores corresponded to the end of second grade because they had been retained. The tests administered at the end of first grade are (a) Phoneme Segmentation Fluency (PSF), (b) Nonsense Word Fluency (NWF), and (c) Or al Reading Fluency (ORF). All these tests ar e timed with the purpose of assessing students level of automaticity. The PSF is a te st of phonological aw areness that evaluates the ability to fluently segment words into phonemes. The test takes approximately 2 minutes to administer. The benchmark goal for the end of first grade is >35. The alternate-form reliability of the PSF is .79. The NWF is a test of the alphabetic principle and the ability to blend letters into words. It takes appr oximately 2 minutes to administer. The benchmark goal for the end of first grade, as well as the end of second grade is >50 correct letter sounds per minute. The alternate-form reliability for the NWF is .83. The ORF is a test of reading accuracy and fluenc y of connected text. It uses a set of passages that students read aloud for one minute. The oral reading fluency rate corresponds to the number of correct words read per minute. The benchmark goal for the end of first grade is >40 correct words per minute and for the end of second grade is >90 correct words per minute. The DIBELS pr ovides four types of risk status that identify need for additional and substantial intervention: (a) above average (at or above the 60th percentile, no need for additional inte rvention), (b) low risk (at grade level, no need for additional intervention, (c) moderat e risk (moderately below grade level, requires additional intervention), and (d) hi gh risk (significantly below grade level, requires substantial intervention; Flor ida Center for Reading Research, 2006).

PAGE 96

96 Invented Spelling Assessment: A Sound Beginning. The Invented Spelling Assessment (Lane & Pullen, 2004) was developed as a measure of phonological awareness and understanding of t he alphabetic principle. In this assessment, the examiner dictates ten unfamiliar words and the student atte mpts to write them. The examiner then assigns point s to each word based on its phonological accuracy. The scoring scale goes from 0 to 4 points based on the level of phonological accuracy of the students spelling. On average, a score below three points per word indicates that the student lacks appropriate phonological skills. The Invented Spelling Assessment offers percentile ranks for fall of second grade based on a sample of 2,000 students in kindergarten to second grade. The assessment has a interscorer relia bility of >.97 (Lane & Pullen, 2004). A score below the 40th percentile indicates a need for reading intervention. Woodcock-Muoz Language Survey Revised (WMLS-R). The WLMS-R (Woodcock, Muoz-Sandoval, Ruef, & Alvarado, 2005) is an individually administered test of oral language, languag e comprehension, reading and wr iting proficiency. It was developed with the purpose of determining language proficiency, language dominance, changes in language ability, eligibility for educational services, readiness for Englishonly instruction, among others. The survey has two parallel forms, one in English and one in Spanish, which can be used with indivi duals ranging from 2 years old to more than 90 years old. Each form takes appr oximately 45 minutes to administer. The WMLS-R English form has seven tests: Test 1-Picture Vo cabulary, Test 2Verbal Analogies, Test 3-Letter-Word I dentification, Test 4-Dictation, Test 5Understanding Directions, Test 6-Story Recall, and Test 7-Passage Comprehension. A

PAGE 97

97 cluster score for oral language-total can be obt ained by combining tests 1, 2, 5, and 6. This cluster represents a broad measure of language competency. The WMLS-R Spanish form has seven equivalent tests: Te st 1-Vocabulario sobre Dibujos, Test 2Analogas Verbales, Test 3-Identificacin de Letras y Palabras, Test 4-Dictado, Test 5Comprensin de Indicaciones, Test 6-Rememoracin de cuentos, and Test 7Comprensin de Textos. The corresponding oral language total (lenguaje oral total) can be obtained by combining tests 1, 2, 5, and 6 (Alvarado, Ruef, & Schrank, 2005). The standardization sample of the WM LS-R English form consisted of 8,818 individuals, ages 2 to over 90 years old, repr esenting different regions of the country, different races, types of schools, etc. T he calibration sample fo r the WMLS-R Spanish form consisted of 1,157 native Spanish-spea king individuals from different regions inside and outside the United States (e.g., Mexico, Argentina, Panama, Costa Rica, Colombia, and Puerto Rico; Alvarado et al., 2005). The split-half test of reliability ranged from .76 to .97 for tests and fr om .88 to .98 for clusters. One of the scores yielded by the WLMS -R is the Cognitive-Academic Language Proficiency (CALP). The CALP measures students proficiency with context reduced and cognitively demanding langu age (Alvarado et al., 2005, p. 61). CALP scores have six levels and two regions of uncertainty fo r the different clusters. In each of these levels, cognitive-academic langu age proficiency is identified as it compares to others of the same age or grade level. Each level has s pecific implications for instruction as it relates to context-reduced and cognitively demanding language learning tasks in each language. At Level 1 students proficiency is negligible (imperceptible). They are likely to find language demands impossible to handle. In Level 2 proficiency is very limited (muy

PAGE 98

98 limitado) and students are ex pected to find language demands extremely difficult. Students in Level 3 have limit ed (limitado) proficiency a nd find language demands very difficult. In Level 4, proficiency is fluent (fluido). Students tend to find language demands manageable. Level 5 is characterized by advanced (avanzado) proficiency, where students are likely to find language demands very easy. Finally, in Level 6 proficiency is very advanced (muy avanzado) and students ar e expected to find language demands extremely easy. In addition to the six leve ls, the two regions of uncertainty include Level 3-4 (3.5), where students have limited to fluent proficiency (limitado a fluido) and Level 4-5 (4.5), where students have fluent to advanced proficiency (fluido a avanzado; Alvarado et al., 2005). In this study, CALP scores for the English and Spanish Oral Language-Total cluster were used. Conners Abbreviated Teacher Rating Scale (C-ATRS). The C-ATRS is an abbreviated form of the Conners Teacher Ra ting Scale, developed to assess teachers perceptions of their students behavior in t he classroom (Conners, Sitarenios, Parker, & Epstein, 1998). The scale has 10 items that describe different childrens behaviors (i.e., restless, excitable, inattentive). For each it em, the teacher is asked to rate how much the child has shown the behavior in the last month. There are four possible options: not at all, just a little, pretty much, or very much. Home language survey. The Home Language Survey was developed by the researcher with the purpos e of obtaining demographic data and information about language practices taking place at the student s homes. The survey was translated to Spanish to provide an option to parents who feel more comfortable reading in Spanish. Demographic information and responses ab out the language practices are reported on

PAGE 99

99 an individual basis to enhance the profile in formation of each participant. Each of the participating students t ook home a survey, accompanied by a reminder letter that was also translated to Spanish (see Appendix B fo r a copy of both reminder letters) that explained the purpose of the su rvey, how to respond, and how to send it back. The response rate was 90%. Grouping Instruments With the purpose of assigning selected participants to an appropriate tutoring group, the researcher conducted measures of reading accuracy by taking running records of students reading leveled books. A r unning record (Clay 1972) is an untimed measure of oral reading accuracy. The pur pose is to analyze reading behaviors that shed light into the type of reading skills and strategies that students use when they interact with text (Denton, Ciancio, & Flet cher, 2006). Running records can be used to (a) estimate the rate of reading accura cy, (b) identify students reading levels (independent, instructional, or fr ustrating), (c) group students fo r instruction, (d) monitor reading progress, and (e) identify areas of strength and weakness, among others. In this study, running records served to establish rates of reading accuracy in order to identify students instructional reading levels. Instruct ional levels refer to texts that are read with 90% to 95% accuracy and that should be the focus of instruction. The percentage of accuracy was calculated by di viding the number of words read correctly by the total number of words read, and mult iplying it by 100. The process was as follows. The researcher selected two sets of books leveled 1 to 20 according to Reading Recovery guidelines. Based on students init ial DIBELS ORF scores, the researcher provided a starting leveled book that approximated their reading ability. Students who read the book at the instructional level were given a second book at that same level for

PAGE 100

100 corroboration purposes. Students who read with an accuracy of more than 95% (independent level) were given upper leve led books until an instructional level was reached. In contrast, students who read the first book with lower than 90% accuracy (frustration level) were given lower leve led books until an inst ructional level was attained. Again, when students read a leve led book with 90% to 95% accuracy, a second book at the same level was prov ided to confirm that it was indeed their instructional level. Students book levels read in each trial and their percentages of accuracy are reported in Table 3-7. Table 3-7. Running records re sults used for grouping purposes Name Trial 1 Trial 2 Trial 3 Trial 4 Amelia 10 (88%) 9 (94%)9 (93%) Pedro 10 (94%) 10 (95%) Jessica 10 (92%) 10 (95%) Viviana 1 (98%) 2 (90%)2 (93%) Ernesto 1 (98%) 2 (93%)2 (93%) Maggie 5 (88%) 4 (89%)3 (94%)3 (91%) Maria 1 (98%) 2 (100%)3 (91%)3 (93%) Jennifer 1 (71%) Aldo 1 (<50%) Jorge 1 (83%) For each trial, the number outside the parenthesis refers to the book level read, while the number in parenthesis corresponds to the percentage of readi ng accuracy. The number of trials varied across students. Once instructional levels were identi fied for each student, those with similar reading levels (within one or two levels of difference) were grouped for intervention. A total of three instructional groups were forme d. Refer to Table 3-1 for the summary of the composition of these groups. Supplemental Preand Post-Intervention Instruments With the purpose of establishing a comprehensive student readi ng profile that could help in measuring and interpreting student s response to the tutoring program, the

PAGE 101

101 researcher conducted preand post-intervent ion measures of reading ability. These included word reading skills, fluency, compr ehension, and attitudes toward reading. In addition, teachers were ask ed to rate their students on five areas of reading and classroom behaviors before and after the inte rvention. What follows is a detailed description of each measure. Test of Word Reading Efficiency (TOWRE). The TOWRE (Torgesen, Wagner, & Rashotte, 1999) is a norm referenced, indi vidually administered, measure of word reading efficiency that assesses the ability to read printed words with accuracy and fluency. The purposes of this test are to help monitor student growth on phonemic decoding and sight word reading, to identify students who might need more instruction, and to help identify reading disabilities as par t of a battery of diagnostic tests. The TOWRE has two subtests: Sight Word Efficiency (SWE) and Phonemic Decoding Efficiency (PDE). The SWE measures the number of real words that can be accurately identified, while the PDE m easures the number of nonwords that can be accurately decoded. Each subtests lasts 45 seconds (Torgesen, Wagner, & Rashotte, 1999). The standardization sample for the TOWR E included 1,500 subjects, ages 6 to 24 years old, from different regions of the count ry. Alternate-form reliability scores are .93 for the SWE, .94 for the PDE, and .96 for the Total Word Reading Efficiency (Torgesen et al., 1999). In this study, raw scores and percentile ranks for SWE, PDE and Total Word Reading Efficiency were reported. Kaufman Test of Educational Achi evement-II (KTEA-II) NWD and LWR. The KTEA-II (Kaufman & Kaufman, 2004) is a norm referenced, individually administered, measure of academic achievement for i ndividuals between the ages of 4 years, 6

PAGE 102

102 months through 25 years. Two subtests were used in this study: the Nonsense Word Decoding (NWD) and the Letter-Word Rec ognition (LWR). The NWD requires the student to decode increasingly difficulty nonsense words, while the LWR requires the student to identify letters and decode increasingl y difficult real word s. Both tests are untimed. It is estimated that the NWD takes 3 minutes and the LWR takes 4 minutes to administer (Kaufman & Kaufman, 2004). The mean split-half reliability coefficient fo r the NWD subtest is .94 for grade and age norms. The mean reliability coefficient for the LWR is .97 for age norms and .96 for grade norms (Kaufman & Kaufman, 2004). In this study, raw scores and percentile ranks for both subtests were reported. DIBELS Nonsense Word Fluency (NWF) and Oral Reading Fluency (ORF). Since the DIBELS scores used for screeni ng purposes corresponded to the end of the previous academic year, the researcher wa s interested in obtaining beginning of year NWF and ORF scores to determine the le vel of performance before and after intervention. Given that the school disc ontinued the use of DI BELS assessments, the researcher administered both tests. The NWF is a test of t he alphabetic principle and the ability to blend letters into words. The test is individually administered and it takes approximately 2 minutes to complete it. T he student is provided with a set of VC and CVC nonsense words in random order. The student is asked to read either individual letter-sounds or whole nonsense words in one minute. Scores represent the number of correct letter-sound correspondences read per minute. The benchmark goal for the beginning and end of second grade is >50 correct letter sound correspondences per minute (Good & Kaminski, 2002).

PAGE 103

103 The ORF is a standardized test of accu racy and rate for reading connected text. The test is individually administered to students with the purpose of identifying those who are in need of additional instruction, as well as to monitor progress. The ORF utilizes a set of standardized grade-leveled passages that the students read aloud for one minute. During this time, an examiner re cords the number of words read correctly and the number of errors. The test yields a rate of oral reading flue ncy that corresponds to the number of words read correctly per mi nute. Passages are available for first grade (middle and end of year), second grade (begi nning, middle, and end of year), and third grade (beginning, middle, and end of year). The benchmark goals for the beginning of second grade are >44 correct words per minut e. As stated previously, the benchmark goal for the end of second grade is >90 correct words per minute. Test-retest reliability coefficient for the ORF ranges from .92 and .97. Alternate-form re liability for passages within a grade level ranges from .89 to .94 (Tindal, Marston, & Deno, 1983). At the beginning of the study, each partici pant was given a passage to read that corresponded to a beginning second grade leve l (i.e., My Handprints). Students who were at low risk or above average were not given further passages. Students who were at moderate or high risk were given a pa ssage corresponding to the end of first grade (i.e., The Sand Castle). From these, student s who still were at moderate or high risk were given a middle of first grade passage (i.e., Ice Cream). A similar process was followed at the end of the study, except that students started reading passages corresponding to the end of second grade (i.e., If I had a Robot). Also, to avoid student fatigue, those who were at moderate risk or high risk on the end of grade passage were first given a beginning second grade passage and if they were at low risk or above

PAGE 104

104 average, then the middle of second grade passage (i.e., Moving Day) was administered. Otherwise, students kept reading first grade passages in the same way described above. This procedure had two purposes: (a) to provide an estimation of the current level of reading fluency of each student, and (b ) to serve as a decision making process for selecting passages from t he QRI-4 to be administered. Qualitative Reading Inventory-4 (QRI-4)-Comprehension. The QRI-4 (Leslie & Caldwell, 2006) is an informal reading inventor y designed to estimate students oral and silent reading levels, as well as comprehens ion skills. The QRI-4 employs narrative and expository passages that span from pre-prim er to high school. Comprehension skills for first and second grade are determined throug h story retell, as well as open-ended questions. Once a student finishes reading a passage, students are asked to retell the story and answer implicit and explicit ques tions. Scores reported include the total number of ideas recalled (retell) and the tota l number of correct answers given for each passage. Based on the total number of correct answe rs, the QRI-4 provides three different levels at which students comprehend text: independent, instructional, and frustration. According to the tests criter ia, students at the independent level are able to read the passage successfully without assistance and are able to answer 90% of the questions correctly. Students at the instructional le vel read passages with assistance and answer 70% of the questions correctly. On the frustrat ion level, students are unable to read the passage and answer less than 70% of the questi ons correctly. Alternate-form reliability coefficients for the QRI-4 are .80 for comp rehension scores on two different passages at the same readability level. Inter-scorer reliability for scoring oral reading miscues is

PAGE 105

105 99%, while reliability for scoring comprehensi on questions is 98% (Leslie & Caldwell, 2006). The implementation of QRI-4 passages wa s contingent on students ORF scores. The decision making process was as follows. If a student was at low risk, or above average on any of the second grade passages (beginning, middl e, or end of year) of the ORF, a QRI level two passage was administered (i.e., What Can I Get for My Toy?). If a student was at moderate risk or high risk on second grade passa ges of the ORF, and at low risk or above average on the end of first grade passage, a QRI level one passage (i.e., The Bear and the Rabbit) was adminis tered. Furthermore, if a student was at moderate or high risk on the end of first grade passage from the ORF, and was at low risk or above average on the middle of firs t grade passage of the ORF, a QRI primer level passage (i.e., Fox and Mouse) was adminis tered. On the other hand, if a student was moderate or high risk on the middle of first grade passage from the ORF, no QRI passage was administered. The purpose of this procedure was to lower students frustration when reading text that is beyond their reading abilities. Reading Ability Rating Scale (RARS). The Reading Ability Rating Scale (RARS) for teachers was created by the researc her to obtain information about students changes in reading skills and classroom behavio rs as a result of the intervention (Appendix C). The checklist has 26 items tota l, 23 related to reading ability (e.g., phonological awareness, phonics, fluency, co mprehension, vocabulary) and three related to classroom behaviors (e.g., partici pation in class, motivation to read, and use of English to communicate in class). Teachers were asked to rate each of their students before and after the intervention, using a 4-poin t Likert scale that ranges from very weak

PAGE 106

106 to very strong. For each of the five areas of reading, a mean score was calculated and reported. For classroom behaviors, raw sco res on each of the three items were reported. Elementary Reading Attitude Survey (ERAS). The ERAS (McKenna & Kear, 1990) is a group or individually administered instrument designed to measure students attitudes toward recreational and academic reading (Kazelskis, Thames, & Reeves, 2004). The survey can be used with students fr om first through sixth grade. According to McKenna and Kear, the survey can be used to monitor the impact of instructional programs on students attitudes. The survey has a total of 20 items, 10 related to recreational reading and 10 rela ted to academic reading. With young students, the tutor reads the items together with the students. Students are asked to circle the picture that best describes how they feel about each item. There are four possibl e pictures that go from very happy to very ups et. Approximate completion time is 10 minutes. The survey yields three scores: total for recreational reading, total for academic reading, and a composite total. Scores can range from 10 to 40 points in each scale, and 20 to 80 points for the composite (McKenna & Kear). For this study, raw scores and percentile ranks were reported. The norming sample consisted of 18, 138 students in grades 1 through 6 from different regions of the country. These samples represented a variety of races. Cronbach alpha reliability coeffi cient for the survey ranged from .74 to .89 across grade levels (McKenna & Kear, 1990). Intervention The UFLI tutoring program is an early literac y intervention designed to help struggling readers develop beginning reading skills. Originally, UFLI was designed as

PAGE 107

107 an education tool to help prepare pre-servic e teachers in the area of reading. By conducting one-on-one tutoring wit h struggling readers, teachers are able to learn about the reading process and how it is acquired by children, the difficulties that many students face when they lear n to read, and the instructional methods that help struggling readers overcome their reading difficulties. The tutoring program was founded on current research regarding (a) early literacy development and (b) effective instructional strategies for struggling r eaders (Lane, Pullen, H udson, & Konold, 2009). Although UFLI was originally designed to be implemented in a one-on-one setting, several small-group modifications of the tutoring model have been successfully implemented (e.g., Pullen, 2000; Pullen, Lane, Lloyd, Nowak, & Ryals, 2005). The focus of this study was on sm all-group implementation. Early Literacy Components Addressed in UFLI Each UFLI tutoring session is designed to promote the develop ment of phonemic awareness, print awareness, decoding, reading fluency, comprehension, and strategy use (Lane et al., 2009). Some of these skills were addressed multiple times throughout the lesson. What follows is a description of how each reading component was addressed in each session. Phonemic awareness. In each tutoring session, phonemic awareness was developed in two ways. First, students we re required to count and identify phonemes using Elkoning boxes. Second, they l earned to blend and s egment phonemes using manipulative letters (Hayes, Lane, & Pullen, 2005; Lane et al., 2009). Print awareness. Print awareness was developed in every lesson through book reading and sentence writing activities. When students read new and familiar books, students were able to handle books, turn pages, and point to different parts of the text.

PAGE 108

108 By doing this, they learned about different parts of the book, reading directionality, and conventions of print (e.g., concept of wo rd, spacing, punctuation). In addition, when students were asked to write a sentence relat ed to the story read, they also practiced directionality, spacing, and punctuation (Hayes et al., 2005; Lane et al., 2009). Decoding. Decoding skills were developed during word work activities with manipulative letters, as well as reading and wr iting practice. In word work activities, students were required to manipulate letters in order to encode and decode real and nonsense words at the onset-rime and phonem e levels. During reading practice, students were required to decode new and unf amiliar words while repeatedly reading familiar books and while reading new and unfam iliar books for the fi rst time. During writing practice, students were asked to identify the phonemes in words, to encode the words using Elkonin boxes, and rewrit e words on a sentence page. With ample opportunities for practice, UFLI promoted decoding accuracy and automaticity (Hayes et al., 2005; Lane et al., 2009). Fluency. Fluency was primarily developed by repeated readings of connected text. Students reread familiar books at an independent reading level (more than a 95% accuracy). In the beginning, the focus of the tutoring was on developing reading accuracy. Once students became more accu rate while reading connected text, the focus shifted to reading automaticity. Finally, when students acquired appropriate reading automaticity the focus of the session s shifted to prosody (Hayes et al., 2005; Lane et al., 2009). Comprehension. Comprehension skills were develop ed before, during, and after reading connected text. Before the students read a new book, the tutor introduced the

PAGE 109

109 book and engaged the group in a picture walk with the goal of activating prior knowledge and creating context for the story. Then, students were encouraged to make connections and predictions related to the st ory. During reading, the tutor modeled and guided students to self-monitor for comprehension. After reading, the tutor and students engaged in a discussion of the st ory through the use of literal inferential, and evaluative questions. Children were also asked to su mmarize part of the story and then write a summary sentence (Hayes et al., 2005; Lane et al., 2005). Strategy use. Throughout the UFLI tutoring progr am, students acquired different reading strategies to identify words and che ck for accuracy. Strategies taught included use of grapho-phonemic information, semantic and syntactic cues to confirm decoding accuracy, monitoring, cross-checking, and se lf-correcting, among others. In addition, the tutor scaffolded the use of strategies by using the ABC mnemonic, which stands for Acquire, Build, Control. First, the tutor demons trated and modeled the use of a new reading strategy that student s needed to acquire. Then, as students built their strategy repertoire, the tutor prompted students to use specif ic strategies. Finally, once children had control over t he strategies, the tutor observed how students applied them to connected text and asked them to explain how they were able to figure out specific words. It was expected that students who we re able to select appropriate strategies without prompting and explain accurately how the strategies worked would be able to use them independently (Hayes et al., 2005; Lane et al., 2009). Effective Instructional Practices for ELLs The UFLI tutoring program included resear ch-based instructional practices that facilitate le arning among E nglish language learners. Thes e were (a) explicit and systematic instruction (Ger sten & Geva, 2003; Helman, 2009; Manyak & Bauer, 2008;

PAGE 110

110 Vaughn et al., 2006), (b) small-group instru ction (Gersten et al., 2007), (c) ample opportunities for practice (Gersten & Geva 2003), (d) assessment of students progress (Gersten & Geva, 2003; Helman, 2009), (e) in teractive teaching (Gersten & Geva, 2003), (f) modeling (Helman, 2009), (g) integration of vocabulary into reading instruction (Antunez, 2002; Helman, 2009), (h) use of vi suals (Gersten & Baker, 2000; Helman & Burns, 2008), (i) connection between oral and wr itten forms of words (Helman & Burns, 2008), (j) self-monitoring (Helman & Burn s, 2008), and (K) use of native language strategically (Ger sten & Baker, 2000). UFLI Sessions The tutoring sessions were designed bas ed on the small-group UF LI model by Lane, Pullen, and Hayes (2007) and modified for second grade English language learners. Each session had four steps: (a) gaining fluency and measuring progress, (b) word work with manipulative letters, (c ) introducing and reading a new book, and (d) writing for reading. The instructor follo wed a session guide t hat outlined each step (Appendix D). Each session had an accompanyi ng session notes sheet to be completed by the tutor (Appendix E). What follows is a brief description of each step, including modifications for ELLs made for the study. Step 1 Gaining fluency a nd measuring progress (10 minutes). The purpose of this step was twofold: (a) to develop reading fluency of connected text, and (b) to measure students reading progress. Fluen cy was developed by targeting word reading accuracy and automaticity, reading rate, and prosody. Students reread one to three books that had been read in previous sessions and that could be read by all group participants with 90% to 100% accuracy. Ea ch student read at least one book (choral reading, partner reading ), with minimal tutor assistance. Books selected for this step

PAGE 111

111 provided opportunities to practice decoding, sight word reading, as well as improve reading rate. Books that were too easy were not selected because they did not offer instructional value. As students reached appropr iate levels of accuracy and rate, the tutor shifted the focus to prosody. During progress monitoring, the tutor took a running record of one group member. The student was asked to read the new book that was introduced in the previous session, while the other group members we re reading familiar books. The running record served two purposes: (a) to determine the level for the new book to be read in step 3, and (b) to plan for instruction. To determine the new book level, the tutor calculated the rate of accuracy (total word s read correctly/total words read x 100). If the accuracy rate was below 90% (frustration leve l), the tutor decided whether to introduce a new book at the previous leve l or try a new one at the same level. If the accuracy rate was between 90% and 95% (instructional level) the tutor introduced a new book at the same level. If the accuracy rate wa s above 95% (independent level), the tutor introduced a new book at a higher level. To pl an for instruction, the tutor examined the students miscues (e.g., attempts, omissions, self-corrections, additions, repetitions, emerging patterns) and strat egy use. Immediately after, the tutor provided feedback on students use of reading strategies to help them develop awareness of how strategies were being used. When appropria te, the tutor asked the studen t to identify strategies that were used to figure out challengi ng words (Lane et al., 2001; Lane et al., 2005). Throughout the first step, the tutor supported the language needs of students by relying on visuals and by providing quick definitio ns of unfamiliar word s (Gersten & Baker, 2000).

PAGE 112

112 Step 2 Word work with ma nipulative letters (10 minutes). The purpose of this step was to develop an understanding of the alphab etic principle, as well as automatic word recognition skills. In this step, manipulat ive letters were used to conduct word work with new and familiar words. The tutor and each participant had a magnetic board and a set of magnetic letters. Students practi ced encoding and decoding real and nonsense words, at the onset-rime and phoneme leve l. When needed, the tutor pointed out the similarities and differences between Span ish and English sounds and word patterns (Linan-Thompson et al., 2003). The tutor pl anned ahead of time by selecting a known word from a familiar book, and then creating a list of real words and nonsense words that could be derived from m anipulating the known word. The list of words followed a sound sequence that moved from easier to diffi cult (e.g., continuous sounds in initial position before stop sounds). The process was as follows. First, the tuto r either pointed out the familiar word in the book or asked students to find it. Then, th e tutor asked students to either spell the word or read the word that was spelled with magnetic lette rs on their own boards. Next, the tutor prompted students to either encode or decode each of the words in the list by making changes to the onset-rime or phonemes in the word. To address the needs of ELLs, the tutor specified whether each word was a real word or a nonsense word. This was important because many ELLs were still building their vocabularies and were not able to distinguish between real and nonsense words (Linan-Thompson et al., 2003). In addition, vocabulary was supported through the use of picture cards, brief definitions, and in some instances, translation of key words. This was relevant given that vocabulary is one of t he biggest challenges ELLs face (Carlo et al., 2004). As

PAGE 113

113 students acquired word reading strategies, t he tutor introduced more complex spellings, including prefixes, suffixes, and other mo rphographic features (Lane et al., 2001; Lane et al., 2009). Step 3 Introducing and reading a new book (10 minutes). The purpose of this step was to learn and practice reading strategi es with progressively more challenging books. The tutor selected a book based on the rate of accuracy obtained during the running record. Then, the tutor introduced th e book to the students with a picture walk and a discussion that highlighted key vocabul ary words, unusual spellings, and/or repeated language patterns. Picture cards and quick definitions were used to support vocabulary development. In addition, students were encouraged to make connections to their personal lives, as well as predictions about the story, which were later confirmed or refuted. This provided participating ELLs wit h opportunities to practice oral skills and acquired vocabulary while learning comprehens ion strategies. After a brief discussion, students started to read the book while the tutor coached them in learning and applying reading strategies. Coaching was more expl icit at the beginni ng when students were learning new strategies and became more imp licit when the student s started acquiring and demonstrating control over them (Lane et al., 2007; Lane et al., 2009). In addition to reading connected text, the tutor engaged the students in word work with new words. Students were prompted to appl y different reading strategies to figure out unknown words. To do this, the tuto r selected one or two words that students struggled with when reading the new book. The tutor used manipulative letters or dryerase boards to (a) demonstrate how the new words were spelled, (b) point out similarities and differences between new words and familiar words, (c) use parts of the

PAGE 114

114 new words as base for practicing word fam ilies, (d) practice segmenting and blending using Elkonin boxes, or (e) practice writi ng multisyllabic words and/or sight words (Lane et al., 2007; Lane et al., 2009). As in previous word work activities, visuals and brief definitions were used to support the vocabulary needs of students. Step 4 Writing for reading (15 minutes). The purpose of this step was to use writing as a means to develop print awareness, phonemic awar eness, decoding, and encoding automaticity, and familiarity with sight words and word patterns. First, the tutor had a brief informal conversation with the st udents about the new book read. During the conversation, the tutor identified one or tw o meaningful sentences from each participant that provided optimal opportunities for inst ruction. The tutor jotted down one sentence for each student on the session notes sheet. Each sentence included at least one to three high frequency words and at least two to four decodable words. The tutor then asked the students to write t heir sentences in their writ ing books. Each writing book, formed of blank sheets of white paper, was turned sideways in order to have a top page for word work and a bottom page for sentence writing. Initially, the students wrote their sentence on the top page, whil e the tutor moved from one student to the other checking for spelling accuracy. As students finished writ ing their sentences, the tutor asked them to either write words that were spelled correctly in the sentence page, or do word work on the top page with words that were misspelled. Word work was conducted in different ways: (a) unfamiliar high frequency words with uncommon spellings (e.g., because, was) were spelled by the tutor first and then written multiple times by the students (i.e., write the word inside this box, write the word as large as you can, write the word as small as you can, write the word as fast as you

PAGE 115

115 can), and (b) unfamiliar words with regular s pelling patterns were practiced with Elkonin boxes first and then rewritten once again bef ore writing it on the sentence page. Uncommon words with unusual spellings were dealt with in one of three ways: (a) the tutor wrote the word on the sentence page, (b) the tutor fi rst wrote the word on the practice page and then the student wrote it on the sentence page, or (c) the tutor helped the student identify parts of the words he or she could spell and completed the word for the student before the student wrote it on the sentence page (Lane et al., 2007; Lane et al., 2009). Word work was conducted individually or as a group if every student was working on the same word. Every time a word wa s added to the sentence page, the child was required to reread what he or she had written to that point. This provided additional reading practice and reinforced grapho-phonemic connections. To support the needs of ELLs, the tutor helped students produce comple te and grammatically correct sentences by reinstating what the students had said and making t hem repeat the complete sentence before writing it. In some instanc es, when students used words in Spanish to express an idea, the tutor translated the wo rd for the student and asked them to repeat the word verbally, before writing it. Materials Based on the UFLI tutoring program, all participating groups had access to the same type of materials. These materials wer e provided by the investigator at no cost to the school or the students. Materials incl uded (a) leveled books, (b) magnetic letters and magnetic boards, (c) picture ca rds, and (d) other materials. Leveled books. The books selected for each lesson followed the Reading Recovery level criteria, which are designed to span from level 1 to level 20. Levels 1 to

PAGE 116

116 16 correspond to first grade, and levels 17 to 20 correspond to second grade. Leveled books match the text to the specific needs of readers (Lane et al., 2009), allowing them to practice decoding skills and sight word reading (Helman, 2009), and to apply reading strategies (Fountas & Pinnell, 1999). Wi thin each reading level, narrative and informational texts that covered a variety of topics were selected to support students interests and backgrounds. Magnetic letters and magnetic boards. Each student had a set of solid blue, lowercase magnetic letters and a magnetic board to manipu late during word work activities. Research shows that the use of magnetic letters supports the development of decoding skills. For example, Pullen (2000) studied the effect of alphabetic word work using manipulative letters on the reading skill s of first grade struggling readers. Results showed that students in the experimental gr oup (lessons with manipulative letters) had better decoding skills than students in the comparison condition (lessons without manipulative letters) and students in the control group (no treatment ). In a different study, Pullen and colleagues (2005) conduct ed a multiple-baseline design across groups study to evaluate the use of mani pulative letters during explicit decoding instruction. Students used magnetic letters to blend and segment, decode and recode target words from books previously read. The authors found that all participating students (N=9) increased the percentage of correct pseudowords read per minute, from baseline (average of 46.5%) to intervention (average of 86.5%). These findings offer additional evidence for the use of manipulative letters to during reading interventions that target word reading skills.

PAGE 117

117 Picture cards. With the purpose of facilitat ing comprehension throughout the tutoring lessons, picture cards of key vocabu lary words were used to help students gain understanding during reading and wo rd work activities. Sets of picture cards were either purchased or created by the re searcher. According to Gersten and Baker (2000), visual aids help students visualize the abstractions of language (p. 463). T herefore, they can effectively support English language learners as they deal with the language demands of each lesson. Other materials. In each tutoring session, th ree additional materials were required. First, session notes for each group to keep a detailed record of their performance. Second, a sentence writing book was given to each participant to complete step 4 of the less on. And third, pencils and a digital timer to ensure that adequate time was spent on each activity. Dependent Variable Two dependent variables were selected and measured throughout the different phases of this study. The variables were (a) rate of correct pseudowords read per minute (CPPM) and (b) rate of correct sight w ords read per minute (CSPM). Pseudoword reading, also known as nonsense word reading or nonword reading, has long been identified as a re liable predictor of readi ng achievement (Gough, 1983). Shankweiler and colleagues (1999) examined the relation between word reading, nonword reading, and reading comp rehension. Findings showed that there is a strong relation between word and nonword readi ng and between these skills and reading comprehension. According to Shankweiler and colleagues, nonwor d reading measures require the use of phonologically analytic decoding processes, which are critical to reading ability (p. 87) Pullen and colleagues ( 2005) also state that using a rate of

PAGE 118

118 pseudoword reading enables us to determine the level of automaticity with which readers decode words, that is, decoding fluency (Hudson et al., 2009). The rate of correct pseudowords read per minute (CPPM) was used in this study to make decisions about movement across phases (baseli ne, intervention, maintenance). In addition to pseudowords, the rate of correct sight words read per minute (CSPM) was monitored throughout the intervention to measure the development of word recognition skills. According to Ehri ( 2005b), a useful way in which readers might read a word is by sight or memory. A wo rd that is well known can be recognized automatically, as a single unit, allowing the readers attention to be focused on meaning while reading connected text (Ehri, 2005b; Eh ri & Snowling, 2004, LaBerge & Samuels, 1974). Sight words are learned by establis hing connections between the graphemes and phonemes in words. These connections are based on knowledge of the alphabetic principle and spelling patterns, which is dev eloping in beginning readers. Each time a word is read, the grapheme-phoneme c onnections are str engthened (Ehri, 2005b). According to Ehri and Snowling (2004), having an extensive sight word lexicon is central to reading. Therefore, measur ing students ability to read sight words accurately and fluently is key in evaluating the effect iveness of beginning reading interventions like UFLI. For the purposes of this study, si ght words are defined as high-frequency words with irregular spelling patterns. To measure the dependent variables, two types of probes were created by the researcher: (a) pseudoword probes and (b) sight word probes. Pseudoword probes were created from a pool of 100 word s that represent common letter-sound combinations and spelling patterns. From this pool of words, 15 different probes of 50

PAGE 119

119 words each were generated by random selecti on in order to prevent practice effects (see Appendix F for all pseudowor d probes). In a similar fash ion, 15 sight word probes were created from a pool of 100 high frequency words selected from Dolchs list of high frequency words for first and second grade. Ea ch probe consisted of 50 randomly selected sight words (see Appendix G for all sight word probes). To estimate the rate of CPPM and CSPM, the researcher conduct ed 1-minute timings with each participant every other day. No feedback was provided to the students at this time. The rate of CPPM and CSPM corresponded to the total number of words read correctly in one minute. Design This study implemented a multiple baseline across groups design (Kazdin, 1982; Kennedy, 2005) to assess the ef fects of the UFLI program on the reading skills of second grade, Spanis h-speaking, English la nguage learners. Single-subject designs are relevant to the field of literacy for various reasons: (a) they emphasize the individual as the unit of concern; (b) they systematically determine if a specific intervention is effective and for whom it is effective, allowing the analysis of responders and nonresponders to treatment; (c) they provide a practica l way to test educational procedures under typical educational conditions; (d) they incorporate ways to assess not only the outcomes of an intervention, but also the proce ss of change and the maintenance of change across time, and (e) they offer a costand time -efficient way to investigate critical instruct ional questions (Horner et al., 2005; Neuman & McCormick, 2000). In 2003, single-subject research was identif ied by the Division of Research of the Council for Exceptional Children as one of th e four research methodologies needed to

PAGE 120

120 identify evidence-based practices (Odom et al., 2005). Through the use of withinand betweensubject comparisons and systematic replication, single-subject design offers ways to control for majo r threats to external and internal validity, a major concern for research in general (Horner et al., 2005). The design for this study consists of th ree phases: (1) baseline, (2) intervention, and (3) maintenance. A detailed description of each phase and the tutor/researcher that conducted each phase is provided below. Follo wing that, procedural information is given on supplemental data collection. Baseline During bas eline, the dependent variables (CPPM and CSPM) were measured for each participant using resear cher created probes. Each data collection session lasted approximately three minutes per student. Once i ndividual rates were calculated, a mean group was estimated and plotted into a graph. All groups started baseline at the same time. When Group 1 showed a stable line on si x continuous data points (no significant increasing or decreasing trend in behavior), they moved to the intervention phase. In the meantime, the other two groups remained in baseline. Intervention During intervention, each group participated in a series of 45-minute, small-group tutoring lessons, two to five times a week Lessons for each group were scheduled in collaboration with the st udents teachers and the school pr incipal to avoid interfering with important instructional time. All less ons took place in a quiet and well illuminated room to avoid distractions. Each session fo llowed a structured format that involved the four steps described previously. In each tu toring lesson, students read familiar and new leveled books, did word work with magnetic letters, and wrote sentences related to the

PAGE 121

121 readings. During this phase, response to intervention was measured using the same set of pseudoword and sight word reading probe s and following the same procedure as in baseline. Data were collected for each parti cipant at the beginnin g of the tutoring session (6 to 10 minutes per group). Accordi ng to Pullen et al. (2005), administering probes at the beginning of a session is a mo re accurate measure since it reduces the chance of students applying what they had just practiced during the lesson. While one group member completed the probes, the other group members sat on the other side of the room and read familiar books. After each individual rate was established, a group mean was calculated and plotted into a graph. When Group 1 showed an increase in the rate of CPPM on four consecutive data points, Group 2 started the intervention phase. In the meantim e, Group 3 remained on baseline. When Group 2 showed an increase on four consecutive data points, Group 3 started intervention. Each group remained in the intervention phase until (a) every member of the group was readi ng at grade level (level 20 books) or (b) when every member of the group completed a minimum of 40 tutoring lessons. Maintenance The maintenance phase began for each group two weeks after the intervention phase ended. The same data colle ction procedures as in baseline were followed. Each data collect ion session lasted approximately th ree minutes per student. Once individual rates were established, a group mean was calculated and plott ed into a graph. The maintenance phase ended for each group when dat a points showed that a stable line was achieved, indicating that they were st ill reading at an appropriate reading level. The total number of data points varied across groups: Group 1 had 5 data points, while Group 2 and 3 had a total of 6 data points.

PAGE 122

122 Tutor/Researcher The researcher conducted all the tutori ng sessions. She was trained in the UFLI program by the programs developer and had experien ce im plementing it with Spanishspeaking English language learners, in one-on-one and small-group formats. The researcher was originally fr om Guatemala and had a Mexic an heritage as well. She was a native speaker of Spanish and spoke English fluently. The tutor had a B.A. in Psychology and Master of Education and Education Specialist degrees in Counselor Education. At the time of the study, the tutor was a doctoral candidate in Special Education. Her areas of expertise were early literacy, early reading difficulties, and English language learners. Data Analysis Data collec ted in each phase were analyzed using systematic visual comparison of responding within and across conditions (H orner et al., 2005, p. 169). In visual analysis, data points are graphed in order to ex plore and observe the types of patterns that arise over time (Kennedy, 2005). Visual comparison of data points allowed the researcher to identify changes in the dependent variable as a function of the independent variable (UFLI tutoring program). According to Horner et al. (2005), when the research participant is a group, and not an individual, the group gener ates a single score for eac h measurement period. In this study, mean rates of CPPM and CSPM we re calculated for each group based on each members individual score. The group mean was plotted in a noncumulative way using a simple line graph (Kazdin, 1982). Da ta were then analyzed by estimating an overall mean for each phase and conducting within group comparisons across phases.

PAGE 123

123 Supplemental Data Supplemental data include preand post-in tervention measures of early literacy skills an d attitudes toward reading, teac hers ratings of st udents reading abilities and classroom behaviors. It also addresses book le vels read by each group throughout the intervention phase. What follows is a brief description of the procedures followed for each of these. Preand PostIntervention Data At the beginning of the study, before baseline data were collected, all students were assessed indiv idually on measures of early literacy skills and attitudes toward reading. At the end of the study, when t he maintenance phase was completed for all groups, students early literacy skills and atti tudes toward reading were assessed again. Assessment times were scheduled with the hel p of the students teachers and the school principal. Pre-intervention assessm ent lasted approximately three hours per student, while post-intervention assessment lasted approximately 90 minutes. The time varied from preto post-intervention becaus e, at the beginning of the study, the Woodcock-Munoz Language Survey-Revised (WMLS-R) was administered for screening purposes. Assessment sessions took pl ace at the school and were completed in multiple sittings to accommodate the needs of young children and to avoid fatigue. Assessment at the beginning of the study was completed by the investigator and three graduate students from the University of Florida. The three graduate assistants were also fluent in English and Spani sh and had background knowledge of reading, second-language learning, and educational assessment. T hey were pursuing graduate studies in School Psychology, and were famili ar with a few of the assessments used in this study. The investigator provided one 2-hour training session where trainees were

PAGE 124

124 able to learn about the assessments procedur es and to practice implementing them. Assessment at the end of the study was completed only by the researcher, given that time requirements were less demanding since the WMLS-R was not administered at that time. In addition to directly measuring students early literacy skills, teachers were also asked to complete a reading behavior rati ng scale to measure students changes as a result of the intervention. The scale addres sed five areas of reading (i.e., phonemic awareness, phonics, fluency, comprehension, and vocabulary), as well as classroom behaviors (i.e., participation in class, mo tivation to read, and use of English to communicate in class). Each teacher rece ived a rating scale before the study started and one at the culminati on of the study. The Book Levels during Intervention During intervention, data were collected for each group regarding the changes in book levels read throughout the tutoring less ons. These data were then plotted into a graph for visual analysis. The graph depicts the starting reading le vel for each group, the number of lessons that each group rema ined at each reading level, and the level reached at the end of the intervention. This data provided additional information on the effectiveness of the intervention. Interobserver Agreement To ensure that the dependent variables (CPPM and CSPM) were measured and recorded with integrity, interobserver agr eements (IOA) were es tablished between the tutor and an observer. The tutor and observer listened to students reading of pseudowords and sight words probes duri ng one-minute timings. Independently, the tutor and the observer recorded the number of words read correctly and calcula te the

PAGE 125

125 rate of accuracy. To calculate the leve l of agreement, a frequen cy-ratio approach was used. In this approach, the researcher calc ulated the total number of responses that each observer recorded. Then, the smaller to tal of responses (S) was divided by the larger total of responses (L) and then was mu ltiplied by 100 (S/L x 100). An acceptable level of agreement is 80% (Kazdin, 1982; Kennedy, 2005). The level of agreement for all pseudoword observations was 93.8% and for sight word observations was 98.40%. These levels of agreement were based on a to tal of five observations for pseudoword and sight words conducted for each group: one during baseline, three during intervention, and one during maintenance. The only exception was an additional observation conducted with one student in Gr oup 1. On that occasion, Amelia was absent and Pedro was not able to speak out loud due to illness. An IOA was conducted based on only one student. Treatment Integrity Several measures were taken to ens ure that the tuto ring program was implement ed in a reliable manner. First, duri ng each lesson the tutor had a copy of the session guide that outlined eac h of the four steps. The session notes sheets also served as an additional guide for the tutor sinc e they outlined each of the steps in the lesson. Second, the tutor conducted self-evaluations using a treatment fidelity checklist (Appendix H). One checklist was completed every week for each of the participating groups. For Group 1 a total of 11 checklists were completed. The mean percentage of adherence was 97.45%. For Gr oup 2, 14 checklists were completed with a mean percentage of adherence of 96. 6%. For group 3, a total of 13 checklists were completed. The mean percent age of adherence was 96.2%.

PAGE 126

126 Third, the tutors adherence to the program was evaluated by one of two observers, who completed a treatment fide lity checklist on each visit. The tutor was observed two times with each group. Observer 1, a professor from the University of Florida who developed the UFLI tutori ng program completed two observations. Observer 2, a doctoral student in Special Educati on at the University of Florida and experienced UFLI tutor, co mpleted four observations. Af ter each session, the observer and the tutor met to discuss the integrity of the intervention. The researcher then calculated the percentage of adherence by dividing the total number of items completed by the total number of lesson components an d multiplying it by 100. The mean adherence percentage for all six observations was 96%. Adherence percentage for each treatment integrity observation is reported in Table 3-8. Table 3-8. Treatment integrity Group number Observation number Number of items completed Percentage of adherence 1 1 39/42 93% 2 40/42 95% 2 1 41/42 98% 2 41/42 98% 3 1 40/42 95% 2 40/42 95% Social Validity Social validity is defined by Kennedy (2005) as the esti mation of the importance, effectiveness, appropriateness, and/or satisf action various people ex perience in relation to a particular intervention (p 219). This is especially im portant in educational settings because it gives the researcher an indication of the level of acceptability of the intervention among its consumers (p. 218). To estimate social valid ity, a subjective evaluation approach (subjective evaluation) was conducted to obtain feedback from

PAGE 127

127 students and teachers about the effectiveness and the acceptability of the intervention (Kennedy, 2005). First, each student participated in a brief in terview at the end of the maintenance phase. Questions focused on students opinions about the UFLI tutoring lesson and its components. Each interview took approximatel y five minutes to be completed. Second, all the teachers of the par ticipating students were invi ted to observe a videotaped tutoring lesson and complete a social validity questionnaire. The goal of the questionnaire was to evaluate the accept ability of the tutoring program and its components and to determine whether teachers w ould see themselves using this lesson or any of its components in their cla ssrooms. The questionnaire was adapted from Pullen (2000) to address the UF LI program components as it re lates to ELLs. A copy of the students interview questions and the teachers questionnaire can be found in Appendix I. The researcher sent home consent forms asking parents for permission to videotape their children during a tutoring le sson (see Appendix A). A total of five students received parental consent to participate in the video, three from Group 2 (Maggie, Maria, and Viviana) and two from Group 3 (Aldo and Jennifer). Because of scheduling difficulties, one studen t from Group 2 (Viviana) and two students from Group 3 (Jennifer and Aldo) were selected to parti cipate. Once the videotape was completed, teachers watched the tape and were asked to complete the questionnaires. Teachers were provided a copy of the lesson guide to help them identif y the different steps of the program as they observed the video. Due to time constraints, teachers opted to complete the questionnaire at a late r time. The response rate was 100%.

PAGE 128

128 Delimitations and Limitations There were several delimitations and limitat ions to the study. First, delimitations and their impact on external validity are addre ssed. Second, the studys limitations are described as well as how each may have influenced the findings and their generalizability. Delimitations The first delimitation pertains to the gr oup of students selected to participate. All students were in second-grade and all spok e Spanish as their first language. This may limit the generalization of findings to other grade levels and other groups of ELLs that might speak another language at home. A second delimitation is related to the par ticular setting in which the program was implemented. This study was conducted in only one school located in a rural area of North Florida. This particular setting has a se t of variables that may differ from other schools (e.g., student population, human resources, instructional resources, scheduling) and these variables may have influenced intervention effects in some way (Kazdin, 1982). For example, the schools schedule allowed for daily 45-minute sessions that facilitated the provision of t he UFLI program to all groups for an extended period of time. While the number of weekly sessions varied across groups due to school activities that conflicted with the establis hed schedule, the length of t he intervention was extended until students reached grade leve l reading or until they completed a minimum of 40 tutoring sessions. Also, since the school serves all second and third grade students in the county, there was a suffici ent number of Spanis h-speaking ELLs in second-grade to group second graders appropriately by reading level. This might not be the case in other

PAGE 129

129 schools, where there may be a smaller number of ELLs Changing group composition could influence the efficacy of the UFLI program and could alter the results. A third delimitation is related to the tutor implementing the intervention. In this study, the intervention was conducted only by the researcher. According to Kazdin (1982), there is a possibility that the characte ristics of the individual implementing the intervention may help attain the intervention e ffects. In this case, the researcher shared a similar heritage with the majority of the participants, was proficient in English and Spanish, had extensive kno wledge about the reading proc ess and literacy development among ELLs, and had experience implementi ng the UFLI progr am with Spanishspeaking ELLs. It is unknown at this point if similar intervention effects can be attained if the UFLI program is implemented with ELLs by other tutors with different sets of characteristics, knowledge, and skills. Limitations The first limitation is inherent to single-subject design and pertain s to sample size. One of the main criticisms against this met hodology is that the results of a particular study may not be generalized to larger groups of subjects due to the limited sample size (Kazdin, 1982; Neuman & McCormick, 1995). To increase the external validity of results, direct and systematic replication is necessary. The second limitation is related to treatm ent interference. In this study, students were participating in the SRA Early Reading Intervention (ERI) program and/or Harcourt Interventions while the study took place. While the use of a multiple baseline design was used to establish control across groups it is unknown to what degree their participation in other programs mi ght have influenced the results.

PAGE 130

130 Description of One-on-One Tutoring Methodology For the student that received one-onone tutoring (Miriam pseudonym), data were collec ted on rates of CPPM and CSPM during baseline, intervention, and maintenance. Preand post-intervention data were collected using the same assessment instruments implem ented with the other par ticipants. Running record trials at the beginning of the study showed that she was reading level 6 books at an instructional level. Table 3-9 shows the rate of accuracy in each running record. What follows is a brief description Miriams demographic data, instructional information, reading scores used for selection, language levels, and classroom behaviors. Table 3-9. Running records results used for Miriam Student Trial 1 Trial 2 Trial 3 Trial 4 Trial 5 Trial 6 Miriam 10 (85%) 9 (88%) 8 (89%) 7 (83%) 6 (92%) 6 (93%) Miriam was a Hispanic female, aged 7 years-11 months, enrolled in second grade for the first time. Her parent s were from Mexico and they only spoke Spanish at home. Her mother reported not finishing high school. Her mother stated t hat Miriam had had four years of instruction in the US. As part of her daily r eading instruction, she received the SRA/EIR program in small groups, five times a week, for 45 minutes. School records showed that Miriam had an LEP stat us, but did not receive LEP classes. Reading scores showed that at the end of first grade, Miriam scored in the 46th percentile on the SAT-10. Her DIBELS scores for the end of first grade were 50 on PSF (low risk), 36 on NWF (moderate risk), and 27 on ORF (moderate risk). At the beginning of second grade, she scored between the 10th and 20th percentil e on the Invented Spelling Assessment. Language scores on the WMLS-R demonstrated that she had limited to fluent English oral language ability (Level 3.5) and limited Spanish oral

PAGE 131

131 language ability (Level 3). On the C-ATRS, her teacher r eported that Miriam had no attitude or behavior problems in class. She affirmed that Miriam was a very easy-going student and that she was almost always focused on her work. The student received the UFLI individual tutoring program, which follows the same principles and addresses the same reading skills as the group program, but follows a different lesson format. In each lesson, a fifth step is included, which focuses on extending literacy practices by exposing students to different types of reading genre. An outline of the steps for one-on-one UFLI tu toring can be found in Appendix J. Pilot Study A pilot study was conducted prior to this research study with the purpose of evaluating the effectiveness of the UFLI tutori ng program on the reading skills of Spanish-speaking English language learners, who we re experiencing reading difficulties. In the pilot study, a mult iple baseline across participants design was implemented with a total of three first-grade students. The par ticipants were selected based on their English language learner status and their atrisk reading status as determined by DIBELS scores. The experi mental procedures (movement across phases) were similar to the ones described in this study. The dependent variable was the rate of correct pseudowords read per minute (CPPM), using progress monitoring probes from DIBELS. The tutoring lessons had all the components addressed in this study plus the addition of a fifth step that fo cused on extending literacy practices. In that step, students were exposed to different types of reading genre with the goal of increasing awareness of different text structures (Lane et al., 2009). Students participated in an average of 22 tu toring lessons, 4 to 5 days a week. Each lesson lasted approximately 45 minutes. Analysis of data showed that all students

PAGE 132

132 made a marked improvement by the end of the intervention. Student A went from a total of 4 CPPM during baseline to 24 CPPM at the end of the in tervention. Student A also improved on book levels read, going from level 3 to level 10 in 27 sessions. Student B went from a total of 10 CPPM during baseline to 34 CPPM at the end of the intervention. In book reading, she went from level 3 to level 17 in only 16 sessions. Student C went from 1 CPPM to 9 CPPM after intervention, and from a level 1 books to a level 5 in 25 sessions. Students in the pilot study varied in their in itial level of oral language proficiency based the WMLS-R CALP levels. Student A and B demonstrated limited English oral proficiency (Level 3), while student C demonstrated very limit ed English oral proficiency (Level 2). Regardless of the level of English oral profic iency, all students responded to the intervention, a finding that was also r eported in studies conducted by Gunn et al. (2000; 2002; 2005), and LinanThompson et al. (2003). As a result of the pilot study, and in agr eement with other res earch studies (LinanThompson et al., 2003; Mohr & Mohr, 2007) a fe w modifications were made to the UFLI tutoring program to address the language needs of English language learners. First, picture cards and quick definitions were us ed to clarify unfamiliar words during reading and word work activities. Second, real words and nonsense words were identified during decoding and encoding activities to help students make connections between written words and their meaning. Finally, complete sentences and standard grammar were modeled by the tutor to help students express complete ideas.

PAGE 133

133 CHAPTER 4 RESULTS The purpos e of this study was to evaluate the effectiveness of a modified version of the University of Florida Literacy In itiative small group tutoring program on the reading skills of second grade Spanish-spea king English language learners who are struggling to read. For this purpose, a multiple baseline across groups design was implemented to measure students response to intervention. The goal of this chapter is to present the results obtained throughout the study. The chapter is divided in several sections: (a) dependent variables, (b) total number of tutoring sessions (c) supplemental data, (d) social validity, (e ) summary of one-on-one tutori ng data, and (f) summary of findings. Dependent Variables Throughout the three phases of the st udy, CPPM and CSPM data were collected using researcher-created probes. After each probe was implemented, the rate of CPPM and CSPM was first calculated for each gr oup member. A few students (i.e., Amelia, Pedro, Jessica, and Maggie) completed some of the CSPM probes within one minute. On those occasions, the time remaining was recorded and t hen used to estimate what the actual number of correct sight word s per minute would hav e been if students had continued reading for the full minute. Once individual rates of CPPM and CSPM were calculated, a group mean was estimated and plotted into a graph. At t he end of each phase, overall mean rates of CPPM and CSPM were calculated for each group. Decisions about movement from one phase to the next were made based on each groups rate of CPPM.

PAGE 134

134 Results for the two dependent variables are organized by phases: (a) baseline, (b) intervention, and (c) maintenan ce. In each phase, group data ar e reported, including the range of mean scores and the overall mean fo r the entire phase. In addition, the difference in overall means between baselin e and intervention, as well as between intervention and maintenance are reported. Individual data can be found in Tables 4-1 to 4-3. Baseline Phase During bas eline, the rate of CPPM and CSPM were calculated until a stable line of response was established for each group. Each data session lasted approximately 6 to 10 minutes per group. Individual da ta is presented in Table 4-1. Group 1 data. The mean rate of correct pseudow ords read per minute for Group 1 ranged from 6 to 7 CPPM with an overall phase mean of 6.33 CPPM. The mean rate of sight words ranged from 48.67 to 52.39 CSPM with an overall phase mean of 50.82 CSPM. Data were collected during a total of six baseline sessions. Group 2 data. The mean rate of pseudoword reading for Group 2 ranged from 4.25 to 5 CPPM, with an overall phase mean of 4.75 CPPM. The mean rate of sight words ranged from 22.5 to 24.3 CPPM, with an overall mean of 23.82 CSPM. Data were collected over a total of seven sessions. Group 3 data. The mean rate of correct pseudowor ds read per minute for Group 3 ranged from 1.66 to 3 CPPM, with a phase m ean score of 2.42 CPPM. For sight word reading, the mean ranged from 1.7 to 3.7 C SPM, with an overall p hase mean of 2.63 CSPM. Data were collected over 11 sessions. Summary. During baseline, all groups displa yed no significant increase or decrease in the trend of response for CPPM or CSPM. Baseline levels differed across

PAGE 135

135 groups. Group 1 had the highest baseline rates for both variables, followed by Group 2 and Group 3. The difference across groups for CPPM was smaller than for CSPM. Between Group 1 and Group 2, there was a di fference of 1.58 CPPM, compared to a difference of 27 CSPM. Similarly, betw een Group 2 and Group 3 there was a difference of 2.33 CPPM, compared to 21.19 CSPM. Table 4-1. Individual rates of CPPM and CSPM during baseline CPPM CSPM Participants Range Mean Range Mean Amelia 12-15 13.5 62.72-68.1866.35 Pedro 1-3 1 49.41-5552.80 Jessica 3-5 3.83 32-34 33.33 Viviana 1-2 1.57 12-15 14.3 Ernesto 2-5 4.14 28-31 29.9 Maggie 7-9 8.28 30-33 31.7 Maria 4-7 5 18-21 42 Jennifer 3-5 3.90 2-5 4 Aldo 0 0 0-1 0.36 Jorge 2-4 3.36 2-5 3.54 Intervention Phase During intervention, probes were given to each participant at the beginning of the tutoring session (6 to 10 minutes per group). T he purpose of this phase was to establish a change in the rate of CPPM and CSPM from baseline, as well as the progress in rate as a result of the interv ention. Individual dat a is presented in Table 4-2. Group 1 data. Once a stable trend line was es tablished in baseline, Group 1 started the intervention phas e. During this phase, the group received a total of 24 pseudoword and sight word probes. The mean rate of pseudoword ranged from 11 to 34 CPPM, with an overall phase mean of 23. 88 CPPM. This represents a positive change of 17.55 CPPM from baseline to intervention. For sight wo rds, the mean rate

PAGE 136

136 ranged from 51.53 to 73.99 CSPM with a phase mean of 62.96 CSPM. This shows an increase of 12.14 CSPM from baseline to intervention. Group 2 data. When Group 1 showed an increase in the rate of pseudoword reading over four consecutive sessions, Gro up 2 started the intervention phase. In the meantime, Group 3 cont inued on baseline. During the in tervention phase, Group 2 received a total of 26 pseudoword and si ght word probes. The mean rate of pseudoword ranged from 7.25 to 18.3 CPPM, with a phase mean of 13.41CPPM. This shows a positive change of 8.66 CPPM from baseline to inte rvention. For sight words, the mean rate ranged from 24.25 to 41.43 CSPM, with a phase mean of 33.42 CSPM. This shows an increase of 9.6 CSPM from baseline to intervention. Group 3 data. When Group 2 showed an increase in the rate of pseudoword reading over four consecutive sessions, Gro up 3 started the intervention phase. During the intervention phase, Group 3 received a total of 25 pseudoword and sight word probes. The mean rate of pseudoword ranged fr om 5 to 13 CPPM, with an overall mean of 8.53 CPPM. This shows a positive c hange of 6.11 CPPM from baseline to intervention. For sight words, the mean rate ranged from 3 to 12.3 CSPM, with an overall mean of 7.14 CSPM. This shows an increase of 4.51 CSPM from baseline to intervention. Summary. During intervention, all groups s howed an increase in the rate of pseudowords and sight words correctly read per minute. The rate of improvement during intervention varied across groups on bot h variables. Based on the overall phase mean for each group, it was observed that Group 1 had the highest rate of improvement on CPPM and CSPM, followed by Group 2 and Group 3. From baseline to intervention,

PAGE 137

137 Group 1 had an increase of 17.55 CPPM and 12.14 CSPM. Group 2 had an increase of 8.66 CPPM and 9.6 CSPM, while Group 3 had an increase of 6.11 CPPM and 4.51 CSPM. Table 4-2. Individual rate of CPPM and CSPM during intervention CPPM CSPM Range Mean Change from baseline Range Mean Change from baseline Amelia 17-42 30.29 16. 7966.66-93.7576.25 9.9 Pedro 11-39 27.04 27.38 51.6-72.6365.23 12.42 Jessica 6-24 14.20 10.3733-57.64 47.45 14.12 Viviana 4-7 11.11 9.5414-31 23.88 9.6 Ernesto 0-16 11.80 7.6630-47 40.38 10.53 Maggie 9-24 17.5 9. 2223-52.63 41.17 9.46 Maria 7-18 13.23 8.2319-38 28.26 8.84 Jennifer 9-19 14.12 10.225-19 10.52 6.52 Aldo 0-6 1.96 1.960-8 3.2 2.84 Jorge 5-14 9.52 6.163-12 7.72 4.18 During this phase, Aldos rate of pseudoword showed a unique trend. He moved up and down between 0 and 2 CPPM for 19 pseudoword probes. On the 20th probe, he moved up to 5 CPPM and by the 25th probe he was reading 6 CPPM. Maintenance Phase The maintenance phase started for each group two weeks after the intervention phase ended, with the purpose of determining whether the rate of CPPM and CSPM were sustained over time. The maintenanc e phase ended for each group when data points showed a stable line at an appropriate reading level. Individual data can be found in Table 4-3. Group 1 data. During the maintenance phase, data were collected over five sessions. The mean rate of pseudowords reading for Group 1 ranged from 36 to 40 CPPM with an overall mean of 37.87 CPPM, an increase of 13.99 CPPM. In addition, the mean rate of sight word reading ranged from 72.71 to 75.56 CS PM, with an overall

PAGE 138

138 phase mean of 74.03 CSPM. This shows a pos itive improvement fr om intervention to maintenance of 11.07 CSPM. Group 2 data. Maintenance data were collected over six sessions. Data showed the rate of pseudowords r anged from 15.25 to 21.8 CPPM, with an overall phase mean of 18.88 CPPM. In comparison to the intervention phase, the maintenance phase showed a positive improvement of 5.47 CPPM. Data also showed that the rate of sight words ranged from 38.75 to 44.17 CSPM, with a phase mean of 41.51 CSPM. This shows in increase of 8.09 C SPM from the previous phase. Group 3 data. During the maintenance phase, dat a were collected over six sessions. Group 3 showed a rate of pseudowor ds that ranged from 13 to 14.7 CPPM, with a phase mean of 13.8 CPPM When compared to the intervention phase, Group 3 showed an increase of 5.27 CPPM. Data gat hered on rate of sight words ranged from 10 to 13 CSPM, with a phase mean of 12.22 C SPM. This shows an increase of 5.08 CSPM from intervention to maintenance. Summary. During maintenance, all groups sustai ned and improved on their rates of CPPM and CSPM. For CPPM, Group 2 showed a larger difference from the previous phase (18.88 CPPM), followed by Group 1 (11.71 CPPM), and Group 3 (5.27 CPPM). For CSPM, Group 1 showed the largest diffe rence from the previous phase (14.56 CSPM), followed by Group 2 (8.09 CSPM), and Group 3 (5.08 CSPM). Summary of Dependent Variables Visual inspection of the data (Figures 4-1 and 4-2) reveals that all groups achieved an immediate improvement in both dependent variables once the intervention started. Group 1, for example, moved from a mean rate of 6 CPPM in baseline to 11 CPPM after only one tutoring session. Group 2 went from 4.8 CPPM in baseline to 7.25 CPPM and

PAGE 139

139 Table 4-3. Individual rate of CPPM and CSPM during maintenance CPPM CSPM Range Mean Change from intervention Range Mean Change from intervention Amelia 40-4342 11.71 88.23-94.8390.81 14.56 Pedro 38-4240 12.96 73. 16-76.9275.03 9.8 Jessica 29-3431.6 17.4 54.23-58.856.25 8.8 Viviana 14-1816.16 5. 05 29.34 31.83 7.95 Ernesto 13-1916.16 4.36 42-47 44.16 3.78 Maggie 19-2924.16 6.66 47-53.5750.39 9.22 Maria 15-2119 5.77 35-45 39.66 11.40 Jennifer 17-1918.33 4.21 16.18 17 6.48 Aldo 9-1110.16 8.2 6-9 7.83 4.63 Jorge 12-1513.5 3.98 8-14 11.83 4.11 Group 3 moved from 2.66 CPPM in baselin e to 5.33. The trend of improvement continued for all groups as the tutoring sessions progressed. For example, the rate of CPPM during intervention ranged from 11 to 34 for Group 1, 7.25 to 18.25 for Group 2, and 5.33 to 13 for Group 3. A comparison of overall group means between baseline and intervention showed a dramatic change ac ross groups. Specific ally, Group 1 moved from a mean rate of 6.33 CPPM to 23.88 CPPM. Similarly, Group 2 mo ved from a mean rate of 4.75 CPPM to 13.41 CPPM, while Group 3 moved from 2.42 CPPM to 8.53 CPPM. Improvement was also noticed for the rate of correct sight word s read per minute. The immediacy of change observed after one tutoring lesson was less dramatic for CSPM. Group 1, for example, moved from a mean rate of 51.27 C SPM in baseline to 51.33 CSPM. Group 2 went from 24 CSPM to 24.25 CSPM. Similarly, Group 3 moved from 2.33 CSPM to 3 CSPM. The trend of impr ovement continued for all groups as the

PAGE 140

140 Figure 4-1. Rate of correct pseud owords per minute (CPPM) across groups.

PAGE 141

141 Figure 4-2. Rate of correct sight words read per minute (CSPM) across groups.

PAGE 142

142 tutoring sessions progressed. For example, the rate of CSPM during intervention ranged from 51.53 to 73.99 CSPM for Group 1, 24.25 to 41.45 CSPM for Group 2, and 3 to 12.33 CSPM for Group 3. A comparison of overall group means between baseline and intervention showed a noticeable change across groups. Specifically, Group 1 moved from a mean rate of 50.82 CSPM to 62.96 CSPM. Similarly, Group 2 moved from a mean rate of 23.82 CSPM to 33.42 CSPM, while Group 3 moved from 2.63 CSPM to 7.14 CSPM. The effect of the UFLI tutoring program on the rate of correct pseudowords and sight words read per minute was sustained tw o weeks after the intervention ceased. Moreover, the overall mean rate of CPPM and CSPM fo r all groups continued to increase during the maintenance phase, dem onstrating that students had acquired the reading skills taught and continued to use them independently. For Group 1, the overall mean rate of CPPM went from 23.88 to 37.87 CPPM, while the rate of CSPM went from 62.96 to 74.03 CSPM. Similarly, for Group 2, the rate of CPPM we nt from 13.41 to 18.88 CPPM, while the rate of CSPM went fr om 33.42 to 41.51. For Group 3, the rate of CPPM went from 8.53 to 13. 8 CPPM, and the rate of CSPM went from 7.14 to 12.22. Total Number of Sessions The number of the tutoring sessions va ried across groups. Two elements were considered in determining the total number of sessions for each group. First, tutoring sessions would be terminated once all mem bers were reading level 20 books with at least 90% accuracy (instructional level). Se cond, additional tutoring sessions would be added to allow all participants wh o had been absent to participate in a minimum of 40 sessions. A summary of the total number of sessions fo r each group is presented in Table 4-4.

PAGE 143

143 Group 1 reached level 20 books on session 35. On that session, Amelia read a level 19 book with 96% accuracy, allowing t he group to advance to level 20. To ensure that all members of the gr oup were reading level 20 books at an instructional level, three more sessions were conducted. By t hat time, everyone in Group 1 was able to read level 20 books at an independent level (98% to 100%). Af ter considering the total number of absences each student had, Amelia participated in a total of 34 sessions. She had six absences but was able to make up two sessions. The two make up sessions were done individually and covered all the steps of t he small-group lesson format. Jessica participated in a total of 37 sessions, while Pedro participated in a total of 38 sessions. Group 2 had a total of 47 tutoring sessions. This group did not reach level 20 at the end of the intervention. Therefore, to ensure that all member s participated in a minimum of 40 sessions, seven extra session s were added to compensate for students absences. Vivian had four absences, so she participated in a total of 43 sessions. Ernesto missed five sessions, so he participat ed in a total of 42 sessions. Maria missed seven sessions, participating in a total of 40 tutoring sessions. Finally, Maggie received the most sessions (47 total) since she did not miss one day of tutoring. Group 3 had a total of 45 sessions to compensate for students absences. This group did not reach level 20 at the end of the intervention. Jennifer did not miss one day of tutoring, so she participated in all 45 sessions. Aldo had two absences, so he participated in a total of 43 sessions, while Jorge, with five absences, participated in 40 sessions.

PAGE 144

144 Table 4-4. Total number of tuto ring sessions received by each student Participants Total absences Total number of sessions attended/group total Amelia 6 34/38 Pedro 0 38/38 Jessica 1 37/38 Viviana 4 43/47 Ernesto 5 42/47 Maggie 7 40/47 Maria 0 47/47 Jennifer 0 45/45 Aldo 2 43/45 Jorge 5 40/45 Supplemental Data In addition to the dependent variable, preand post-intervention data were collected o n students early literacy skills and attitudes toward reading, as well as teachers ratings of students reading abilitie s and classroom behaviors. Also, data were gathered on book levels read by each group during intervention. What follows is detailed description of findings. Results can be found in Tables 4-5 to 4-13. Measures of Early Literacy Skills and Attitudes to wards Reading Several measures of reading ability were measured before and after intervention. These included: (a) decoding accuracy and fl uency, (b) word recognition accuracy and fluency, (c) reading fluency, and (d) reading co mprehension. Students attitudes towards both recreational and academic reading were measured as well. Decoding accuracy and fluency. Decoding accuracy was assessed by the Nonsense Word Decoding subtest (NWD KTEA-II), while decoding fluency was measured by the (a) Phonemic Decoding Efficiency subtes t (PDE TOWRE) and the (b) Nonsense Word Fluency test (NWF DIBELS). Students made a marked improvement in both skills after interv ention. The NWD showed that all students

PAGE 145

145 improved on decoding accuracy, with a mi nimum gain of 6 nonsense words read correctly and a maximum of 21 words read correctly. The PDE test showed improvement in reading fluency with gains that ranged from 4 correct nonsense words to 16 nonsense words. A dramatic change was also observed on the NWF (DIBELS), where students gains ranged from 11 to 68 correct letter-sound correspondences per minute. According to risk levels, before inte rvention there were tw o students at high risk, six at moderate risk, two at low risk, and none at above average. In contrast, after intervention, there were no st udents at high or moderate risk, while six were at low risk and 4 at above average. Individual raw sco res and percentile ranks can be found in Tables 4-5 to 4-7. Word recognition accuracy and fluency. Word recognition fluency was measured by the Sight Word Efficiency te st (SWE TOWRE), while accuracy was measured by the Letter Word Recognition subtest (LWR KTEA-II). Results showed that all students made positive gains in both skills. Students gains in word reading accuracy ranged from 5 to 15 words read corre ctly, while gains in word reading fluency ranged from 0 to 15 words per minute. In dividual data, including raw scores and percentile ranks can be found in Tables 4-5 and 4-6. Table 4-5. Preand post-interventi on TOWRE raw scores and percentiles ranks Participants Pre-SWE Post-SWEPre-PD E Post-PDE Pre-total Post-total Amelia 47 (73%) 62 (84%) 20 (64%) 35 (89%) 70% 91% Pedro 46 (70%) 57 (74%) 14 (45%) 24 (66%) 61% 74% Jessica 41 (61%) 41 (39%) 13 (42%) 22 (61%) 52% 50% Viviana 25 (27%) 33 (25%) 2 (12%) 15 (39%) 14% 29% Ernesto 34 (45%) 45 (48%) 4 (16%) 17 (45%) 25% 42% Maggie 36 (50%) 46 (48%) 7 (21%) 18 (48%) 32% 48% Maria 27 (32%) 38 (32%) 11 (32%) 15 (39%) 29% 32% Jennifer 11 (8%) 29 (19%) 8 (25%) 24 (66%) 10% 39% Aldo 5 (3%) 18 (6%) 0 (10%) 12 (29%) 3% 10% Jorge 13 (10%) 34 (25%) 10 (29%) 22 (61%) 14% 39% SWE= sight word efficiency. PDE = phonemic decoding efficiency. Total = total word reading efficiency.

PAGE 146

146 Table 4-6. Preand post-intervention KT EA-II raw scores and percentiles ranks Participants Pre-LWR Post -LWR Pre-NWDPost-NWD Amelia 44 (50%) 49 (53%) 4 (25%) 23 (75%) Pedro 35 (30%) 50 (58%) 8 (45%) 23 (75%) Jessica 40 (42%) 49 (53%) 9 (50%) 15 (58%) Viviana 24 (5%) 36 (18%) 2 (16%) 16 (58%) Ernesto 29 (16%) 40(27%) 2 (16%) 13 (53%) Maggie 39 (30%) 44 (37%) 3 (21%) 21 (70%) Maria 31 (21%) 41 (30%) 0 (3%) 21 (70%) Jennifer 23 (4%) 30 (10%) 3 (21%) 14 (55%) Aldo 15 (0.5%) 29 (8%) 0 (3%) 11 (45%) Jorge 20 (2%) 31 (30%) 4 (5%) 14 (55%) LWR stands for letter-word recognition. NW D stands for Nonsense Word Decoding. Table 4-7. Preand post-intervention DIBELS Nonsense Word Fluency scores Participants Preintervention Postintervention Amelia 67(LR) 116 (AA) Pedro 45 (MR) 113 (AA) Jessica 47 (MR) 112 (AA) Viviana 46 (MR) 57 (LR) Ernesto 46 (MR) 63 (LR) Maggie 66 (LR) 64 (LR) Maria 35 (MR) 54 (LR) Jennifer 42 (MR) 99 (AA) Aldo 3 (HR) 51 (LR) Jorge 38 (HR) 59 (LR) Scores refer to the number of correct lette r-sound correspondences read per minute. AA = above average, LR = low risk, MR = moderate risk, and HR = high risk. Reading fluency. The Oral Reading Fluency test (ORF DIBELS) was used to measure students fluency of connected tex t. Results showed that all students made gains from the beginning of second grade passage to the end of second grade passage, with gains ranging from 9 to 54 correct word s per minute. On grade-level passages, at the beginning of the interventi on, six students were at high risk and 4 were at low risk. In contrast, at the end of the intervention, five students were at high risk, three at moderate risk, and two at low risk. Students who were at moderate risk or high risk on any of the passages were administered easier passages until a low risk or ab ove average status

PAGE 147

147 were achieved, or until passages corres ponding to the middle of first grade were reached. Individual data can be found in Tables 4-8 and 4-9. Table 4-8. Pre-intervention DI BELS Oral Reading Fluency scores Participants Beginning-second grade ORF passage End-first grade ORF passage Mid-first grade ORF passage Amelia 55 (LR) Pedro 63 (LR) Jessica 61 (LR) Viviana 14 (HR) 11 (HR) 19 (MR) Ernesto 25 (HR) 34 (MR) 47 (AA) Maggie 45 (LR) Maria 16 (HR) 21 (MR) 18 (MR) Jennifer 7 (HR) 10 (HR) 6 (HR) Aldo 3 (HR) 1 (HR) 3 (HR) Jorge 7 (HR) 7 (HR) 3 (HR) Scores correspond to the total number of words read correctly per minute. Passages with no scores were not administered to those particular students. AA = ab ove average, LR = low risk, MR = moderate risk, and HR = high risk. Table 4-9. Post-intervention DI BELS Oral Reading Fluency scores ORF Passages Participants End-2nd grade Mid-2nd grade Beginning 2nd grade End-1st grade Mid-1st grade Amelia 102 (LR) Pedro 94 (LR) Jessica 75 (MR) 73 (LR) Viviana 49 (HR) 48 (HR)46 (LR) Ernesto 79 (MR) 65 (MR)74 (AA) Maggie 79 (MR) 66 (MR)62 (LR) Maria 56 (HR) 52 (MR)48 (LR) Jennifer 33 (HR) 34 (MR)28 (MR) 33 (LR) Aldo 12 (HR) 11 (HR)11 (HR) 17 (MR) Jorge 32 (HR) 35 (MR)34 (MR) 31 (LR) Scores correspond to the total number of correct word s read per minute. Passages with no scores were not administered to those particular students. AA = ab ove average, LR = low risk, MR = moderate risk, and HR = high risk. Reading comprehension. The Qualitative Reading Inventory-4 (QRI-4) was administered to measure reading comprehension in two ways: (a) ideas recalled, and (b) questions answered. The passage se lected to read was contingent on ORF (DIBELS) risk status. Before intervention, four students read level two passages, one

PAGE 148

148 student read a primer pass age, and four students did not read any passages. In contrast, after intervention, seven students read level two passages, two read primer passages, and only one student did not read any passages. Results showed that all students increased the number of ideas recalled. For the total number of questions answered correctly (explicit and implicit), scores remained the same for one student and increased for the other nine students. Individual data c an be found in Tables 4-10. Attitudes toward reading. The Elementary Reading Attitude Survey (ERAS) was used to assess students attitudes toward reading. This survey measures attitudes toward academic reading and recreational readi ng. It also provides a total attitude toward reading score. Results showed that all students made a marked improvement in attitudes toward reading in general, with gai ns in raw scores ranging from 15 to 38. Changes in attitudes toward recreational reading ranged from 4 to 19, and toward academic reading ranged from 5 to 23. A su mmary of raw scores and percentile ranks is presented in Table 4-11. Reading Ability Rating Scale. Before and after the intervention, teachers evaluated the reading ability of their students using the Read ing Ability Rating Scale. This scale addressed five key areas of r eading: (i.e., phonological awareness, phonics, fluency, comprehension, and vocabulary), and one area related to classroom behaviors (i.e., participation in class, use of English to communicate in cla ss, and motivation to read). Mean scores were calculated and report ed for each of the five reading areas and raw scores were reported for classroom behav iors. Preand post-intervention scores showed equal or higher ratings for all students on phonological awareness, phonics, fluency, and comprehension. In contrast, in the area of vocabul ary, six students

PAGE 149

149 Table 4-10. Pre/Post-intervent ion QRI-4 comprehension raw scores Passage Level Ideas Recalled Explicit Questions Implicit Questions Total Questions Reading Level Student Pre Post Pre Post Pre Post Pre Post Pre Po st Pre Post Amelia 2 2 5 12 3 4 3 2 6 6 Inst Inst Pedro 2 2 5 19 4 4 2 4 6 8 Inst Ind Jessica 2 2 4 8 1 3 1 1 2 4 Frus Frus Viviana N/A 2 5 2 0 2 Frus Ernesto P 2 0 9 1 3 0 2 1 5 Frus Frus Maggie 2 2 8 9 3 4 0 2 3 6 Frus Inst Maria N/A 2 10 4 3 7 Inst Jennifer N/A P 5 2 1 3 Frus Aldo N/A N/A Jorge N/A P 4 2 1 3 Frus Students read one of three passages: P = Primer, 1 = Level 1, 2 = Level 2. Ideas recalled, explicit questions, implicit questio ns, and total questions answered correctly correspond to the passage read by each student. Reading levels were identified based on the number of total correct answers based on QRI-4 norms. Inst = Instructional, Ind = Independent, Frus = Frustration. Students with no scores did not take the QR I-4 due to risk status on DIBELS ORF.

PAGE 150

150 Table 4-11. ERAS Pre-and post-interv ention raw scores and percentile ranks Recreational Reading Academ ic Reading Total Reading Participants Pre Post Pre Post Pre Post Amelia 27 (31%) 33 (68%)25 (32%)36 (85%)52 (29%) 69 (79%) Pedro 23 (11%) 29 (44%)28 (46%)37 (89%)51 (26%) 66 (71%) Jessica 21 (6%) 30 (50%)25 (32%)34 (78%)46 (1%) 64 (66%) Viviana 20 (4%) 39 (94%)21 (15%)36 (85%)41 (6%) 75 (92%) Ernesto 30 (50%) 34 (74%)19 (8%)32 (67%)49 (20%) 66 (71%) Maggie 21 (6%) 36 (84%)23 (23%)37 (89%)44 (9%) 73 (88%) Maria 24 (15%) 38 (92%)26 (37%)35 (81%)50 (23%) 73 (88%) Jennifer 23 (11%) 30 (50%)21 (15%)37 (89%)44 (9%) 67 (74%) Aldo 16 (1%) 31 (56%)14 (1%)37 (89%)30 (0%) 68 (77%) Jorge 13 (0%) 21 (8%)19 (8%)27 (41%)32 (1%) 48 (18%) obtained higher ratings, three obtained equal ratings, and one obtained a lower rating. For classroom behaviors, the ma jority of students received equal or higher ratings on participation in class, use of English in cl ass, and motivation to read. For each of these behaviors, only one or two students obtained lower rates from their teachers. Data can be found in Tables 4-12 and 4-13. Table 4-12. Preand post-intervention mean rate scores for reading skills on the RARS Phonemic Awareness Phonics Fluency Comprehension Vocabulary Name Pre Post Pre Post Pr e Post Pre Post Pre Post Amelia 2.3 4 1.8 3.3 1.8 3 1.6 3 4 4 Pedro 3 3 3.2 3 2.2 2.2 2.8 3.2 3.66 4 Jessica 2 4 2.5 3.5 2 3.2 1.6 2.6 2 3 Viviana 1 2 1 1.8 1 1.2 1 1.4 1 1 Ernesto 1 1 1.5 1.6 1 1.2 1 1.6 1.6 1.6 Maggie 2.6 3 2.5 3.331.3 2.831.2 2.6 3.3 3.66 Maria 3 4 2.5 3.662.2 2.5 1.6 2 2.6 2.66 Jennifer 1 2 1 1.6 1 1 1 1.2 1 1.3 Aldo 1 3.33 1 2.831 1.5 1 1.4 1 2.66 Jorge 2.3 3 1.3 2.661 1.831 2.2 3.66 2.66 Scores represent mean rates that range from 1 (very weak) to 4 (very strong). Book Levels Read during Intervention As a result of the tutori ng programs, all groups showed a marked improvement in book levels Group 1 started at a higher level than the other groups. At the beginning of

PAGE 151

151 Table 4-13. Preand post-intervention ra te scores for classroom behaviors on the RARS Participation in Class Use of English in Class Motivation to Read Pre Post Pre Post Pre Post Amelia 2 2 4 4 3 3 Pedro 4 4 4 4 3 3 Jessica 3 4 3 4 3 4 Viviana 2 2 4 3 2 3 Ernesto 2 3 2 3 4 4 Maggie 3 4 4 4 4 4 Maria 3 2 3 4 3 2 Jennifer 3 3 3 3 3 2 Aldo 1 3 2 3 1 4 Jorge 3 3 4 4 1 2 Scores for classroom behaviors repres ent the raw score for each of the three items included in this area. Scores go from 1(very weak) to 4 (very strong). the intervention (early November), students were reading books at level 9, corresponding to the middle of first grade. After 38 tutoring sessions (mid-January), all members of the group were reading books at level 20 with 98 to 100% accuracy. According to Reading Recovery book levels, students in second grade should reach level 20 by the end of the school year. This means that students in Group 1 were able to catch up to and surpass the expected reading levels for the specific time of year. Group 2 also showed notable im provement in book reading. At the beginning of the intervention (beginning of November), the group start ed reading books at level 2, which corresponded to the beginning of first grade. After 47 sessions the group reached level 16, corresponding to the end of firs t grade. While these students were still behind their peers, it is important to note that in approximately three and a half months, they had made the equivalent of one years progress according to Reading Recovery book levels. For Group 3, the improvement was less drastic, but still relevant. All members of this group started the intervention phase reading level 1 books, corresponding to the

PAGE 152

152 beginning of first grade. Level 1 books contain the most basic level of text, and the text is heavily supported by the pictures. Children ar e expected to read level 1 books in late kindergarten or at the beginning of first grade. At the end of 45 sessions (three and a half months), the group progre ssed to level 7. These students remained in each book level far longer than those in the other groups showing marked difficulties acquiring the decoding strategies being taught. Yet, with systematic and explicit instruction, they were able to make more improvement in three and a half months of intervention than they had made in all of their schooling before t he intervention began. It is important to consider that all three mem bers were repeating second gr ade, so their progress during intervention can be considered their first ac tual reading growth. A graph depicting the progression of book levels by ea ch group can be found in Figure 4-3. Figure 4-3. Book levels read by each group during intervention.

PAGE 153

153 Summary of Supplemental Data Supplemental data showed that st udents made marked improvements on all measures of early literacy skills, on book le v els read, and on attitudes toward reading. Some of the largest improvements were on decoding accuracy (NWD from KTEA-II) and decoding fluency (PDE from TOWRE and NWF from DIBELS). Word recognition accuracy (LWR from KTEA-II) and fluency (SWE from TOWRE) were also impacted but to a lesser degree. The effect of the inte rvention was also observed on students oral reading fluency (ORF DIBELS) where students were able to read more words per minute in connected text. Furthermore, at th e end of the intervention, QRI-4 scores demonstrated that students were able to comprehend texts, by recalling more ideas from the stories read and by answering more explicit and implicit questions. Improvements in students reading skills we re also observed by teachers who rated students higher on phonemic awareness, phoni cs, fluency, comprehension, and vocabulary. Results also showed that students in ev ery group made considerable advances on the levels of books they read during intervention making at least one halfs year progress. In addition, a much more positive outlook on reading was observed as students exhibited dramatic improvements in their attitudes toward academic and recreational reading. Social Validity Data collec ted through the students interviews and the teachers questionnaire showed that the program was regarded as im portant, effective, and feasible. What follows is a summary of students and teac hers responses. A visual depiction of teachers answers can be found in Figures 4-4 and 4-5.

PAGE 154

154 Student Interview At the end of the maintenanc e phase, students participat ed in a social validity interview that consisted of seven questions. The interviews took approximately five minutes and were conducted individually. Results showed that all ten participants had positive feelings about the tutoring lessons. They all thought the lessons helped them become better readers and that other students should participate in the program. Nine students affirmed that they will us e w hat they learned when they read. In relation to students favorite part of the lesson, six out of ten said they preferred doing word work with magnetic letters, while four said they liked reading books. When asked about their least favorite part of t he lesson, only four st udents stated disliking a part of the lesson. They all r eported that writing was their least favorite. Finally, nine out of ten students said that they enjoyed working in small groups. Teacher Questionnaire At the end of the intervent ion, teachers completed a social validity questionnaire after observing a videotaped tutoring session The questionnaire was divided in six different sections. The first five sections focused on t he different components of the UFLI program (e.g., repeated reading of familiar books, progr ess monitoring, word work with manipulative letters, reading a new book wr iting for reading). In each of the five sections, teachers had to answer yes or no to four questions: (1) Do you feel that this part of the lesson will help EL Ls develop reading skills? (2) Do you believe that this component is easy to implement with ELLs? (3) Currently, do you use a similar strategy when you work with ELLs? (4) Would you be likely to use a strategy like this one with ELLs during your reading instruction? Besides answering yes or no, teachers had the option of explaining the reason behind their answers.

PAGE 155

155 The sixth section focused on the overall tutoring lesson. Teachers had to answer six questions: (1) Do you feel that the tutoring lesson can help ELLs develop reading skills? (2) Do you believe that this lesson is easy to implement? (3) Would you be willing to implement the lesson in your classr oom? (4) Would you be willing to implement components of the lesson in y our classroom? (5) Would you be interested in learning how to implement this tutoring program? (6) Do you think other teachers of ELLs might be interested in learning to implement this program? Teachers also had the option of explaining the reas on for their answers. Results showed that all teachers believe d that each of the components and the UFLI lesson as a whole can help ELLs dev elop their reading skills. They also agreed that the UFLI lesson and all its component s would be easy to implement in the classroom, except for one teac her who stated that progre ss monitoring and writing for reading would be difficult to execute with ELLs. According to her, it gets difficult with a full class. When asked if they were using similar stra tegies with ELLs, they all reported using repeated reading in the classroom. Four out of six teachers said they were conducting progress monitoring, and using a strategy simila r to reading for writing, while five out of six reported doing word work and reading ne w books. All teachers stated that they would be likely to incorpor ate the UFLI lesson or any of its components in their classrooms, but only four out of six were interested in learning how to implement it. Also, five out of six agreed that other teacher s of ELLs might be interested in learning how to implement the UFLI program. Teachers answers for each question related to

PAGE 156

156 UFLI tutoring components are depicted in Figure 4-4, while answers for questions related to the UFLI tutoring program as a while are represented in Figure 4-5. Figure 4-4. Teachers responses to the social validity questi onnaire about the UFLI components Figure 4-5. Teachers responses to the social validity questi onnaire about the UFLI program Summary of One-on-One Tutoring Data In this secti on, a summary of findings is provided for the student receiving one-onone tutoring. First, changes in the tw o dependent variables (CPPM and CSPM) are

PAGE 157

157 reported for each of the phases. Then, supplemental data is reported, including preand post-intervention measures of early readi ng skills, attitudes toward reading and teachers rating scales, as well as changes in leveled book reading. Dependent Variables The rates of CPPM and CSPM were calculat ed for each probe and plotted into a graph for visual analysis (see Figure 4-6). Du ring baseline, Miriam received a total of five pseudoword and sight word probes. The rate of pseudowords ranged from 3 to 5 CPPM with a mean score of 4.66 CPPM, while the rate of sight word reading ranged from 13 to 16 CSPM, with a mean rate of 14.66 CSPM. Once a stable trend line was established during baseline, Miriam started the intervention phase. During this phase, she received a total of 26 pseudoword and sight word probes, one for each tutoring session. The rate of pseudoword reading ranged from 10 to 28 CPPM, with a mean score of 17.65 CPPM, which shows an in crease of 12.99 CPPM from baseline to intervention. Furthermore, the rate of si ght word reading ranged from 16 to 46 CSPM, with a mean score of 30.1 CSPM. This show s a positive change of 15.44 CSPM from baseline to intervention. Intervention ended after 26 tutoring lessons when she was able to read books at level 20 with more than 95% accuracy. Two weeks after the intervention ceased, maintenance data were co llected using the same set of probes. At that time, Miriams rate of pseudoword reading ranged from 24 to 27 CPPM, with a mean score of 25.4 CPPM, which shows an in crease of 7.75 from t he previous phase. Furthermore, the rate of sight word reading ranged from 43 to 46 CSPM, with a mean score of 44.2 CSPM. This shows a positiv e change of 14.1 CSPM from intervention to maintenance.

PAGE 158

158 Figure 4-6. Rate of correct pseudoword and sight word read per minute for Miriam. Supplemental Data Before intervention, Miriams raw score on Sight Word E fficiency (SWE) was 28, which placed her at the 32nd percentile, while her score on Phonemic Decoding Efficiency (PDF) was 6, which placed her on the 19th percentile. Miriams percentile score on the Total Word Readi ng Efficiency was 21%. By the end of the intervention, she obtained a score of 54 on the SWE (67th percentile) and 20 on the PDF (84th percentile). Her Total Word Readi ng Efficiency placed her at the 64th percentile. The KTEA-II scores improved after intervention as well. On Letter Word Recognition (LWR) she went from the 36 correct words read (32nd percentile) to 43 words (34th percentile). On Nonsense Word Decoding (NWD), she went from 6 nonsense words (37th percentile) to 20 nonsense words read correctly (68th percentile). On DIBELS, Miriams NWF scores went from 51 (low risk) to 94 (above average), showing an improvement of

PAGE 159

159 43 correct letter-sound correspondences per minute. ORF scores showed an increase of 48 words, going from 18 (high risk) to 66 words per minute (high risk). Given her high risk status, easier passages were administer ed at both times. At the beginning of the study, she read 18 words on t he end of first grade passage (h igh risk), and 20 words on the middle of first grade passage (low risk) At the end of the study, Miriam read 68 words on the middle of second grade passage (low risk). Comprehension scores on the QRI-4 befor e intervention were based on a primer level passage. Miriam was able to recall two ideas, and answer one explicit question correctly. This showed that she was reading prim er level passages at a frustration level. After intervention, she was given a level two passage. Miriam was able to recall 14 ideas from the story and answe red four explicit and three im plicit questions. A score of seven points indicated that she was reading level two passages at an instructional level. Attitudes toward reading showed a mark ed improvement after intervention. Initially, Miriam obtained 20 points (4th percentile) on recreational reading and 22 points (18th percentile) on academic reading. Overa ll, she scored a total of 42 points (7th percentile). After intervention, recreation reading score went up to 32 points (62th percentile) and academic went up to 35 points (81st percentile). Her total score was 67, which corresponded to the 74th percentile. The teachers ratings on the RARS increas ed after intervention for most areas of reading ability. The mean score for phonological awareness remained at 3 (strong). In contrast, phonics mean rates went from 2.66 (weak) to 3.3 (strong) fluency rates from 1.83 (very weak) to 2.3 (weak), comprehens ion from 2.4 (weak) to 3.2 (strong), and vocabulary from 3 (strong) to 4 (very strong). Classroom behaviors remained the same

PAGE 160

160 for class participation with a score of 3 (str ong) and use of English to communicate in class with a score of 4 (very strong). Moti vation to read increased from a score of 3 (strong) to 4 (very strong). Miriam also showed rapid progress in leveled books throughout the intervention. At the beginning of the study, she was reading books at level 6, which correspond to early first grade. On session 25th, she moved up to level 20, which corresponds to the end of second grade. On session 26th, she read a level 20 book with more than 95% accuracy. Figure 4-7 depicts Miriams progr ess compared to the changes in groups book reading. Figure 4-7. Book levels read by Miriam during intervention. This figure compares the changes in book levels during one-onone tutoring compared to the progress made by students in small-group tutoring and illustrates the faster rate of growth for the student rece iving one-on-one tutoring. Summary of Findings The findings described in this chapter rev eal that the UFLI tu toring program was an effective small-group reading intervention that produced drasti c changes in students

PAGE 161

161 reading abilities. Improved rates of CPPM and CSPM were reached during the intervention phase and sustained two wee ks after it ended. Changes in pre-and postintervention measures, as well as changes in book levels read throughout the lessons provide additional evidence of the programs effectiveness. Fu rthermore, social validity data demonstrated that the UFLI tutoring progr am was considered effective, efficient, and important by students and t eachers. Chapter 5 provides interpretation of these findings, including their implicatio ns for practice and research.

PAGE 162

162 CHAPTER 5 DISCUSSION The purpos e of this study was to evaluate the effectiveness of a modified version of the University of Florida Literacy Initiati ve (UFLI) small-group tutoring program on the reading skills of second-grade Spanish-speaking, English language l earners, who were struggling to read in English. The aim of th is chapter is to summarize and interpret the results obtained in this study. The chapter is organized in several sections: (a) overview of the study, (b) summary of findings, (c) interpretation of findings, (d) interpretation of supplemental data, (e) social validity, (f) e ffectiveness of the UFLI tutoring program, (g) implications for future res earch, (h) implications for practice, and (i) conclusions. Overview of the Study A single-subject, multiple baseline across groups des ign was implemented to examine the effects of the UFLI tutoring program on the reading s kills of 10 Spanishspeaking, English language le arners who were struggling to read in second grade. Participants ranged in age from 7 years 8 mont hs to 10 years 1 month at the beginning of the study. School personnel identified st ruggling second-grade ELLs, whose native language was Spanish, and who were performing below level on their latest end-of-year SAT-10 and DIBELS scores. Ba sed on these two scores, as well as beginning-of-year Invented Spelling Assessment scores, student s were selected to participate in the small-group tutoring program designed to address word reading skills and modified to support their language needs. The level of English and Spanish oral language proficiency of the participants va ried from negligible to fluent. Students were grouped based on the level of book they were able to read with 90% to 95% accuracy, as determined by runni ng records and using Reading Recovery

PAGE 163

163 book levels. A total of three groups were formed. Group 1 was composed of three students (1 male and 2 female) reading level 9 books. Group 2 was composed of four students (1 male and 3 female) reading leve l 2 books. And Group 3 was composed of three students (2 male and 1 fema le) reading level 1 books. The focus of this study was an examination of the effects of the UFLI program on two reading skills: decoding and sight word reading. Decoding was measured by assessing the rate of correct pseudowords read per minute, while sight word reading was measured by assessing the rate of corr ect sight words read per minute. For this purpose, data were collected during three phases : (a) baseline, (b) intervention, and (c) maintenance. During baseline, students were given pseudoword and sight word probes with the goal of establishing the rate of co rrect words each was able to read per minute. When Group 1 demonstrated a stable line over six data points, the group moved to the intervention phase. Meanwhile, the other groups remained in baseline. When Group 1 showed an increase in four consecutive data points, Group 2 moved from baseline to intervention. The cycle was repeated for Group 3. During intervention, 10 student s participated in smallgroup tutoring sessions for 45 minutes, 2 to 5 times a week. The tota l number of sessions varied across groups based on students attendance and book level reached by the group. The maintenance phase of the study was designed to examine w hether the effects of the UFLI program were sustained two weeks after the intervention concluded. One additional student participated in one-on-one tutoring sessions using the same instructional methods. In addition to rate of pseudowords and sight words read per minute, supplemental data were collected, including preand postmeasures of reading skills (e.g., decoding,

PAGE 164

164 word recognition, fluency, comprehension, attitudes toward reading) as well as teachers ratings of students reading ab ilities and classroom behaviors. Also, changes in book reading throughout the intervention we re recorded for each group. Finally, social validity information was collected thr ough student and teacher social validity questionnaires. Summary of Findings An examination of the data gathered showed several key findings. First, the UFLI small-group tutoring program was effectiv e in producing substantial improvement on decoding and word recognition skills of student s, as measured by the rate of correct pseudowords read per minute and the rate of correct sight words read per minute. Furthermore, students sustained these improved rates two weeks after the intervention ceased, demonstrating the effectiveness of the UFLI program in producing a lasting effect on students reading skills. Second, in addition to the two main variables, all groups showed a marked improvement when reading leveled books. Group 1 made the equivalent of one and a half years progress in 38 sessions, going from level 9 to level 20 books. Group 2 made a years progress in 47 sessions, going from le vel 2 to level 16 books. Finally, Group 3 showed approximately one-half y ears progress in 45 sessions, moving from level 1 to level 7 books. In other words, all groups made more rapid progress during the intervention than would be expected duri ng typical classroom instruction. Third, supplemental data showed t hat participants improved on pre-and postintervention measures of decoding, wo rd recognition, or al reading fluency, comprehension, and attitude toward r eading. In particular, students showed improvement in decoding and word recognition accuracy and fluency. Changes in

PAGE 165

165 students abilities were also observed by their teachers, who ra ted them higher on phonemic awareness, phonics, fluency, vocabul ary, comprehension, and classroom behaviors. Fourth, all participants benefitt ed from the intervention, r egardless of their initial level of English oral language proficiency. It is important to note that not all students in the higher reading performing group had the highes t level of oral proficiency. On the contrary, there were some students in lowe r groups who had equal or higher levels of oral proficiency. Thus, the rate of impr ovement differed across groups with different reading levels and not different proficiency levels. Fifth, the student who received one-on-one intervention using the same instructional methods also demonstrated considerable reading growth on the two dependent variables. In fact, her progress was more rapid than that of the students receiving small-group instruction. Supp lemental data also showed a marked improvement in reading ability after intervention. Finally, social validity data showed that the UFLI program was well received by students and teachers who noticed its effectiveness in improving students reading skills. The program was not only deemed effect ive, but also feasible. Students believed that other ELLs might benefit from the program, and teac hers reported their willingness to implement it in their classrooms. Interpretation of Findings Findings from this study provide evidenc e to support the use of early reading interventions in Englis h for ELLs who are st ruggling to learn to read in their second language. Specifically, the UF LI tutoring program was effe ctive in promoting word reading skills among second-grade, Spanish -speaking ELLs who were receiving

PAGE 166

166 reading instruction and who had been performing below grade level in measures of reading achievement. As a result of daily, small-group tutoring sessions focusing on word reading skills, partici pating students showed a mark ed improvement in their decoding skills, as measured by the rate of correct pseudowords read per minute (CPPM), and word recognition, as measur ed by correct sight words read per minute (CSPM). These findings are consistent with previous studies reporting the effectiveness of word-level reading interventions on t he reading skills of Spanish-speaking ELLs, particularly on word attack (Gunn et al., 2000, 2002, 2005; Linan-Thompson, Vaughn et al., 2003; Vaughn, Cirino et al., 2006; Vaughn, Linan-Thompson, & Hickman, 2003; Vaughn, Mathes et al., 2006) and word identification skills (Al Otaiba, 2005; Gunn et al., 2000, 2002, 2005; Quiroga et al., 2002; Vaughn, Cirino, et al., 2006). Decoding and automatic word recognition are critical skills in the early stages of reading development. Beginning readers need to develop the ability to analyze and manipulate the sounds in speech, as well as to recognize the letter-sound correspondences so that they can effectiv ely decode unfamiliar words in print (Ehri, 2004). Furthermore, students must develop the ability to read words accurately and automatically to achieve fluent reading (Stahl 2004). In order to promote these reading skills, researchers have underscored the impor tance of providing instruction in phonemic awareness and phonics not only for native speakers of English (Adams, 1990; Ehri, 2004), but also for English langu age learners (Ehri & Roberts, 2006; Helman & Burns, 2008). The present study provides further evidence to support the importance of such instruction.

PAGE 167

167 Progression through Phases of Word Recognition Development Based on recent findings suggesting that native Englis h speakers and ELLs develop word recognition skills in similar ways (Chiappe & Siegel, 2006; Chiappe, Siegel, & Gottardo, 2002; Chiappe, Siegel & Wade-Wolley, 2002; Geva et al., 2000; Jackson & Lu, 1992; LeSaux, Koda, Siegel, & Shanahan, 2006; Limbos & Geva, 2001; Verhoeven, 1990, 2000; Wade-Woolley & Siegel 1997), students response to the UFLI intervention can be described following Ehr is phases of word reading development. According to Ehri and McCormick (1998), reader s progress through five phases of word reading development: (a) pre-alphabetic, (b) partial-alphabetic, (c) full-alphabetic, (d) consolidated-alphabetic, and (e) automatic (Ehri & McCormick, 1998). Each phase is characterized by the learners underst anding and use of the alphabetic system (p. 140). Before the UFLI intervention, some of the second-grade par ticipants displayed reading behaviors typical of the partialalphabetic phase, while others displayed behaviors of an incipient full-alphabetic phase. According to Ehri (2005b), the partial alphabetic phase is commonly seen among kinder garten and first grade students, while the full-alphabetic phase is co mmonly seen among first-graders. Nevertheless, it is not unusual to find older struggli ng readers at any of these phases. Students who have not developed strong skills in the full-alphabetic phase, particula rly struggling readers, need systematic phonics and phonemic awareness instruction to help them develop the ability to read words accurately and automatic ally (Ehri & McCormick, 1998). Consistent with this premise, after participating in t he UFLI tutoring program, all ten participants exhibited a more advanced word reading abi lity, signaling a movement towards more proficient reading.

PAGE 168

168 For example, evidence from the reading behaviors exhibited by students in Group 1 indicated that they had begun transitioning into the fullalphabetic phase prior to the intervention. They had started developing a full working knowledge of the alphabetic system as well as more advanced segmentat ion ability, but they were still lagging substantially behind their average performi ng peers. Their reading was slow and arduous, but they had started to represent full sight words in memory, to use decoding and analogy to read unfamiliar words, and to s pell words more accurately. According to Ehri and McCormick (1998), to help students move from slow deliberate decoding to faster decoding it is impor tant to provide ample opportuni ties to practice (p. 152). Specifically, instruction should give opport unities to practice analyzing within-word grapho-phonemic associations, dividing words into smaller units (e.g., syllable, onsetrime, phonogram), working with word fa milies, applying decoding and analogy strategies when reading connected text, and using prediction to confirm reading accuracy (Ehri & McCormick). Given that the UFLI tutoring program is adapted to the indivi dual needs of the students, instruction was geared towards providing this type of instruction and practice. As a result, students in Group 1 were able to move into the consolidated-alphabetic phase by the end of the intervention, demonstrated by the chunking strategies they were using during lessons. By the end of t he intervention, student s had not only showed a marked improvement in the rate of CPPM and CSPM, but also the level of reading accuracy during level book reading. At that time, students were reading books at level 20, which corresponds to the end of second grade. This showed that they had caught up to and surpassed their second-grade peers based on the book levels expected for

PAGE 169

169 that grade. By this time, students were able to read more words accurately and automatically, using a more complete k nowledge of the alphabet ic system, recognizing and using consolidated units (i.e., rimes, syllables, root words, morphemes) to read multisyllabic words. They continued using decoding and analogy to read unfamiliar words, while their sight vocabulary continued to grow (Ehri & McCormick, 1998). Before the intervention, students in Group 2 performed considerably below students in Group 1 on measures of decoding and word recognition yet their reading behaviors indicated that each had started the transition into the full-alphabetic phase. One reading behavior characteristic of this trans ition relates to the type of reading errors made when students decode unfamiliar words. While younger students tend to produce errors that do not resemble target word s, students advancing in the development of word reading ability tend to produce errors that resemble the target words, signaling the use of a decoding strategy (Ehri, 2005a). In order to complete the transition in to the fullalphabetic phase, students need instruction in decoding and sight word learning. Specifically, instruction s hould guide students in processing all the letters in words, instead of just initial or final letters (Ehri, 2005a; Ehri & McCormick, 1998). With an intervention tailored to their needs, students in Group 2 showed a marked improvement in their rate of CPPM and CSPM as well as the level of accuracy during book reading activities. At t he end of the intervention, students were reading books at level 16, which corresponds to the end of first grade. This shows that the group made the equivalent of one years progress after 47 sessions ( approximately 3 months). By this time, it seemed that students in Gr oup 2 had completely moved into the fullalphabetic phase and had begun dem onstrating reading behaviors indicative of a

PAGE 170

170 transition into the conso lidated-alphabetic phase. They had a better knowledge of grapho-phonemic correspondences and were consis tently using decoding skills to read and spell unfamiliar words (Ehri, 2005a). Thei r growing sight vocabulary was apparent during reading and writing activities as we ll. For example, during reading, students began to cover word parts and read other parts as chunks. Nevertheless, it is important to note that their reading was still slow and laborious. Thus, continued instruction in decoding and sight word readi ng (Ehri, 2005a) is recommended for them to complete the transition into the cons olidated-alphabetic phase. Prior to the intervention, students in Group 3, the lower performing group, exhibited reading behaviors characteristic of the partial-alphabetic phase. Given their incipient knowledge of letters and sounds an d their developing phonemic segmentation ability, they were unable to use basic reading strategies, like decoding, analogy, or sight word reading. This is exemplified by their initial low rates of CPPM and CSPM, as well as their low level of reading accuracy during book reading activities. Ehri and McCormick (1998) identified common mistakes exhibited by students in the partialalphabetic phase: (a) confusing words that had similar spelli ngs (e.g., think and thank), (b) misreading words by using letter names inst ead of sounds (e.g., jo instead of go), (c) guessing words based on partial letter-sound relations and context cues (e.g., house instead of home), (d) reading words backwards due directionality problems (e.g., was instead of saw), (e) misreading words with graphem es (i.e., /c/ /h/ /e / /s/ /t/ instead of /ch/ /e/ /s/ /t/), and (f) writ ing words using only salient sounds (e.g., mret instead of market). Students in Group 3 exhibited all th ese behaviors before the intervention. To

PAGE 171

171 develop their word reading skills, these students needed to transition into the fullalphabetic phase. Juel, Griffith, and Gough (1986) identified f our elements needed for this transition: (a) phonemic awareness (segmenting, blending, s ubstitution), (b) exposu re to print (text level being read), (c) cipher knowledge ( nonword decoding), and (d) sight word knowledge (recognition of misspellings). Given that the UFLI tutoring program addresses these four components, students in Group 3 started making the transition to the full alphabetic phase as the intervention progressed. At first, Jennifer and Jorge were advancing at a faster rate than Aldo. Jennifer and Jorge were acquiring a more complete knowledge of the alphabetic system and were using this knowledge to decode unfamiliar words and to acquire sight words. Initially, their reading was extremely slow and laborious, but with practice it started to become more fluent. Their rate of CPPM and CSPM started increasing at a similar pace and their level of text reading accuracy started to improve. In contrast, Aldo continued having difficulty making letter-sound connections and identifying all the sounds in words. Many of hi s attempts to read unfamiliar words and high frequency words tended to be unsuccessful. His rate of CPPM and CSPM remained close to zero. Torgesen (2000) stated that many struggling readers or treatment resisters need more intensive intervention. Torgesen further states that many struggling students who are facing severe risk factors may require more than 1 or 2 years of intervention. In a study conducted by Vaughn, Linan-Thomps on and Hickman (2003), researchers found that when more intervention was provided a nd instruction was modified to meet the specific needs of each individual, students responded positively. In the present study,

PAGE 172

172 intervention was modified for Aldo. Star ting on session 20, Aldo received 15 extra minutes of instruction, in the form of one-on-one tutoring th at took place before his small-group session, focu sing on the letter-sound correspondences to be practiced during the small group session. Immediately, he showed a small improvement on CPPM and CSPM. Instead of reading a 0 to 1 C PPM, he started reading between 0 and 2 CPPM. Similarly, he went from reading 0 to 3 CSPM to reading between 2 and 4 CSPM. After 19 sessions of additional tutoring, Aldo made a marked improvement. His rate of pseudoword reading went from a range of 0 to 2 CPPM to a rate of 4 to 6 CPPM. His rate of sight word reading w ent from a range of 2 to 4 CSPM to a range of 4 to 8 CSPM. This change was also visible during book r eading and writing activities. He started segmenting and blending sounds more accura tely, and he started showing independent use of self-monitoring and self-correction. Th ese changes were also noticed outside the tutoring sessions. Aldos teachers reported that not only was he using reading strategies in the classroom, but for the fi rst time, he was also enthusiast ic about reading in class. By the end of the intervention, Aldo started showing reading behaviors ty pical of the fullalphabetic phase, just like J ennifer and Jorge, which incr eased his self-confidence. Despite their substantial improvements, it is important to recognize that students in Group 3 were still performing significantly below their second-grade peers. All three demonstrated urgent need for continued reading intervention, given that they were repeating second grade. Also, given Aldos overall performance, admi nistration of other measures could help uncover additional characteristics or the presence of a disability, which might be influencing his progress. Some suggested areas commonly associated

PAGE 173

173 to learning disabilities include phonological pr ocessing, working memory, word retrieval, and naming speed (Altmann, Lom bardino, & Puranik, 2008). Prior research makes it clear that students who continue struggling require systematic, explicit instruction in phonem ic awareness and phonetic decoding (Ehri & McCormick, 1998; Torgesen, 2000). Further more, teachers should provide daily opportunities to read connected text using easy materials that can be read with at least 96% accuracy (Allington, 1983, p. 555). Unfortunately, many struggling readers do not have these opportunities in their regular classrooms, as their peers move onto more advanced skills. Without continued intervention that is tailored to the individual needs of these students, it is unlikel y that struggling readers wil l catch up to their higher performing peers. In contrast to the groups response to intervention, Miriam moved to a more advanced phase of word recognition development in considerably less time. At the start of the intervention, she showed reading behav iors typical of an incipient full-alphabetic phase, where she started using decoding strategies to read unfamiliar words, as well as recognizing high frequency words by sight. Still, her reading was slow and laborious. This was demonstrated not only by low rates of CPPM and CSPM, but also by connected text reading during running record s. In response to the intense one-on-one intervention, Miriam rapidly moved into the consolidated-alphabetic phase, where she was able to decode larger word units and us e those to decode mult isyllabic words with accuracy and speed. Her full knowledge of letter-sound correspondences translated into an enhanced sight word reading ability and a more accurate and fluent reading of connected text.

PAGE 174

174 Overall, all students made subs tantial progress as a resu lt of the intervention. Students in Group 1 are likely to be able to succeed in reading without additional intervention. Students in Group 2 improved considerably but still will likely require additional support to continue making progress. Students in Gr oup 3, despite significant gains, remained well below expected performanc e for their grade level and will require continued intensive intervention. Similarly to Group 1, Miriam is likely to continue thriving in the classroom without additional intervention. Relation between CPPM and CSPM Two important findings concerning the relation between CPPM and CSPM merit further discussion. First, there was a marked difference between the CPPM rate and the CSPM rate for Groups 1 and 2. Throughout t he study, both groups consistently had a higher rate of CSPM than CPPM. This finding is consistent with t hat of Ehri and Wilce (1983), who found that less skilled readers were less able to decode nonsense words than real words. Ehri and Wilce stated that these students are inaccurate and excessively slow in decoding nonsense words, and that while practice improved their pseudoword reading speed, it remained slower than real words (p. 15). One possible explanation is that students who lack decoding skills are relying on visual features of words (Gough & Juel, 1991), or they are still relying on visual cue reading strategies characteristic of the pre-al phabetic phase (Ehri, 2002) to compensate for their lack of decoding skills. Second, it was noticeable in a ll groups that the rate of CSPM increased as the rate of CPPM improved, even t hough the emphasis of instruction was on decoding. According to Ehri (2005a), this re lation is possible because decoding helps build students sight vocabulary. When readers develop a working knowledge of the alphabetic system, they use th is knowledge to acquire and retain words in memory

PAGE 175

175 (Ehri, 2005b). This was certainly the case for Group 3, who initially decoded high frequency words and later started recognizing words by sight. Interpretation of Supplemental Data In addition to data collected during each phase of the si ngle-subject design, supplemental data were collected to provi de additional information about the effects of the intervention. Data were collected on student s early reading skills, attitudes toward reading, and teachers ratings of students reading ability. What follows is detailed analysis of these data. Early Literacy Skills Preand post-intervention data provide addi tional support for t he effectiveness of the UFLI program on students decoding an d wo rd recognition skills. These two skills were assessed under untimed conditions (KT EA-II) to evaluate reading accuracy and under timed conditions (TOWRE and DIBELS-NWF) to evaluat e reading fluency. The largest improvement in all participants wa s on decoding accuracy and fluency. This is most likely the result of t he programs prime focus on decodi ng instruction. Results also showed that students ability to recognize words by sight in creased after intervention. Instruction also targeted sight word reading, but to a lesser degree. It is possible that the newly acquired decoding ski lls impacted students ability to read words by sight. According to Ehri (2005a), decoding is a sel f-teaching mechanism that supports the acquisition of sight words (p. 149). W hen students decode unfamilia r words, those words are retained in memory, adding to their growing sight vocabulary. Another important finding from this study relates to reading fluency. Preand postintervention data on the ORF (DIBELS) showed that all the participants made drastic improvements on their rate of words read per minute. Aldo (Group 3) for example, who

PAGE 176

176 was reading three words per minute at t he beginning of the school year and who had difficult decoding basic VC words, was able to read 12 words per minute after intervention. Although this was still very weak reading, this change represents a substantial improvement considering that he was repeating second grade and the previous year he had read 4 words per minute on the end of year ORF passage. Dramatic improvements were also observed in other students. Ernesto (Group 2), for example, went from 25 words per minute to 79 words per minute, while Amelia (Group 1) went from 55 to 102 words per minute. Reading fluency is a complex construct that encompasses multiple processes and skills, of which decoding fluency is a signifi cant one (Hudson, Pullen, Lane, & Torgesen, 2009). According to Hudson and colleagues, an i nefficient decoding process interferes with the readers ability to read text fluently, given that a c onsiderable amount of energy is expended trying to decipher unfamiliar word s. This was certainly the case for the participants in this study. Their initial lack of decoding fluency interfered with their ability to read connected text with accuracy and autom aticity. However, once students started acquiring more efficient decoding processe s their ability to read connected text improved. This was not only reflected in t heir post-intervention fluency scores, but also throughout the lessons as they increased thei r level of reading accuracy of leveled books. The benefit of enhanced word reading skills does not stop with improved reading fluency. It is well known that a close relation exists between fluency and comprehension. When a reader fails to dec ode and recognize words automatically, reading fluency is compromised (Hudson et al., 2009). When reading fluency is

PAGE 177

177 compromised, comprehension suffers. Acco rding to LaBerge and Samuels (1979), this is because students are not able to focus their energy on higher order skills, so they are unable to derive meaning from text. Students comprehension scores in this study are a perfect example of this connection. Before t he intervention, half of the students did not take the QRI-4 because they were not abl e to read second-grade or even first-grade passages fluently. One student was given a primer level passage and due to his extreme difficulty with decoding and word recognition, he was not able to understand what he was reading. His comprehension sco res showed that he was reading at a frustration level. Furthermore, three of t he four students whose ORF scores showed low risk for second grade passages, and who were given QRI-4 level two passages, also had difficulties reading words accurately and fluently. His comprehension scores also showed that they were reading at a frustrat ion level according to QRI-4 norms. After intervention, when word level skills improved and fluency rates increased, a development in comprehension ability was readily observed. For example, this time, seven out of ten students were able to r ead level two passages. Of these students, three were reading at an instructional leve l and one at an independent level. Moreover, four of the five students who we re not given the QRI-4 before intervention, were able to take it after. Two of t hese students read level two pass ages and two read primer level passages. This shows that the transition toward a more effective and efficient reading ability had initiated. Students ability to derive meaning from te xt was also influenced by the programs high emphasis on reading comprehension deve lopment. Each tutori ng lesson included activities designed to help students acqui re effective comprehension skills. For

PAGE 178

178 example, before the students read a new book, the tutor introduced the book and engaged the group in a discussion of the pict ures, with the goal of activating prior knowledge and creating context for the story. Then, as students read the books, they were encouraged to make connections and predictions related to the story. The tutor also modeled and guided students to self-monito r for comprehension. After reading, the tutor and students engaged in a discussion of the story through t he use of literal, inferential, and evaluative questions. Children were also asked to summarize part of the story and then write a summary sentence (H ayes et al., 2005; Lane et al., 2005). Finally, the English vocabulary support pr ovided by the tutor throughout each lesson contributed to students under standing of what they re ad (Carlo et al, 2004). Attitudes toward Reading Another change obser ved after intervention was an increased positive attitude toward reading. Results from the ERAS showed that at the beginning of the study students did not have strong positive attitudes toward reading, particularly recreational reading. This meant that st udents preferred reading activi ties associated with school more than reading for fun outside of school. These findings are most likely associated with students history of r eading difficulties. Students wh o have not acquired basic reading skills and who have experienced repeat ed failure during reading activities are less likely to become motivated and self-r egulated readers who are willing to read outside of school (Quirk & Schwanenflugel, 2005). It is also important to note that despit e showing more positive attitudes toward academic than recreational readi ng at the beginning of t he study, their scores placed them considerably below second-grade norms. Quirk and Schwanenflugel (2005) stated that, typically, students who are in rem edial reading interventions have already

PAGE 179

179 experienced at least a year of repeated failu re associated with reading. Thus, students low academic reading attitude scores might have been a reflection of this occurrence. For example, six of the ten students who were repeating second grade when the study started were still having trouble reading grade-level books. Unless these students acquired the necessary reading skills to be su ccessful readers, they would continue to experience negative attitudes toward reading. Fortunately, not only did reading skills improved after intervention, but their attitudes toward academic and recreational r eading improved, as well. It is likely that with successful reading came a heightened sense of self-efficacy, which in turn translated into a positive disposition for reading (Quirk & Schwanenflugel, 2005). This connection between reading abilit y, self-efficacy, and attitudes toward reading was reflected not only on students scores, but al so on students performance in the tutoring program and in the classroom. For example, the tutor noted that, initially, there was a tendency among students to complain and avoi d work that was too challenging for them. As their skills started to improv e, students began to participate more and complain less. For example, at the beginn ing, Aldo complained when asked to read a page that had more than one sentence. Toward the end of the intervention, he would ask to read the whole book by himself. This trend was observed in the other groups, as well. Students would ask to read more than one page at a time or to read more books. Teachers also reported that students were vo lunteering to read in class and were eager to read more books. Students improved attitudes toward reading may contribute to the lessening of the Matthew Effect. Stanovich ( 1986) states that good readers continue to improve more

PAGE 180

180 rapidly than poor readers at least in part because they read more, thereby getting more practice in the skills necessary for improvement. It was evident from their preintervention reading achievement and attitude scores that the students in this study were not practicing reading nearly enough. With improved skills and accompanying improved attitudes toward reading, perhaps these students will continue to read more and get the practice they need to close the gap between their reading and that of their peers. Teacher Reading Ability Rating Scale (RARS) Chang es in reading ability were also not iced by students classroom teachers. Overall, teachers rated students reading abilitie s higher after intervention. This is very important because it indi cated that students newly acquired skills were being generalized to the classroom setting. Teacher s observations of students progress were also corroborated through their informal co mmunications with the researcher. A few times, teachers expressed their contentment with students increased reading ability and increased levels of participation in class. According to Perry and Meisels (1996), teachers are in a good position to evaluat e their students achievements because they interact with them on a regular basis. Still, a closer look at specific tasks withi n each of the five reading skills showed that, for some students, teachers failed to recognize a change despite students marked improvement on the measurements used in the study. Interestingly, some of these teachers expressed on different occasions that their students were still struggling or that they still could not read. This raises the ve ry important issue of how teachers measure progress. In these classrooms, for example, struggling r eaders were required to read grade-level texts. Students who were performing significantly below grade level before

PAGE 181

181 intervention (e.g., reading primer level passages) and who made considerable growth as a result of the intervention (e.g., readi ng first grade passages), may not be able to demonstrate this progress when teachers provide reading materials that are still above their reading ability. Furt hermore, it is likely that for some teachers it might be difficult to recognize students progress because, as Allington (1983) acknowledged, many struggling readers tend to work arduously thr ough any text, even the ones they can read accurately. Allington recommended that str uggling readers be given opportunities to read easy materials so that they get to read more like good readers (p. 555). Social Validity Single-subject research helps identify inte rventions that target socially relevant outcomes (Horner et al., 2005). D emonstrati ng the social validity or appropriateness (Wolf, 1978) of an intervention is import ant when working in educational settings (Kennedy, 2005). According to Horner and colle agues, the social validity of a singlesubject study is enhanced when teachers and other intervention agents report that procedures are acceptable, feasible, and effectiv e. In this study, the goal was to receive feedback from students and teachers. Students Responses The acceptability of the interventi on by direct consumers (Kennedy 2005), students in this case, is very important becaus e it can influence how they respond to the intervention. The fact that students had positive feelings towa rd the program might have contributed to their positiv e disposition throughout the lessons. Most students were eager to work and worked hard to complete all the tasks. Their positive feelings were also reflected in the excitement they showed when the tutor would pick them up in their classroom. Interestingly, Jorge, who refu sed to work on many occasions and created

PAGE 182

182 distractions for other students, reported enjoying tutoring. One possible explanation for his positive response is that tutoring prov ided an escape from the regular classroom where he frequently got in trouble. Neverthel ess, when the work from the tutoring started to get difficult for him, he would get frustrated a nd immediately engage in covert and overt misbehavior (e.g., making fun of ot hers, ripping books, throwing his pencil or magnetic letters, doing minimal work, not answ ering questions). For students like Jorge, who are in critical need of s upport it is important to devise appropriate ways to also help him take control of his emotions and find better ways to express his frustration. An important finding from the intervie ws is that students recognized that participating in the tutoring program helped them become better readers. Some students stated that they were reading mo re, reading better, and reading faster. Jennifer, in particular, said she actually l earned to read as a resu lt of the tutoring program. Aldo recognized that he was reading on his own. He said: Now I read books. You no help much. The fact that the Aldo was able to sound out words on his own and read faster than before was very motivating for him. Toward the end of the intervention, he repeatedly said that he was reading without t he tutors help. This finding is relevant given the tendency of struggli ng readers to have low confidence in their ability to read (Guthrie & Davis, 2003). In relation to the different components of a UFLI lesson, students had different preferences. Some students liked doing word work with magnetic letters and others preferred reading books. Yet, when asked about their least favorite part, the only activity that was identified by students was writing sentences (Writing for Reading). Some of the reasons given were that it was too hard, too much work, and too tiring. Writing for

PAGE 183

183 Reading involves a lot of work for thes e students: (a) encoding familiar and unfamiliar words by segmenting and blendi ng, (b) writing sight words by practicing correct spellings, (c) figuring out the appropriate me chanics of writing like capitalization and punctuation, and (d) rereadi ng the sentence created after each work is added. Unfortunately, many of these students were not accustomed to persisting until they could produce a correct sentence, and this probabl y is why it felt like too much work for them. Fortunately, despite being a challe nging program, all par ticipants agreed that other struggling readers should parti cipate in the tutoring lessons. Teachers Responses Given that the intervention was implemented by the rese archer, it was important to know what indirect consumers (Kennedy, 200 5), in this case the teachers, thought about the intervention. Teachers opinions about the acceptability feasibility, and effectiveness of the intervention (Horner et al., 2005) can influence th eir decisions about adopting the program and implementing it with ELLs. All teachers agreed that the UFLI program and each of its components can hel p ELLs become better readers, and that they would be likely to incorporate the UFLI lesson or any of its components in their classrooms. This is relevant given that teachers beliefs about the importance of an intervention influence their disposition to adopt and implement new practices (Sparks, 1988). One teacher recognized that one of her students who couldnt put two sounds together at the beginning of the school year started reading as a result of his participation in the program. Similar comme nts were expressed by other teachers throughout the study, who st arted noticing an improv ement not long after the intervention started.

PAGE 184

184 Despite all stating that t hey were willing to implem ent the program in their classrooms, two teachers reporte d not being interested in l earning how to implement it. One stated that she had a good understanding of it, while th e other said that she had done that type of program befor e. Unfortunately, just being familiar with it or having a good understanding is not enough to ensure appropriate implem entation. Participating in professional development is critical fo r teachers to acquire the necessary knowledge and skills to properly implement the interventi on, especially as it addresses the needs of ELLs. It is well documented in the literature that professional development is associated not only with improved instructional practices, but also with students learning (Borko, 2004). Finally, it is important to acknowledge that all teachers reported using similar strategies in their classrooms, particul arly repeated reading. Some teachers did progress monitoring, while others did word wo rk, or writing activities. Many of them stated that the SRA/EI R program that they use in their classrooms has some similar activities. Yet, they stated that word work ac tivities did not inclu de manipulative letters and writing activities did not go as in -depth as the UFLI program does. These distinctions could explain the differences in relative effectiveness of the two intervention programs. Overall, the UFLI program was we ll received among st udents and teachers. Moreover, other school personnel (e.g., school principal, r eading coach) reported their satisfaction with the program and their desire to use it in the future. They stated that other teachers had heard about students progress and were interested in learning

PAGE 185

185 about it. Professional development will be provi ded by the researcher to all interested teachers. Effectiveness of the UFLI Tutoring Program There are three plausible explanations or the effectiveness of the UFLI tutoring program. These include: (a) ear ly intervention and language prof ic iency, (b) provision of instructional practices for ELLs, and (c) consistency with the RTI approach. What follows is a brief description of each explanat ion and the theoretical foundation behind it. Early Intervention and Language Proficiency The role of early reading interventions in the prevention of later reading difficulties has been well documented in the literature (E hri, Dreyer, Flugman, & Gross, 2007; Gersten & Dimino, 2006; Tor gesen et al., 2001; Vaughn, Cir ino et al., 2006). Early interventions are parti cularly important giv en the tendency of struggl ing readers to fall further and further behind their higher performing peers. According to Gunn and colleagues (2000), students who have not learned to read by the end of third grade are less likely to successfully develop their literacy skills. While Chall (1983) stated that up until third grade the focus of the curriculum is on learning to read and that the focus shifts from learning to read to reading to learn in fourth grade, in practice and with increased accountability, it seems that this shi ft occurs much earlier. A close look at the Florida Sunshine State Standards, for exampl e, shows that students in third grade require reading proficiency to fully benefit from the curriculum (FLDOE, 2010d). Therefore, it is critical that struggling readers develop the necessary reading skills in order to successfully complete third grade tasks. Unfortunately, there is a tendency for educ ators to delay interventions until nonnative speakers develop their English skill s or to exclude students with limited English

PAGE 186

186 proficiency because they have lower probab ilities of benefitting from interventions (Ashdown & Simic, 2000). These notions have been supported by available literature that underscores how ELLs struggle to unders tand the words they read (Helman, 2009), and consequently, struggle to comprehend texts (Peregoy & Boyle, 2005). Unfortunately, these approaches lim it the opportunities for ELLs to receive the help they need. Furthermore, when educators delay inte rvention, children get farther behind, making it more difficult for t hem to catch up to their peers. Consequently, the initial gap that existed between Spanish-speaking ELLs and their English-speaking peers continues to widen, forcing educators to fi nd and implement more intense interventions. Fortunately, there is sound evidence t hat ELLs benefit from English reading interventions regardless of t heir level of English oral l anguage proficiency (Gunn et al., 2000, 2002, 2005; LeSaux & Siegel, 2003; Quir oga, Lemos-Britton, Mostafapour, Abbott, & Berninger, 2002; Roberts, 2003; Stuart, 1999). The present study provided additional support for t he use of early reading interventions for ELLs with different levels of oral language proficie ncy. It also provided evidence that delaying intervention while language proficien cy develops is unnecessary. Data showed that all partici pating students made a marked im provement by the end of the intervention, including students who scored negligible in oral language proficiency on the WMLS-R. Out of the ten participants, six had a CALP score of 3, equivalent to limited oral proficiency. Of the six students, three were in Group 1 (higher performing), two were in Group 2 (lower performing), and one was in Group 3 (lowest performing). Furthermore, the two lower performing group s (2 and 3) had at least one member who outscored the higher performing group in level of oral proficiency. Maria, in Group 2,

PAGE 187

187 had a CALP score of 4 (fluent) and Jorge, in Group 3, had a CALP sco re of 3.5 (limited to fluent). This finding shows that reading ability is not entirel y contingent on oral language proficiency. Therefor e, there is no compelling reason for delaying literacy instruction and interventions for struggling readers who are learning to read in their second language (Ehri & Roberts, 2006). Perhaps the UFLI program would have been even more effective if it had been implemented earlier. Yet, providing it in second grade was still early enough to remediate and a ccelerate their learning process. It should be noted, however, that the fact that students with limited oral language proficiency improved as a result of this reading intervention does not preclude the need for English language development. Oral language proficiency is critical for students overall achievement. Developing proficiency in English oral skills should remain a key component of ELL instruction. Effective Instructional Practices for ELLs Approximately 60% of ELLs in this count ry are receiving reading instruction in English (August, 2006; Goldenber g, 2008). Rec ently, the NLP repor ted that students who have no access to native language instruction can succeed if high quality instruction is provided in their second l anguage (Snow, 2006). According to Gersten and Geva (2003) there are six inst ructional strategies that ar e highly effective when working with ELLs: (a) explicit teaching (e.g., model skills and strategies, provides prompts, adjust own use of English during lesson, ma kes relationships overt); (b) systematic instruction in phonemic awareness, lette r-sound correspondence, and decoding; (c) English learning (e.g., uses visuals or m anipulatives to teach content, encourages elaborate responses, uses gestures or facial expressions to clarify meaning); (d) vocabulary development (e.g., teaching vo cabulary prior to and during a lesson,

PAGE 188

188 provides opportunities to speak English, engages students in meaningful interactions about text); (e) interactive teaching (e .g., maintains students engaged during lessons, incorporates students ideas and experiences in to lessons, provides sufficient time for students to respond); and (f) instruction geared toward low performers (e.g., promotes response accuracy, opportunities to practice independent practice, ongoing monitoring of understanding and performance, modifies instruction according to students needs). Furthermore, Helman and Burns (2008) recommend giving students ample opportunities to hear and discuss novel words in context, to connect oral and written forms of words, and to use novel words in students own sentences. The effectiveness of the UFLI tutoring program in promoting the reading skills of ELLs can also be attributed to its compli ance with the aforement ioned practices. The UFLI tutoring program, as it was origin ally designed, addressed many of the instructional practices identified by Gerst en and Geva (2003) as well as Helman and Burns (2008), given that many of these st rategies have also been identified in the literature as effective practices for struggling readers in general (Torgesen, 2002; Vaughn, Gersten, & Chard, 2000). In additi on, the UFLI program was modified to support the language needs of Spanish-speaking ELLs. Based on the findings from the pilot study and recommendations from Linan-Thompson and colleagues (2003), the researcher made sure to distinguish bet ween real and nonsense words, use picture cards, and provide quick definitions of unfamiliar words when needed. In addition, based on Mohr and Mohrs (2007) recommendation, the researcher extended students interactions by accepting phrases or parti al answers and modeling complete sentences with standard grammar.

PAGE 189

189 These practices were consistently impl emented throughout the tutoring session with all groups, and while all students benefitted from these practices, it was observed that students with lower levels of oral language proficiency relied more heavily on them. Anecdotal data showed that Jennifer (CAL P 1 negligible) and Viviana (CALP 2 very limited) had limited E nglish vocabulary and relied heavily on the use of picture cards and quick definitions. These strategies were consistently used with them during word work, reading, and writing activities. Further more, during writing activities, Jennifer, Viviana, and Aldo tended to provide incomplete or grammatically incorrect sentences. In response, the tutor supported them by acc epting their answers, rephrasing their ideas into complete and grammatically correct sentences, and asking them to repeat the rephrased sentence before attempting to write it. Consistency with the Respons e to Intervention Approach Respons e to Intervention (RTI) is a multi-tiered educational approach that promotes early identification and early inte rvention for students who may be at risk for reading difficulties (NRCLD, 2007). This approach provides increasingly intense levels of service (Mellard, 2004) for students who do not respond well to the instruction provided (Vaughn, Wanzek, Woodruff, & Linan-Thompson, 2007). At each level, students benefit from research-based interventions, continuous progress monitoring, and data-driven instruction (NJCLD, 2005). Tier 1 consists of research-based core reading programs that address t he needs of the majority of students. Tier 2 offers more strategic and intense interventions in addi tion to the core reading instruction (approximately 30 minutes more). Tier 3 provides sustained, intensive and scientifically based interventions that are tailored to the s pecific needs of each individual (McMaster, Fuchs, Fuchs, & Campton, 2005; Vaughn et al., 2007).

PAGE 190

190 In this study, the UFLI program was impl emented in addition to the core reading program; thus, it can be considered a Tier 2 intervention. Cons istent with the RTI approach, the UFLI program incorporated continuous progr ess monitoring and datadriven instruction. Based on students running records, the tutor determined the area of focus for word work and the level of book reading for that particular lesson. Furthermore, running records and the use of leveled books assisted the tutor in establishing students current reading level to make decisions about the need for further intervention. For example, students in Gro up 1 were able to catch up to grade level (level 20 books) after 38 sessi ons. At that time, students exited the intervention program and moved back to receiving ju st Tier 1 instruction. The RTI model also allows students who do not respond to Tier 2 intervention to receive a more intensive Tier 3 program. It is estimated that between 5% and 10% of students will require Tier 3 interventions (V aughn et al., 2007). According to Torgesen (2002) available research suggests that mu ltilayered interventions can help reduce the number of students who fail to show adequate reading growth. In this study, for example, one participant (Aldo) remained nonresponsive for the majority of the intervention. As a result of this lack of responsiveness, the researcher decided to provide additional one -on-one instruction for that st udent. These extra 15 minutes, which could be considered Tier 3 instructi on, provided the necessary support to start producing a change in Aldos re ading ability. This change was observed in an increased CPPM and CSPM, as well as an increased accuracy level in book reading. Furthermore, this improvement was sustai ned two weeks after the intervention had ended.

PAGE 191

191 Data obtained through Miriams one-on-one tutoring also supported the use of UFLI as more intensive Tier 3 interventi on for students who are performing below grade level. While groups made substantial progre ss, Miriams advancement was considerably faster than that of t he small-group participant s, including that of students who started at a higher book level. Specifically, Miriam made the equivalent of a year and a halfs progress in only 26 sessions. This finding is crit ical for educators that face the challenge of helping students who are significantly below grade level make quick progress in order catch up to their higher performing peers. It should be noted that, wit hout replication of the one-on-one approach with simila r students, it is impossible to know for certain the extent of the role that UFLI played in Miriams rapid im provement. However, given that other aspects of her schooling were the same as students in the small-group intervention and given the result s of the pilot study and other prior UFLI studies that employed one-on-one instructi on, the assumption that one-on-one tutoring produces more rapid gains than small-group tutoring is a reasonable one. It should be noted that in a true RTI model the Tier 2 and Tier 3 interventions would have been initiated with these strugglin g students much earlier in their school career, as soon as progress-monitoring data indicated that they were not keeping up with their peers. Also, it is unc lear whether the core reading instruction provided at the school would meet the criteria for research-based instru ction. Still, even under less than ideal circumstances for implementation, the UFLI intervention proved to be an effective tool that would be appropriate for Tier 2 or Tier 3 in an RTI model. Implications for Future Research The results of this study showed that t he UFLI tutoring program is an effective English reading interv ention for sec ond-grade Spanish-speaking ELLs, who are

PAGE 192

192 struggling to read. Furthermore, these findings are consistent with previous research demonstrating the efficacy of English inte rventions for struggling English language learners. This study holds several import ant implications for future research. One of the main issues with any single-s ubject design has to do with sample size. Given the limited number of participants in this study, it is important to ensure its external validity; that is, the extent to which the results can be generalized to other participants, places, and conditions (Kennedy 2005). For this purpose, direct and systematic replications of the study are necessary. Dir ect replications should be conducted to make sure that these findings generalize to other students who share similar traits with the participants in this study Also, it is critical to conduct systematic replications by purposely sele cting new but dissimilar subjects, settings, and contexts. In particular, systematic replication will help determine whether the findings of this study extend across different school settings, parti cipants, and tutors. Finally, examining the intervention in an experimental control-group design with a larger sample size would be a logical next step to enhance th e generalizability of findings. Since the study was conducted in only one sc hool located in a rural area of North Florida, it is recommended that the UFLI program be examined in different school settings. The effectiveness of the program may be impacted in some way by the characteristics of each particular school. For example, schools may have different schedules that may not allow for daily 45-mi nute sessions or for extended interventions. Reducing the duration of each session, the frequency of sessions, or the length of the intervention may greatly impact the results.

PAGE 193

193 Similarly, given that the participants in this study were in one school and shared similar traits, it is critical that other studies be conducted to examine how the UFLI program can impact the reading skills of ot her Spanish-speaking ELLs. There might be within-group differences that ca n impact students response to intervention. Therefore, it is important to know how other groups wit h different heritages, socioeconomic status, immigration status (i.e., migrants, U.S. residents, U.S. nat ionals), etc., respond to the UFLI tutoring program. Future research should also examine t he effectiveness of the UFLI program as different types of educators implement it (i .e., mainstream teachers, special education teachers, ESOL teachers, reading specialist, paraprofessionals). According to Kazdin (1982), the characteristics of those implementing a program may influence the intervention effects. For example, in a st udy conducted by Ehri, Dreyer, Flugman, and Gross (2007) about the effectiveness of a tutoring intervention for language minority students, the researchers compared result s across reading specialists, credentialed teachers, and paraprofessionals One of the findings showed that paraprofessionals did not affect students ability to decode pseudowords as well as reading specialists. Ehri and colleagues hypothesize that lack of backg round knowledge in phonics instruction might have been the reason behind their lower per formance. At this point, it is unknown if students response to the intervention would differ if the UFLI program was implemented by other professionals, with different levels of knowledge about reading. Besides replicating the study using single-subject designs, between-group comparisons using quantitative analysis should be conducted to further investigate the effectiveness of the UFLI pr ogram in promoting the readi ng skills of beginning readers

PAGE 194

194 who are learning to read in English. For ex ample, given that students were able to maintain the decoding and word recognition rates two weeks after the intervention ceased, it is also important to conduct fo llow-up studies to determine if gains can be sustained for longer periods of time. In addition, it is important to determine whether supplemental data continues to show an improved reading ability. It is also critical to examine how effect ive the program is wit h other ELLs whose native language is not Spanish. While Spanish is the most common language spoken by ELLs, other languages are also represented in the classroom. Klindler (2002) reports that for the year 2000-2001, approximately 10% of students were ELLs and that the most common languages after Spanish we re Vietnamese (2%), Hmong (1.6%), Cantonese (1%), and Korean (1%). Therefore, a logical next step would be to test the UFLI program with speakers of other languages. Furthermore, since many ELLs are being instructed in English alongside their native-English speaking peers (Goldenberg, 2008), it would be important to examine the impact of the UFLI program when ELLs and native-English speaker s are grouped together. In this study, data showed that all participants improved in decoding and word recognition skills as a result of the interv ention, independent of their initial level of English oral language profici ency. While these results are consistent with previous findings (Gunn et al., 2000, 2002, 2005; LeSaux & Siegel, 2003; Quiroga, LemosBritton, Mostafapour, Abbott, & Berninger 2002; Roberts, 2003; Stuart, 1999; Vaughn et al., 2003), examining the UFLI program with a larger samp le would allow researchers to further examine the role of oral proficiency in students response to intervention. One

PAGE 195

195 option would be to compare across groups that are not only grouped by reading level, but also by oral language proficiency. The recent emphasis on the RTI model highlights the import ance of identifying effective interventions that can address the growing needs of struggling readers, especially treatment resist ers or nonresponders (Torge sen, 2000). Given that the UFLI program was originally designed for one-on-one tutoring and later adapted for small-group intervention, exam ining its effectiveness as part of an RTI model has great implications for the fields of reading and special education. The program should be examined as a multilayered approac h, first as a Tier 2 inte rvention (i.e., small group) and then as a Tier 3 intervention (one-on-one; Torgesen, 2002). Consistent with RTI and the use of more int ensive interventions, it is important to evaluate the impact that group size has on the effectiveness of the UFLI program. Given the rapid response of the student receivi ng one-on-one tutoring compared to students receiving small-group tutoring (almost half the number of sessions as Group 2), it is important to do a cost-benefit analysis. In a cost benefit analysis, researchers can directly compare student achievement in diff erent size groups and the cost of providing interventions in smaller groups or one-on-one. RTI also highlights the importance of preventing severe reading difficulties by providing early interventions; thus, the UF LI program should also be examined with younger ELLs, starting in first grade. This is pivotal given that first grade struggling students tend to remain poor readers in later grades (Juel, 1988). Furthermore, it is well known that reading problems in older students are more difficult to remediate even when intensive interventions are provided (Torgesen et al., 2001).

PAGE 196

196 Since the UFLI program was originally designed with the intent of training preservice teachers to become better reading teachers, it would be valuable to investigate the effect that UFLI training and UFLI implementation would have on teachers knowledge about reading. Also, it would be helpful to determine if it has an impact on teachers instructional practi ces in the general classroom. In particular, an examination of the effects of the modified version of UF LI implemented in this study on teachers knowledge of effective reading interv ention for ELLs would be warranted. It is also crucial that studies be conduct ed to examine the most efficient ways to expand the implementation of the UFLI program at the school, district, state, and eventually national level. According to Klingner, Ahwee, Pilonieta, and Menendez (2003), educators face many barriers that can keep them from consistently implementing research-based interventions. So me of these barriers include lack of time, lack of support from administrators, mismat ch between the interventions, other methods mandated by the district, among ot hers. Therefore, future research should also focus on identifying what works best, for whom, and under what conditions (McDonald, Keesler, Kauffman, & Schneider, 2006) so that any scaling-up efforts become feasible. Implications for Practice The findings from this study have signific ant implications for educators working with Spanish-speaking ELLs. The UFLI tutori ng program is an effective English reading intervention that helps develop the skills of beginning readers who are struggling to read in their second language. Teachers can impl ement this small-group intervention to supplement their core reading program and address the needs of many students. It is also important that teachers of E LLs provide interventions for beginning readers that emphasize word reading ski lls, particularly phonemic awareness and

PAGE 197

197 decoding. Providing explicit and systematic instruction in t hese two skills are necessary for the development of students reading fluency. Interventions should also provide daily opportunities for ample practice reading con nected text at an appropriate reading level (Allington, 1983) in order to develop reading automaticity. The fact that all participants benefitted fr om the intervention regardless of their initial level of oral language proficiency indicates t hat educators can provide early interventions to those students who are experiencing difficulties lear ning to read. This has important implications for practice given the tendency for educators to postpone reading interventions until students develop su fficient oral language proficiency. In accordance with Ehri and Roberts (2006) sugges tion, there is no compelling reason for delaying reading interventions for ELLs. Students can learn to read while they continue developing English oral proficiency. Another implication relates to gro uping practices and language support. Since students reading improvement was closely related to their init ial reading level and not to oral proficiency level, t eachers can group students with diffe rent levels of English proficiency to receive intervention. In doing so, it is also impor tant to incorporate instructional practices that support the varying language needs of ELLs so that everyone attains the maximum benefit. Findings from this study have particular relevance for schools, given the recent emphasis on RTI. Results show that UFLI is an effective small-group tutoring program that can be used as a Tier 2 interventi on for Spanish-speaking ELLs who are not responding adequately to Tier 1 instruction. Its emphasis on progress monitoring and data-driven instruction makes this interv ention optimal for RTI. Furthermore, given

PAGE 198

198 Aldos response to additional instructional ti me following the UFLI protocol, teachers can consider using the UFLI program as a Tier 3 intervention, as well. This suggestion is also supported by previous data from the pilo t study that examined the UFLI program as a one-on-one tutoring program for Spanish-speak ing ELLs. In that study, three firstgrade students showed marked improvement on decoding and leveled book reading as a response to intervention. Conclusion Children in the United States whose native language is not Englis h are at particular risk for reading diffi culties (Snow et al., 1998). While the number of language minority students in the U.S. is growing (Klingner, Artiles, & Barletta, 2006), access to effective bilingual programs (August, 2006; Gomez-Bellenge, Chen, & Schulz, 2008) is limited. Up to 60% of ELLs in the nation ar e receiving reading instruction in English (August, 2006; Goldenberg, 2008). Because the level of achievement for ELLs is drastically below their m onolingual peers, there is a need for effective reading interventions at an early stage (Cartl edge et al., 2009; Denton & Mathes, 2003). Unfortunately, many schools do not provide early interventions because they believe that literacy instruction should be delayed until ELLs develop sufficient oral language proficiency (Ehri & Roberts, 2006; Tabors & Snow, 2002). Fort unately, there is evidence that ELLs who participate in English readi ng interventions originally designed for monolingual students respond positively despite having limited levels of oral proficiency (Gunn et al., 2000, 2002, 2005; LeSaux & Siegel, 2003; Quiroga, Lemos-Britton, Mostafapour, Abbott, & Berninger, 2002; Roberts, 2003; Stuart, 1999). These initial findings highli ght the importance of identifying English interventions that have proven to be effective with str uggling monolingual students and evaluate how

PAGE 199

199 effective they are in promoting the reading skills of developing ELLs. For this purpose, a multiple-baseline design across three groups was used to evaluate the effectiveness of the University of Florida Literacy Initiati ve (UFLI) small-group tutoring program for Spanish-speaking ELLs who were strugg ling with reading. The results of this investigation demonstrated that the UFLI tu toring program is an effective reading intervention for second-grade, Spanish-speaking ELLs who are struggling to read in English. Providing an intervent ion that promotes word re ading skills of young learners effectively improves the reading skills of EL Ls. Furthermore, the program proved to be effective despite students varying levels of English oral language proficiency. These findings contribute to the growing literatur e on the effectiveness of English reading intervention for struggling ELLs of varying levels of language proficiency. As Ehri and Roberts (2006) state, there is no compelling reason for postponing reading interventions for struggling ELLs. This study demonstrates that well-des igned intervention can support the reading development of struggling students who are also ELLs. Given the evidence generated by this study and others like it, it is essentia l that schools use what is known to provide appropriate intervention for students with r eading and language difficulties. Making this happen may require reallocation of resources, rethinking of intervention models, or additional professional developm ent for teachers, but the ne ed is too great to ignore.

PAGE 200

200 APPENDIX A IRB DOCUMENTATION

PAGE 201

201

PAGE 202

202

PAGE 203

203

PAGE 204

204

PAGE 205

205

PAGE 206

206

PAGE 207

207

PAGE 208

208

PAGE 209

209

PAGE 210

210 Childs Intervention Assent Form (English and Spanish versions) Hi, my name is Stephanie Arriaza, and I'm from the University. I would like to do some reading activities with you and work with sounds and letters. We will be working together for a few weeks. You can earn sti ckers for participating and you can stop working at any time. I also want you to k now that whatever you decide, this will not affect your grades. Would you like to work with me? Hola, mi nombre es Stephanie Arriaza y vengo de la Universidad. Me gustara hacer algunas actividades de lectura contigo y tr abajar con letras y sus sonidos. Estaremos trabajando juntos por algunas semanas. Tendrs la oportunidad de ganar calcomanas y podrs parar de trabajar cuando tu lo desees. Tambin quiero que sepas que si decides dejar de participar esto no afecta ra tus calificaciones. Quieres trabajar conmigo? Childs Assessment Assent Form (English and Spanish versions) Hi, my name is Stephanie Arriaza, and I am from the University. We will work together for about an hour today and maybe an hour some other day. We will do some reading activities and respond to some questions in English and Spanish. You can earn stickers for participating and you can stop working at any time. I also want you to know that you can stop working at any time. I also want you to know that whatever you decide, this will not affect your grades. Would you like to work with me? Hola, mi nombre es Stephanie Arriaza y vengo de la Universidad. Vamos a trabajar juntos alrededor de una hora al da de hoy y posiblemente una hora otra da. Vamos a hacer algunas actividades de lectura y contestaras algunas preguntas en Ingles y Espaol. Tendrs la opor tunidad de ganar calcomanas y podrs parar de trabajar cuando tu lo desees. Tambin quiero que seas que si decides dejar de trabajar, esto no afectara tus calificaciones Quieres trabajar conmigo?

PAGE 211

211 APPENDIX B HOME LANGUAGE SURVEY AND REMINDER LETTERS Hom e Language Survey Childs name: _________________________ Childs age: ___________________________ Your name: ___________________________ Todays date: __________________________ Household Demographics What is your relationship to the child? mother father other _________________ What is your ethnicity/nationality? ________________________________________________________ How many people live in your household? What are their ages? ___________________ How long have you lived in your current home? ______________________________________________ How many years has your child been in school in this country? __________________________________ What is your highest level of education? Did not complete high school Completed high school or equivalent Community college Obtained undergraduate degree Other Home Language Practices How many languages are spoken at home? ___________ Specify which language are spoken at home ________________________________________________ Which language do you use to interact with your child? _______________________________________ Do you use different languages for different purpo ses? Yes ____ No ____. Specify ________________ ____________________________________________________________________________________ Do you consider yourself a fluent speaker of English? Yes ___ No ___ Is there anything else about your childs language or literacy development that youd like to add?

PAGE 212

212 Lenguaje en el Hogar Encuesta Nombre del nio(a):_____________________ Edad del nio(a):________________________ Nombre del Padre/Madre:_________________ Fecha de hoy: __________________________ Informacin General Cul es su parentesco con su hijo/hija? madre padre otro__________________ Cul es su nacionalidad? ________________________________________________________________ Cuantas personas viven en su hogar? Que edades tienen? ____________________ Cuanto tiempo ha vivido en su actual casa? _________________________________________________ Cuanto aos ha estado su hijo en la escuel a en este pas? _______ ________________________________ Cul es el nivel ms alto de educacin que usted tiene? No complete el bachillerato/preparatoria/high school Complete el bachillerato/preparatoria/high school Estudio Tcnico Licenciatura Otro Actividades de Lenguaje en el Hogar Cuantos idiomas o lenguajes se habla en su hogar? ___________ Especifique que idiomas o lenguajes se hablan en su hogar______________________________________ Con que idioma o lenguaje usted se comunica con su hijo/hija?__________________________________ Usa usted diferentes lenguajes para dife rentes propsitos? Si ____ No ____. Especifique___________________________________________________________________________ Considera usted que habla el idioma Ingles? Si ___ No ___ Hay algo ms que usted quisiera agregar acerca del desarrollo del lenguaje y escritura de su hijo o hija?

PAGE 213

213

PAGE 214

214

PAGE 215

215 APPENDIX C TEACHERS READING ABILITY RATING SCALE (RARS) READING ABILITY RATING SCALE Student Teacher Date Please rate your students ability to perform th e following reading tasks by circling one of the numbers on the scale (1 = very weak 4 = very strong) for each item. Phonological Awareness very weak.....very strong Detect individual sounds in words 1 2 3 4 Segment sounds in words 1 2 3 4 Blend sounds to form words 1 2 3 4 Phonics very weak.....very strong Decode simple words (e.g., cat, boat) 1 2 3 4 Decode more complex words (e.g., multisyllable) 1 2 3 4 Spell simple words 1 2 3 4 Spell more complex words 1 2 3 4 Read high frequency words by sight 1 2 3 4 Use context clues to figure out words 1 2 3 4 Fluency very weak.....very strong Read words accurately in connected text 1 2 3 4 Read words quickly in connected text. 1 2 3 4 Read text with inflection and expression. 1 2 3 4 Read connected text on grade level. 1 2 3 4 Self-monitoring to check for accuracy 1 2 3 4 Self-correcting reading errors 1 2 3 4 Comprehension very weak.....very strong Make predictions about a story. 1 2 3 4 Make personal connections to the story. 1 2 3 4 Summarize text. 1 2 3 4 Engage in discussions of a story or text read. 1 2 3 4 Comprehend grade level texts. 1 2 3 4 Vocabulary very weak.....very strong Comprehend English words while reading 1 2 3 4 Use English vocabulary in speech 1 2 3 4 Use English vocabulary in writing 1 2 3 4 Classroom behaviors very weak.....very strong Participate in class 1 2 3 4 Use English to communicate in class 1 2 3 4 Motivation to read 1 2 3 4 Additional comments about your studen ts reading or language abilities:

PAGE 216

216 APPENDIX D UFLI SMALL GROUP SESSION GUIDE UFLI Small-Group Session Guide STEP 1: GAINING FLUENCY & MEASURING PROGRESS Select and read 1-3 familiar books Have students pair-read a famili ar book while taking running record Present book from prev ious session to student Take running record while child reads Provide feedback regar ding self-corrections you observed STEP 2: WORD WORK WITH MANIPULATIVE LETTERS Hand out magnetic boards and magn etic letters to each student Do word work with familiar words STEP 3: INTRODUCING & READING A NEW BOOK Introduce the new book Coach students through the text Discuss the story Work with new words Record strategies used and focus of word work STEP 4: WRITING FOR READING Elicit and record sentences Ask each student to write their own sentences (check for spelling) Work on each word, while coaching students

PAGE 217

217 APPENDIX E UFLI SMALL-GROUP SESSION NOTES Date: __________________ Session time ________ to _______New book level:_______ Absentees: _____________ Step 1: Gaining Fluency & Measuring Progress Familiar Book Level/Title ________________________________________ ___________________________________ Running Record book title/level ___________________________________________________ Running Record Student ___________________________________ Accuracy _________% Observations Step 2: Word Work Areas of focus for word work: words decoded words encoded ELL accommodations: Step 3: Introducing and Reading a New Book Book Level/Title Strategies introduced/practiced: Working with new words: ELL accommodations:

PAGE 218

218 S tep 4: Writing for Reading Student Sentence Student Sentence Student Sentence Student Sentence ELL accommodations: S ession Observations:

PAGE 219

219 APPENDIX F PSEUDOWORD PROBES jorp dasp tig croob faf gake sape tham taz stup doil vop fike korsh dod nuz lurt jath rast keef zeck beash zid joad shan sev keb snill yabe tusp huz bipe buk bim doon heesh yud meab vift jit waib mip zow fump rofe weff trin vake pef naz

PAGE 220

220 marp nep parb zeck naz zow huz waib jorp hup rofe meab moz rast slig puzz heesh zid shog tusp crin snop mox meez lav rox trin lurt sev tep fump sarb fent tham wut cham yabe sape doil beash keef tig stup dod belp yin jath lish keb cack

PAGE 221

221 rast waf cham nuz marp rell sarb pash keef pef flaf bipe doon puzz dod sev zow keb wut vame doil jorp joad faf yin lav yabe delk dush buk shog zeck tusp mip tep plup tig fump vup parb tane voy sape fent baf shan vake dosh gerg fike

PAGE 222

222 mox doon vake belp doil groob korsh drup meab cack sarb jath tane tusp shog delk keef tep fump zeck fent beash vame joad gerg waib dasp pash jorp mip snop cag mox leat trin zeck plup beal lurt fike huz tham vope yabe parb snill bipe nep wope dush

PAGE 223

223 cag jorp wope flaf korsh zeck tham boz bipe jath croob lish snop vift beal cham tusp beash belp sarb dasp vake yud voy doil plup vame parb grup waib snill bipe lurt stup fike gerg shog keb dasp belp wope dreck rast yabe goft trin rofe bipe tane yud

PAGE 224

224 dreck keef boz fent rofe lav wope sarb zid trin plup shog weff jorp pef wut cham snop flaf buk croob zot meab tane heesh beal gerg jit tep vope beash tusp vop zow tham marp nep tig sev waib rox korsh slig baf voy taz stup parb dosh puzz

PAGE 225

225 b af fike jath meez waf d reck joad cack nuz crin l ish yin cag stup tane b uk moz huz dasp flaf r ast rofe mox parb snill p lup tham tig vame zot c roob shog nep hup rox c ham gerg weff korsh doon v up sev fay gake sape m ip beal beash snop marp

PAGE 226

226 cham sarb lish vake mox delk veek rofe fent boz keb jit slig sape doil buk plup jath dasp cag croob hup nep baf zow puzz doon vame bim pef beal yud stup meez flaf weff dush voy dreck snop parb goft shan joad snill tusp rox vift zeck nuz

PAGE 227

227 jath jorp drup vift lish sape rell mox nep gake snill flaf lurt meez waib bim bipe vame marp heesh moz taz zid sarb baf boz tig zeck voy wope dasp joad huz pef rofe croob delk yud mip fike cag korsh goft belp tane beal keef plup naz yabe

PAGE 228

228 pef dush tane yin snop jit trin weff vup grup doon gerg zeck slig marp jath cack rast vame beal fent mox croob parb korsh lurt dreck stup flaf huz doil fay rox gake dod drup faf yabe goft vope nuz waf plup heesh lav moz sev veek snill tig

PAGE 229

229 goft puzz parb cack delk doon trin fay mox vope vift lish joad dosh waf rofe lurt vame yin korsh shog taz drup bim voy keef tusp yabe tep gerg vop bipe plup keb baf rast lav sarb mip shan sape vake stup waib leat nep rell tig gake grup

PAGE 230

230 sarb heesh lurt grup dreck zid shan tep doil faf rox pash pef tig voy lav goft rell baf beash tane parb zow beal leat sape plup shog moz vame nuz slig nep korsh drup flaf yabe wope yin dod vake waib wut belp keb jath dosh doon trin sev

PAGE 231

231 buk yin tane croob korsh doon parb dosh nep fump plup vup weff marp vope dush yud dod fike leat sev dasp crin grup keb moz gake delk snop slig goft veek bim rast drup waib fent gerg rox beash voy beal tep tusp vop pash lish doil fay zid

PAGE 232

232 snill puzz vake vop dosh plup naz meab snop rox keef jit huz taz gake hup boz pef leat dush korsh flaf shan tig pash waib cham mox stup zot slig joad beash tusp yabe gerg doil yud sev bipe zid nuz zeck grup rell shog bim fay wut dreck

PAGE 233

233 weff sarb lurt snop boz vift bim trin pef jit tig doon nep vake vop marp jath meez stup shog voy joad dod cack zow wut tep fike delk gerg dasp hup slig snill huz drup fay lish vup puzz cham mip vope zeck naz mox goft doil faf lav

PAGE 234

234 APPENDIX G SIGHT WORD PROBES are big first know some their our because any going had saw where right again may did take upon very why this does could ask by yes always your read stop under please must did came all good she there under help gave many her ate here new saw they

PAGE 235

235 live had these like four wash going both after give would their which please your come one off first myself about some done around white many never keep thank her how before use them over because pull right found best would think have funny once what its came just under

PAGE 236

236 look funny live its could work these green bring after many read because open gave would give very found had let once over always one does together wish four about upon them own under done stop cold wash use ask goes white were again pretty take round what her would

PAGE 237

237 upon goes would those live give pull your both had off some sleep why just one dont jump work come very funny let buy think again first have once going green pretty them got wish sit could brown found much cold would ask best been keep what like know write

PAGE 238

238 round their much because like fast were have had upon wash any those walk off which again buy open myself give over take sing under let wish use before been why got funny these work five stop how pull where right pretty both together about does done best sleep keep

PAGE 239

239 her could because pull where best look goes first funny cold made take green dont have always much their again from what right let pretty please done gave there like around live think open thank been upon were which found walk any many had call after some once them jump

PAGE 240

240 does your around write those together look found wash ate every keep any why brown had five cold their much think before own away which upon round going gave work white bring where sing what jump after many made ask pull its got were buy four them very read know

PAGE 241

241 its open buy made gave four again which from around goes look work would your sleep before off what many never over best myself five very their funny right cold white give one walk first please going bring pretty where dont them got how pull always have sing once these

PAGE 242

242 round had work what look use wish those its call sing these five together gave green many off walk goes wash pretty going bring open found before ate why first her best some myself were under thank came cold does upon every there ask know just got buy brown always

PAGE 243

243 look came always going their open once best keep never funny please live why gave from them those first wash bring these call sleep which jump ate does let come read stop thank walk your her pretty much green brown got cold just where its would been work because dont

PAGE 244

244 green live them buy after work four done wash their wish round from five about stop please own have made does goes one once always around cold these going jump were never both what know ask ate bring upon sit come came sing open off gave could pull why because

PAGE 245

245 know their best stop away been together does never thank like read after open both once dont let think around before her why found your going buy pretty write these four under brown look call any give about please upon ate work over from much right very always pull myself

PAGE 246

246 stop five over could open have sing goes around just read from which sleep keep would write let her much give think round wash gave bring call own funny about upon these them know there pretty its had sit every does any myself got before wish both very brown never

PAGE 247

247 could much both just jump white funny how got what come these buy any why were ate always write away after take sleep stop came before ask read goes going work think pull look four every once off cold its use where please done your wish their upon thank first

PAGE 248

248 some about these open why going green let found could bring round just off upon right after again goes which would had never myself always many cold write funny together how please gave there jump every those from live its five got very once dont your their work before read

PAGE 249

249 APPENDIX H TREATMENT INTEGRITY CHECKLIST Treatment Integrity Checklist Group No._____________________________ Date: _________________________________ Observer ______________________________ School: _______________________________ STEP 1: GAINING FLUENCY & MEASURING PROGRESS YES NO NA The tutor provided appropriate books (within a 2-level range from the current new book level; read with 90% accuracy or better). Students read familiar books. Students were paired up for read ing while tutor took a running record Tutor completed a running record with one student using yesterdays new book. Tutor analyzed students strategies/errors. Tutor determined book level based on accuracy. Tutor provided feedback about strategies used by the child. STEP 2: WORD WORK WI TH MANIPULATIVE LETTERS YES NO NA Tutor handed out a magnetic board and ma gnetic letters to each student Tutor led students in word work with familiar words/spelling patterns Decoded words at the onset/rime level Encoded words at the onset/rime level Decoded words at the phoneme level Encoded words at the phoneme level Used real words Used nonsense words Tutor helped students locate word(s) from their word work in the text. Tutor identified which words were real words and which words were nonsense words. Tutor made comparisons between English and Spanish words and sounds. Tutor used picture cards to illustrate unfamiliar word meanings.

PAGE 250

250 STEP 3: INTRODUCING & READING A NEW BOOK YES NO NA The tutor introduced the book and discussed the story. Tutor led students in discussion about illustrations. Tutor promoted student involvement in discussion. Tutor pointed out repetitive language. Tutor used vocabulary from story. Tutor used picture cards or gave quick definitions of unfamiliar words. Tutor helped students make predictions about the story. The tutor coached studen ts through the book. Tutor prompted strategy use. Tutor encouraged students to re -read sentences after decoding a word. Tutor used word work (with manipulat ive letters or dry-erase boards) to introduce new words or spelling patterns. STEP 4: WRITING FOR READING YES NO NA Tutor elicited a sentence from each student based on the new book. Tutor recorded each sentence. Tutor asked each student to write their own sentence on the writing book Student coached each student Sight words were practiced appr opriately (1-3 words) Elkonin boxes were used appropriately (2-4 words) Students read and reread their sentence as each word was added. PROCEDURES YES NO NA Tutor worked with every student in the group. Tutor started the session on time Session did not exceed 45 minutes. Tutor had all the materials ready ahead of time. Tutor handled materials with ease. Observations:

PAGE 251

251 APPENDIX I SOCIAL VALIDITY INSTRUMENTS Student Interview 1. How do you feel about participa ting in the tutoring lessons? 2. Do you think that these lessons helped to become a better reader? How? 3. What part of the lesson was your favorite? Why? 4. What part of the lesson was your least favorite? Why? 5. How do you feel about worki ng in small groups? 6. Will you use what you learned from the lesson in your reading? 7. Do you think other students should pa rticipate in the tutoring lessons?

PAGE 252

252 Teachers Social Validity Questionnaire Thank you for participating in th e evaluation of a UFLI tutoring lesson. Your opinion about the tutoring program and procedures will help us improve the lessons and procedures followed with English language learners (ELLs). Your ident ity will be kept confident ial. If you have any questions, please cont act Stephanie Arriaza at (352) 870-7056 or send an email to satienza@ufl.edu As you observe the tutoring lesson, please use the lesson hand out as a guide and respond to the following questions. Repeated Reading of Familiar Books 1. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 2. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 3. Currently, do you use a similar strategy when you work with English language learners? Yes No 4. Would you be likely to use a strategy like this one with ELLs during your reading instruction? Please explain. Yes No Progress monitoring 5. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 6. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 7. Currently, do you use a similar strategy when you work with English language learners? Yes No 8. Would you be likely to use a strategy like this on e with ELLs during your read ing instruction? Please explain. Yes No

PAGE 253

253 Word work with magnetic letters 9. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 10. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 11. Currently, do you use a similar strategy when you work with English language learners? Yes No 12. Would you be likely to use a strategy like this on e with ELLs during your read ing instruction? Please explain. Yes No Reading a new book 9. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 10. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 11. Currently, do you use a similar strategy when you work with English language learners? Yes No 12. Would you be likely to use a strategy like this on e with ELLs during your read ing instruction? Please explain. Yes No

PAGE 254

254 Writing for Reading 13. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 14. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 15. Currently, do you use a similar strategy when you work with English language learners? Yes No 16. Would you be likely to use a strategy like this on e with ELLs during your read ing instruction? Please explain. Yes No Overall tutoring lesson 17. Do you feel that the tutoring lesson can help ELLs develop reading sk ills? Why or why not? Yes No 18. Do you believe that this lesson is easy to implement? Why or why not? Yes No 19. Would you be willing to implemen t the lessons in your classroom? Yes No 20. Would you be willing to implement com ponents of the lesson in your classroom? Yes No 21. Would you be interested in learning how to implement this tutoring program? Yes No 22. Do you think other teachers of ELLs might be in terested in learning to implement this program? Yes No

PAGE 255

255 APPENDIX J UFLI INDIVIDUAL TUTO RING SESSION GUIDE

PAGE 256

256 LIST OF REFERENCES Adams, M. J. (1990). Learning to read: Thinking and lear ning about print. Cambridge, MA: MIT Press. Adams, M. J., Foorman, B. R., Lundberg, I., & Beeler, T. (1998). Phonemic awareness in young children: A classroom curriculum. Baltimore, MD: Brookes Publishing. Al Otaiba, S. (2005). How effective is code-bas ed reading tutoring in English for English language learners and preser vice teacher-tutors? Remedial and Special Education, 26 (4), 245-254. Allington, R. L. (1983). The reading instru ction provided readers of differing reading abilities. The Elementary School Journal, 83(5), 548-559. Altmann, L. J. P., Lombardino, L. J., & Pu ranik, C. (2008). Sentence production in students with dyslexia. International Journal of Language and Communication Disorders, 43 (1), 55-76. doi :10.1080/13682820701284522 Alvarado, C. G., Ruef, M. L., & Schrank, F. A. (2005). Woodcock-Muoz Language SurveyRevised: Comprehensive Manual. Itasca, IL: Riverside Publishing. Anderson, R. C., Hiebert, E. H., Scott, J. A., & Wilkinson, I.A. (1985). Becoming a nation of readers: The report of the commission on reading. Washington, DC: The National Institute of Education. Antunez, B. (2002). Implementing Reading First with English language learners. Directions in Language and Education, 15, 1-12. Ashdown, J., & Simic, O. (2000). Is early intervention effective for English language learners? Evidence from Reading Recovery. Literacy Teaching and Learning, 5 (1), 27-42. August, D. (2006). Demographic overview. In D. August & T. Shanahan (Eds.), Developing literacy in second-language lear ners: Report of the national literacy panel on language-minority children and youth (pp. 43-49). Mahwah, NJ: Lawrence Erlbaum Associates. August, D., Carlo, M., Dressler C., & Snow, C. (2005). The cr itical role of vocabulary development for English language learners. Learning Disabilities Research & Practice, 20 (1), 50-57. August, D., & Hakuta, K., (1997). Improving schooling for language minority children: A research agenda. Washington, DC: National Research Council. August, D., & Shanahan, T. ( 2006). Developing literacy in second-language learners: Report of the national literacy panel on language-minority children and youth. Mahwah, NJ: Lawrence Erlbaum Associates.

PAGE 257

257 Beck, I. L., Farr, R. C., St rickland, D. S., Amor, A. F., Hudson, R. F., McKeown, M. G.,Washington, J. A. (2007). StoryTown (K to 6th grade Core Reading Program). Orlando, FL: Harcourt School Publishers. Bear, D. R., Helman, L., Templeton, S. Invernizzi, M., & Johnston, F. (2007). Words their way with English learners: Word study for phonics, vocabulary, and spelling instruction. Upper Saddle River, NJ: Pearson. Berninger, V. W., Abbott, R. D., Vermeulen, K., Ogier, S ., Brooksher, R., Zook, D., & Lemos, Z. (2002). Comparison of fast er and slower responders to early intervention in reading: differentiating features of thei r language profiles. Learning Disabilities Quarterly, 25 (1), 59-76. Borko, H. (2004). Professiona l development and teacher l earning: Mapping the terrain. Educational Researcher, 33 (8), 3-15. Bravo, M. A., Hiebert, E. H., & Pearson, P. D. (2006). Tapping the linguistic resources of Spanish-English bilinguals: The role of cognates in science. In R. K. Wagner, A. E. Muse, & K. R. Tannenbaum (Eds.), Vocabulary acquisition: Implications for reading comprehension (pp. 140-154). New York, NY: The Guilford Press. Campbell, M. L. (1984). Corrective Reading program evaluated with secondary students in San Diego. Direct Instruction News, 7 (4), 15-17. Cartledge, G., Gardner, R., & Ford, D. Y. (2009). Diverse learners with exce ptionalities: Culturally responsive teaching in the inclusive classroom. Upper Saddle River, NJ: Pearson. Cardoso-Martins, C., Rodrigues, L. A., & Ehri, L. C. (2003) Place of environmental print in reading development: Evi dence from nonliterate adults. Scientific Study of Reading, 7 (4), 335-355. doi: 10.1207/S1532799XSSR0704_2 Carlo, M. S., August, D., Maclaughlin, B., Snow, C. E., Dressler, C., Lippman, D. N.,White, C. (2004). Closing the gap: Addressing the vocabulary needs of English-language learners in bilingual and mainstream classrooms. Reading Research Quarterly, 39 (2), 188-215. doi:10.1598/RRQ.39.2.3 Carnine, D., Silbert, J., & Kameenui, E. (1990). Direct instruction reading. Columbus, OH: Merrill Publishing Company. Center,Y., & Wheldall, K. ( 1992). Evaluating the effectivene ss of Reading Recovery: A critique. Educational Psychology, 12 (3/4), 263275. Center, Y., Wheldall, K., Freeman, L., Outhred, L., & McNaught, M. (1995). An evaluation of Reading Recovery. Reading Research Quarterly, 39 (2), 240-263. Chall, J. (1983). Stages of reading development. New York, NY: McGraw-Hill.

PAGE 258

258 Chhabra, V., & McCardle, P. (2004). Contributions to evidence-based research. In P. McCardle & V. Chhabra (Eds.), The voice of evidence (pp. 3-11) Baltimore, MA: Paul H. Brookes Publishing. Chiappe, P., Siegel, L. S., & Gottardo, A. (2002). Reading related skills in kindergartners from diverse linguistic backgrounds. Applied Psycholinguistics, 23, 95-116. doi:10.101 7.S014271640200005X Chiappe, P., Siegel, L. S., & Wade-Woolley, L. (2002). Linguistic diversity and the development of reading skills. Scientific Studies of Reading, 6 (4), 369-400. doi:10.1207/S1532799XSSR0604_04 Clay, M. M. (1972). Sands: The concepts about print tests. Auckland, New Zealand: Heinemann. Clay, M. M. (1985). The early detection of reading difficulties. Auckland: Heinemann. Clay, M. M. (1987). Reading Recovery: S ystemic adaptations to an educational innovation. New Zealand Journal of Educational Studies, 22, 35-58. Clay, M. M. (1993a). An observation survey of early literacy achievement. Auckland, New Zealand: Heinemann. Clay, M. M. (1993b). Reading Recovery: A gui de for teachers in training. Portsmouth, NH: Heinemann. Collier, V. P. (1989) How long? A synthesis of resear ch on academic achievement in a second language. TESOL Quarterly, 23 (3), 509-531. Conners, C. K., Sitarenios, G., Parker, J. D., & Epstein, J. N. (1998). Revision and restandardization of the Conners Teacher Rating Scale (CTRS-R): Factor structure, reliability, and criterion validity. Journal of Abnormal Child Psychology, 26(4), 279-291. doi:10.1023/A:1022606501530 Cummins, J. (1981). Age on arrival and immigr ant second language learning in Canada: A reassessment. Applied Linguistics, 2 (2), 131-149. doi:10.1093/applin/II.2.132 DeFord, D. E. (1995). Early wr iting: Teachers and children in Reading Recovery. In S. Swartz & A. Klein (Eds.), Research in Reading Recovery (pp. 148-172). Portsmouth, NH: Heinemann. Denton, C. A., Anthony, J. L., Parker, R., & Hasbrouck, J. E. (2004). Effects of two tutoring programs on the English readi ng development of Spanish-English bilingual students. The Elementary School Journal, 104 (4), 289-305. Denton, C. A., Ciancio, D. J., & Fletcher, J. M. (2006). Validity, reli ability, and utility of the Observation Survey of Early Literacy Achievement. Reading Research Quarterly, 41 (1), 8-34.

PAGE 259

259 Denton, C. A., & Mathes, P. G. (2003). Interventions for struggling readers: Possibilities and challenges. In B. R. Foorman (Ed.), Preventing and remediating reading difficulties: Bringing science to scale (pp. 229-252). Timonium, MD: York Press. Ehri, L. C. (1989). The development of s pelling knowledge and its role in reading acquisition and reading disability. Journal of Learning Disabilities, 22 (6), 356-365. Ehri, L. C. (1991). Development of the ability to read words. In R. Barr, M. Kamil, P. Mosenthal, & P. Pearson (Eds.), Handbook of reading research (Vol. II, pp. 383417). New York, NY: Longman. Ehri, L. C. (1997). Learning to read and learning to spell are one and the same, almost. In C. A. Perfetti, L. Rieben, & M. Fayol (Eds.), Learning to spell: Research, theory, and practice across languages (pp. 237-269). Mahwah, NJ: Erlbaum. Ehri, L. C. (1998). Grapheme-phoneme knowledge is essential for learning to read words in English. In J. Me tsala & L. C. Ehri (Eds.), Word recognition in beginning literacy (pp. 3-40). Mahwah, NJ: Erlbaum. Ehri, L. C. (2002). Phases of acquisition in learning to read words and implications for teaching. In R. Stainthor p & P. Tomlinson (Eds.), Learning and teaching reading (pp. 7-28). London, England: British Journal of Educational Psychology Monograph Series II. Ehri, L. C.(2004). Teaching phonemic awar eness and phonics: An explanation of the national reading panel metaanalysis. In P. McCardle & V. Chhabra (Eds.), The voice of evidence in reading research (pp. 153-186). Baltimore, MA: Brookes Publishing Ehri, L.C., (2005a). Development of sight word reading: Phases and findings. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 135-154). Malden, MA: Blackwell Publishing. Ehri, L. C. (2005b). Learning to read words: Theory, findings, and issues. Scientific Studies of Reading, 9 (2), 167-188. doi: 10.1207/s1532799xssr0902_4 Ehri, L. C., Dreyer, L. G., Flugman, B., & Gross, A. (2007). Reading Rescue: An effective tutoring intervention model for language minority students who are struggling readers in first grade. American Educational Research Journal, 44 (2), 414-448. doi:10. 3102/0002831207302175 Ehri, L. C., & McCormick, S. (1998). Phases of word learning: Implications for instruction with delayed and disabled readers. Reading & Writing Quarterly, 14(2), 135-163. doi :10.1080/1057356980140202

PAGE 260

260 Ehri, L. C., & Roberts, T. ( 2006). The roots of learning to read and write: Acquisition of letters and phonemic awareness. In D. K. Dickinson & S. B. Neuman (Eds.), Handbook of early literacy research (Vol. II, pp. 113-131). New York, NY: The Guilford Press. Ehri, L. C., & Snowling, M. (2004). Developmental variation in word recognition. In A. Stone, E. Silliman, B. Ehren, & K. Apel (Eds.), Handbook of language and literacy: Development and disorders (pp. 433-460). New York: The Guilford Press. Ehri, L. C., & Wilce, L., ( 1983). Development of word identification speed in skilled and less skilled beginning readers. Journal of Educational Psychology, 75 (1), 3-18. doi:10.1037/00220663.75.1.3 Ehri, L. C., & Wilce, L. (1985). Movement into reading: Is the first stage of printed word learning visual or phonetic? Reading Research Quarterly, 20 (2), 163-179. Engelmann, S., & Bruner, E. (1988). Reading Mastery. Chicago, IL: Science Research Associates. Engelmann, S., Carnine, L., & Johnson, G. (1988). Corrective Reading: Word-attack basics. Chicago: Science Research Associates. Engelmann, S., & Engelmann, K. (2004). Impediments to scaling up effective comprehensive school reform models. In T. Glennan, S. Bodilly, J. Galegher, & K. Kerr (Eds.), Expanding the reach of education reforms: Perspectives from leaders in the scale-up of educational interventions (pp. 107-133). Santa Monica, CA: RAND. Escamilla, K. (1994). Descubriendo La Lectu ra: Un programa en Espaol de lectoescritura para las primeras etapas. Literacy, Teaching and Learning, 1(1), 72-85. Fitzgerald, J. (1995). English-as-a-s econd-language learners cognitive reading processes: A review of the research in the United States. Review of Educational Research, 65 (2), 145-190. Florida Center for R eading Research (2006). DIBELS risk levels chart. Retrieved May 28th, 2010 from http://www.fcrr.org/pmrn/docs/ dibles_risklevels_k3_0 607_0809.pdf Florida Department of Education (2009). Student performance results: School reading demographic report. Retrieved May 28th, 2010 from https://app1.fldoe.org/FCATDemographics/Selections.aspx?lev el=School&subj= Reading Florida Department of Education (2010a). Florida school indicators report 2008-2009: Membership by category, school data. Retrieved May 29th, 2010 from http://www.fldoe.org/eias/eiaspubs/0809fsir.asp

PAGE 261

261 Florida Department of Education (2010b). Florida school indicators report 2008-2009: Membership by category, district and state data. Retrieved May 29th, 2010 from http://www.fldoe.org/eias/eiaspubs/0809fsir.asp Florida Department of Education (2010c). Florida school indicators report 2008-2009: Membership by grade, school data. Retrieved May 29th, 2010 from http://www.fldoe.org/eias/eiaspubs/0809fsir.asp Florida Department of Education (2010d). Next generation Suns hine State Standards Retrieved July 15th, 2010 from http://www.floridastandards.org /Standards/FLStandar dSearch.aspx Fountas, I. C., & Pinnell, G. S. (1999). Matching books to readers: Using leveled book in graded reading, K-3. West port, CT: Heinemann. Francis, D., Lesaux, N., & August, D. (2006). Language of instruction. In D. August & T. Shanahan (Eds.), Developing literacy in second-l anguage learners: Report of the national literacy panel on language-minority children and youth (pp. 365-413). Mahwah, NJ: Lawrence Erlbaum Associates. Fuchs, L. S., & Deno, S. L. (1992). Effects of curricu lum within curriculum-based measurements. Exceptional Children, 58, 232-243. Fuchs, L. S., & Fuchs, D. (2007). A model for implementing responsiveness to intervention. Teaching Exceptional Children,39 (5), 14-20. Genesee, F., Lindholm-Leary, K., Saunders, W ., & Christian, D. (2005). English language learners in U.S. Schools: An overview of research findings. Journal of Education for Students Placed at Risk, 10 (4), 363-385. doi:10.1207/s 15327671espr1004_2 Gerber, M., & Eng lish, J. (2003). Project La Pater: Final report. Santa Barbara, CA: Center for Advanced Studies of Individual Differences. Gerber, M., Jimenez, T., Leafsted t, J., Villaruz, J., Richards, C., & English, J. (2004). English reading effects of small-group intensive intervention in Spanish for k-1 English learners. Learning Disabilities Research & Practice, 19 (4), 239-251. doi:10.1111/j.1540-5826.2004.00109.x Gersten, R., & Baker, S. ( 2000). What we know about effect ive instructional practices for English language learners. Exceptional Children, 66 (4), 454-470. Gersten, R., Baker, S., Shanahan, T., Linan-Thompson, S., Collins, P., & Scarcella, R. (2007). Effective literacy and Eng lish language instruction for English learners in the elementary grades (NCEE 2007-4011). Washington DC: National Center for Education Evaluation and Regional Assistance. Retrieved October 14th, 2009 from http://ies.ed.gov/ncee/wwc/publications/practiceguides

PAGE 262

262 Gersten, R., & Geva, E. (2003). Teachi ng reading to early language learners. Educational Leadership, 60 (7), 44-49. Geva, E., Yaghoub-Zadeh, Z., & Sc huster, B. (2000). Understan ding differences in word recognition skills of ESL children. Annals of Dyslexia, 50 123-154. doi:10.1007/s 11881-000-0020-8 Glynn, T., Crooks, T., Bethune, N., Ballard, K., & Smith, J. (1989). Reading Recovery in context. Wellington: New Zealand Department of Education. Goldenberg, C. (2008). Teaching English la nguage learners: What the research does and does notsay. American Educator, Summer. Good, R. H., & Kaminski, R. A. (2002). Dynamic indicators of basic early literacy skills (6th edition). Eugene, OR: Institute for the Development of Educational Achievement. Retrieved August 17th, 2009 from https://dibels.uoregon.edu/measur es/files/admin_and_s coring_6th_ed.pdf Gough, P. (1983). Context, form, and inte raction. In K. Rayner (Ed.), Eye movement in reading (pp.331-358). Cambridge MA: MIT Press. Gough, P. (1996). How children le arn to read and why they fail. Annals of Dyslexia, 46 1-20. Gough, P., & Juel, C. (1991). The first stages of word recognition. In L. Reiben, &. C. A. Perfetti (Eds.), Learning to read: Basic re search and its implic ations (pp. 47-56). Hillsdale, N.J: Erlbaum. Greene, J. P. (1997). A meta -analysis of the Rossell and Baker review of bilingual education research. Bilingual Research Journal, 212 (2/3), 1-22. Gregory, R., Hackney, C., & Gregory, N. (1982). Corrective Reading programme: An evaluation. British Journal of Educational Psychology, 52, 33-50. Gunn, B., Biglan, A., Smokow ski, K., & Ary, D. (2000). The efficacy of supplemental instruction in decoding skills for His panic and non-Hispanic students in early elementary school. The Journal of Special Education, 34 (2), 90-103. Gunn, B., Smolkowski, K., Bigl an, A., & Black, C. (2002). Supplemental instruction in decoding skills for Hispanic and non-Hispa nic students in early elementary school: A follow-up. The Journal of Spec ial Education, 36(2), 67-79. Gunn, B., Smolkowski, K., Bi glan, A., Black, C., & Blai r, J. (2005). Fostering the development of reading skill through s upplemental instruction: Results for Hispanic and non-His panic students. The Journal of Special Education, 39 (2), 66-85.

PAGE 263

263 Guthrie, T. J., & Davis, M. H. (2003). Motivating struggling readers in middle school through an engagement model of classroom practice. Reading & Writing Quarterly, 19 (1), 59-85. doi :10.1080/10573560308203 Hasbrouck, J. E., Ihnot, C., & Rogers, G. H. (1999). Read Naturally: A strategy to increase oral reading fluency. Literacy Research and Instruction, 39 (1), 27-37. Hayes, L. F., Lane, H. B., & Pullen, P. C. (2005). University of Florida Literacy Initiative: Tutoring for beginning readers. Unpublished manual, Univer sity of Florida, Gainesville, Florida. Helman, L. (2009). Opening door s to texts: Planning effect ive phonics instruction with English learners. In L. Helman (Ed.), Literacy development with English learners (pp. 138-155). New York, NY: The Guilford Press. Helman, L., & Burns, M. ( 2008). What does oral language have to do with it? Helping young English-language learners acqui re a sight word vocabulary. The Reading Teacher, 62 (1), 14-19. Hickman, P., Pollard-Durodola, S., & Vaughn, S. (2004). Storybook reading: Improving vocabulary and comprehension for English language learners. The Reading Teacher, 57 (8), 720-730. Hobsbaum, A. (1995). R eading Recovery in England. In S. Swartz & A. Klein (Eds.), Research in Reading Recovery (pp. 132-147). Portsm outh, NH: Heinemann. Hoover, W. A., & Gough, P. B. (1 990). The simple view of reading. Reading and Writing, 2 127-160. Horner, R. B., Carr, E. G., Halle, J., McGee, G., Odom, S., & Wolery, M. (2005). The use of single-subject research to ident ify evidence-based practice in special education. Exceptional Children, 71 (2), 165-179. Hudson, R. F., Pullen, P. C., Lane, H. B., & Torgesen, J. K. (2009). The complex nature of reading: A multidimensional view. Reading & Writing, 25 (1), 4-32. doi:10.1080/10573560802491208 Individuals with Disabilit ies Education Act of 2004, Pub. L. No.108-446. Jitendra, A., Edwards, L., St arosta, K., Sacks, G., Jacobs on, L., & Choutka, C. (2004). Early reading instruction for children with reading difficulties: Meeting the needs of diverse learners. Journal of Learning Disabilities, 37 (5), 421-439. Jordan, G. (2006). What is SRA Early Interventions in Reading Level 1? Florida Center for Reading Research, Retrieved November 24th, 2008 from http://www.fcrr.org/FCRRReports/PDF/SRAEIR.pdf

PAGE 264

264 Juel, C. (1988). Learning to read and write: A longitudinal study of 54 children from first through fourth grades. Journal of Educational Psychology, 80 (4), 437-447. doi:10.1037/00220663.80.4.437 Juel, C., Griffith, P. L., & Gough, P. B. (1986). Acquisition of literacy: A longitudinal study of children in first and second grade. Journal of Educational Psychology, 78(4), 243-255. doi:1 0.1037/0022-0663.78.4.243 Kamps, D., Abbott, M., Greenwood, C., A rreaga-Mayer, C., Wills, H., Longstaff, J.,Walton, C. (2007). Use of evidence-based, small group reading instruction for English language learners in element ary grades: Secondary-tier intervention. Learning Disabilities Quarterly, 30 (3), 153 168. Kaufman, A. S., & Ka ufman, N. L. (2004). Kaufman Test of Educational Achievement-II, Comprehensive form manual. Minneapolis, MN: Pearson Assessments. Kazdin, A. E. (1982). Single-case research designs: Methods for clinical and applied settings. New York, NY: Oxford University Press. Kazelskis, R., Thames, D., & Reeves, C. (2004). The Elementary Reading Attitude Survey: Factor invariance across gender and race. Reading Psychology, 25 (2), 111-120. doi:10. 1080/02702710490435682 Kelly, P. R., Gomez-Bellenge, F. X., Chen, J., & Schulz, M. M. (2008). Learner outcomes for English language learner lo w readers in an early intervention. TESOL Quarterly, 42 (2), 235-260. Kennedy, C. H. (2005). Single-case designs for educational research. Boston, MA: Allyn and Bacon. Kindler, A. (2002). Survey of the states limited English pr oficient students and available educational programs and services 2000-2001 summary report. Klingner, J. K., Ahwee, S., Pilonieta, P., & Menendez, R. (2003). Barriers and facilitators in scaling up research-based practices. Exceptional Children, 69 (4), 411-430. Klingner, J. K., Artiles, A., & Barletta, L. (2006). English language learners who struggle with reading: Language acquisition or LD? Journal of Learning Disabilities, 39 (2), 108-128. Krashen, S. D. (1991). Bilingual education: Arguments for and (bogus) arguments against. In J. E. Alat is & A. Tan (Eds.), Round table on languages and linguistics (pp. 111-127). Washington, D. C.: Georgetown University Press. LaBerge, D., & Samuels, J. (1974). Toward a theory of automatic information processing in reading. Cognitive Psychology, 6 (2), 293-323. doi:10.1016/00100285(74)90015-2

PAGE 265

265 Lane, H. B., & Pullen, P. C. (2004). Phonological awareness assessment and instruction: A sound beginning. Boston, MA: Allyn and Bacon. Lane, H. B., Pullen, P. C ., & Hayes, L. F. (2007). UFLI: Small group intervention handbook: Kindergarten & first grade. Unpublished manual, University of Florida, Gainesville, Florida. Lane, H. B., Pullen, P. C., Hudson, R. F., & Konold, T. R. (2009) Identifying essential instructional components of literacy tutoring for st ruggling beginning readers. Literacy Research and Instruction, 48 (4), 1-20. doi: 10.1080/19388070902875173 Leafstedt, J. M., Richards, C. R., & Gerber, M. M. (2004). Effectiveness of explicit phonological-awareness instruction fo r at-risk English language learners. Learning Disabilities Research & Practice, 19(4), 252-261. Lee, J., Grigg, W., & Donahue, P. (2007). The nations report card: Reading 2007. Washington, DC: National Center for Educational Statistics. Lemos-Britton, Z., & Mostafapour, E. ( 1997). Spanish phonological awareness assessment and training for ESL students (SPAAT). Concord, CA: Author. Lesaux, N. K., & Siegel, L. S. (2003). The development of reading in children who speak English as a second language. Developmental Psychology, 39 (6), 1005-1019. doi:10.1037/00121649.39.6.1005 Lesaux, N. K., Koda, K., Siegel, L. S., & Shanahan, T. (2006). Development of literacy. In D. August & T. Shanahan (Eds.), Developing literacy in second-language learners: Report of the national litera cy panel on language-minority children and youth (pp. 75-122). Mahwah, NJ: Law rence Erlbaum Associates. Leslie, L., & Caldwell, J. (2006). Qualitative Readi ng Inventory-4. Boston, MA: Allyn and Bacon. Limbos, L. M., & Geva, E. (2001). Accuracy of teac her assessments of secondlanguage students at risk for reading disability. Journal of Learning Disabilities, 34(2), 136-151. Linan-Thompson, S., Vaughn, S., Hickman-Davis, P., & Kouzekanani, K. (2003). Effectiveness of supplemental reading in struction for second-grade English language learners with r eading difficulties. The Elementary School Journal, 103 (3), 221-238. Linan-Thompson, S., Vaughn, S., Prater, K., & Cirino, P. (2006). The response to intervention of English language lear ners at risk for reading problems. Journal of Learning Disabilities, 39(5), 390-398.

PAGE 266

266 Lindsey, K. A., Manis, F. R., & Bailey, V. E. (2003). Predicti on of first-grade reading in Spanish-speaking Eng lish-language learners. Journal of Educational Psychology, 95(3), 482-492. doi:1 0.1037/0022-0663.95.3.482 Lonigan, C. J. (2006). Devel opment, assessment, and promot ion of preliteracy skills. Early Education & Development, 17 (1), 91-114. doi:10.1207/s15566935eed1701_5 Manyak, P. C., & Bauer, E. B. (2008). Expl icit code and comprehension instruction for English learners. The Reading Teacher, 61(5), 432-434. Marchand-Martella, N., Slocum, T., & Martella, R. (2004). Introduction to direct instruction. Boston, MA: Pearson Education. Martson, D. (1989). Curriculum-based measurem ent: What is it and why do it? In M. Shinn (Ed.), Curriculum-based measurement: Assessing special children (pp. 1878). New York, NY: Guilford Press. Mathes, P. G., Denton, C. A., Fletcher, J. M., Anthony, J. L., Francis, D. J., & Schatschneider, C. (2005). The effects of theoretically differ ent instruction and student characteristics on the skills of struggling readers. Reading Research Quarterly, 40 (2), 148-182. doi:10.1598/RRQ.40.2.2 Mathes, P. G., & Torg esen, J. K. (2005). SRA Early interventions in reading: Level 1. Columbus, OH: SRA/McGraw-Hill. McCardle, P., Mele-McCarthy, J., Cutting, L., Leos, K., & DEmilio T. (2005). Learning disabilities in English language learners: Identifying the issues. Learning Disabilities Research & Practice, 20(1), 1-5. doi:10.1111/j.15405826.2005.00114.x McDonald, S. K., Keesler, V. A., Kauffm an, N. J., & Schneider, B. (2006). Scaling-up exemplary interventions. Educational Researcher, 35 (3), 15-24. McKenna, M. C., & Kear, D. J. (1990). Measuring attitude to ward reading: A new tool for teachers. The Reading Teacher, 43 (9), 626-639. McMaster, K. L., Fuchs, D., Fuchs, L. S., & Compton, D. L. ( 2005). Responding to nonresponders: An experimental field tr ial of identification and intervention methods. Exceptional Children, 71 (4), 445-463. Mohr, K. A., & Mohr, E. S. (2007). Ext ending English-language learners classroom interactions using the Response Protocol. The Reading Teacher, 60 (5), 440-450. doi:10.1598/RT.60.5.4 Mortensen, E. (1984). Reading achievement of native S panish-speaking elementary students in bilingual vs. monolingual programs. Bilingual Review, 11 (3), 31-36.

PAGE 267

267 National Center for Educat ion Statistics. (2001). Entering kindergarten: Findings from the condition of education 2000. Retrieved November 24th, 2008 from http://nces.ed.gov/pubs2001/2001035.pdf National Commission on Exc ellence in Education. (1983). A nation at risk: The imperative for educational reform. Washington, DC: U.S. Department of Education. National Data Evaluation Center. (2008). Technical report: Reading Recovery and Descubriendo La Lectura national report 2006-2007. Retrieved November 15th, 2008 from http://www.ndec.us/Documentation.asp National Reading Panel. (2000). Teaching children to read: An evidence-bas ed assessment of the scientific research literature on reading and its implications for reading instruction. Washington, DC: Na tional Institute of Child Health and Human Development. Neal, J. C., & Kelly, P. R. (1999). The success of Reading Recovery for English language learners and Descubriendo La Lect ura for bilingual students in California. Literacy Teaching and Learning, 4 (2), 81-108. Neuman, S. B., & McCormick, S. (2000). A case for single-subject experiments in literacy research. In M. L. Kamil, P. D. Pearson, R. Barr, & P. B. Mosenthal (Eds.), Methods of literacy research (pp. 181-194). Mahwah, NJ: Erlbaum. Nicholson, T. (1989). A comm ent on Reading Recovery. New Zealand Journal of Educational Studies, 24 95-97. No Child Left Behind Act of 2001, PL 107-110, 20 U.S.C. 6301 et seq. OConnor, R. E., Notari-Syverson, A., & Vadasy, P. F. (1998). Ladders to literacy: A kindergarten activity book. Baltimore, MA: Brookes Publishing. Odom, S. L., Brantlinger, E., Ge rsten, R., Horner, R. H., T hompson, B., & Harris, K. R. (2005). Research in special education: Scientific methods and evidence-based practices. Exceptional Children, 71 (2), 137-148. Peregoy, S. F., & Boyle, O. F. (2005). Eng lish learners reading Eng lish: What we know, what we need to know. In Z. Fang (Ed.), Literacy teaching and learning: Current issues and trends (pp. 18-27). Upper Saddle Riv er, NJ: Pearson Education. Perry, N. E., & Meisels, S. J. (1996). How accurate are teachers judgments of students academic performance? (Working Paper No. 9608,National Center for Educational Statistics).W ashington, DC: U.S. Department of Education. Pinnell, G. S., Lyons, C. A., De Ford, D. E., Bruk, A. S., & Se ltzer, M. (1994). Comparing instructional models for the literacy education of high-risk first graders. Reading Research Quarterly 29(1), 8-39.

PAGE 268

268 Planty, M., Husar, W., Sny der, T., Kena, G., KewalRamani A., Kemp, J., Bianco, K., Dinkes, R. (2009). The condition of education 2009 (NCES 2009-081) Washington, DC: National Center for Educational Statistics. Retrieved on April 19th, 2010 from the http://nces.ed.gov/pubs2009/2009081.pdf Pollar-Durodola, S. D., Mathes, P G. Vaughn, S., Cardenas-Hagan, E., & LinanThompson, S. (2006). The role of or acy in developing comprehension in Spanish-speaking Eng lish language learners. Top Language Disorders, 26(4), 365-384. Pullen, P. C. (2000). The effe cts of alphabetic word work with manipulative letters on the reading acquisition of struggling fi rst-grade students (Unpublished doctoral dissertation). University of Florida, Gainesville. Pullen, P. C., Lane, H. B., Lloyd, J. W., Nowak. R., & Ryals, J. (2005). Effects of explicit instruction on decoding of struggling fi rst grade students: A data-based case study. Education & Treatment of Children, 28 (2), 63-76. Quirk, M. P., & Schwanenflugel, P. J. (2005). Do supplemental remedial reading programs address the motiva tional issues of struggling readers? An analysis of five popular programs. Literacy Research & Instruction, 43 (3), 1-19. Quiroga, T., Lemos-Britton, Z ., Mostafapour, E., A bbott, R., & Berni nger, V. (2002). Phonological awareness and beginning reading in Span ish-speaking ESL first graders: Research into practice. Journal of School Psychology, 40 (1), 85-111. doi:10.1016/S00224405(01)00095-4 Richards, C., Leafstedt, J. M., & Gerber, M. M. (2006). Qua litative and quantitative examination of four low-performi ng kindergarten English learners. Remedial and Special Education, 27(4), 218-234. Richards, C., Leafstedt, J. M., Gerber, M. M., Jimenez, T., & Filippini, A. (2003). La Patera phonological awareness test (LP-PAT). Santa Barbara: University of California, Center for the Advanc ed Study of Individual Differences. Roberts, T. A. (2003). Effects of alphabet-letter instruction on young childrens word recognition. Journal of Educational Psychology, 95 (1), 41-51. doi:10.1037/00220663.95.1.41 Samuels, J. (1979). The method of repeated readings. Reading Teacher, 32, 403-308. Santoro, L., E., Jitendra, A. K., Staros ta, K., & Sacks, G. (2006). Reading well with Read Well: Enhancing the reading performanc e of English language learners. Remedial and Spec ial Education, 2(2), 105-115.

PAGE 269

269 Scarborough, H. S. (2 003). Connecting early language and literacy to later reading (dis) abilities: Evidence, theory, and practice. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (pp. 97110). New York, NY: The Guilford Press. Schwartz, R. M. (2005). Literacy learni ng of at-risk first-grade students in Reading Recovery early intervention. Journal of Educational Psychology, 97 (2), 257-267. doi:10.1037/00220663.97.2.257 Schirmer, B. R., Casbon, J., & Twiss, L. L. (1996). Innovative literacy practices for ESL learners. The Reading Teacher, 49 (5), 412-414. Seymour, P., & Duncan, L. (2001) Learning to read in English. Psychology: The Journal of the Hellenic Psyc hological Society, 8 281-299. Seymour, P., & Elder, L. (1986). Beginning reading without phonology. Cognitive Neuropsychology, 3 (1), 1-36. doi :10.1080/02643298608252668 Shanahan, T., & Barr, R. (1995) Reading Recovery: An i ndependent evaluation of the effects of an early instructional intervention for at-risk learners. Reading Research Quarterly, 30 (4), 958-996. Shanahan, T., & Beck, I. (2006). Effectiv e literacy teaching for English-language learners. In D. Augus t & T. Shanahan (Eds.), Developing literacy in secondlanguage learners: Report of the national literacy panel on language minority children and youth (pp. 415-488) Mahwah, NJ: Lawrence Erlbaum Associates. Shankweiler, D., Lundquist, E., Katz, L., Stu ebing, K. K., Fletc her, J. M., Brady, S.,Shaywitz, B. A. (1999). Comp rehension and decoding: Patterns of association in children wit h reading difficulties. Scientific Studies of Reading, 31(1), 69-94. doi:10. 1207/s1532799xssr0301_4 Siegler, R. S., & Crowley, K. (1991). A microgenetic me thod: A direct means for studying cognitive development. American Psychology, 46 (6), 606-620. doi:10.1037/0003066X.46.6.606 Silliman, E., Wilkinson, L., & Brea-Spahn, M. (2004). Policy and practice imperatives for language and literacy learning: Who will be left behind? In C.A. Stone, E.R. Silliman, B.J. Ehren, & K. Apel (Eds.), Handbook of language and literacy (pp. 97-129). New York, NY: The Guilford Press. Slaving, R. & Cheung, A. (2005). A synthesis of res earch on language of reading instruction for English language learners. Review of Educational Research, 75(2), 247-284. Smith, S. (2004). What is Corrective Reading? Florida Center for Reading Research. Retrieved on November 23rd, 2004 from http://www.fcrr.org/FCRRReports/PDF /corrective_reading_fina l.pdf

PAGE 270

270 Snow, C. (2006). Cross-cutting themes and future research directions. In D. August & T. Shanahan (Eds.), Developing literacy in second-l anguage learners: Report of the National Literacy Panel on langua ge-minority children and youth (pp. 631-651). Mahwah, NJ: Lawrence Erlbaum Associates. Snow, C., Burns, M., & Griffin, P. (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press. Sparks, G. M. (1988). Teachers a ttitudes toward change and subsequent improvements in classroom teaching. Journal of Educational Psychology, 80(1), 111-117. doi:10.1037/0022-0663.80.1.111 Sprick, M. M., Howard, L. M., & Fidanque, A. (1998). Read Well: Critical foundations in primary reading. Longmont, CO: Sopris West. Stahl, S. (2004). What do we know about fluen cy? Findings from the National Reading Panel. In. P. McCardle & V. Chhabra (Eds.), The voice of evidence in reading research (pp. 187-211). Baltimore, MA: Brookes Publishing. Stanovich, K. E. (1980). Toward an inte ractive-compensatory model of individual differences in the development of reading fluency. Reading Research Quarterly, 16(1), 32-71. Stanovich, K. E. (1986). Matt hew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21 (4), 360407. Stanovich, K. E. (1990). Concepts in developm ental theories of reading skill: Cognitive resources, automaticity, and modularity. Developmental Review, 10 (1), 72-100. doi:10.1016/02732297(90)90005-O Stuart, M. (1999). Getting ready for r eading: Early phoneme awareness and phonics teaching improves reading and spelling in inner-city second language learners. British Journal of Educational Psychology, 69 587-605. Swartz, S. L., & Kl ein, A. F. (1995). Research in Reading Recovery. Portsmouth, NH: Heinemann. Tabors, P. O., & Snow, C. E. (2002) Young bilingual childr en and early literacy development. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (pp. 159-178). New York, NY: The Guilford Press. Tindal, G., Marston, D ., & Deno, S. L. (1983). The reliability of direct and repeated measurement (Research Report 109). Minneapolis, MN: University of Minnesota Institute for Research on Learning Disabilities.

PAGE 271

271 Torgesen, J. K. (1998). Catch them before they fall: Identification and assessment to prevent reading failure in young children. American Educator, 22, 32-39. Torgesen, J. K. (2000). Indivi dual differences in response to early interventions in reading: The lingering problem of treatment resisters. Learning Disabilities Research & Practice, 15 (1), 55-64. Torgesen, J. K. (2002). The prev ention of reading difficulties. Journal of School Psychology, 40(1), 7-26. doi:10.1016/S0022-4405(01)00092-9 Torgesen, J. K., Alexander, A. W., Wagner, R. K., Rashotte, C. A., Voeller, K. K., & Conway, T. (2001). Intensive remedial in struction for children with severe reading disabilities: Immediate and long-term outcomes from two instructional approaches. Journal of Learning Disabilities, 34 (1), 33-58. Torgesen, J. K., Wagner, R. K. & Rashotte, C. A. (1999). Test of word reading fluency. Austin, TX: Pro-Ed. Vaughn, S., Cirino, P. T., Li nan-Thompson, S., Mathes, P. G. Carlson, C. D., Hagan, E. C.,Francis, D. J. (2006). Effectiv eness of a Spanish intervention and an English intervention for English-languag e learners at risk for reading problem. American Educational Research Journal, 43 (3), 449-487. Vaughn, S., Gersten, R., & Chard, D. J. (2000). The underlying message in LD intervention research: Findings from research syntheses. Exceptional Children, 67(1), 99-114. Vaughn, S., Linan-Thompson, S. & Hickman, P. (2003). Res ponse to instruction as a means of identifying students with reading/learning disabilities. Exceptional Children, 69 (4), 391-409. Vaughn, S., Linan-Thompson, S., Kouzekanani, K., Bryant, D. P., Dickson, S., & Blozis, S. A. (2003). Reading instruction grouping for students with reading difficulties. Remedial and Spec ial Education, 24 (5), 301-315. Vaughn, S., Linan-Thompson, S. Pollard-Durodola, S., Ma thes, P. G., & Hagan, E. (2006). Effective interventions for Eng lish language learners (Span ish-English) at risk for reading difficulties. In D. Dickinson & S. Neuman (Eds.), Handbook of early literacy research (pp. 185197). New York, NY: The Guilford Press. Vaughn, S., Mathes, P. G., Linan-Thompson, S., Cirino, P., Carlson, C., PollardDurodola, S.,Francis, D. (2006). Effectiveness of an English intervention for first grade English lan guage learners at risk fo r reading problems. The Elementary School Journal, 107 (2), 153-180. doi:10.1086/510653

PAGE 272

272 Vellutino, F. R., Scanlon, D. M., Sipay, E. R., Small, S., Pr att, A., Chen, R., & Denckla, M. B. (1996). Cognitive profiles of diffi cult-to-remediate and readily remediated poor readers: Early intervention as a vehi cle for distinguishing between cognitive and experiential deficits as basic causes of specific reading disabilities. Journal of Educational Psychology, 88 (4), 601-638. doi:10. 1037/0022-0663.88.4.601 Velluntino, F. R., Scanlon, D. M., & Tanzman, M. (1994). Components of r eading ability: Issues and problems in operationalizing wo rd identification, phonological coding, and orthographic coding. In G. R. Lyon (Ed.), Frames of reference for the assessment of learning disabilitie s: New views on measurement issues (pp. 279324). Baltimore, MA: Brookes Publishing. Verhoeven, L. T. (1990). Acquisition of reading in a second language. Reading Research Quarterly, 25 (2), 90-114. Verhoeven, L. T. (2000). Components in ear ly second language reading and spelling. Scientific Studies of Reading, 4 (4), 313-330. doi:10.120 7/S1532799XSSR0404_4 Wade-Woolley, L., & Siegel, L. S. (1997). The spelling pe rformance of ESL and native speakers of English as a function of reading skill. Reading & Writing, 9 (5), 387406. doi:10.1023/A:1007994928645 Wagner, R. K., Torgesen, J. K., & Rasho tte, C. A. (1994). De velopment of reading related phonological processing abilities: New evidence of bidirectional causality from a latent variable longitudinal study. Developmental Psychology, 30 (1), 7387. doi:10.1037/0012-1649.30.1.73 Wanzek, J., & Vaughn, S. (2008). Response to varying amounts of time in reading interventions for students with low response to intervention. Journal of Learning Disabilities, 41(2), 126-142. doi :10.1177/0022219407313426 Watts-Taffe, S., & Truscott, D. M. (2000). Using what we know about language and literacy development for ESL students in the mainstream classroom. Language Arts, 77 (3), 258-265. Whal, M. (2007). What is Read Well? Florida Center for Reading Research. Retrieved November 23rd, 2008. from http://www.fcrr.org/fcrrrepor ts/PDF/Read_W ell_Report.pdf What Works Clearinghouse. (2007). Intervention: Reading Recovery. Retrieved November 20th, 2008 from http://ies.ed.gov/ncee/wwc/reports/ beginning_reading/reading_recovery/ Whitehurst, G. & Lonigan, C. (2002). Emergent literacy: Development from prereaders to readers. In S. Neuman & D. Dickson (Eds.), Handbook of early literacy research (pp. 11-29). New York, NY : The Guilford Press.

PAGE 273

273 Wolf, M. M. (1978). Social validity: The case for subjective measurement on how applied behavior analysis is finding its heart. Journal of Applied Behavior Analysis, 11 (2), 203-214. doi: 10.1901/jaba.1978.11-203 Woodock, R. W., Muoz-Sandoval, A. F., R uef, M. L., & Alvar ado, C. G. (2005). WoodcockMuoz Language Survey Revised, English. Itasca, IL: Riverside Publishing. Woodock, R. W., Muoz-Sandoval, A. F., R uef, M. L., & Alvar ado, C. G. (2005). Woodcock-Muoz Language Survey Revised, Spanish. Itasca, IL: Riverside Publishing.

PAGE 274

274 BIOGRAPHICAL SKETCH Stephanie Arriaza de Allen was born and ra ised in Guatemala city, Guatemala in 1975. She attended a bilingual k-12 school a nd graduated with an English/Spanish High School degree in 1992. She earned her bachel ors degree in psychology from the Universidad del Valle de Guatemala in 2000. She came to the United States in 2000 as an international student and completed one year of English as a second language courses at the University of Florida and Santa Fe College in order to prepare for her graduate studies. In 2001, she was accepted into the Ma ster of Education and Educat ional Specialist degree in counselor ed ucation at the University of Florida. She completed the program in 2004 with a focus on school counse ling. Towards the end of the program, she completed three field experiences in elementary schools where she worked with students who were experienci ng emotional and behavioral diffi culties. Many of these students were English language learners. As she conducted individual and small-group counseling, she realized that many of their issues stemm ed from academic difficulties, particularly in the area of reading. An inte rest in learning about struggling readers and effective ways to serve them developed from these observations. She decided to pursue a masters degree in Special Education at the University of Florida to learn how to help students become successful readers. At the time, her plan was to return to Guatemala and serve the specia l education community in that country. As the masters program progressed and she learned about effe ctive reading intervention for struggling readers and the particular needs of English l anguage learners, her passion for helping grew stronger and stronger. Toward the end of the masters progr am, she decided to

PAGE 275

275 apply to the Special Education do ctoral program at the University of Florida. In 2006, she was accepted and awarded an alumni fe llowship to complete her studies. During the four year program, she focused on literacy development, reading disabilities, early interventions, and Englis h language learners. While completing her course work, she had the opportunity to co-teach and teach undergraduate and graduate level courses in liter acy, reading disabilities, bi lingual special education, and family involvement. She worked in research projects related to literacy and teacher education. She has published an article in a peer reviewed journal. She has been a reviewer and a presenter in several state and national conferences. In 2007, she was awarded second place for best paper present ed at the Language and Globalization: Policy, Education, and Media Conference at Georgetown Universi ty. Other awards received include the Outstanding Internati onal Student Award for the College of Education, as well as certific ates for excellence in academi c achievement offered by the International Student Center at the University of Florida. In 2010, she graduated with a Ph.D. in Specia l Education. Her professional goals include teaching at the university level and becoming a national and international consultant for the prevention of reading di fficulties and the promotion of effective reading interventions for all struggling readers. She also plans to open a tutoring center for underprivileged children who are in need of reading support.



PAGE 1

1 EVALUATING THE EFFECTS OF THE UNIVERSITY OF FLORIDA LITERACY INITIATIVE (UFLI) ON THE READING SKILLS OF SPANISH-SPEAKING ENGLISH LANGUAGE LEARNERS By STEPHANIE LISET A RRIAZA DE ALLEN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORID A IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Stephanie Liset Arriaza de Allen

PAGE 3

3 To my loving husband Matt, dad Luis Francis co, mom Yolanda and loving family Thank you for all your love and support

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my gratitude to everyone who su pported me in the completion of my doctoral program. Most im portantly, I thank God and Jesus Christ for all the blessings I have received in my life. May this accomplishment be used for Their glory. I also want to thank our Holy Mo ther for Her uncondition al love and constant prayers. A special thanks goes to my husband, Matt, whose love and support encouraged me every step of the way. I am thankful for the many hours he stay ed by my side while I worked through the night, and for the long walks we took when I was stressed. I am also thankful for the words of enc ouragement he gave me when I felt exhausted, and for the countless times he held me close when I felt overwhelmed. I also want to thank my mom and dad for in stilling in me a passion for learning, for believing in me, and for always supporting my endeavors. They taught me the value of hard work and perseverance. Thanks go to my family and friends, as well, for their guidance and unceas ing encouragement. I thank Holly Lane, chair of my doctora l committee, for being a mentor and a dear friend. She was a source of inspiration to me because of her love of learning and her passion for helping children. I appreciate the sincere kindness and immeasurable generosity she showed me as I completed my doctoral program. I want to thank Hazel J ones, co-chair of my doctora l committee, for guiding me and helping me develop my knowledge of single-subject design. I am also grateful for my other committee members, Roxanne Hudson, Linda Lombardino, and Erica McCray. Their knowledge and expertise greatly contributed to my dissertation.

PAGE 5

5 Special thanks go to Penny Cox who was t he first person to suggest that I should pursue a doctoral degree and who guided me ever y step of the way. I also want to thank Shaira, Michell and Vicki for all the help they provided me throughout the years. They opened their hearts to me and became part of my life. I will always cherish their friendship. I also want to express my appreciati on to Donna Long, principal of Suwannee Elementary School and all t he school personnel, for welcoming me to the school and facilitating the completion of this study. T hanks go to all the wonderful students I had the privilege of tutoring. Their hard work and ent husiasm for learning were a source of inspiration. Finally, I want to thank all the graduat e students who helped me complete data collection, including Susan Helvenston, Maria Rodriguez, Yulia Aguilar, and Lindsay Bell. They always had a positive dispositi on and a great attitude toward the students. Their assistance was priceless.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDG MENTS .................................................................................................. 4LIST OF TABLES............................................................................................................ 9LIST OF FIGURES........................................................................................................ 10ABSTRACT................................................................................................................... 11CHAPTER 1 INTRODUCTION.................................................................................................... 13Statement of the Pr oblem....................................................................................... 14Ehris Phases of Word Reading Development........................................................ 17Pre-Alphabeti c Phase....................................................................................... 18Partial-Alphabet ic Phase.................................................................................. 18Full-Alphabeti c Phase...................................................................................... 19Consolidated-Alph abetic P hase....................................................................... 20Automatic Phase.............................................................................................. 21The Role of Instruction in Prom oting Word Readi ng Development......................... 21Purpose of the Study.............................................................................................. 242 REVIEW OF TH E LITERA TURE ............................................................................ 25Reading Re covery.................................................................................................. 26Direct Instruction..................................................................................................... 34Core Intervention Model................................................................................... 35Reading Mastery and Co rrective R eading........................................................ 38Proactive Reading............................................................................................ 43Read We ll......................................................................................................... 52Multiple Direct Instruction Inte rventions............................................................ 57Other Reading In tervent ions................................................................................... 59Word Reading In tervent ions............................................................................. 60Comprehensive Readi ng Interv entions............................................................ 62Discuss ion.............................................................................................................. 70Instruction in Wo rd Reading Skills.................................................................... 71Language and Reading Instruct ion................................................................... 72Strategies for E nglish Language Learners....................................................... 74Implications for Res earch................................................................................. 743 METHODS.............................................................................................................. 77Setti ng..................................................................................................................... 77Participants............................................................................................................. 79

PAGE 7

7 Assessment In strume nts......................................................................................... 94Screening In struments..................................................................................... 94Grouping Inst ruments....................................................................................... 99Supplemental Preand Post-I ntervention In struments................................... 100Intervent ion........................................................................................................... 106Early Literacy Components Addressed in UFLI.............................................. 107Effective Instructional Practices for ELLs....................................................... 109UFLI Se ssions................................................................................................ 110Materi als......................................................................................................... 115Dependent Variable.............................................................................................. 117Design................................................................................................................... 119Baseli ne.......................................................................................................... 120Intervent ion..................................................................................................... 120Maintenance................................................................................................... 121Tutor/Researcher........................................................................................... 122Data A nalys is........................................................................................................ 122Supplemental Data............................................................................................... 123Preand PostIn terventi on Data.................................................................... 123Book Levels duri ng Interv ention..................................................................... 124Interobserver Agr eement...................................................................................... 124Treatment Integrity................................................................................................ 125Social Va lidity....................................................................................................... 126Delimitations and Limita tions................................................................................ 128Delimita tions................................................................................................... 128Limita tions...................................................................................................... 129Description of One-onOne Tutoring Methodology............................................... 130Pilot Study............................................................................................................. 1314 RESULT S............................................................................................................. 133Dependent Variables............................................................................................ 133Baseline Phase.............................................................................................. 134Intervention Phase......................................................................................... 135Maintenance Phase........................................................................................ 137Summary of Depende nt Variables.................................................................. 138Total Number of Se ssions..................................................................................... 142Supplemental Data............................................................................................... 144Measures of Early Literacy Skills and Attitudes to wards Reading.................. 144Book Levels Read dur ing Intervention............................................................ 150Summary of Suppl emental Data..................................................................... 153Social Va lidity....................................................................................................... 153Student Inte rview............................................................................................ 154Teacher Ques tionnaire................................................................................... 154Summary of One-on-O ne Tutoring Data............................................................... 156Dependent Variables...................................................................................... 157Supplemental Data......................................................................................... 158Summary of Findi ngs............................................................................................ 160

PAGE 8

8 5 DISCUSSION....................................................................................................... 162Overview of the Study........................................................................................... 162Summary of Findi ngs............................................................................................ 164Interpretation of Fi ndings...................................................................................... 165Progression through Phases of Word Recognition Development................... 167Relation between CPPM and CSPM.............................................................. 174Interpretation of Supplemental Data..................................................................... 175Early Lite racy Sk ills........................................................................................ 175Attitudes towa rd Reading............................................................................... 178Teacher Reading Ability Rating Scal e (RARS)............................................... 180Social Va lidity....................................................................................................... 181Students Re sponses..................................................................................... 181Teachers Respons es..................................................................................... 183Effectiveness of the UF LI Tutori ng Progr am......................................................... 185Early Intervention and Language Profic iency................................................. 185Effective Instructional Practices for ELLs....................................................... 187Consistency with t he Response to Inte rvention A pproach.............................. 189Implications for Future Re search.......................................................................... 191Implications for Practice........................................................................................ 196Conclu sion............................................................................................................ 198APPENDIX A IRB DOCUME NTATION ....................................................................................... 200B HOME LANGUAGE SURVEY A ND REMINDER LETTERs ................................. 211C TEACHERS READING ABILITY RATING SC ALE (R ARS) ................................. 215D UFLI SMALL GROU P SESSION GUIDE .............................................................. 216E UFLI SMALL-GROU P SESSION NOTES ............................................................. 217F PSEUDOWORD PROBES ................................................................................... 219G SIGHT WORD PROBES ....................................................................................... 234H TREATMENT INTEGR ITY CHECK LIST ............................................................... 249I SOCIAL VALIDITY INSTRUM ENTS ..................................................................... 251J UFLI INDIVIDUAL TUTO RING SESSION GUIDE ................................................ 255LIST OF RE FERENCES............................................................................................. 256BIOGRAPHICAL SKETCH.......................................................................................... 274

PAGE 9

9 LIST OF TABLES Table page 3-1 Group com position ............................................................................................. 813-2 Summary of participants demographic in formation............................................ 823-3 Summary of participants instructional informa tion............................................. 843-4 Participants reading scores used for sele ction pur poses................................... 853-5 Students English and Spanish language levels based on the WMLS-R............ 853-6 Conners Abbreviated Teac her Rating Scal e (C-ATRS)..................................... 863-7 Running records result s used for groupi ng pur poses....................................... 1003-8 Treatment integrity............................................................................................ 1263-9 Running records re sults used fo r Miriam.......................................................... 1304-1 Individual rates of C PPM and CSPM dur ing basel ine....................................... 1354-2 Individual rate of CPPM and CSPM during intervent ion................................... 1374-3 Individual rate of CPPM and CSPM duri ng maintenance................................. 1394-4 Total number of tutoring se ssions received by each student............................ 1444-5 Preand post-intervention TOWR E raw scores and per centiles ranks............. 1454-6 Preand post-intervention KTEA-II raw scores and per centiles ranks............. 1464-7 Preand post-intervention DIBE LS Nonsense Word Fluency scores............... 1464-8 Pre-intervention DIBELS Oral Reading Fl uency scores................................... 1474-9 Post-intervention DIBELS Oral Reading Fl uency scores.................................. 1474-10 Pre/Post-intervention QR I-4 comprehensio n raw scores.................................. 1494-11 ERAS Pre-and post-intervention raw scores and perce ntile ranks................... 1504-12 Preand post-intervention mean rate scores for reading skills on the RARS... 1504-13 Preand post-intervention rate scores for classroom behaviors on the RARS. 151

PAGE 10

10 LIST OF FIGURES Figure page 4-1 Rate of correct pseudowords per minute (C PPM) across groups..................... 1404-2 Rate of correct sight words r ead per minute (CSPM) across groups................ 1414-3 Book levels read by eac h group during in tervention......................................... 1524-4 Teachers responses to the social validity questionna ire about the UFLI component s...................................................................................................... 1564-5 Teachers responses to the social validity questionna ire about the UFLI progr am............................................................................................................ 1564-6 Rate of correct pseudoword and sight word read per minute for Miriam.......... 1584-7 Book levels read by Mi riam during in tervention................................................ 160

PAGE 11

11 Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy EVALUATING THE EFFECTS OF THE UNIVERSITY OF FLORIDA LITERACY INITIATIVE (UFLI) ON THE READING SKILLS OF SPANISH-SPEAKING ENGLISH LANGUAGE LEARNERS By Stephanie Liset Arriaza de Allen August 2010 Chair: Holly B. Lane Cochair: Hazel A. Jones Major: Special Education Many English language learners (ELLs) in schools experience reading difficulties, particularly Spanish-speaking st udents. This is a serious problem given the tendency for struggling readers to fall further behind as they advance in school. This issue becomes more pressing as the number of ELLs enrolled in schools rises and the availability of native language instruction dimi nishes. Fortunately, there is evidence that English reading interventions that have proven to be effective for struggling native English speakers are also effective for ELLs with diffe rent levels of English oral proficiency. Given that the literature is scarce, it is important to investigate the impact of other reading interventions on the reading outcomes of ELLs. The purpose of this study was to examin e the effectiveness of a modified version of the University of Florida Literacy Initiati ve tutoring program in promoting the reading skills of second-grade Spanish-speaking ELLs who are struggling to read. The program was modified to include small-group instruction and practices that supported the

PAGE 12

12 language needs of ELLs. A multiple-base line across groups design was used to examine students response to the interventi on, as measured by the rate of correct pseudowords (CPPM) and sight words (CSPM) read per minute. Results showed that all groups improved from baseline to intervention in their rates of CPPM and CSPM and maintained appropriate rates tw o weeks after the intervention ceased. In addition, all groups showed a marked improvement in book reading accuracy of at least one halfs year progress and up to a year and a halfs progress in a limited number of sessions. St udents improved reading skills were further corroborated by preand post-intervention measures of decoding, word recognition, fluency, and comprehension, as well as teac hers ratings of students reading abilities and classroom behaviors. After intervention, students also demonstrated more positive attitudes towards reading. Most importantly, the intervention proved effective for all students regardless of their level of English oral proficiency. Social validity measures showed that the UFLI program was regarded as important, effective, and feasible among participants and teachers. These results are consistent with available literature that supports the implement ation of early English interventions for ELLs.

PAGE 13

13 CHAPTER 1 INTRODUCTION Learning to read is an important milestone in every ch ilds life. Due to its close relationship with academic achiev ement (Whi tehurst & Lonigan, 2002), proficient reading holds lifelong implications. Accordi ng to Adams (1990), reading is the key to education, and education is the key to succe ss for both individuals and a democracy (p. 13). The ability to read not only opens a wealth of professional opportunities for individuals, but also facilitates the completion of basic daily tasks, such as paying bills, reading the newspaper, reading medicine labels, reading inst ructions, and filling out forms (Anderson, Hiebert, Scott, & Wilk inson, 1985; Chhabra & McCardle, 2004). The recognition of the importanc e of reading has prompted m any initiatives throughout the years with the purpose of better understanding the processes involved in skilled reading, as well as identifying and setting in place effective ways to promote its development (Adams, 1990; Anderson et al., 1985; National Commission on Excellence on Education, 1983; National Reading Panel, 2000; No Child Left Behind, 2001; Snow et al., 1998). Unfortunately, there are many children in school who struggle with reading. In 2007, the National Center for E ducation Statistics (NCES) r eported that 33% of fourth grade students in public schools scored below basic levels on reading (Lee, Griff, & Donahue, 2007). The percentages across states showed great variance, from 19% in Massachusetts to 66% in the District of Co lumbia. The report also showed a 27-point gap between Blacks and Whites, and a 26point gap between Hispanics and Whites. There is also evidence that a gap already exists when children enter kindergarten. For example, the Early Childhood Longitudi nal Study examined a group of 19,000

PAGE 14

14 kindergarten students from a representative sample of 940 private and public schools in the nation and found that 18% of students were not familiar wit h the conventions of print (i.e., directionality of text, book structure), while 34% were unable to name the letters of the alphabet (NCES, 2001). Sc arborough (2003) stated that between 65% and 75% of children who experience reading difficulties in the early years continue to show poor reading skills during the course of their school years. These numbers are troublesome, given the tendency for struggling readers to fall further and further behind their higher performing peers (Whitehurst & Lonigan, 200 2), a phenomenon identified by Stanovich (1986) as the Matthew Effect in Reading (p. 380). That is, children who read well tend to read more, and as a result become bette r readers. On the ot her hand, students who struggle with reading tend to read less and, as a consequence, fail to become skilled readers. Statement of the Problem Children in the United States whose native language is not Englis h are at particular risk for experienci ng reading difficulties and falli ng behind in school (Snow et al., 1998). The growing number of language minority students in the U.S. (Klingner, Artiles, & Barletta, 2006) and the limited availability of bilingual programs (August, 2006; Gomez-Bellenge, Chen, & Schulz, 2008) make this problem even more pressing. The term English language le arner, or ELL, has been defined by the National Literacy Panel on Language-Minority Child ren and Youth (NLP; August, 2006) as language-minority students who are limited English profici ent (p. 44). Other names used in the literature include (a) limited Engl ish proficient (LEP) (IDEA, 2004), (b) English as a second language learners (ESL) (Fitzgerald, 1995), and (c) culturally and linguistically diverse (CLD) (Cartledge, &Gar dner, & Ford, 2009). According to Silliman,

PAGE 15

15 Wilkinson, and Brea-Spahn (2004 ), there are marked diff erences in achievement between English language learners and those w hose first language is English. This is significant given the changing demographics in schools today. According to the NCES (Planty et al., 2009), in 2007, 20% (10.8 millio n) of children between the ages of 5 and 17 spoke a language other than E nglish at home, marking a 2% increase from the year 2000. Among this group, 25% of children spoke English wit h difficulty. Furthermore, 75% of children who spoke English with difficu lty spoke Spanish as their first language (approximately 2.1 million). This is reflected in Florida, as well, where 78.8% of students who speak English with difficulty speak Spanish. As the number of ELLs increases in schools, educators face more challenges because many ELLs experience academic failure (Klingner, Artiles, & Barletta, 2006; McCardle, Mele-MacCarthy, Cutting, Leos, & DEmilio, 2005), performing below level in reading achievement tests (Bravo, Hiebert, & Pearson, 2006). According to the Survey of the States Limited Engl ish Proficient Students and Ava ilable Educati onal Programs and Services 2000-2001 Summary Report (Kin dler, 2002), only 18.7% of ELLs across 41 states scored above state norms on measur es of reading comprehension. According to Snow et al. (1998), Hispani c students are twice as likely as non-Hispanic Whites to perform below level on measur es of reading. Furthermore, these gaps are observed early on and tend to persist across the years. In order to help these student s catch up to their higher performing peers, schools need to implement effective reading interven tions at an early stage (Cartledge et al., 2009; Denton & Mathes, 2003). The existing lit erature in the field of reading has highlighted the importanc e of developing effective reading programs that include critical

PAGE 16

16 reading elements, like phonol ogical awareness, phonics, fl uency, vocabulary, and comprehension (National Read ing Panel, 2000; Snow et al., 1998). In particular, instruction in phonemic awareness and phonics becomes crucial because they play a key role in the early stages of reading dev elopment among native speakers of English (Adams, 1990; Ehri, 2004) and English l anguage learners (Ehri & Roberts, 2006; Helman & Burns, 2008). When students begin to read, they need to develop the ability to analyze and manipulate the sounds in speech, as well as to recognize the lettersound correspondences so that they can ef fectively decode unfamiliar words in print (Ehri, 2004). Furthermore, students must devel op the ability to read words accurately and automatically in order to achieve fl uent reading (Stahl, 2004). When students are able to read words automatically, they are ab le to focus their attention on higher order skills like comprehension (Adams, 1990; LaBerge & Samuels, 1974). This connection between word reading skills and reading comprehension has been well documented in monolingual readers (Gough, 1996; Stanovich, 1990; Vellutino, Scanlon, & Tanzman, 1994) and language minority student s (Hoover & Gough, 1990; Lesaux, Koda, Siegel, & Shanahan, 2006). Recently, the NLP (August & Shanahan, 2006) reported that word reading skills develop in similar ways between native spea kers of English and ELLs. Evidence of these similarities came from research studies that compared native speakers and ELLs on different measures of word reading skills, including pseudoword reading (Chiappe, Siegel, & Gottardo, 2002; Chiappe, Siegel & Wade-Wolley, 2002; Geva et al., 2000; Jackson & Lu, 1992; Limbos & Geva, 2001; Verhoeven, 1990, 2000; Wade-Woolley & Siegel, 1997). For example, Chiappe et al. (2002) found that ELLs and native speakers

PAGE 17

17 of English in kindergarten had similar perform ance on measures of letter identification and word reading and that by the end of fi rst grade no differences existed on measures of word and pseudoword reading, despite ELLs initial lower oral language proficiency. In another study, LeSaux and Siegel (2003) found that the va riables that accurately identified struggling be ginning readers who are native speak ers of English, also served to identify struggling ELLs. Since ELLs and na tive speakers of English learn to read in similar ways (Ehri & Roberts, 2006; Helman, 2009), Ehris explanation on how children learn to read can serve as a theoretic al foundation on which to build reading interventions. What follows is a description of Ehris theory of word reading development, which delineates how children learn to read words by sight (Ehri, 2005b). Ehris Phases of Word Reading Development According to Ehri (2005a), a skilled r eader is able to read isolated words and words in text accurately and quick ly. There are four ways in which words are read (Ehri & Snowling, 2004): (a) decoding (using grap ho-phonemic correspondence to identify unfamiliar words); (b) analogy (recognizing familia r words that have similar spellings to known words), (c) prediction (predicting a wo rd based on text cues like initial letters, surrounding words, pictures, etc.), and (d) si ght (automatically recognizing words that have been read before). Developi ng the ability to read words by sight is advantageous since it allows readers to focus their att ention on meaning (Ehri, 2005a). This ability to read words is acquired gradually. According to Ehri and McCormick (1998) readers go through five phases of word reading dev elopment, each one characterized by the learners understanding and use of the alphabetic system (p 140). These phases are: (a) pre-alphabetic, (b) partial-alphabetic, (c) fu ll-alphabetic, (d) cons olidated-alphabetic, and (e) automatic-alphabetic (Ehri & McCormick, 1998).

PAGE 18

18 Pre-Alphabetic Phase In this phase, learners read words based on nonalphabetic visual cues (Ehri & Snowling, 2004) that ar e unrelated to the sounds in the words. Arbitrary associations are made between visual aspects of the written word and the spoken form of a word, or between the written form and it s meaning (Ehri & McCormick, 1998). For example, the child who encounters a stop sign recognizes t he red, octagonal shape, and says stop, and the child who recognizes the familiar swirl of the Coca Cola logo says Coke. Yet, when the visual cues are taken away, the child is unable to read the words. This happens because in this phase children hav e limited knowledge of the alphabetic system. Therefore, they are unable to make connections between the letters and the sounds in words when they are trying to read (Ehri, 2005b). Since students rely on ineffective reading strategies, they are unabl e to read connected text (Ehri & Snowling, 2004). According to Ehri and Snowling (2004), this phase is char acteristic of children in preschool or kindergarten who have not received reading instru ction. Yet, older students with significant reading difficu lties might also be reading at a pre-alphabetic level (Ehri & McCormick, 1998). Partial-Alphabetic Phase The partial-alphabetic phase or phonetic cue reading Eh ri & Wilce (1985) is characterized by the partial knowledge of the alphabetic system. Students possess an incipient knowledge of letters and sounds, and use this knowledge to guess words. At this time, children are unable to use decoding or analogy as reading strategies (Ehri & Snowling, 2004). Furthermore, their phonemic s egmenting ability is still developing, so they are unable to identify all the s ounds in words (Ehri & Snowling, 2004). Consequently, when they encounter a word, only partial letter-sounds connections are

PAGE 19

19 made (Ehri & McCormick, 1998). For this reason, this phase has also been called rudimentary alphabetic (Ehri & Snowling, 2004). Some common reading behaviors characteristic of this phase include (a) writing words using only salient sounds and leaving out middle sounds (Ehri, 2005b), (b) misreading words as other words that have similar letters (i.e., tall for tell), (c) usi ng letter names to spell words (Ehri, 1989), (d) misspelling words that have sounds not included in the names of the letters (i.e., /g/ as in go), and (e) processing only partial grapheme-phoneme connections to learn sight words (i.e., remembering word book by t he /b/ and /k/ sounds) (Ehri & Snowling, 2004). Phonetic cue reading is characteristic of children in kindergarten and first grade, once letters are learned and reading instructi on begins (Ehri, 2005b). Yet, as with the pre-alphabetic phase, older st udents with learning disabilitie s might also be reading at this level (Cardoso-Martins, Rodrigues, & Ehri, 2003; Ehri & McCormick, 1998). While phonetic cue reading is more re liable than visual cue reading, it is not sufficient for skilled reading (Ehri, 1991, 1998). Full-Alphabetic Phase The full-alp habetic phase or cipher r eading (Gough & Hillinger, 1980) typifies students who have a full working knowledge of the alphabetic system, as well as phonemic segmentation ability. Learners are able to decode new words and learn sight words by using complete grapho-phonemic cues (Ehri & Snowling, 2004). With the ability to represent full sight words in memory, reading words by analogy becomes possible. Initially, word reading is slow and laborious because children are working to segment and blend all the sounds in the word. It seems that readers ar e glued to print (Chall, 1983). Yet, as students have opportunities to practice decoding and learning sight words, their reading becomes more fl uent (Ehri & McCormick, 1998). At this time,

PAGE 20

20 the ability to spell words correctly shows a marked improvement. Using their knowledge of the alphabetic system and their ability to identify sounds in word s, students are able to spell words more accurately (Ehri,, 1997, 2001). Thus, at this point children are acquiring the foundation for skilled reading and writ ing. However, for th is to take place, children need to receive systematic phonemic awareness and phonics instruction (Ehri & Snowling, 2004). Typically, this phase ta kes place in first grade, when children develop their alphabetic knowledge. A ccording to Ehri and McCormick (1998), once children master cipher reading, they are able to advance into the next phases of development. Consolidated-Alphabetic Phase The consolidated phas e or ort hographic phase (Ehri, 1991) initiates during the full alphabetic phase when children use their knowledge of grapheme-phoneme relations to read larger word units (Ehri & Snowling, 2004). As they enc ounter common letter patterns, readers recognize and store them as consolidated units. These larger units include rimes, syllables, morphemes, and root words (Ehri, 2005b). As children are able to read and store these larger units in memory, they gain accuracy and speed. They are able to remember multi-letter combinat ions, which helps decode and remember multisyllabic words (Ehri & Snowling, 2004). As children practice reading words in connected text, their word reading becomes automatic, facilitating comprehension (Ehri & McCormick, 1998). Readers in this phase of development are commonly found in second grade, but can also be found in hi gher grade levels among struggling readers (Ehri & Snowling, 2004).

PAGE 21

21 Automatic Phase Children who are considered proficient readers are typi cally in the automatic phase. According to Chall (1983) profic ient readers are able to read familiar and unfamiliar words with automatic ity and speed. At this point in time, readers have a copious sight vocabulary and multiple reading strategies that help them recognize most words in text and effectively decode uncommon words (Ehri & McCormick, 1998). Readers fluent word recognition allows them to allocate their attention on meaning when they are reading connected text. The Role of Instruction in Promoting Word Reading Development While Ehris phases describe the process in which norm ally developing students acquire the ability to read, it is important to recognize that not everyone moves through these phases with ease. According to Ehri and McCormick (1998) one reason why struggling readers fail to advance to the next phase is lack of adequate instruction. In general, instruction for childr en at risk for readi ng difficulties needs to be explicit, intensive (i.e., more time, smaller in structional groups), and supportive (e.g., scaffolding) (Torgesen, 2002). In addition, inte rvention needs to be provided in a timely manner to prevent reading di fficulties from getting worse (Lonigan, 2006; Stanovich, 1986; Torgesen, 1998). In relation to Spanish-speaking English language learners, there are two factors that researchers have considered when designing effective reading interventions: (a) language of instruction (Vaughn et al., 2006), and (b) level of oral English proficiency (Ehri & Roberts, 2006). Regarding language of instruct ion, many researchers believe that ELLs should receive instruction in their native language because it facilitates the acquisition of literacy skills by taking advantage of students oral langu age development (Francis,

PAGE 22

22 Lesaux, & August, 2006). In bilingual programs, students first learn to read in their native language before introduci ng literacy instruction in English (Tabors & Snow, 2002). According to the NLP (Francis, LeSau x, & August, 2006), supporters of nativelanguage instruction base their claims on evidence that (a) bilingual students can transfer procedural and declarative knowledg e from their first la nguage (L1) to their second language (L2), (b) bilingualism does not affect learning in either language, and (c) reading proficiency in ch ildrens native lan guage can predict reading proficiency in the second language. A synthesis of the literature on lang uage of reading instruction conducted by Slavin and Cheung (2005) and Gr eene (1997) reported posi tive results for bilingual programs compared to English-only reading program s. It is important to recognize, however, that many ELLs do not have access to native-language instruction due to geographic, political, and economic constr aints (Kelly et al., 2008; Goldenberg, 2008). Some reports show that approximately 60% of ELLs in the nation are receiving reading instruction in English (August, 2006; Goldenberg, 2008), while others report that up to 85% are in mainstream classroom s (Schirmer, Casbon, & Twiss, 1996). Furthermore, the percentage of students that have access to Spanish instruction has declined in recent years from 40.1% in 1993 to 20.4% in 2003 (August, 2006). Fortunately, the NLP (August & Shanahan, 2006) reported that ma ny of the instructional components that are effective fo r monolingual learner s appear to be effective for ELLs, and that those who have no access to first-language support can succeed when highquality instruction is provided in their second language (Snow, 2006). Regarding level of oral language profic iency, some researchers and educators believe that when students are learning in Eng lish, literacy instruction should be delayed

PAGE 23

23 until students develop sufficient English l anguage proficiency (E hri & Roberts, 2006; Tabors & Snow, 2002). One reason for this view is that students who have limited English proficiency lack the necessary vocabulary to understand the words they read (Helman, 2009), and consequently, are not able to make sense of texts (Peregoy & Boyle, 2005). The problem with this approach is that by the time interventions are implemented, children are already far behind and will need more intensive interventions to catch up to their peers. Fo rtunately, there is evidence that ELLs who participate in English reading interventions respond positivel y, despite having limited levels of English oral proficiency (Gunn et al., 2000, 2002, 2005; LeSaux & Si egel, 2003; Quiroga, Lemos-Britton, Mostafapour, Abbott, & Bern inger, 2002; Roberts, 2003; Stuart, 1999). Many of these studies implemented reading programs that had proven to be effective with monolingual str uggling readers (i.e., Proactive Reading, Reading Mastery, Corrective Reading, Reading Recovery). In some cases, researchers implemented the programs as they were designed (Gunn et al., 2000, 2002, 2005; Al Otaiba, 2005) and sometimes they modified the original vers ion to include activities and strategies designed to support the language needs of ELLs (Linan-Thompson et al., 2003; Vaughn, Linan-Thompson, Mathes et al., 2006; Vaughn, Mathes et al., 2006; Vaughn, Cirino et al, 2006). In all cases, results showed that students with differ ent levels of oral language proficiency made significant gains on measures of phonol ogical awareness, decoding, oral reading fluency, and passage comprehension. Theref ore, there is no compelling reason why literacy instructi on needs to be postponed until students develop sufficient oral English profic iency, and there is substantial support for providing early intervention (Ehri and Roberts, 2006; Gunn et al., 2006; Linan-Thompson et al., 2003).

PAGE 24

24 Purpose of the Study The findings from previous research prov ide guidance on how to effectively serve Spanish-speaking ELLs who are being instruct ed in English and w ho are struggling with reading. Unfortunately, the lit erature on effective intervent ions is still scarce (Gunn et al., 2000; Shanahan & Beck, 2006; Snow, 2006). There is a great need to identify other English int erventions that have proven to be effective with struggling monolingual students and evaluate how effective they ar e in promoting the reading skills of developing readers who are ELLs. Therefore, t he purpose of this study is to evaluate the effectiveness of a modified version of the University of Florida Literacy Initiative (UFLI) tutoring program with second-grade Spanish-speaking ELLs who are struggling with reading. The effectiveness of the pr ogram was examined through a multiple baseline across groups design. The findings fr om this study will not only contribute to the literature on effective r eading interventions, but also provide more options for teachers who are struggling to support ELLs as they teach them to read in English. In Chapter 2, a critical anal ysis of the literature on Engl ish reading interventions for struggling Spanish-speaking ELLs will be pr esented. Next, Chapter 3 will provide a detailed description of the met hodology used in this study. This is followed by Chapter 4, which provides a detailed description of the results obtained. Finally, a discussion of findings will be addressed in Chapter 5, along with an overview of limitations, implications for future research and implications for practice.

PAGE 25

25 CHAPTER 2 REVIEW OF THE LITERATURE Developin g the ability to read is vital for a child to succeed academically and in life, yet learning to read is not an easy task. M any young learners have di fficulties acquiring the skills necessary to read with fluency and to comprehend what they read (Gunn et al., 2000, 2005). English language learners ar e at particular risk for experiencing reading difficulties (Snow et al., 1998). In recent years, reports have showed that the majority of ELLs have received reading instruction in English (August, 2008; Goldenberg, 2008). Unfortunately, reports also show that this group of students is performing below average on measures of readi ng achievement (Kindler, 2002). As the number of ELLs in schools across the nati on continues to rise, particularly Spanishspeaking ELLs (Genesee, Lindhol m-Leary, Saunders, & Christian, 2005; Kindler, 2002), the need to identify effective English readi ng interventions becom es prominent (Vaughn, Linan-Thompson, et al., 2006). What follows is a critical synthesis of the literature on English reading interventions for Spanishspeaking ELLs who are struggling to read. The studies selected were identified th rough an extensive review of three databases: PsycInfo, EBSCO, and Google Scholar. The following key words were used in different combinations: English languag e learners, Spanish-s peaking, bilingual learners, second language learners, reading in terventions, supplemental interventions, and early interventions. In addition, a hand search of the latest issues in relevant journals in the field of educ ation was conducted (e.g., Reading Research Quarterly The Elementary School Journal Learning Disabilities Research and Practice ), as well as a ancestral search using key articles and reports (e.g., August & Shanahan, 2006; Vaughn, Linan-Thompson, et al., 2006). Of the available literat ure, studies that met the

PAGE 26

26 following criteria were selected: (a) at l east 40% of the partici pants were Hispanic or spoke Spanish as their first language; (b ) participants were considered English language learners or limited English proficient; (c) participant s were at risk for reading difficulties or performing below level of measures of reading ability; (d) included a supplemental reading intervent ion separate from the core reading program; (e) instruction was given in English; and (f) the intervention had a strong focus on foundational reading skills (e.g., phonological awareness, phonics, fluency). Only peer reviewed articles were included. A total of 18 studies met the inclusion criteria. The studies selected were grouped by type of inte rvention: (a) Reading Recovery, (b) Direct Instruction (Core Intervention Model, Proac tive Reading, Reading Mastery/Corrective Reading, and Read Well), and (c) other reading interventions (word reading interventions and comprehensiv e reading interventions). First, each study is described, addressi ng the purpose of the researchers, participants, intervention, dependent measures and results. Each section ends with a summary and analysis of findings from that particular type of interventions. The chapter ends with a general discussion of findings across studies, along with implications for research. Reading Recovery Reading Recovery (RR) is an intensive early intervention program designed to help first graders who are having difficultie s in reading and writ ing (Shanahan & Barr, 1995), and who have been identified by their teachers as the lowest reading achieving in their classrooms (N eal & Kelly, 1999; Swar tz & Klein, 1995). The main goal of this program is to accelerate the rate of progress of struggling reade rs in order to help them catch up to their cla ssmates (Clay, 1993b). The program lasts from 12 to 20

PAGE 27

27 weeks and provides daily, 30-minute individu alized lessons. Each lesson covers seven reading and writing activities that are individually des igned based on students daily progress: (a) rereading of two or more familiar books, (b) independent reading of previous days new book while the tuto r observes and records miscues, (c) letter identification tasks and/or word work breaki ng and making, (d) writing a story the child has created while listening and recording the sounds in words, (e) reassembling a cutup story created by the student, (f) intr oducing a new book, and (g) reading the new book using learned strategies (Center, Whendall, Freeman, Outhred, & McNaught, 1995; Clay, 1993b). Participants are selected based on two conditions: teachers recommendations and results from An Observation Survey of Early Literacy Achiev ement (lowest 20%; Clay, 1993a). The survey includes six measures related to early reading and writing: (a) letter identification (identify upper and lower case letters), (b) word test (read a list of 20 frequently used words), (c) concepts about pr int (tasks related to book reading, including directionality, concepts about letters and words), (d) writin g vocabulary (write as many words as possible in a ten-minute period), (e) hearing and recording sounds in words (writes a sentence read by the tutor) and (f) text reading level (highest leveled book read with a 90% or higher accuracy) (Clay, 1993a; Swartz & Klein, 1995). Students are discontinued form t he program when they show accelerated progress, when their scores on the Observation Survey are within the average range for first grade, and when they demonstr ate independent use of st rategies, becoming selfsustaining learners (Clay, 1993b; Neal & Kelly, 1999; Shanahan & Barr, 1995). In Reading Recovery, children who receive sixty or more lessons or who are discontinued

PAGE 28

28 with less than 60 lessons are referred to as p rogram children (Swartz & Klein, 1995, p. 3). Reading Recovery was developed by Ma ry Clay in the mid-70s and adopted nationwide in New Zealand in 1983. Various evaluations in New Zealand provided support for the effectiveness of the program in accelerating students progress (Clay,1985, 1987; Glynn, Croo ks, Bethune, Ballard, & Smit h, 1989). The program was brought to the United States by researchers at The Ohio State University in 1984, and since its inception, over 1.7 millio n students have been served around the nation (National Data Evaluation Center [NDE C], 2008). Research conducted by program developers in the United Stat es has found evidence of the effectiveness of the program in helping struggling readers (D eFord, 1995; Pinnell, Lyons, DeFord, Bryk, & Seltzer, 1994; Schwartz, 2005). A recent 3-year independent evaluatio n of the program conducted by the What Works Clearinghouse (2007) also reported positive effects on alphabetic skills and general reading achievement. This evaluation also reported potentially positive results on fluency and co mprehension. According to the NDEC (2008), in the year 2006-2007 almost 100,000 students were served. From these, 73% of the ones who successfully completed the program reached average levels in reading and writing. The rest were re ferred for further evaluation. In response to the growing number of E nglish language learners who do not have access to instruction in their native langu age, some researchers have evaluated the extent to which Reading Recovery produces positive gains when inst ruction is provided in English. Findings from a study conduc ted in England showed no significant differences between monolingual English s peakers and ELLs on a ll subtests of the

PAGE 29

29 Observation Survey after participating in RR (Hobsbaum, 1995). In the United States, two studies examined the impact of RR on English language learners (Ashdown & Simic, 2000; Neal & Kelly, 1999). Neal & Kelly (1999) reported data for students receiving RR in English and Descubriendo La Lectura (Spanish version of RR; Escamilla, 1994). Yet, in this review, onl y data for students participating in English interventions were addressed. What follo ws is a description of these studies. Ashdown and Simic (2000) evaluated the effectiveness of Reading Recovery on the reading achievement of struggling first grade native and non-native English speakers, from 37 Reading Recovery sites aff iliated with the New York University. Data from a six year period (1992-1998) were obtained from the Reading Recovery Data Sheet of the National Data Evaluation Center Three types of participants were selected based on their language status: (a) native E nglish speakers, (b) non-native speakers with limited English proficiency LEP, and (c) fluent non-native speakers ESL. English proficiency was determined by results from a language proficiency test, when available, or teacher judgment, as reported in t he RR national questionnaire (NDEC). These participants were then assigned to one of thre e groups based on their performance in the Observation Survey of Early Literacy Achievement : (a) RR group (natives = 20,863, LEP = 1814, ESL= 2924), representing the lowest 20% of first graders who received RR intervention; (b) comparison group (natives 8845, LEP = 995, ESL = 1427), representing at risk-students who did not re ceive intervention; and (c) random sample group (natives = 15,595, LEP = 731, ESL = 2037), representing the top 80% of first graders in Reading Recovery classrooms. A ll students received core reading instruction in English. The majority of non-native s peakers spoke Spanish as their first language

PAGE 30

30 (54% of LEP and 74% of ESL). Analysis sh owed that 63% of st udents in the RR group successfully exited the program. Chi-square analysis found significant differences in success rates among the language groups, point ing out that fluent ESL students had a higher success rate (66.3%) than native spea kers (62.2%) and LEP students (61.7%). Further analysis demonstrated that all gr oups, regardless of language background and English proficiency had equal op portunity to receive the fu ll Reading Recovery program (at least sixty lessons). Nevertheless, non-native speakers (ESL and LEP) were significantly underrepresented in the RR gr oup compared to native-English speakers and this may be a consequence of teacher bias in the selection process. The authors suggest that some educators might delay in tervention until non-native speakers develop their English skills or might select students who have higher probabilities of benefitting from the intervention, in this case, native speakers. To compare the reading achievement at the end of first grade for all groups, an analysis of variance was conducted with la nguage (native, ESL and LEP) and sample group (RR, comparison, or random sample) as fixed factor s, and Text Reading Level as the dependent variable. The analysis showed significant interaction between sample and language, with mean differences in the expected direction. The differences between the three language groups varied significant ly among the three samples, with nativespeakers outperforming the nonnative groups. Yet, the di fferences between language groups in the RR sample were much sma ller. The gap between LEP and ESL in the RR sample was narrower than those in the co mparison and random sample groups. ESL and LEP students in the RR sample signif icantly outperformed other ESL and LEP students at risk who did not receive intervent ion (comparison sample). Furthermore, the

PAGE 31

31 LEP students in RR outperformed the LEPs in the random sample, who were in the higher 80% of their classrooms (random sample ). The authors concluded that RR is an effective intervention for native and non-native speakers, independent of their level of English proficiency. Prov iding intensive, one-on-one in tervention in English helped promote the reading outcomes of non-native speakers who did not have access to bilingual programs. Additional support for the use of RR with ELLs comes from the work of Neal and Kelly (1999). They conducted a study with fi rst grade ELLs receiving Reading Recovery. The study compared data from the years 1993 to 1996 in the state of California. In this review, data are reported for three groups t hat received instruction in English. For analysis, the authors reported findings for three groups: (a) RR-ELL (n = 3992), comprising English language lear ners receiving instruction and intervention in English; (b) RR-English (n = 18,787), comprising all English-speaking chil dren participating in RR (including native speakers and ELLs); (c ) English random sample (typical first graders instructed in English, but not receiving RR), agai nst which the RR-ELL and the RR-English would be compared. The total number of participants for the random samples was not reported. Groups were compared based on three of the six subtests of the Observation Survey: writing vocabulary, hearing and recording sounds in words, and text reading levels. These three subtests were selected because they were considered valid indicators of childrens gr owth, with no ceiling effects. Results showed that RR groups made significant gains from preto postmeasures on the three subtests. Furthermore, students who were discontinued from the programs (reached average reading level compared to their classroom peers) outperformed the random

PAGE 32

32 sample (typical first graders) on all measures, while those who were not discontinued performed below their typically performing peers. It was al so estimated that 72% of children in the RR-ELL reached average perfo rmance; thus, they were discontinued from the program after an av erage of 67.69. These findings are comparable to the proportion of students in RR-En glish (75.2%) who were discontinued after 63.27 lessons on average. The author s concluded that RR was e ffective in helping at-risk children catch up to typical achieving first grader peers, regardless of language status. In regard to children who were not discontin ued, the authors stated that they performed at 8 to 10 text reading levels below childr en who were discontinued, an indication that they had not developed a system of literacy learning and t hat they would need more extensive interventions. An interesting finding related to not-discontinued children is that their progress in text reading was slower and that they stalled around a level five, equivalent to a pre-primer 2 level on basal series. This finding points out the need for more intensive interventions (Torgesen et al ., 2001). It is interesting to note that the studies that used RR with ELLs did not state making language accommodations, showing that the same intervention that worked for native speakers also worked for ELLs. In summary, these studies revealed that RR was effective in improving the reading outcomes of struggling first grade, Spanish-s peaking ELLs. In particular, students who were successfully discontinued from the program demonstrated average or above average performance compar ed to their first grade peers. These findings are comparable to the findings of studies that evaluated the effectiveness of RR with nativeEnglish speaking children ( Clay, 1985; Glynn et al., 1989; Pinnell et al., 1994). Of

PAGE 33

33 particular interest are the positive gains made by English language learners when instructed in English. While many researc hers highlight the benefits of using students native language during reading instruction (August & Hakuta, 1997; Greene, 1997; Krashen, 1991; Mortensen, 1984), it is certai n that many struggling readers do not have access to bilingual education programs (G oldenberg, 2008). These students are in need of effective early interventions that can s upport their literacy development in English. Reading Recovery can be considered a promis ing approach for this particular group of children. Notwithstanding, before any concluding stat ements are made about the effectiveness of RR, more research should be conducted. Particularly, it should address how effective is RR for different languag e groups and how students response to the intervention is impacted by leve l of English language proficiency. In the two studies that examined RR for ELLs, only Ashdown and Si mics (2000) reported data on the language composition of the sample. Yet, it did not disaggregate the data for each particular group. On the other hand, they reported that RR helped improve the reading outcomes of students with limited English proficiency as well as fluent English speakers. Despite the positive outcomes reported fo r RR, there are some areas of concern regarding the data collection proc edures, as well as the res earch designs. First, the two studies based their main analysis of student s performance on the Observation Survey (English and Spanish version), which was de signed by the develo per of the Reading Recovery program. According to Center and Wheldall (1992), the survey may be biased because it focuses on tasks that are explicit ly taught in the program, increasing the likelihood that students are identified as successful when t hey might not have acquired the necessary skills to thrive in the general classroom. Second, st udies did not report

PAGE 34

34 data on the long lasting effects of the inte rvention. There is a need for longitudinal studies that evaluate students ability to th rive once they leave the program. Third, none of the studies compared the effectiveness of RR to other individualized tutoring programs that address early reading skills. T hus, we cannot assure that RR is more effective for English language le arners than other available pr ograms. Fourth, no study included a control group with r andom assignment. This is a common criticism against Clays research design (Center & Wheldall, 1992). According to Nicholson (1989), by not having random assignment, progress on part of the lowest performing students may not be explained with certainty by participatio n in RR. It is possible that students who are performing at low levels at pre-test make progress even in the absence of an intervention. Fifth, according to Shanahan and Barr (1995), the empirical evaluations of this program have been limited to a set of unpublished technical reports. While this review did not include technical reports, it is important to note that both studies were published in Literacy Teaching and Learning a journal published by the Reading Recovery Council of North America. This calls attention to the need for independent evaluations of RR for Spanish-speaking English language learners, given its widespread implementation. Direct Instruction Direct Instruction (DI) i s an approach desig ned to accelerate student learning by the careful control of curriculum and inst ruction delivery (Engelmann & Engelmann, 2004; Marchand-Martella, Slocum, & Marte lla, 2004). There are three critical components of this approach: (a) organization of instruction (scheduling and material organization that promotes reading-engaged ti me); (b) program design (specifying clear objectives, teaching learning st rategies, clear instructional formats that specify what

PAGE 35

35 teachers and students will do, prov iding an optimal sequence of skills, provision of appropriate examples, and opportunities to pr actice the new skills); and (c) teacher presentation techniques (small group inst ruction, appropriate pacing, progress monitoring, teaching to mastery, motivation through high levels of student success (Carnine, Silbert, & Kameenui, 1990; Ma rchand-Martella et al., 2004). Various programs have been developed following the principles of Dir ect Instruction: Corrective Reading (Engelmann, Carni ne, & Johnson, 1988), Reading Mastery (Engelmann & Bruner, 1988), Proactive Reading or Earl y Interventions in Reading (Mathes & Torgesen, 2005), Core Intervention Model (Gerber & English, 2003), and Read Well (Sprick, Howard, & Fidanque, 1998). Recently researchers have evaluated the impact of these programs on the reading outcomes of Spanish-speaking ELLs. Core Intervention Model The Core Intervention Model (Gerber & E nglish, 2003) is based on principles of Direct Instruction (Carnine, Silber t, & Kame enui, 1990) and it focuses on the systematic reteaching of phonological skills through an exp licit correction process called correction staircase (Gerber, Jimenez Leafstedt, Villaruz, Richards & English, 2004, p.241). Correction staircase differentiates instruct ion by breaking down complex tasks into simpler steps and providing a scaffolding se quence of instruction to ensure that students provide the correct answer. Each s equence of instruction st arts with a supply question (e.g., What word rh ymes with cat?), where students construct and formulate an answer. If a student is unable to provi de an answer or makes a mistake, the instructor takes the student down the stair case to a less demanding question. In this case, a binary choice question is introduc ed, where students sele ct from the two options provided (e.g., What rhymes with cat, calf or bat?). If the student provides the

PAGE 36

36 right answer, the instructor moves up the staircase and ask the original question, allowing the student to answer it independently. If the student fails to answer the binary question, the next step down is to model-lead, where the instructor models and then elicits the right answer (e.g., Bat rhymes with cat. What word rhymes with cat?). If the student fails again, instructor moves down to the model-imitation step, where students are asked to imitate the correct answer (e .g., Say bat.). When students provide the correct answers, the instructor guides them back up the staircase until they answer the supply question again. Once students ans wer the supply question correctly, the instructor either asks a similar supply questi on to ensure that learning has taken place or moves to a different task (Gerber et al., 2004; Leafstedt, Richards, & Gerber, 2004; Richards, Leafstedt, & Gerber, 2006). Leafstedt et al. (2004) init ially tried the CIM model with a class of Spanishspeaking kindergarten students rece iving instruction in English (n=16) and found that students made significant grow th in phonological awarene ss and word reading after 10 weeks of intervention (15-minute sessions 2 times a week). The instruction incorporated a phonological awareness curriculu m, called Early Reading Project, which included developmentally sequenced skills (onset-ri me before phonemes) and tasks (identification before manipulation, and manipulation before production). Leafstedt and colleagues found that students ac ross ability levels responded to the CIM model, but that the lowest performing gr oup would have benefited from a longer intervention. While this model was used as a whole class intervention, Richards and colleagues (2006) decided to try the model as a supplement al intervention for Spanish-speaking students identified as struggling readers.

PAGE 37

37 The study conducted by Richards et al. (2006) examined the impact of the CIM model, following the developmental sequence of skills and tasks from the Early Reading Project. They examined the response of four at-risk Spanish-s peaking students in kindergarten to a 10-week intervention (15minute sessions, 2 times a week). Students were identified based on their low scores on the word identification and word attack subtests of the Woodcock Johnson-Achie vement (WJ-ACH). Based on a microgenetic methodology, used to analyze the process of change during a specific period of time (Siegler & Crowley, 1991), the authors collected weekly measures for fluency and strategy use. Fluency was examined with the nonsense word fluency and segmentation fluency measures of the Dynamic Indicators of Basic Early Literacy Skills (DIBELS). Strategy use was examined with three phonolog ical awareness tasks (e.g., onset, rime, and segmentation) from La Patera Phonologi cal Awareness Test (LP-PAT; Richards, Leafstedt, Gerber, Jimenez, & Filippini, 2003) Specifically, the instructors asked students a series of questions t hat allowed them to explain their responses (i.e., How do you know that? What letter do they st art with? How many sounds do you hear?). Strategies were coded based on a hypothes ized model of strategy development for kindergarten. Strategies were either coded as lower level (e.g., student says words sound the same, they repeat a word, or se gment one phoneme) or higher level (e.g., student says individual phonemes sound the sa me, segments two, three, or four phonemes). Results showed that, in fluency, 3 out of 4 students (responders) met midyear kindergarten benchmarks on nonword fluenc y and segmentation fluency. These students seemed to use more strategies from the higher level for all three phonological

PAGE 38

38 tasks. The fourth student progressed at a slower pace, failing to use higher level strategies (i.e., non-responder). Notwithstandi ng, the difference between the responders and the non responder was in timing, not in st rategy development. This means that with more time, the non responder would probably start using higher level strategies. Variation in response to explicit and system atic interventions has been reported by other researchers (Torgesen, 2000; Wanzek & Vaughn, 2008). This study not only provides additional support for the use of CIM as a promising phonological awareness intervention, but also highlights the importance of studying strategy use as an effective measure of response to intervention. The authors suggest replicating this study for generalization purposes. In replic ating, it is important to consider providing longer interventions to allow all students to develop reading strategies. Reading Mastery and Corrective Reading Reading Mastery is a reading progra m designed for beginning readers that provides explic it instruction in phonemic awareness, sound-letter correspondence, and decoding (Gunn, Biglan, Smolkowski, & Ary, 2000). Corrective Reading was designed for older students reading below grade level and struggling with basic decoding skills. It addresses reading accuracy (decoding), reading fluency, and reading comprehension (Smith, 2004). Following the principles of DI, both programs group students according to instructional needs, provide am ple opportunities to practice teach students to mastery, and provide frequent teacher modeling and feedback (Gunn et al., 2000). Research has found both programs to be effective in improv ing the reading outcomes of beginning and struggling readers (Campbell, 1984; Gregory, Hackney, & Gregory, 1982). Three studies conducted by Gunn and colle agues evaluated the effectiveness of Corrective Reading and Reading Mastery on the reading outcomes of Hispanic

PAGE 39

39 (Spanish-speaking) and non-Hispanic (European Am erican, English-speaking) students at risk for reading difficulties (Gunn, Biglan, Smolkowski, & Ary, 2000; Gunn, Smolkowski, Biglan, & Black, 2002; Gunn, Smolkowski, Bi glan, Black, & Blair, 2005). The first study by Gunn and colleagues (200 0) included a total of 256 students (158 Hispanics and 98 non-Hispanics) from K to 3rd grade. Students were screened in the spring of the year prior to intervention and were selected based on teachers rating of aggressive behavior and/or based on below grade level reading performance (158 selected through reading criter ia; 98 selected through aggre ssive/reading criteria). Students were matched and randomly assigned to one of two conditions: (a) Intervention including st ruggling beginning readers in first and second grade who received Reading Mastery and struggling thir d and fourth grade students who received Corrective Reading (73 Hispanics and 50 non-Hispanics); and (b) comparison including first to f ourth grade struggling r eaders who did not receive intervention (79 Hispanics and 45 Non-Hispan ics). The intervention students were tested three times: Time 1, in the fall of the fi rst year (year after screening), bef ore intervention began; Time 2, in the fall of the first year, after 6 to 7 months of intervention; and Time 3, in the fall of the second year, after 9 months of additional intervention. Students in the intervention condition partici pated in daily, small group instruction (1-3 members) for 25 to 30 minutes per lesson fo r 2 years. At the end of the first year of intervention (Time 2), analysis of variance showed that students in the intervention group significantly outperformed the compar ison group on measures of word attack (WJ-ACH), but not on word i dentification (WJ-ACH) or Or al Reading Fluency (Martson, 1989). No significant interaction between et hnicity and instructional condition were

PAGE 40

40 found. Effects due to ethnicity showed that Hispanics perform ed significantly lower than non-Hispanics on measures of oral reading fl uency. At the end of the second year of instruction (Time 3), students in the intervention group significantly outperformed the comparison group on four measur es of the WJ-ACH: word a ttack, word identification, reading vocabulary and passage comprehensi on. In addition, oral reading fluency approached significance (p<.056). Further anal ysis showed no significant differences between Hispanics and non-Hispanics on word ident ification, word attack, fluency, or passage comprehension. Non-Hispanics m ade greater gains in vocabulary than Hispanics. Additional analysis showed that Hispanics students who started the program with limited English proficien cy (n=19) benefitted as much as the rest of Hispanics. Gunn and colleagues analyzed the relationship between decoding, reading fluency, and comprehension. They found that improvement in decoding from Time 1 to Time 3 was associated to improvement in oral r eading fluency and passage comprehension. The strongest predictor of comp rehension in this study was oral reading fluency. The authors concluded that Corrective Reading and Reading Mastery were effective in improving the reading skills of Hispanic and non-Hispanic struggling readers, and that supplemental instruction is e ffective for ELLs regardless of language proficiency. Providing explic it and systematic instruction in alphabetic reading skills improved students decoding ski lls, which in turn improved fluent reading, and comprehension. Based on the greater gains made by students at the end of the second year, the authors state that providing longer interventions is necessary to make an impact on the reading skills of struggling readers.

PAGE 41

41 In a follow-up study, Gunn and colleagues (2002) examined the sustaining effects of Corrective Reading and Read ing Mastery one year afte r the intervention had ended (Time 4). From the original 256 students incl uded in the first study, the researchers collected data on 195 students. Analysis of vari ance showed that students that received the supplemental intervention conti nued to outperform the comparison group on measures of word attack and oral reading fl uency. Based on a significant interaction between condition and ethnicity, further analysis showed that Hispanics made significant gains in word attack, com pared to non-Hispanics. While no significant differences were found between conditi ons in relation to vocabulary and comprehension, the authors st ated that the effects approached significance for both. Analysis of the performance of Hispanics with different levels of English language proficiency showed no significant differenc e between students who spoke English at the onset of the intervention and students w ho did not. Yet, these results should be approached with caution because of the limited number of student data available related to language proficiency (n=16). The analysis also showed that Hispanics that received intervention made significant gains not only in word attack and oral reading fluency, but also in comprehension. These results provided additional support fo r the effectiveness of Reading Mastery and Corrective Reading as effective intervent ions for struggling readers, Hispanic and non-Hispanic. According to Gunn and colleagues, these findings go against a common belief that providing English language learners with English instruction is harmful. Quite the opposite, interventions that offer direct instruction, modeling, immediate feedback, and practice opportunities are very helpful in supporting their reading skills. In their

PAGE 42

42 concluding remarks, Gunn and colleagues pointed out that while students made significant gains after 2 years of intervent ion, they continued to perform below grade level. They hypothesize that reducing the duration of each lesson to 30 minutes instead of the recommended 40 to 50 minutes could have had a negative impact on the outcomes. In 2005, following the same format as the first study, Gunn and colleagues compared once again the effects of the two reading programs on the reading skills of kindergarten through third grade Hispanic and non-Hispanic stude nts at-risk for reading difficulties. This time, they included a larger sample (n= 299; 159 Hispanics and 140 Non-Hispanics) and an extended period of analysis (2 years of intervention and a 2 year follow-up). Students were matched and randomly assigned to one of two groups: (a) intervention (Reading Mastery for first and second graders and Corrective Reading for third and fourth graders) and (b) comparison group (struggling first to fourth graders with no intervention). In this st udy, the authors conducted an addi tional assessment, Time 5, which was conducted two years a fter the intervention had ended. Results demonstrate that at the end of the intervention, students who received the two reading programs made faster gains in letter-word identification (WJ-ACH) and Oral Reading Fluency than the comparison group Significant differences between intervention and comparison groups were also found for word attack, reading comprehension, and vocabulary. Two years after the intervent ion was over, significant differences between conditions were found for letter-word identification, oral reading fluency, and reading comprehension, with vocabul ary falling just below the chosen .05 alpha level. The authors found once again that Hispanics benefited from the intervention

PAGE 43

43 as much as non-Hispanics, and that language pr oficiency did not affect their response to the intervention. Furthermore, the progr ams continued to produce sustaining effects two years after the intervention had ended. In contrast to the prev ious study, Gunn and colleagues found that those w ho participated in the intervention approached national average levels on word identification (42nd percentile) and exceeded national levels on word attack (53rd percentile). Nevertheless, their performance in vocabulary and comprehension continued to lag behind the national standar ds (18th and 25th percentiles, respectively). To make an impact, the authors suggest that early interventions should emphasize t hese areas to a greater degree. All three studies provide support for the use of Reading Mastery and Corrective Reading as supplemental reading intervent ions for struggling readers that speak English as their second language. The tw o programs not only improved students outcomes at the end of the intervention, but produced sustained effects one and two years after the intervention was over. Gunn and colleagues also made evident the benefit of providing longer interventions to improve the outcomes of students at risk for reading difficulties. In addition, these studies highlight the import ance of not delaying reading interventions until second language learners develop their oral English language skills. English language learners can benefit from instruction provided in English that explicitly addresses critical early reading skills, like decoding, reading fluency, and comprehension. Proactive Reading Proactive Reading or SRA Early Interv entions in Reading is a small-group intervention program derived fr om DI (Carnine, Silbert, & Kameenui, 1990), designed to help struggling firstand second-grade st udents become competent readers (Mathes,

PAGE 44

44 Denton, Fletcher, Anthony, Francis, & Sc hatschneider, 2005). The program has 120 40minute lessons devised to promote phonemic awareness, phonetic decoding, reading fluency, and comprehension (Jordan, 2006). Participants complete daily lessons (6-10 short activities) in small homogeneous groups. For every activity included, instructors model the new content (very little new c ontent addressed in each lesson), facilitate guided and independent practi ce, monitor students progre ss, and provide immediate positive and corrective feedback (Mathes et al., 2005). The program has a specific scope and sequence of tasks to pr omote the gradual acquisition of knowledge and skills (e.g., phonetic elements in isolation before applying them in words, decoding words in isolation before applying them to connected text). Movement from one lesson to the other is dependent on mastery of content (1 00% accuracy on an activity) by every student in the group (Mathes et al., 2005; Jor dan, 2006). A typical lesson addresses five reading components: (a) phonemic awareness: phoneme segment ation, blending, and discrimination (from initial s ounds, to final and medial so unds, to consonant blends); (b) letter knowledge: saying and writing le tter-sound correspondences (a new letter-sound correspondence introduced every 2 to 3 days) ; (c) word recognition: sound out regular and irregular, high frequency words (moving from CVC words to more complex, multisyllabic words); (d) connected text fl uency: repeated reading of decodable text to improve rate and accuracy (2-3 times; c horal, paired, and individual reading); (e) comprehension: before reading activities (prediction, activation of background knowledge, establishing purpose for reading) and after reading activiti es (story-retell, sequencing, summarizing, story grammar elements; Jordan, 2006; Vaughn, Cirino et al., 2006; Vaughn, Mathes et al., 2006).

PAGE 45

45 Proactive Reading was first examined wit h monolingual Engl ish-speaking students with reading difficulties and was found to im prove many critical reading elements. Mathes et al. (2005), for exampl e, found that after two years of intervention, first grade students who received Proactive Reading si gnificantly improved in measures of phonological awareness, word reading accuracy, word reading fluency, word attack, word identification, spelling, and reading flue ncy. Yet, no significant differences were found in measures of reading comprehension. Four studies have examined the effect s of Proactive Reading on the reading outcomes of Spanish-speaking ELLs. Three studies implemented a modified version of Proactive Reading (Linan-Thompson, Vaughn, Prater, & Cirino, 2006; Vaughn, Mathes et al., 2006; Vaughn, Cirino et al., 2006), and one study implemented Proactive Reading as it was originally designed (Al Otaiba, 2005). Linan-Thompson et al. (2006) and Vaughn, Cirino et al. (2006) also reported data for students receiving Lectura Proactiva (Spanish version of Proactive Reading). Fo r the purpose of this review, only data addressing English interv entions will be presented. The modified version of Proactive Reading had an additional 10-minute component addressing vocabulary and oracy development. According to PollardDurodola et al. (2006), the parti cipants in these studies woul d greatly benefit from this added component due to their limited vocabulary and listening comprehension skills. The vocabulary and oracy component included a five-day cycle of shared book reading and vocabulary review (Hickman, Pollard -Durodola, & Vaughn, 2004). On day 1, students participated in a five-step routine: (1 ) introduction of a story and three to four Tier 2 vocabulary words; (2) read-aloud of a passage from an informational text or

PAGE 46

46 storybook; (3) rereading of text focusi ng on vocabulary words; (4) focus on deep understanding of vocabulary kn owledge; and (5) summarization. On days 2-4, the instructor reviewed the words and passage from the day before, and introduced a new passage with new target words, following t he same format as day one. On day 5, instructors reviewed five words students had difficulty with, reread and processed the whole text (Pollard-Durodola et al., 2006). In one study, Vaughn, Mathes et al. ( 2006) examined how effective Proactive Reading (modified version) was in impr oving the reading outcomes of first-grade Spanish-speaking stud ents experiencing reading difficu lties. Proactive Reading was selected to match the students language of instruction used in the schools. Participants who scored below the 25th percentile on the letter-word identificat ion subtest of the Woodcock Language Proficiency Battery-Revised (WLPB-R) and failed to read more than one word from a five-word experimental reading list in English and Spanish were selected to participate. Based on their pre-test scores, students were matched and randomly assigned to either a treatment group that received the intervention (n= 22) or a comparison group that did not received the intervention (n = 19). The lessons had an additional piece: embedded language support activities (3-8) designed to facilitate comprehension of ta sks and concepts. These activities, supported by available literat ure on effective English language learning strategies, included instructional scripts with pi ctures, use of gestures and fa cial expressions, explicit instruction of English language use, among others (Gerst en & Baker, 2000). Results showed that after 7 months of intervention (average 96.55 hours of instruction), students in the treatment group significantly outperformed the comparison

PAGE 47

47 group on various English measures, including rapid letter-naming (CTOPP, effect size 0.88), phonological awareness composite (CTO PP, effect size 1.24), letter sound identification (untimed, effect size 1.10), word attack (W LPB-R, effect size 1.09), passage comprehension with cloze procedure (WL PB-R; effect size 1.08), as well as verbal analogies (WLPB-R; effect size 0.77) No significant differences were found in oral reading fluency (DIBELS), picture vo cabulary and listening co mprehension (WLPBR). Results for Spanish measures showed significant difference for phonological awareness composite (CTOPP; effect size 0.76), Spanish oral language (WLPB-R, effect size 0.01), word atta ck (WLPB-R; effect size 0. 87), and comprehension (WLPB-R; effect size 0.81), but not for oral reading fluency (DIBELS). The intervention was effective for Spanis h-speaking students l earning to read in English. Furthermore, transfer of skills to Spanish was obs erved as a result of the intervention. It is possible that transfer from English to Spanish is easier because Spanish has a more transpar ent orthography and because students already posses oral language proficiency. Of particular interest is the fact that student s responded to the program despite initial low reading and language pr oficiency in English and Spanish. This supports the conclusion reached by Gunn et al. (2005) to not delay reading interventions until students develop their oral language skills. Another important finding from this study is the significant gains made in passage comprehension. Many early intervention programs tend to improve foundational skills (word attack and word identification) but tend to have limited impact on comp rehension. Comprehension strategies like story retell and the inclusion of the vocabulary and oracy component may

PAGE 48

48 be responsible for the outcome s. Vaughn and colle agues affirmed that more gains are attained when English as second language strategies are used. In a follow-up study, Linan-Thompson, Vaughn, Prater, and Cirino (2006) examined the number of student s from the previous stud y (Vaughn, Mathes, et al., 2006) who benefited from the in tervention provided at the end of 1st grade and who continued to benefit at the end of 2nd grade. Follo wing the Response to Intervention (RTI) model, students who received resear ch-based interventions and made expected gains based on predetermined criteria were identified as responders and were expected to continue to thrive when the suppl emental intervention was removed. On the other hand, students who at the end of an inte rvention failed to meet the established criteria or made minimal gains were identified as non-responders and were considered to be at-risk for reading difficulties. Responders and nonresponders were identified based on their end-of-year scores on two subtests of theWLPB-R: word attack and passage comprehension. Students who scor ed above 85 on either measure were considered responders, and those who scored below 85 were considered nonresponders. Data were available for 39 par ticipants at the end of first grade and 29 at the end of second grade. Results showed that 91% of students met criteria at t he end of first grade and 94% met criteria at the end of se cond grade. No data were available related to the specific subtest in which students met criteria (wor d attack or passage comprehension, or both). This shows that the interventions were effect ive for the majority of students and that the gains were sustained one year after t he intervention had ended. Linan-Thompson and colleagues also found that m any students in the comparis on group reached criteria at

PAGE 49

49 the end of first and second grade, even though they did not receive supplemental intervention. Yet, the percentages were not as high as wit h the treatment groups. For example, only 42% of students met criteria at the end of fi rst grade and 44% at the end of second grade. The authors stated that the par ticipating schools were considered exemplary or recognized because of the high percentage of students passing the third grade state-level reading exams. Their core reading programs were effectively serving the majority of student s. Therefore, it was expected that m any students would benefit from core reading in struction itself. The low percentage of student in the comparison group that reached criteria at the end of se cond grade suggests that students learning in English may need early, intense and sustained interventions. A word of caution: one of the reasons why this study possibly showed higher percentage of responders is t hat the measures utilized we re untimed. According to Linan-Thompson and colleagues, fluency measures (timed) are some of the hardest to influence through intervention (Torgesen et al., 2001; Vellutino et al., 1996). It is recommended that future research includes timed and untimed measures to see if different results are obtained. Vaughn, Cirino, et al. (2006) wanted to replicate the findings obtained with the modified versions of Proactive Reading du e to the limited research available on effective interventions for English language learners. Following the same methodology and format as the first study (Vaughn, Mat hes, et al., 2006), Vaughn and colleagues evaluated the program with a different sample of Spanish-speaking first-grade students. As before, students were selected, matc hed, and randomly assigned to one of two groups: intervention (n=43) or comparison (n = 48). Students in the intervention group

PAGE 50

50 participated in 50-minute lessons, five da ys a week, for 7 months (average of 115 sessions). Results showed that the interventi on group significantly outperformed the comparison group on letter-sound identification ( untimed; effect size 0.36), phonological awareness composite (CTOPP, effect size 0. 38), word attack (WLPB-R, effect size 0.42), word reading efficiency (TOWRE, effect size 0.41), oral reading fluency (DIBELS), and spelling (words spelled corre ctly from a list of 25 words in an experimental test, effe ct size 0.35). No significant di fferences were found for English language related measures, including listening comprehension, passage comprehension, picture vocabulary, and verbal analogies. This might be related to the extremely low English oral language sco res that this group had at pre-test. In sum, the study replicated the positiv e findings from the first study on many beginning reading skills, particularly in phonological awareness composite. Many significant gains were obtained, even though their level of performance was lower than those in the first study. It is important to note that the modified ve rsion of the programs had the vocabulary and oracy component to hel p students develop their vocabulary and oral language proficiency; yet, none of these were significant ly affected at the end of the intervention. Compared to the previous studies, passage comprehension was not significantly different from the comparison group. These mixed results suggest that further research is necessary before maki ng concluding claims about the need for the added component. While the previous studies focused on the modified version of Proactive Reading, Al Otaiba (2005) evaluated the effectiveness of Proactive Reading as it was designed,

PAGE 51

51 with no added vocabulary or oracy component. Using a case study design, the author studied the language and reading outcomes of eight ELLs (6 Spanish-speaking and 2 Arabic speaking). Of these, three students we re in kindergarten (2 Spanish and 1 Arabic), three in first grade (all Spanish), and two in third grade (1 Spanish and 1 Arabic). Students were identif ied as at high risk based on their limited progress in reading. Students participated in 15 sessi ons over a period of 10 weeks. Al Otaiba conducted t-tests on vocabular y (PPVT-R), sound matching (CTOPP), as well as word attack, word identific ation, and passage comprehension (Woodcock Reading Mastery R). For the comparisons, raw scores and standard scores were used. Results showed significant gains on ra w scores for sound matc hing, word attack, and passage comprehension, as well as gains in standard scores for word attack. The author reports a standard score gain of .38 on word attack, .18 on wo rd identification, and 0.30 on passage comprehension for each session. The response to intervention was not affected by the students English language proficiency. The author observed variation in the responses, where some made sharp gains, while others made slower progress. All of this was obtained in just 15 lessons, compared to the previous studies that lasted 7 months and had an average of 115 lessons. This investigation provided initial support for the use of Proactive Reading in its or iginal version as a supplemental program for struggling English language learners. In summary, the studies described support for Proactive Reading as a promising intervention for struggling Spanish-speakin g English language learners. Many basic reading skills were positively impacted by the intervention, particularly phonological awareness, word attack, word reading effici ency, letter-sound identification. Initially,

PAGE 52

52 Vaughn, Mathes, et al. (2006) attributed t he positive comprehension outcomes to the added vocabulary and oracy component. These findi ngs were contrary to what Mathes et al. (2005) found for monolin gual English-speakers. Not withstanding, the studies with English language learners did not include a condition where the original and the modified version of the program were co mpared. Therefore, the improvement in comprehension cannot be solely attribut ed to the added component. This is of relevance, since the replication study by Vaughn, Cirino et al. (2006) did not find significant differences in passage compr ehension for either program. Moreover, the added component, with its emphasis on vocabulary development, failed to produce significant gains in picture vocabulary. These results suggest the need for further research, including various comparisons to determine what elemen ts of the program have a greater impact on oral and reading outcomes, and what elements need to be intensified. For example, t he authors suggested focusing on fluency, which is critical for future academic achievement. Read Well Read Well (Sprick, Howard, & Fidanque, 1998) is a reading program for beginning readers in kindergarten and in first grade, as well as for struggling readers in need of remediation (Wahl, 2007). The program provides small group instruction focusing on the explic it and systematic teachi ng of English decoding, as well as sustained practice in reading of decodable texts, di scussion of vocabulary, and concepts in the text (Denton, Anthony, Parker, & Hasbrouck, 2004). Each lesson typically lasts 30 minutes (15 minutes on decoding and 15 minutes on story r eading). Lessons are organized by units that are thematically based. Each lesson starts with a new letter sound, which is related to the words in decoding activities, which, in turn, are related to the stories read. In each

PAGE 53

53 lesson, students read connected text, where sounds that have been previously learned are practiced. One particular characteristic of this program is the ty pe of texts that are read: duet stories and solo st ories. In duet stories, te xts have decodable segments for the students, as well as more advanced segm ents for the teacher. The solo stories include fully decodable texts that are r ead by the students (Wahl, 2007). A study conducted by Jitendra et al. (2004) evaluated the effectiveness of Read Well with diverse learners (e.g., learning disabilities, attenti on-deficit disorder, second-language learner) and found positive effects on reading, spelling, and comprehension for most students. Two studies have evaluated its effectiveness as a supplemental intervention program for struggling ELLs (Denton et al., 2004; Santoro, Jitendra, Starosta, & Sacks, 2006). The first study conducted by Denton et al. (2004) examined the effectiveness of two tutoring programs, Read Well and a revis ed version of Read Naturally (Hasbrouck, Inhot, & Rogers, 1999), on the reading achiev ement of Spanish-speaking bilingual students. A total of 93 students from sec ond to fifth grade with similar reading performance were selected based on the schools standardized tests and teachers recommendations. To be included, students had to show adequate oral English proficiency and basic reading proficiency in their native language as measured by the Language Assessment Scales Oral in Englis h and Spanish, as well as the Language Assessment Scales Reading and Writing. Students were then assigned to one of two programs based on their performance on the word attack subtest of the Woodcock Reading MasteryR. Those who scored below grade one (e.g. students in need of decoding instruction) were assigned to Read Well, and those who scored equal or

PAGE 54

54 above grade one (e.g. students with established decoding but needing practice in fluency and comprehension) were assigned to Read Naturally. Within each program, students were matched and randomly assig ned to one of two conditions: treatment (Read Well = 19; Read Naturally = 32) or co mparison group (not receiving intervention; Read Well = 14; Read Naturally = 28). Read Well and Read Naturally were not compared with each other, just with their comparison groups. Student s in the treatment conditions participated in small group (1-4 members), 40-minute sessions, 3 times a week, for 10 weeks (average of 22 sessions). To address the specific needs of Spanishspeaking students, the Read Well program focused on the differences between Spanish and English language in their activities. The Read Naturally program modified the lesson to include vocabulary instruction (di scussion of meaning for two words in each passage), decoding (teaching one high fr equency word per passage and teaching of unknown words), and compr ehension (through discussion). At the end of 10 weeks, Denton and colle agues compared students outcomes on three subtests of the Woodcock Readi ng Mastery Test-Revised (WRMT-R): word attack, word identification, and passage comprehension. They found significant differences between Read Well treatment and comparison groups only on word identification (WRMT-R), whic h showed the impact that a small amount of explicit English phonics instruction can have on struggling English language learners, particularly when attention is called to the differences between their first and second language. On the Read Naturally condition, no significant differences were found between treatment and comparison groups on any of the subtests. The authors speculated that in the Read Well condition, students did not have sufficient time to

PAGE 55

55 develop the reading automaticity needed to a ffect comprehension. They also suggested that the lack of systematic vocabulary t eaching might have hindered their ability to comprehend what they were reading. T hus, they recommended providing longer interventions and including a systematic vocabulary component when working with English language learners. In relation to the comprehens ion outcomes of the Read Naturally condition, Denton and colleagues also suggested the need for systematic vocabulary instruction. In addition, they st ated that increased fluency could have had an adverse effect on English l anguage learners, who might have needed additional time to process what they were read in English. Notwithstanding, they did not provide any evidence to support this claim. Little research has been conducted on the fluency skills of English language learners (Antunez, 2002), but available data on the fluency skills of struggling readers have demonstr ated that improved fluency is related to improved comprehension (Samuels, 1979). Most of th e studies dealing with the comprehension skills of English language learners point to the critical role of vocabulary (August, Carlo, Dressler, & Snow, 2005; Carlo et al., 2004; Gersten & Baker, 2000). Therefore, investigating the effect of word reading ski lls on fluency outcomes should be addressed in future studies. The second study, conducted by Santoro, Jitendra, Starosta, and Sacks (2006), examined the impact of Read Well on the reading skills of second-grade English language learners, two of which were Hispa nic and spoke Spanish as their native language. Using a multiple probe across par ticipants design, Santoro and colleagues were able to make evident the functional relationship between the intervention and the reading outcomes. The length of intervention varied for each participant. As specified in

PAGE 56

56 the program, each lesson lasted an average of 30 minutes and was provided individually by a trained tutor. Results demonstrated that the positive effect of Read Well was replicated across participants. The Test of Oral Reading Flue ncy (Martson, 1989) was used to report the number of correct words per minute read in a grade-level passage. All students reading rate increased from baseline to interventio n to follow up (2 weeks later). Lourdes, a Spanish-speaking student, comp leted 14 units in 14 weeks. She went from 36 cwpm at baseline to 61 at intervention, to 79 at follow up. Juan, the other Spanish-speaking student, completed 5 units in 7 weeks. He w ent from 2 cwpm at baseline, to 10 at intervention, to 21 at follow up. Despite the fact that both started at different levels, Read Well supported the development of fluency skills in both students. Similar patterns of improvement were reported for phoneme segmentation fluency, letter naming fluency, letter sound fluency, and non word fl uency (DIBELS). Preto posttest measures of word attack (WRMT-R) show ed gains for Lourdes (from 71 to 84) and Juan (39 to 53). Yet, measures of Word Identification (WRMT-R) did not show improvement from either student. This showed that wh ile Read Well was effective in supporting students ability to decode (Word Attack), it was not sufficient to impact students ability to read real words. In the passage co mprehension subtest of the WRMT-R, Lourdes went from 95 to 103, while J uan went down from 53 to 48. This can be attributed to Juans difficulties with reading fluency. T he passage comprehension requires the child to read a short passage and identify a key word that is missing from the passage. Thus, Juan might not have reached the level of aut omaticity necessary to understand the passages and select an appropriate answer. The authors suggested t hat the vocabulary

PAGE 57

57 and comprehension components of the program did not provide connections to students background and personal interests, and that this might have negatively impacted some of the outcomes. The studies presented showed mixed resu lts about the effectiveness of Read Well on the reading performance of Spanish-spea king English language learners. While Denton and colleagues (2004) found that only t he word identification skills of students receiving Read Well were significantly hi gher than those in a comparison group, Santoro and colleagues (2006) found that the two Spanis h-speaking students improved in oral reading fluency, non word fluency, phoneme segmentation fluency, letter naming fluency, letter-sound fluency and word attack, exc ept for word identification. Denton and colleagues stated that the short duration of their intervention may have limited the impact of the program. Probably, with longer interventio ns, researchers can more effectively measure the effectiveness of Read Well. In regard to comprehension, Denton and colleagues, as well as Santoro and colleagues agreed that including systematic teaching of vocabulary can improve the co mprehension outcomes of English language learners. More research is needed to evaluate the value of this program. Multiple Direct Instruction Interventions A study by Kamps, et al (2007) compared the effect of three Direct Instruction interventions (Read Well, R eading Mastery and Early Interventions in Reading or Proactive Reading) with an English as a second langua ge/balanced literacy intervention. The curriculums targeted phonem ic awareness, letter sound recognition, decoding, fluency and comprehension through ex plicit instruction, repeated practice, and small group instruction (3-6 students). The ESL/balanced intervention focused on word study, group and individual reading, and wr iting activities through small group

PAGE 58

58 instruction (5-12 students). A total of 230 first and second grade students receiving supplemental interventions participated in the study. Two groups were formed for comparison purposes: a) experimental group (n=117, 84 ELLs and 33 English speakers), representing at -risk students receiving one of the three supplemental programs mentioned above; and b) comparison group (n=113, 60 ELLs and 53 English speakers), representing at-ri sk students participating in the English as a Second LanguageESL/balanced literacy appr oach. The majority of ELLs in the sample were Spanish speaking (58%), with the rest speaking Somalian, Sudanese, or Vietnamese. The authors did not report t he total number of sessions that students participated in each of the programs, but reported pre-test data from the fall semester and post-test data from the spring semester. Analysis of students performance showed greater gains for students participating in the direct instru ction interventions. Specifically, an analysis of variance repeated-measures test showed that on nonword fluency (NWF; DIBELS), there were significant differences between experimental and comparison groups (in favor of experimental), but not between En glish-speakers and ELLs. This means that ELLs and English-speakers responded similarly to direct instruction interventions. In Oral Reading Fluency (DIBELS), there were significant differences between experimental and comparison group, as well as between English-speakers and ELLs. Pairwise comparisons showed that the diff erences were between English-speakers and ELLs in the comparison group, and not in t he experimental group. This finding further supports the positive effect of direct inte rventions on both language groups. Additional analysis (ANOVA) focusing solely on ELLs found significant differences between experimental and comparison gr oups. Specifically, those in the direct instruction

PAGE 59

59 condition showed larger gains from preto post-interventions in first grade on NWF (effect size .879) and second grade on ORF (e ffect size .947). Simi lar findings were reported for three subtests of the Woodcock Reading Master y Test (word attack, word identification, and passage comp rehension). At the end of firs t grade, students in direct instruction interventions showed larger gains in word attack (effect size 1.78), word identification (effect size 1.54), and pass age comprehension (effect size 1.04). At the end of second grade, they obtained larger gains in word identification (effect size 1.39) and passage comprehension (effect size 1.35) The authors also report that ELLs participating in direct instruction interventions demonstrated a faster rate of growth than those in ESL/Balanced approach in the firs t grade. This difference was no longer significant in second grade, but the authors state that 16 out of 19 students in the ESL/balanced approach had received direct instru ction in first grade, thus improving their performance in second grade. Kamps and colleagues concluded that providing supplemental, small group instruction is important for at-risk Engl ish language learners; and that programs like Read Well, Reading Mastery, and Early Interventions in Reading are effective first grade interventions. The authors suggested that students who participated in the ESL/balanced intervention did not make si gnificant progress due to the lack of systematic phonemic and phon ics instruction. They also speculate that the large group size had a negative impact on their performance. Other Reading Interventions This section addresses studies that focu sed on specific reading int erventions, but that were not identified as following the Direct Instruction approach. The section is

PAGE 60

60 divided in two sections. The first section addresses word reading interventions, and the second section addresses comprehensive reading interventions. Word Reading Interventions Word reading skills p lay a key role in beginning reading. According to Ehri and Snowling (2004), the ability to read words with accuracy in isolation or in context is crucial for skillful reading. Struggling readers take longer time to read words and make more mistakes because they have diffi culty making phonemegrapheme connections (Ehri, 2004).To be able to read words accurately and fluently, readers need to develop phonological awareness and decoding skills (G unn et al., 2005). The following study evaluates a reading intervention that focuses on word reading skills at the word and text level. Quiroga, Lemos-Britton, Mostafapour, A bbott, and Berninger (2002) examined whether an English intervention addr essing phonological aw areness, phonics, and fluency improved the reading outcomes of Spanish-speaking first-grade students learning to read in English. Participants we re identified based on their low scores (2 standard deviations below the mean) on the word attack and word identification subtests of the Woodcock Reading Mastery Test-R (WRMT-R). Each student received a total of 12 individual sessions over a 6-week period (30-minute sessions, 2 times a week). Each lesson included phonological awar eness activities in English and Spanish, as well as reading instruction at the subword, word, and text level. At the beginning of each session, students completed a short phonological awareness lesson from the Spanish Phonological Awareness Training Program (Lemos-Britton & Mostafapour, 1997). In the first six sessions, instruction focused on syllable segmentation, and on the last six sessions, on phoneme segmentation. Next, they participated in English sound

PAGE 61

61 games where students counted syllables an d phonemes in words selected from the text to be read later on the session. A fter that, students learned English phonemegrapheme correspondences from the Talking Letters Program. Using a set of 92 correspondences cards (summary of the spel ling unit together with pictured words that contain the spelling units), st udents were exposed to the spe lling units four times in the course of the intervention. Next, student s practiced the learned phoneme-grapheme correspondences in single words and c onnected text through repeated readings. Throughout the lessons, instructors monitore d for comprehension by asking students what they liked about the text or by predict ing what would happen next. However, this was not an area of focus for the intervention. At the end of the interv ention, the group made significant gains in word identification, but not in word attack. I ndividual scores showed that all participants improved their real word reading skills, while only six improved pseudoword reading skills. Furthermore, their performa nce in word reading at the end of the intervention was considered low average, which is more t han expected based on their low levels of oral language proficiency. Thus, the provision of an empirically-sound intervention that addressed phonological awareness in both languages, phoneme-grapheme correspondences in English, and repeated readi ng of connected text was effective in improving the reading outco mes of Spanish-speaking st udents who performed way below their peers. The authors stated that reading interventio ns should not be delayed because of limited language profic iency. On the contrary, st udents should learn to read while they continue to develop oral proficiency. Additional support for this intervention is

PAGE 62

62 needed, particularly studies that include fluen cy measures to see if the skills targeted also promote fluent reading. Comprehensive Reading Interventions Following recommendations from t he NRP (2000) and the NLP (August & Shanahan, 2006), the next four st udies inc luded reading interv entions that incorporated comprehension instruction in addition to word reading skills. The studies also examined the effect of increasing the in tensity of the intervention. So me studies reduced the group size, while others increased t he amount of intervention. Vaughn, Linan-Thompson, Kouzekanani et al. (2003) evaluated the effectiveness of an English intervention on the reading out comes of struggling second-grade English language learners. A total of 77students (39 monolingual English speakers, MES; 38 English language learners, ELL) were assigned to one of three groups: (a) 1:1 (15 ELLs and 12 MES), who received one-on-one intervent ion; (b) 1:3 (15 ELLs and 14 MES), who received instruction in groups of 3; and 1:10 (8 ELLS and 13 MES), who received intervention in groups of 10. Participants we re originally select ed based on teachers observations of failed reading or students low performance on the second-grade statelevel Texas Primary Reading Inventory (T PRI). The ELL group included all Hispanic students, while the MES group included 17 African Americans, 3 Whites, and 19 Hispanics who spoke English as the primary language. The students participated in a total of 58 sessions over a 13-week period (30 minutes, 5 times a week). Each session incl uded five components: (a) fluent reading (6 minutes), through repeated reading of familiar text in pairs and individually; (b) phonological awareness (6 minutes), oral activities focusing mainly on phonemic awareness selected from Ladders to Literacy (OConnor, Notary-Syverson, & Vadasy,

PAGE 63

63 1998) and Phonemic Awareness in Young Chil dren: A Classroom Curriculum (Adams, Foorman, Lundberg, & Beeler, 1998); (c) inst ructional-level reading (10 minutes), including reading of a new book while t eacher provided support in decoding and modeled comprehension strategies through think-alouds, while students provided the main idea or a summary of the text at the end of the reading; (d) word study (6 minutes), including explicit instruction in phonem e-grapheme correspondences and word patterns, as well as a 1-minute writing activity where students had to write as many words as they could with an opportunity to correct their ap proximations afterwar ds; and (e) progress monitoring (20 minutes weekly) through letter naming, phoneme segmentation and nonsense word probes from DIBELS, as well as connected text reading from Read Naturally. Results showed that all gr oups made significant gains in passage comprehension (WRMT-R), phoneme segmentati on (DIBELS), and fluency (Test of Oral Reading Fluency) from post-test to follow-up (4 wee ks later). Furthermore, in comprehension, groups of 1:1 and groups of 1:3 outperformed gr oups of 1:10. No significant differences were found between groups of 1 and groups of 3. On phonem e segmentation (DIBELS) and reading fluency (Test of Reading Fluency), results were somewhat different. For example, students who were in groups of 1:1 significantly outperform groups of 1:10. No significant differences were found betw een groups of 1:1 and groups of 1:3, and between groups of 1: 3 and groups of 1:10. Vaughn and colle agues observed that 1:1 grouping was superior to 1:10 grouping in phoneme segmentation and comprehension for MES and ELLs. In fluency, MES performed better in groups of 1:1 than 1:10, while ELLs performed better on groups of 1:3 than groups of 1:10. Ov erall, groups of 1:1 did

PAGE 64

64 not outperform the groups of 1:3 on any measure, showi ng that both sizes were effective for this particular intervention. This finding is relevant because many schools have limited resources, so small group (group of 3) can be chosen over one-one-one instruction as a way to effectively se rve larger number of students. Vaughn and colleagues also stated that E nglish language learners benefited from these explicit and intensive intervention as much as monolingual English speakers, and that this finding is noteworthy due to the high number of ELLs who are struggling to read. Yet, one limitation identified by the authors was the lack of a com parison group that would have included struggling readers who did not receive the intervention. Without this comparison, the association between student gains and the program it self is somewhat limited. Further research is recommended in which the different components of the program are evaluated to see which ones make a bigger impact. A second analysis based on the same cohort included in the previous study was conducted by Linan-Thompson, Vaughn, Hickm an-Davis, and Kouzekanani (2003). This analysis included data for 26 English language learners. These data were disaggregated and further examin ed, including two follow-ups (4 weeks and 4 months). The purpose was to determine the effectiv eness of small group intervention (1-3 students) over time. As mentioned previous ly, students participated in a total of 58 sessions over a 13-week period (30 minutes, 5 days a week). The authors affirmed that many accommodations were made for English language learners, including the use of picture cards to exemplify the meaning of un known words, identification of real and nonsense words, provision of qu ick definitions of unfamiliar words, activation of prior knowledge, opportunities to practice oral skills during instructional level reading,

PAGE 65

65 compare/contrast of Spanish and English phonemes and word patterns, and ample opportunities for practice. Based on a series of analyses of varianc e (ANOVA) results showed that students made significant gains from preto post-test on word attack (WRMT-R), passage comprehension (WRMT-R), phoneme segment ation (DIBELS), and fluency (TORF), with fluency showing the greatest gains. Gains were sustained four weeks later on passage comprehension and phoneme segmentation, but not on word attack or fluency. From follow-up 1 (4 weeks) to follow-up 2 (4 months ov er the summer) students maintained their gains on word attack passage comprehension, but only fluency increased significantly. In addition, it wa s observed that student s responded to the intervention regardless of their level of English oral proficiency. These findings provide support for small group interventions that focus on the critical components of reading and that make accommodations for English l anguage learners. Of particular interest are students significant improvem ent in fluency and comprehens ion. Research has shown that comprehension is closely related to fluent reading (LaBerge & Samuels, 1974). Thus, it can be implied that students improved fluency at the end of the intervention facilitated their performance in passage comprehension. Another study, conducted by Vaughn, Linan-Thompson, and Hickman (2003), examined the same reading program descri bed above with a new sample of secondgrade students. In addition, they altered the duration of the intervention to accommodate the needs of students who failed to respond (increments of 10 weeks). A total of 45 students (35 Hispan ics, 6 Whites, and 4 African Americans) participated in the intervention. They were identified bas ed on their low performance in the screening

PAGE 66

66 portion of the TPRI (five or less words r ead correctly out of 8 possible words). All students started the intervention in groups of three. After 10 weeks of intervention, students were assessed to determine response to intervention based on preestablished exit criteria: (a) five or more words read corr ectly from the screening portion of the TPRI, (b) above 55 correct words per minute (cwpm) on second-grade passage from the Test of Reading Fluency (TORF) and (c) 50 cwpm on three consecutive measures of second-grade fluency from Read Naturally. At the end of 10 weeks, 10 students met criteria (early exit) and exited the program Students who did not meet criteria (n=35) were regrouped and received 10 more weeks of intervention. At the end of the next 10 weeks, 14 students exited the program (mid exit) and 21 regrouped and received 10 additional weeks with a modifi ed version of the program. The modified version was designed to meet the individual needs of t he students, including more assessment of basic skills, more time on word study and fluency, and less work on phonological awareness. The researc hers based these changes on observed improvement in phonological awareness and increased need for word attack skills. At the end of the next 10 weeks, 10 students exit ed the program (late exit), while the 11 who failed to meet criteria (no exit) were referred for further evaluation. Assessment took place in two ways: (1) preand post-test (after 30 weeks) measures using the WRMT-R (word attack and passage comprehension) and the CTOPP (composite score for phonological awareness and rapid naming), and (2) four measures of the TORF (before intervention and after each of the three 10-week periods). Results for the TORF showed that all students made significant gains from preto post-test (30 weeks) Effect sizes reported for all groups were large: 3.18 for

PAGE 67

67 early exit group, 2.84 for mid exit group, and 6.06 for late exit. The no-exit group made significant gains (effect size 2.66) despite not meeting exit criteria A closer look at student data showed that the no exit group started at a much lower place in fluency compared to the early-exit group (10.55 and 32.50 mean score, respectively). This meant that progress was obs erved even in the no-exit group, but based on the exit criteria established, students were identified as unresponsive. Fuchs and Fuchs (2007) recommend that a child be identified as unr esponsive when they fail to make adequate progress and when they score below the norm at the end of the intervention or fail to meet a stipulated benchmark. Results also showed that all groups made significant gains at the end of 30 weeks in word attack, passage comprehension, phonological awareness, and rapid naming with effect sizes ranging from 0.47 to 2.22. This showed t hat the intervention was as effective for English language learners (His panics) as for English language speakers (Whites and African Americans). In fact, Vaughn and colleagues report that all ELLs exited the program (6 in early exit, 6 in mid exit, and 3 in late exit). A closer look at language proficiency showed t hat students from the early -exit group were more proficient in English than those from mid and late exit and that might have facilitated their learning. Notwithstanding, these findings show that even students with low oral language proficiency can benefit from the program; they might just need more time. The findings from this study are of vital importanc e to the field of readi ng, particularly with the amount of students who continue to fail despi te intensive interventions (Torgesen et al., 2001). It seems that increas ing the intensity by providin g longer interventions is an effective way of supporting struggling readers.

PAGE 68

68 Wanzek and Vaughn (2008) evaluated the e ffects of a similar reading program, focusing on word reading skills, as well as comprehension. They also examined whether increasing the amount of interv ention had an impact on students who had been identified as low responders. This inve stigation was conducted over two years and focused on first grade struggling readers who had already failed to meet exit criteria at the end of a 13-week program. The exit criteria established for the original intervention was set at more than 30 correct sounds per minute on the nonsense word fluency and more than 20 words per minute on oral reading fluency subtests form DIBELS. The first year (Study 1) had a total of 50 students (21 from the interv ention group and 29 from the comparison group no intervent ion) and the second year (Study 2) had a total of 30 students (14 from the intervention group and 22 from the com parison group). The majority of students in the samples were Hispanics (36 in year 1 and 23 in year 2). In each study, students who were selected from the intervention group remained in the intervention group, and students selected from the comparison group remained in the comparison group. In study 1, the intervention group rece ived an additional 13 weeks of instruction with one 30-minute lessons a day (single dose) five days a week. In study 2, the intervention group received 13 weeks of interv ention with two 30-minute lessons a day, five days a week (double dose). Each lesso n included the following components: (a) phonics and word recognition (15 minutes), where students practiced letter names, letter sounds, spelling of regular and irregular words, word family patterns, and word building activities; (b) fluency (5 minut es), where students participated in fluency activities at the subword, word, and te xt level; and (c) passage reading and reading

PAGE 69

69 comprehension (10 minutes), where students read short passages that included the patters learned previously and responded comprehension questions. Results for study 1 showed no significant differences between intervention and comparison group on all pretest measures (oral readi ng fluency from DIBELS, and word attack, word identification, and passa ge comprehension from WRMT-R), except for nonword fluency from DIBELS (in favor of treatment group). At post test, no significant differences were found on any measure between intervention and comparison group. Furthermore, all student s scored below first grade end-of-year benchmark for oral reading fluency set at 40 correct words per minute. This finding demonstrated that the additional dose of the program was not sufficient for this group of low responding students. Results for study 2 showed no significant differences between groups on all pretest measures. At post-test, significant differences were found only for word attack, but data revealed that more students in the tr eatment group made gains on word attack, word identification and comprehension. Ye t, all students in the treatment group continued to score below grade level on endof-year oral reading fluency (40 cwpm), even though 50% of the students increased their fluency by 10 cwpm. Wanzek and Vaughn (2008) concluded that the additional dos es of intervention did not seem to effectively support the reading skills of low responders, and suggested that low responders might need a smaller group size, more intense instructional routines, or a different intervention. The last suggesti on is a very important point, since the intervention was not compared against ot her programs that might benefit low responders when given in varying amounts of time.

PAGE 70

70 In summary, comprehensive interventions that include word r eading instruction as well as comprehension instruction proved to be effective for many struggling English language learners, except for the program implemented by Wanzek and Vaughn (2008). Notwithstanding, the limited re sponse might be associated to the sample itself. Wanzek and Vaughn stated that in a response to intervention model, these students would be identified and considered for special educat ion. In addition, the studies reported provided mixed results about the use of mo re intensive interv entions with Spanishspeaking students through smaller group size or increased intervention time. For example, Vaughn et al. (2003) found that small group interventions (1:1 and 1:3) were more effective than large group interventions (1:10), and Linan-Thompson et al. (2003) found positive effects of small group interv ention when data were disaggregated for English language learners. On the other hand, provision of extended intervention proved very effective for students in the study conducted by V aughn, Linan-Thompson and Hickman (2003), but not for students in Wanzek and Vaughn s (2008) study. This variation in results is consis tent with previous findings (B erninger et al., 2002; McMaster et al., 2005; Vellutino et al., 1996) and only hi ghlights the need for further research. In particular, these studies should be replic ated using a comparison group receiving a different intervention program in order to determine whether or not the lack of response of some students is due to the intervention itself. Discussion The review of the literat ure included 18 studies of reading interv entions for Spanish-speaking ELLs learning to read in English. Overall, these studies showed that Spanish-speaking ELLs who are struggling to read in English can benefit from English reading interventions that address word reading skills. The following discussion

PAGE 71

71 addresses (a) instruction in word reading sk ills, (b) language and reading instruction, (c) strategies for English l anguage learners, and (d) implications for research. Instruction in Word Reading Skills While the ultimate goal of reading is co mprehension, research has highlighted the important role of word reading skills in beginning readin g, specif ically phonological awareness, decoding, and word reading fluency (Ehri & Snowling, 2004; Stanovich, 1990; Wagner, Torgesen, & Rashotte, 1994). Fluent word recognition is closely related to reading comprehension (Fuchs & Deno, 1992; Stanovich, 1990) because it allows readers to allocate resources to higher order skills (LaBerge & Samuels, 1974). Conversely, when word reading is slow and laborious, readers expend cognitive resources on decoding instead of deriving meaning from text (Ehri & Snowling, 2004; Gunn et al., 2000; LaBerge & Samuels, 1974). This is true for native English speakers as well as ELLs, both of whic h learn to read in similar wa ys (Lindsey, Manis, & Bailey, 2003). In addition to word reading skills, the NRP (2000) also suggests that any comprehensive reading program needs to address vocabulary development and reading comprehension. In this review, most intervention programs focused to a greater degree on word reading skills (e.g., phonologic al awareness, phonics, and fluency). The emphasis on vocabulary and comprehension in struction varied across studies, with some briefly addressing them in each lesson (Denton et al., 2004; Santoro et al., 2006), while others dedicating a significant amount of time to them in each lesson (Vaguhn, Linan-Thompson, & Hickman, 2003; Vaughn, Linan-Thompson, et al., 2006; Vaughn, Linan-Thompson, Kouzekanani, et al., 2003; Vaughn, Mathes, et al., 2006; Vaughn, Cirino, et al., 2006; ). Findings from these studies showed that an emphasis on word

PAGE 72

72 reading skills produced a positive effect in the reading outcomes of Spanish-speaking struggling readers. This was demonstrated by significant gains made on measures of phonological awareness, word attack, word identification, and fluency. Furthermore, improvements in word reading skills were also associated with improvements in reading comprehension (Gunn et al., 2000, 2005). In relation to comprehension, most studies that added a comprehension component reported a positive e ffect on that skill, including the modified version of Proactive Reading and the comprehensive r eading interventions described last (LinanThompson, Vaughn et al., 2003; Vaughn, Linan-Thompson, & Hickman, 2003; Vaughn, Linan-Thompson, Kouzekanani et al., 2003; Vaughn, Linan-Thompson et al., 2006; Vaughn, Mathes et al., 2006) Therefore, it can be c oncluded that addressing word reading skills should be a major component of any early reading interventions. Also, adding comprehension instruction to r eading intervention enhances the reading outcomes of ELLs. Language and Reading Instruction Two important issues constantly addressed in the literature concerning effective literacy instruction for Span ish-speaking ELLs are (a) language of instruction (Vaughn et al., 2006), and (b) level of Englis h oral proficiency (Ehri & Roberts, 2006). Many researchers state that reading development is facilitated when instruction is given in the students native language (Slaving & Cheung, 2005; Greene, 1997). Yet, national data show that approximately 60% of all ELLs are being instructed in English (August, 2006; Goldenberg, 2008). Since Spanish-speaking ELLs are performing below level on measures of reading achievement (Klingner et al., 2006), it is important to identify interventions that match the students l anguage of instruction and that effectively

PAGE 73

73 support students reading development in English. This review of the literature provides ample support for the use of English interventions. Across studies, ELLs improved on different measures of readi ng, including word reading skills (i.e., phonemic awareness, decoding), fluency, and comprehension. Interest ingly, students responded positively to interventions designed origin ally for monolingual Englishspeaking students, some of which included language accommodations (i.e., Linan-Thompson et al., 2006; Vaughn, Cirino et al., 2006; Vaughn, Mathes et al., 2006) and some of which did not (i.e., Al Otaiba, 2005; Kamps et al., 2007; Neal & Kelly, 1999). Regarding level of language prof iciency, there is a common view that in order to benefit from reading instructi on students first need to acquire oral English proficiency (Ehri & Roberts, 2006). In fa ct, many educational program s choose to delay reading interventions for ELLs until they develop suffici ent English oral skills (Gunn et al., 2005). This practice is extremely harmful for young ch ildren given that it takes at least 5 to 7 years for a student to becom e a fluent speaker of English (Collier, 1989; Cummins, 1981). According to Torgesen (1998), the cons equences of a slow start in reading become monumental as they accumulate exponent ially over time (p. 1). Statistics show that a gap already exists between ELLs and English native speakers when they enter school (Lee, Griff, & Donahue, 2007). Therefor e, delaying instruction only places ELLs at a higher risk for reading difficu lties. Fortunately, the stud ies reviewed indicated that delaying instruction is not necessary, gi ven that even students with limited English proficiency responded positively to reading instruction (i.e., Gunn et al., 2000; 2002; 2005; Linan-Thompson et al., 2006; Vaughn, Ma thes et al., 2006). According to Gunn and colleagues (2000), since limited English proficien cy does not limit students

PAGE 74

74 response to intervention schools can help ELLs succeed by intervening as early as first grade. Strategies for English Language Learners Despite the fact that many ELLs improv ed their reading skills without receiving any type of language accommodations, researchers recommend including strategies that support the language needs of Spanish-speaki ng ELLs (Vaughn, Mathes, et al., 2006). For example, Linan-Thompson et al. (2003) used picture card s and quick definitions to explain unfamiliar words, distinguished betw een real and nonsense words during word work activities, and contrasted Spanish and English phonemes and word patterns to help students distinguish betwe en the two. The interventions reviewed also provided other important elements of effective practice for ELLs: (a) explicit and systematic instruction (Helman, 2009), (b) ample opportuni ties for practice (Gersten & Geva, 2003), (c) small group instruction (Gersten et al. 2007), and (d) scaffoldi ng (Watts-Taffe & Truscott, 2000). According to Goldenberg (2008) other strategies for ELLs who are being instructed in English include (a) visual cues and physical gestur es, (b) clarification of words or passages to support comprehension, (c) providing opportunities to interact with the instructor and with peers, and (d) adjusting rate and complexity of speech to accommodate the language abilit y of the students. Implications for Research The findings from this l ite rature review provide an empirical foundation for the present study. Results showed that Spanish-speaking ELLs who are instructed in English and who are struggling to read can benefit from English reading interventions that target word reading skills. Furthermore interventions originally developed for monolingual struggling readers offer the kind of instruct ion that supports the reading

PAGE 75

75 development of ELLs with different levels of oral language proficiency. Due to the high percentage of Spanish-speaking ELLs receiving reading instruction in English, it is important to identify other English interventions that have already proven to be effective for monolingual struggling readers and evaluate their effectiveness with this population. In particular, it is important to cl osely examine how students respond to intervention. While many studies have shown significant gains for students who participate in intensive reading programs, almost every study reported that some students fail to respond and continue to struggl e in reading (Wanzek & Vaughn, 2008). According to Torgesen (2000), between 2% and 6% of students who participate in reading interventions continue to struggle afterwards. Torgesen suggested that these students may benefit from more intensive inte rventions. To make interventions more intensive, some researchers have decr eased the group size and/or increased the amount of instruction provi ded (Berninger et al., 2002; McMa ster et al., 2005; Torgesen et al., 2001), finding some positive resu lts. Vaughn, Linan-Thompson and Hickman (2003) found that when more intervention wa s provided and instruction was modified to meet the specific needs of each individual, students responded positively. In contrast, Wanzek and Vaughn (2008) found that double doses of intervention were not effective for low responders. Therefore, it is important to not only exam ine what interventions are effective, but also what interventions are effective for tre atment resisters and under what conditions (Torgesen, 2000). For this reason, single su bject design becomes a valuable research method because it allows researchers to identify the specific factors that either inhibit or enhance the efficacy of a particular intervention (Jitendra et al., 2004). Therefore, in the

PAGE 76

76 study presented here, a multiple-baseline across groups design was used to evaluate the effectiveness of a modified version of the University of Florida Literacy Initiative (UFLI) tutoring program in developing the word reading skills of Spanish-speaking ELLs in second grade who are struggling to read.

PAGE 77

77 CHAPTER 3 METHODS This study implemented a multiple base line across groups design to examine the effects of a modified version of the Universit y of Florida Literacy Initiative (UFLI) smallgroup tutoring program on the reading ski lls of Spanish-speaking English language learners in second grade, who are experiencing reading difficulties. The purpose of this chapter is to provide a detailed description of the methods used to complete the study. The chapter addresses the following components: (a) setting, (b) participants, including selection criteria, (c) assessment instruments (d) interv entions, (e) dependent variable, study design, and data analysis, (f) supplemental data, (g) interobserver agreement, treatment integrity, and social validity, (h) delimitations and limitations, (i) one-on-one tutoring methodology, and (j) pilot study data. Setting This study was conducted at Suwannee Elementary School in Suwanne e County School District in Florida. Suwannee County Sc hool District is a rural district located in North Florida and is composed of seven school s and one technical center. For the year 2008-2009, the racial/ethnic composition of t he district was approxim ately 73.9% White, 13.9% Black, 9.5% Hispanic, 0.5% Asian, 0. 3% Indian, and 1.8% mu ltiracial. Of these students, approximately 4% we re identified as English language learners and 1.8% as migrant students. The district had approximately 60.3% of students participating in the free/reduced price lunch program (FLDOE, 2010b). Suwannee Elementary is a Title I school that serves only second and third grade students. Approximately 73% of the schools students participated in the free/reduced price lunch program in the year 2008-2009. That same year the school had a total of

PAGE 78

78 778 students enrolled. Of t hese, 392 students were in second grade and 385 students in third grade (FLDOE, 2010c). The racial co mposition of the school was 64.9% White, 17.5% Black, 13.6% Hispanic, 0.8% Asian, 0.8% Indian, and 2.4% Multiracial. The school identified 7.6% of the students as English language learners and 2.3% as migrant (FLDOE, 2010a). According to the Fl orida Department of Education (FLDOE, 2009), the percentage of third grade students in each achi evement level of the FCAT reading for the school in the year 2008-2009 was as follows: 16% at Level 1, 11% at Level 2, 35% at Level 3, 31% at Level 4, and 6% at Level 5. FCAT reading data for ELLs identified 53% at Level 1, 24% at Level 2, 18% at Level 3, 6% at Level 4, and 0% at Level 5. Because Suwannee Elementary School ha d only second and third grades, there were more classrooms at t hese two grade levels than at a typical elementary school. According to school records, on the year 2009-2010, there were 21 second-grade classrooms with a total of 374 st udents. Six of these classrooms (12 to 16 students per class) had been designated for lower performi ng students, and all participants in the study were from these classrooms. The lower performing classrooms used a double dose approach to reading in an effort to hel p students catch up to grade level. This meant that these classroom s had 180 minutes of reading instruction daily. The school used the Harcourt StoryTown Core Reading Program (Beck et al., 2007), which is one of the approved programs for adopt ion in Florida because it meets the criteria outlined by the State. Within the180 minutes of reading instruction, teachers also provided 45 minutes of small group inte rvention using one of two programs: Harcourt Reading Interventions or SRA Early Interventions in Reading (SRA/EIR; Mathes & Torgesen,

PAGE 79

79 2005). Teachers had the option of selecting whic h program to use in their classrooms. All of the participants in the study received at least one of these interventions. All the teachers of the participating students had Flor ida ESOL endorsement on their teaching certificates. Participants The purpos e of this section is to provide a description of the participants, including selection criteria and selection process. As required by the University of Florida Institutional Review Board (IRB), the inve stigator obtained inform ed consent from the parents of children partic ipating in the study, as well as informed assent from each child. For copies of IRB documentation, including the study protocol, consent letters, and assent form, see Appendix A. The following criteria were used to select participants for this study: 1. The child was identified by the school as being an English language learner whose native language was Spanish. This criterion was established because the focus of the study is on the effects of the UF LI intervention with this population. 2. The child was in second grade. This criterion was established because the UFLI program recommends that intervention beginning during the fall should target second graders. First gr aders tend to need a semester of reading instruction before they are ready to benefit from UFLI tutoring. 3. The child was performing below level based on three measures of reading ability. Although teacher judgment wa s the initial identifying fa ctor, this criterion was established to ensure that the judgment of the teacher c ould be corroborated with assessment data. 4. The childs record for the past year showed regular school attendance and low percentage of tardiness. This criterion was established to ensure that the intervention could proceed without excessive interruption and to minimize participant attrition. The selection process proceeded accordi ng to several prescribed steps. School personnel (i.e., principal, reading coach, and teachers) were asked to identify second

PAGE 80

80 grade English language learners, whose nat ive language was Spanish, and who were performing below grade level based on their la test end-of-year Stanford Achievement Test-10 (SAT-10) and the Dynamic Indicators of Basic Early Literacy Skills (DIBELS) scores. A total of 24 students were identified. To corroborate that these students were still performing below grade level, teachers were asked to administer the Invented Spelling Assessment (Lane & Pull en, 2004). The spelling sheets were given back to the researcher to be graded. In order to prot ect the identity of the students, teachers assigned a number to each students spelling sheet. Based on these results, a total of 16 students who scored below the 40th percentile on Invented Spelling were selected to participate in the study. Consent forms were s ent by the school principal to the parents of these students asking their permission to conduct further assessment and to participate in the study. Up until this poin t, the researcher did not have access to students names. A total of 13 students received parental consent, but one of them was dropped from the study because his mot her stated that the child did not speak Spanish. The new pool of participants consisted of 12 students. Next, the researcher took running records from each child using Reading Recovery books to identify their instructional reading level at which students were reading with 90 % to 95% of accuracy. Out of the 12 students, one was dropped because he read a level 19 book with 96% reading accuracy, when the goal for the beginning of second grade is level 17. Thus, based on this criterion, the student was not in need of reading intervention and would not have benefitted from the study. Later, when groups were fo rmed, a second student was dropped from the group intervention because she could not be placed in any of the

PAGE 81

81 three reading groups. This student was readi ng at a book level that was too high for Group 2 and too low for Group 1. Still, she received the UF LI tutoring program on a oneon-one basis, but not as part of the multip le-baseline design. A description of the methods used with this student is provided at the end of this chapter, and results obtained are presented at the end of chapter 4. The final pool of students who received group instruction included 10 participants. Students data on group composition and initial book levels can be f ound in Table 3-1. A detailed description of the procedure used to identify initial book levels for eac h participant and the percentages of accuracy can be found in the assessment section, under Grouping Inst ruments. A summary of this procedure for each participant is also provided in Table 3-7. Table 3-1. Group composition Group number Number of members Males Females Beginning book level 1 3 1 2 9 2 4 1 3 2 3 3 2 1 1 What follows is a description of each par ticipants (a) demographic information, (b) instructional information, (c) reading scores used for selection purposes, (d) English and Spanish oral language pr oficiency, and (e) classroom attitudes and behaviors. Throughout this report of the study, pseudonyms are used to protect the identity of the participants. Demographic information was obtained from parents through a home language survey (Appendix B), teachers, and students when needed. This information includes age, gender, parent nationality, languages spoken at home, language used to communicate with child, year s of schooling in US, and parent al level of education. Demographic data can be found in Table 3-2. Instructional information was gathered

PAGE 82

82 Table 3-2. Summary of participants demographic information Name Age Gender Parents nationality Language(s) spoken at home Language used to communicate with child Years of schooling in US Parents level of education Amelia 78 F Mexico Spanish/ English Spanish/ English 3 years Not Available Pedro 8-8 M Cuba Spanish/ English Spanish 4 years Community College Jessica 8-3 F Mexico Spanish/ English Spanish 6 years No schooling Viviana 8-5 F Guatemala Spanish/ Conjoval Spanish 5 years No schooling Ernesto 8-10 M El Salvador Spanish Spanish 5 years High school Maggie 8-8 F Mexico Spanish Spanish 5 years No schooling Maria 8-8 F Mexico Spanish Spanish 5 years High school Jennifer 10-1 F Guatemala Spanis h Spanish 1 years No schooling Aldo 9-3 M Mexico Spanish/ English Spanish 4 years No schooling Jorge 9-5 M Mexico Spanish Spanish 5 years No schooling

PAGE 83

83 from teachers and school personnel (e.g., readi ng coach) about the grades in which students have been retained, the amount of readi ng instruction students receive daily, the core reading program covered in the classroom, additional reading interventions students received, language status (i.e., LEP /limited English profic iency, ELL/English language learner), LEP classes provided, migrant status, ESE classification, and ESE services rendered. A summary of this information can be found in Table 3-3. Reading scores are reported based on the latest scores from the end-of-year SAT10 and the DIBELS Phoneme Se gmentation Fluency (PSF), Nonsense Word Fluency (NWF) and Oral Reading Fluency (ORF) subtests. For students that were retained in second grade, only NWF and ORF scores are reported. Reading scores also included beginning-of-year Invented Spelling Assessm ent scores for all students (Lane & Pullen, 2004). These scores can be found in Table 3-4. English and Spanish oral language proficiency levels were obtained using t he Woodcock-Munoz Language Survey-Revised (WMLS-R; Woodcock, Munoz-Sandoval, Ru ef, & Alvarado, 2005). Language levels are reported in Table 3-5. Finally classroom attitudes and behaviors were rated by teachers using the Conners Abbreviated Teacher Rating Scale (C-ATRS). A summary of results can be found in Table 3-6. Refer to the asse ssment section later in this chapter for a detailed description on each of these measures. Amelia Amelia was a Hispanic female, aged 7 years-8 months, enrolled in second grade for the first time. Given that Amelias parents did not return the home language survey, the demographic information reported is based on data gathered from Amelia and her teacher. Amelia stated that her parents are from Mexico and they speak

PAGE 84

84 Table 3-3. Summary of partici pants instructional information Name Grades Retained Other reading interventions Migrant status Language status Received LEP classes ESE classification Received ESE services Amelia None HR No ELL/LEP No None N/A Pedro 2nd SRA/EIR No ELL No None N/A Jessica K SRA/EIR No ELL/LEP No None N/A Viviana K HR Yes ELL/LEP Yes (30 min/3 times week) Language disability None Ernesto K SRA/EIR HR No ELL/LEP No Language disability Speech therapy Maggie k, 2nd SRA/EIR No ELL/LEP As needed None N/A Maria 2nd SRA/EIR No ELL/LEP No None N/A Jennifer 2nd HR Yes ELL/LEP Yes (30 min/3 times a week) None N/A Aldo k, 2nd SRA/EIR Yes ELL/LEP Yes (30 min/3 times a week) Language disability Speech therapy Jorge k, 2nd SRA/EIR No ELL Yes (30 min/3 times a week) None N/A HR = Harcourt Reading Interventions, SRA/EIR = SRA Early Interv entions in Reading, ELL = English language learners, and LEP = l imited English proficient. For ESE services, N/A stands for not applicable. This was used fo r students that did not have an ESE classi fication. None was used when a student had an ESE classifica tion, but did not receive ESE services.

PAGE 85

85 Table 3-4. Participants reading scores used for selection purposes Name SAT-10 (percentiles) DIBELS/PS F (number of phonemes per minute) DIBELS/NWF (number of correct lettersounds read per minute) DIBELS/ORF (number of correct words per minute) Invented Spelling (percentiles) Amelia 42 56 (AA)67 (LR)48 (LR) 20-30 Pedro 35 89 (AA)68 (HR) 20 Jessica 46 36 (LR)52 (LR)57 (LR) <10 Viviana 9 49 (LR)49 (MR)26 (MR) <10 Ernesto 13 43 (LR)46 (MR)26 (MR) 10-20 Maggie 13 58 (LR)39 (HR) <10 Maria 15 58 (LR)23 (HR) 20-30 Jennifer 49 (MR)2 (HR) <10 Aldo 7 4 (HR)4 (HR) <10 Jorge 11 17 (HR)5 (HR) <10 Scores for SAT-10 and DIBELS correspond to the end of the previous academic year. For Amelia, Jessica, Viviana, and Ernesto, those scores correspond to the end of first grade. For the rest of the students, the scores correspond to the end of second grade, since they were retained the previous year. Those students do not have PSF scores because that test is not implemented at the end of second grade. No SAT-10 score was available for Jennifer. Invented spelling scores correspond to the beginning of the current year. Table 3-5. Students English and Span ish language levels based on the WMLS-R Name English CALP English language ability Spanish CALP Spanish language ability Amelia Level 3 Limited Level 3 Limited Pedro Level 3 Limited Level 4 Fluent Jessica Level 3 Limited Level 3 Limited Viviana Level2 Very limited Level 3 Limited Ernesto Level 3 Limited Level 3 Limited Maggie Level 3 Limited Level 3 Limited Maria Level 4 Fluent Level 3 Limited Jennifer Level 1 Negligible Level 3 Limited Aldo Level 3 Limited Level 2 Very limited Jorge Level. 3.5 Limited to fluentLevel 3.5 Limited to fluent CALP scores correspond to the level of cognitive-ac ademic language proficiency. CALP scores go from Level 1 (negligible English ability) to Level 6 (very advanced English ability). Spanish and English at home. According to Amelia, she uses both languages to communicate with her parents. Information about her parents level of education is unknown. School records showed that she had attended school in the US at least since kindergarten. Her teacher r eported that Amelia receiv ed the Harcourt Reading

PAGE 86

86 Table 3-6. Conners Abbreviat ed Teacher Rating Scale (C-ATRS) Name Restles or Overactive Excitable/ Impulsive Disturbs others Fails to finish things Constant Fidgeting Inattentive Easily frustrated Cries often Quick mood changes Temper outburst/ unpredictable behavior Amelia NA NA JL JL JL JL PM JL JL JL Pedro PM JL PM NA JL JL NA NA NA NA Jessica NA NA NA NA NA NA NA NA NA NA Viviana NA NA NA NA NA NA NA NA NA NA Ernesto VM PM NA JL VM VM JL NA NA NA Maggie NA NA NA NA NA JL NA JL NA NA Maria NA NA NA NA NA NA NA NA NA NA Jennifer NA NA NA NA NA NA NA NA NA NA Aldo PM PM JL PM JL VM NA NA NA NA Jorge PM JL PM VM VM VM JL NA NA NA For the C-ATRS, NA stands for not at all, JL stands for just a little, PM stands for pretty much, and VM stands for very much. Intervention program 45-minutes a day, five days a week. Amelia was identified by the school as being LEP (limited English profic ient) and enrolled in LEP classes, but her teacher stated that Amelia di d not receive these services. At the end of first grade, Amelia scored in the 42nd percentile on the SAT-10. Her DIBELS scores for the end of first grade were 56 on PSF (above average), 67 on NWF (low risk), and 48 on ORF (low risk). At the beginning of second grade, she scored in the 20th to 30th percentile on the Invented Spelling Assessment. Based on the WMLS-R, Amelia demonstrated limited English and Spanish oral language proficiency (Level 3). On the C-ATRS, her teacher reported that Amelia would get easily frustrated and tended to be a little distracted in class. She st ated that Amelia does not like to not know what to do. She also affirmed that Amelia withdraws if s he is unsure of something. One area of concern reported by her teacher on different occasions was that Amelia had no support from home. Her teacher also reported that Amelias mother had spent

PAGE 87

87 extensive time in jail, and attempts for wr itten or oral communica tion with her guardian tended to be unsuccessful. Pedro. Pedro was a Hispanic male, aged 8 years-8months, enrolled in second grade for the second time. His parents were from Cuba and they spoke Spanish and English, but only used Spanish to communica te with Pedro. Pedro s parents stated that their highest level of educati on corresponded to community college. They also reported that Pedro had had four years of schooling in the US. His current teacher stated that Pedro received SRA Early Reading Interventi ons (SRA/EIR) Level 1 (first grade) for 45 minutes a day, five days a week. The school considered Pedro to be an English language learner, but he no longer had an LEP status. He did not receive LEP classes. At the end of second grade from the previous academic year, Pedro scored in the 35th percentile on the SAT-10. His DIBELS scor es for the end of second grade were 89 on NWF (above average) and 68 on ORF (high risk). At the beginni ng of second grade of the current year, he scored in the 20th percentile on the Invented Spelling Assessment. Language scores based on the WMLS -R, showed that Pedro had limited English oral language ability (Level 3) and fl uent Spanish oral language ability (Level 4). According to teachers report on the C-ATRS Pedro is a good student but likes to talk to other students. The teacher also reported that he was a little bit impulsive and easily distracted. Jessica Jessica was a Hispanic female, aged 8 years-3months, enrolled in second grade for the first ti me. Her parents were from Mexico and they spoke English and Spanish. At home, they used Spanish to communica te with Jessica. Her parents reported not having any schooling experience. Jessicas parents stated that she had

PAGE 88

88 received six years of school ing in the US. School records showed that she was previously retained in kindergarten. In second grade, Jessica participated in the SRA/EIR program Level 1 (fir st grade) for 45 minutes a day five days a week. Jessica had an LEP status and the school reported that she was enrolled in LEP classes; yet, her teacher affirmed that she di d not receive those services. Reading scores for the end of first gr ade showed that Jessica scored in the 46th percentile on the SAT-10. Her DIBELS at the time were 36 on PSF (low risk), 52 on NWF (low risk), and 57 on ORF (low risk). At the beginning of second grade, she scored below the 10th percentile on the Invented Spelling Assessment. Language scores on the WMLS-R showed that Jessica had limited Eng lish and Spanish oral language ability (Level 3). On the C-ATRS, her teacher reported that Jessica did not have any attitude or behavior problems in class. Viviana. Viviana was a Hispanic female, aged 8 years-5 months, enrolled in second grade for the first ti me. Her mother was from Guatemala and they spoke Spanish and Conjoval at hom e. Conjoval is a Mayan or al language, with no written form. To communicate at home, they used Spanish. Her mother r eported not finishing high school. Her mother stated that Viviana had had five years of instruction in the US. School records showed that she was retained in kindergarten. In second grade, she received Harcourt Reading Interventions in sm all groups, five times a week, for 45 minutes. Viviana had both migrant and LEP status. She received LEP classes 3 times a week for 30 minutes. She was also identif ied as having a language disability. Her teacher reported that Viviana did not receive ESE services.

PAGE 89

89 Reading scores showed that at the end of first grade, Viviana scored in the 9th percentile on the SAT-10. Her DIBELS scores for the end of first grade were 49 on PSF (low risk), 49 on NWF (moderate risk), and 20 on ORF (moderate risk). At the beginning of second grade, she scored below the 10th percentile on the Invented Spelling Assessment. Language scores on the WMLS-R demonstrated that she had very limited English oral language ability (Level 2) and li mited Spanish oral language ability (Level 3). The teachers report on the C-ATRS sh owed that Viviana had no attitude or behavior problems in the class. Her teacher stated on many occasions that Viviana was a very sweet girl, very respectful and that she always smiled. Ernesto. Ernesto was a Hispanic male, aged 8 years-10months, enrolled in second grade for the first time. He was previously retained in second grade. His parents were from El Salvador and they only spok e Spanish at home. The highest level of education that parents reported having was high school. Ernesto had had five years of schooling in the US. His t eacher reported that he rece ived the SRA/EIR program Level 1 in small groups, for 30 minutes a day, five days a week. He also received the Harcourt Interventions small group program fi ve times a week for 45 minutes. Ernesto had an LEP status but did not receive LEP classes, despite records showing that he was enrolled in them. He was identified as having a language disability and he received speech therapy two times a week for 20 mi nutes. Ernesto was also referred to the school psychologist to be tested for autism, but result s are still pending. At the end of first grade, Ernesto scored in the 13th percentile on the SAT-10. His DIBELS scores for the end of first grade were 43 on PSF (low risk), 46 on NWF (moderate risk) and 26 on ORF (m oderate risk). At the beginnin g of second grade, he

PAGE 90

90 scored between the 10th and 20th percentile on the Invented Spelling Assessment. Language results on the WMLS-R showed that Ernesto had limited E nglish and Spanish oral language ability (Level 3). On the CATRS, his teacher reported that he is sweet and loves to learn, but tends to be restless, impulsive and inattentiv e. She stated that Ernesto lacked focus in a whole group but did much better on one-on-one or small group. Maggie. Maggie was a Hispanic female, aged 8 years-8months, enrolled in second grade for the second time. She was re tained once in kindergarten as well. Her parents were from Mexico and they only sp oke Spanish at home. They reported not having any schooling experience. They also stated that Maggie ha d had five years of schooling in the US. Her t eacher reported that Maggie participated in the SRA/EIR intervention for 45 minutes, five days a we ek. She was identified as being LEP and receiving LEP classes, but her teacher stated that those services were provided only as needed. Reading scores showed that at the end of second grade from the previous year, she scored in the 13th percentile on the SAT-10. Her DIBELS scores for the end of second grade were 58 on NWF (low risk), an d 39 on ORF (high risk). At the beginning of second grade of the current year, she scored below the 10th percentile on the Invented Spelling Assessment. Based on t he WMLS-R, Maggie demonstrated limited English and Spanish oral language ability (Level 3). Her teacher reported on the CATRS that Maggie would get easily frustr ated and tended to cry often and easily. She also affirmed on different occasions that sh e was insecure and constantly looked for her approval.

PAGE 91

91 Maria. Maria was Hispanic female, aged 8 year s-8 months, enrolled in second grade for the second time. Her parents were from Mexico and they only spoke Spanish at home. Their highest level of education reported was high school. Maria had had five years of schooling in the US. In second gr ade, she received 45 minutes of SRA/EIR small group intervention, five times a week School records showed that Maria had an LEP status and was enrolled in LEP classes; still, her teacher repor ted that she did not receive those services. Maria was the only st udy participant who did not participate in the free or reduced-pr ice lunch program. Marias end-of-second-grade scores from the previous year showed that she scored in the 15th percentile on the SAT-10. Her DIBELS scores were 58 on NWF (low risk) and 23 on ORF (high risk). At the beginni ng of second grade of the current year, she scored between the 20th and 30th percentile on the Invented Spelling Assessment. Language scores on the WMLS-R demonstrated that Maria had fluent English oral language ability (Level 4) and limited Spanish oral language abi lity (Level 3). On the CATRS, her teacher stated that Maria did not have any attitude or behavioral problems in class. Jennifer. Jennifer was a Hispanic female, aged 10 years-1 month, enrolled in second grade for the second time. Her mot her was from Guatemala and they only spoke Spanish at home. Her mother report ed not having any schooling experience. She also stated that Jennifer had had one year of schooling in the US. Her teacher reported that she received 45 minutes of Harcourt In tervention five times a week. She had both migrant and LEP status. School records showed that she received LEP classes three times a week for 30 minutes.

PAGE 92

92 Reading scores at the end of second grade from the previous year showed that Jennifer scored 0 on the SAT-10. Her DIBELS scores for the end of second grade were 49 on NWF (moderate risk), and 2 on ORF (high risk). At the beginning of second grade of the current year, she scored below the 10th percentile on the Invented Spelling Assessment. The WMLS-R showed that Jenni fer had negligible Engl ish oral language ability (Level 1) and limited Spanish oral language ability (Level 3). On the C-ATRS, her teacher reported that Jennifer had no attit ude or behavioral problems in class. On one occasion, her teacher affirmed that one of Jennifers biggest challenges was her limited English oral skills, but t hat she was eager to learn. Aldo. Aldo was a Hispanic male, aged 9 years-3months, enrolled in second grade for the second time. He was also previously retained in kindergarten. His parents were from Mexico and were able to speak Engl ish and Spanish. At home, they only used Spanish to communicate with Aldo. They r eported not having any schooling experience. Aldos parents reported that he had had 4 years of schooling in the US. Aldo participated in the SRA/EIR program Level 1, 45 minutes a day, five days a week. School records showed that Aldo had migrant and LEP status. He participated in LEP classes three times a week for 30 minutes Aldo was identified as having a language disability and received speech therapy one time a week for 30 minutes. At the end of second gr ade the previous year, he scored in the 7th percentile on the SAT-10. His DIBELS scores for the end of second grade were 4 on NWF (high risk) and 4 on ORF (high risk). At the beginning of second grade of the current year, he scored below the 10th percentile on the Invented Spelling Assessment. Based on the WMLS-R, Aldo demonstrated limited English oral language ability (Level 3) and very

PAGE 93

93 limited Spanish oral language abi lity (Level 2). Aldos teac her reported on the C-ATRS that he had a short attention span, and t hat he was restless, impulsive, and easily distracted. She also stated t hat Aldo did not make much pr ogress the previous year and that he was the lowest perfo rming student in her class. Jorge. Jorge was a Hispanic male, aged 9 year s-5 months, enrolled in second grade for the second time. He was also prev iously retained in kindergarten. Jorges parents were from Mexico and they only spoke Spanish. They reported not having any schooling experience. Jorges parents stated that he had had 5 years of schooling in the US. His teacher affirmed that Jorge participa ted in the SRA/EIR program for 45 minutes a day, five days a week. School records showed that Jorge no longer had an LEP status, but his teacher affirmed that LEP classes were provided as needed. At the end of second grade of the pr evious year, he scored in the 11th percentile on the SAT-10. His DIBELS scores for the end of second grade were 17 on NWF (high risk) and 5 on ORF (high risk). At the beginni ng of second grade of the current year, he scored below the 10th percentile on the Invented Spelling Assessment. Based on the WMLS-R, Jorge demonstrated limited to fl uent English and Spanish oral language ability (Level 3.5). On t he C-ATRS, his teacher reported that he was impulsive and restless in class, and that he tended to avoi d work and to disturb his classmates. She also affirmed that Jorge had a short attenti on span and was very inattentive in class. According to her, one of Jorges biggest ch allenges was his behavior. He had been referred and suspended for stealing, for being disrespectful, and for making threatening statements toward his peers.

PAGE 94

94 Assessment Instruments There were four types of assessments instruments given to the participants: (a) screening instruments, (b) grouping instrument s, (c) supplemental instruments, and (d) probes of student progress on the dependent variabl e. The following sections contain a description of each of the first three types of assessments. A description of the probes is presented in the Dependent Variables section. Screening Instruments As part of the selection process, students scores on three measures of reading performance were analyzed to help determine e lig ibility for this study: (a) SAT-10, (b) DIBELS, and (c) Invented Spel ling Assessment. Additional measures were given to selected participants to establish their level oral language proficiency (WMLS-R), and their classroom attitudes and behaviors (C-ATR S). What follows is a brief description of each measure. Stanford Achievement Test-10 (SAT-10). The SAT-10 is a group-administered, standardized measure of reading achievement for grades K to 12, currently used in Florida Reading First schools. The test is untimed and uses a multiple-choice format. The reading portion assesses phonemic aw areness, decoding, phonics, vocabulary, and comprehension. Scoring is conducted by the test publisher. Student scores are reported in percentiles. Scores for SAT -10 were obtained from the school. Dynamic Indicators of Basic Early Literacy Skills (DIBELS). DIBELS (Good & Kaminski, 2002) is a set of tests designed to screen and monitor students reading skills from K to 3rd grade. In this study, the latest end-of-year scores were obtained from the school for each of the 10 participants. For Am elia, Jessica, Viviana, and Ernesto the DIBELS scores reported corresponded to the end of first grade. For all other

PAGE 95

95 participants, end-of-year scores corresponded to the end of second grade because they had been retained. The tests administered at the end of first grade are (a) Phoneme Segmentation Fluency (PSF), (b) Nonsense Word Fluency (NWF), and (c) Or al Reading Fluency (ORF). All these tests ar e timed with the purpose of assessing students level of automaticity. The PSF is a te st of phonological aw areness that evaluates the ability to fluently segment words into phonemes. The test takes approximately 2 minutes to administer. The benchmark goal for the end of first grade is >35. The alternate-form reliability of the PSF is .79. The NWF is a test of the alphabetic principle and the ability to blend letters into words. It takes appr oximately 2 minutes to administer. The benchmark goal for the end of first grade, as well as the end of second grade is >50 correct letter sounds per minute. The alternate-form reliability for the NWF is .83. The ORF is a test of reading accuracy and fluenc y of connected text. It uses a set of passages that students read aloud for one minute. The oral reading fluency rate corresponds to the number of correct words read per minute. The benchmark goal for the end of first grade is >40 correct words per minute and for the end of second grade is >90 correct words per minute. The DIBELS pr ovides four types of risk status that identify need for additional and substantial intervention: (a) above average (at or above the 60th percentile, no need for additional inte rvention), (b) low risk (at grade level, no need for additional intervention, (c) moderat e risk (moderately below grade level, requires additional intervention), and (d) hi gh risk (significantly below grade level, requires substantial intervention; Flor ida Center for Reading Research, 2006).

PAGE 96

96 Invented Spelling Assessment: A Sound Beginning. The Invented Spelling Assessment (Lane & Pullen, 2004) was developed as a measure of phonological awareness and understanding of t he alphabetic principle. In this assessment, the examiner dictates ten unfamiliar words and the student atte mpts to write them. The examiner then assigns point s to each word based on its phonological accuracy. The scoring scale goes from 0 to 4 points based on the level of phonological accuracy of the students spelling. On average, a score below three points per word indicates that the student lacks appropriate phonological skills. The Invented Spelling Assessment offers percentile ranks for fall of second grade based on a sample of 2,000 students in kindergarten to second grade. The assessment has a interscorer relia bility of >.97 (Lane & Pullen, 2004). A score below the 40th percentile indicates a need for reading intervention. Woodcock-Muoz Language Survey Revised (WMLS-R). The WLMS-R (Woodcock, Muoz-Sandoval, Ruef, & Alvarado, 2005) is an individually administered test of oral language, languag e comprehension, reading and wr iting proficiency. It was developed with the purpose of determining language proficiency, language dominance, changes in language ability, eligibility for educational services, readiness for Englishonly instruction, among others. The survey has two parallel forms, one in English and one in Spanish, which can be used with indivi duals ranging from 2 years old to more than 90 years old. Each form takes appr oximately 45 minutes to administer. The WMLS-R English form has seven tests: Test 1-Picture Vo cabulary, Test 2Verbal Analogies, Test 3-Letter-Word I dentification, Test 4-Dictation, Test 5Understanding Directions, Test 6-Story Recall, and Test 7-Passage Comprehension. A

PAGE 97

97 cluster score for oral language-total can be obt ained by combining tests 1, 2, 5, and 6. This cluster represents a broad measure of language competency. The WMLS-R Spanish form has seven equivalent tests: Te st 1-Vocabulario sobre Dibujos, Test 2Analogas Verbales, Test 3-Identificacin de Letras y Palabras, Test 4-Dictado, Test 5Comprensin de Indicaciones, Test 6-Rememoracin de cuentos, and Test 7Comprensin de Textos. The corresponding oral language total (lenguaje oral total) can be obtained by combining tests 1, 2, 5, and 6 (Alvarado, Ruef, & Schrank, 2005). The standardization sample of the WM LS-R English form consisted of 8,818 individuals, ages 2 to over 90 years old, repr esenting different regions of the country, different races, types of schools, etc. T he calibration sample fo r the WMLS-R Spanish form consisted of 1,157 native Spanish-spea king individuals from different regions inside and outside the United States (e.g., Mexico, Argentina, Panama, Costa Rica, Colombia, and Puerto Rico; Alvarado et al., 2005). The split-half test of reliability ranged from .76 to .97 for tests and fr om .88 to .98 for clusters. One of the scores yielded by the WLMS -R is the Cognitive-Academic Language Proficiency (CALP). The CALP measures students proficiency with context reduced and cognitively demanding langu age (Alvarado et al., 2005, p. 61). CALP scores have six levels and two regions of uncertainty fo r the different clusters. In each of these levels, cognitive-academic langu age proficiency is identified as it compares to others of the same age or grade level. Each level has s pecific implications for instruction as it relates to context-reduced and cognitively demanding language learning tasks in each language. At Level 1 students proficiency is negligible (imperceptible). They are likely to find language demands impossible to handle. In Level 2 proficiency is very limited (muy

PAGE 98

98 limitado) and students are ex pected to find language demands extremely difficult. Students in Level 3 have limit ed (limitado) proficiency a nd find language demands very difficult. In Level 4, proficiency is fluent (fluido). Students tend to find language demands manageable. Level 5 is characterized by advanced (avanzado) proficiency, where students are likely to find language demands very easy. Finally, in Level 6 proficiency is very advanced (muy avanzado) and students ar e expected to find language demands extremely easy. In addition to the six leve ls, the two regions of uncertainty include Level 3-4 (3.5), where students have limited to fluent proficiency (limitado a fluido) and Level 4-5 (4.5), where students have fluent to advanced proficiency (fluido a avanzado; Alvarado et al., 2005). In this study, CALP scores for the English and Spanish Oral Language-Total cluster were used. Conners Abbreviated Teacher Rating Scale (C-ATRS). The C-ATRS is an abbreviated form of the Conners Teacher Ra ting Scale, developed to assess teachers perceptions of their students behavior in t he classroom (Conners, Sitarenios, Parker, & Epstein, 1998). The scale has 10 items that describe different childrens behaviors (i.e., restless, excitable, inattentive). For each it em, the teacher is asked to rate how much the child has shown the behavior in the last month. There are four possible options: not at all, just a little, pretty much, or very much. Home language survey. The Home Language Survey was developed by the researcher with the purpos e of obtaining demographic data and information about language practices taking place at the student s homes. The survey was translated to Spanish to provide an option to parents who feel more comfortable reading in Spanish. Demographic information and responses ab out the language practices are reported on

PAGE 99

99 an individual basis to enhance the profile in formation of each participant. Each of the participating students t ook home a survey, accompanied by a reminder letter that was also translated to Spanish (see Appendix B fo r a copy of both reminder letters) that explained the purpose of the su rvey, how to respond, and how to send it back. The response rate was 90%. Grouping Instruments With the purpose of assigning selected participants to an appropriate tutoring group, the researcher conducted measures of reading accuracy by taking running records of students reading leveled books. A r unning record (Clay 1972) is an untimed measure of oral reading accuracy. The pur pose is to analyze reading behaviors that shed light into the type of reading skills and strategies that students use when they interact with text (Denton, Ciancio, & Flet cher, 2006). Running records can be used to (a) estimate the rate of reading accura cy, (b) identify students reading levels (independent, instructional, or fr ustrating), (c) group students fo r instruction, (d) monitor reading progress, and (e) identify areas of strength and weakness, among others. In this study, running records served to establish rates of reading accuracy in order to identify students instructional reading levels. Instruct ional levels refer to texts that are read with 90% to 95% accuracy and that should be the focus of instruction. The percentage of accuracy was calculated by di viding the number of words read correctly by the total number of words read, and mult iplying it by 100. The process was as follows. The researcher selected two sets of books leveled 1 to 20 according to Reading Recovery guidelines. Based on students init ial DIBELS ORF scores, the researcher provided a starting leveled book that approximated their reading ability. Students who read the book at the instructional level were given a second book at that same level for

PAGE 100

100 corroboration purposes. Students who read with an accuracy of more than 95% (independent level) were given upper leve led books until an instructional level was reached. In contrast, students who read the first book with lower than 90% accuracy (frustration level) were given lower leve led books until an inst ructional level was attained. Again, when students read a leve led book with 90% to 95% accuracy, a second book at the same level was prov ided to confirm that it was indeed their instructional level. Students book levels read in each trial and their percentages of accuracy are reported in Table 3-7. Table 3-7. Running records re sults used for grouping purposes Name Trial 1 Trial 2 Trial 3 Trial 4 Amelia 10 (88%) 9 (94%)9 (93%) Pedro 10 (94%) 10 (95%) Jessica 10 (92%) 10 (95%) Viviana 1 (98%) 2 (90%)2 (93%) Ernesto 1 (98%) 2 (93%)2 (93%) Maggie 5 (88%) 4 (89%)3 (94%)3 (91%) Maria 1 (98%) 2 (100%)3 (91%)3 (93%) Jennifer 1 (71%) Aldo 1 (<50%) Jorge 1 (83%) For each trial, the number outside the parenthesis refers to the book level read, while the number in parenthesis corresponds to the percentage of readi ng accuracy. The number of trials varied across students. Once instructional levels were identi fied for each student, those with similar reading levels (within one or two levels of difference) were grouped for intervention. A total of three instructional groups were forme d. Refer to Table 3-1 for the summary of the composition of these groups. Supplemental Preand Post-Intervention Instruments With the purpose of establishing a comprehensive student readi ng profile that could help in measuring and interpreting student s response to the tutoring program, the

PAGE 101

101 researcher conducted preand post-intervent ion measures of reading ability. These included word reading skills, fluency, compr ehension, and attitudes toward reading. In addition, teachers were ask ed to rate their students on five areas of reading and classroom behaviors before and after the inte rvention. What follows is a detailed description of each measure. Test of Word Reading Efficiency (TOWRE). The TOWRE (Torgesen, Wagner, & Rashotte, 1999) is a norm referenced, indi vidually administered, measure of word reading efficiency that assesses the ability to read printed words with accuracy and fluency. The purposes of this test are to help monitor student growth on phonemic decoding and sight word reading, to identify students who might need more instruction, and to help identify reading disabilities as par t of a battery of diagnostic tests. The TOWRE has two subtests: Sight Word Efficiency (SWE) and Phonemic Decoding Efficiency (PDE). The SWE measures the number of real words that can be accurately identified, while the PDE m easures the number of nonwords that can be accurately decoded. Each subtests lasts 45 seconds (Torgesen, Wagner, & Rashotte, 1999). The standardization sample for the TOWR E included 1,500 subjects, ages 6 to 24 years old, from different regions of the count ry. Alternate-form reliability scores are .93 for the SWE, .94 for the PDE, and .96 for the Total Word Reading Efficiency (Torgesen et al., 1999). In this study, raw scores and percentile ranks for SWE, PDE and Total Word Reading Efficiency were reported. Kaufman Test of Educational Achi evement-II (KTEA-II) NWD and LWR. The KTEA-II (Kaufman & Kaufman, 2004) is a norm referenced, individually administered, measure of academic achievement for i ndividuals between the ages of 4 years, 6

PAGE 102

102 months through 25 years. Two subtests were used in this study: the Nonsense Word Decoding (NWD) and the Letter-Word Rec ognition (LWR). The NWD requires the student to decode increasingly difficulty nonsense words, while the LWR requires the student to identify letters and decode increasingl y difficult real word s. Both tests are untimed. It is estimated that the NWD takes 3 minutes and the LWR takes 4 minutes to administer (Kaufman & Kaufman, 2004). The mean split-half reliability coefficient fo r the NWD subtest is .94 for grade and age norms. The mean reliability coefficient for the LWR is .97 for age norms and .96 for grade norms (Kaufman & Kaufman, 2004). In this study, raw scores and percentile ranks for both subtests were reported. DIBELS Nonsense Word Fluency (NWF) and Oral Reading Fluency (ORF). Since the DIBELS scores used for screeni ng purposes corresponded to the end of the previous academic year, the researcher wa s interested in obtaining beginning of year NWF and ORF scores to determine the le vel of performance before and after intervention. Given that the school disc ontinued the use of DI BELS assessments, the researcher administered both tests. The NWF is a test of t he alphabetic principle and the ability to blend letters into words. The test is individually administered and it takes approximately 2 minutes to complete it. T he student is provided with a set of VC and CVC nonsense words in random order. The student is asked to read either individual letter-sounds or whole nonsense words in one minute. Scores represent the number of correct letter-sound correspondences read per minute. The benchmark goal for the beginning and end of second grade is >50 correct letter sound correspondences per minute (Good & Kaminski, 2002).

PAGE 103

103 The ORF is a standardized test of accu racy and rate for reading connected text. The test is individually administered to students with the purpose of identifying those who are in need of additional instruction, as well as to monitor progress. The ORF utilizes a set of standardized grade-leveled passages that the students read aloud for one minute. During this time, an examiner re cords the number of words read correctly and the number of errors. The test yields a rate of oral reading flue ncy that corresponds to the number of words read correctly per mi nute. Passages are available for first grade (middle and end of year), second grade (begi nning, middle, and end of year), and third grade (beginning, middle, and end of year). The benchmark goals for the beginning of second grade are >44 correct words per minut e. As stated previously, the benchmark goal for the end of second grade is >90 correct words per minute. Test-retest reliability coefficient for the ORF ranges from .92 and .97. Alternate-form re liability for passages within a grade level ranges from .89 to .94 (Tindal, Marston, & Deno, 1983). At the beginning of the study, each partici pant was given a passage to read that corresponded to a beginning second grade leve l (i.e., My Handprints). Students who were at low risk or above average were not given further passages. Students who were at moderate or high risk were given a pa ssage corresponding to the end of first grade (i.e., The Sand Castle). From these, student s who still were at moderate or high risk were given a middle of first grade passage (i.e., Ice Cream). A similar process was followed at the end of the study, except that students started reading passages corresponding to the end of second grade (i.e., If I had a Robot). Also, to avoid student fatigue, those who were at moderate risk or high risk on the end of grade passage were first given a beginning second grade passage and if they were at low risk or above

PAGE 104

104 average, then the middle of second grade passage (i.e., Moving Day) was administered. Otherwise, students kept reading first grade passages in the same way described above. This procedure had two purposes: (a) to provide an estimation of the current level of reading fluency of each student, and (b ) to serve as a decision making process for selecting passages from t he QRI-4 to be administered. Qualitative Reading Inventory-4 (QRI-4)-Comprehension. The QRI-4 (Leslie & Caldwell, 2006) is an informal reading inventor y designed to estimate students oral and silent reading levels, as well as comprehens ion skills. The QRI-4 employs narrative and expository passages that span from pre-prim er to high school. Comprehension skills for first and second grade are determined throug h story retell, as well as open-ended questions. Once a student finishes reading a passage, students are asked to retell the story and answer implicit and explicit ques tions. Scores reported include the total number of ideas recalled (retell) and the tota l number of correct answers given for each passage. Based on the total number of correct answe rs, the QRI-4 provides three different levels at which students comprehend text: independent, instructional, and frustration. According to the tests criter ia, students at the independent level are able to read the passage successfully without assistance and are able to answer 90% of the questions correctly. Students at the instructional le vel read passages with assistance and answer 70% of the questions correctly. On the frustrat ion level, students are unable to read the passage and answer less than 70% of the questi ons correctly. Alternate-form reliability coefficients for the QRI-4 are .80 for comp rehension scores on two different passages at the same readability level. Inter-scorer reliability for scoring oral reading miscues is

PAGE 105

105 99%, while reliability for scoring comprehensi on questions is 98% (Leslie & Caldwell, 2006). The implementation of QRI-4 passages wa s contingent on students ORF scores. The decision making process was as follows. If a student was at low risk, or above average on any of the second grade passages (beginning, middl e, or end of year) of the ORF, a QRI level two passage was administered (i.e., What Can I Get for My Toy?). If a student was at moderate risk or high risk on second grade passa ges of the ORF, and at low risk or above average on the end of first grade passage, a QRI level one passage (i.e., The Bear and the Rabbit) was adminis tered. Furthermore, if a student was at moderate or high risk on the end of first grade passage from the ORF, and was at low risk or above average on the middle of firs t grade passage of the ORF, a QRI primer level passage (i.e., Fox and Mouse) was adminis tered. On the other hand, if a student was moderate or high risk on the middle of first grade passage from the ORF, no QRI passage was administered. The purpose of this procedure was to lower students frustration when reading text that is beyond their reading abilities. Reading Ability Rating Scale (RARS). The Reading Ability Rating Scale (RARS) for teachers was created by the researc her to obtain information about students changes in reading skills and classroom behavio rs as a result of the intervention (Appendix C). The checklist has 26 items tota l, 23 related to reading ability (e.g., phonological awareness, phonics, fluency, co mprehension, vocabulary) and three related to classroom behaviors (e.g., partici pation in class, motivation to read, and use of English to communicate in class). Teachers were asked to rate each of their students before and after the intervention, using a 4-poin t Likert scale that ranges from very weak

PAGE 106

106 to very strong. For each of the five areas of reading, a mean score was calculated and reported. For classroom behaviors, raw sco res on each of the three items were reported. Elementary Reading Attitude Survey (ERAS). The ERAS (McKenna & Kear, 1990) is a group or individually administered instrument designed to measure students attitudes toward recreational and academic reading (Kazelskis, Thames, & Reeves, 2004). The survey can be used with students fr om first through sixth grade. According to McKenna and Kear, the survey can be used to monitor the impact of instructional programs on students attitudes. The survey has a total of 20 items, 10 related to recreational reading and 10 rela ted to academic reading. With young students, the tutor reads the items together with the students. Students are asked to circle the picture that best describes how they feel about each item. There are four possibl e pictures that go from very happy to very ups et. Approximate completion time is 10 minutes. The survey yields three scores: total for recreational reading, total for academic reading, and a composite total. Scores can range from 10 to 40 points in each scale, and 20 to 80 points for the composite (McKenna & Kear). For this study, raw scores and percentile ranks were reported. The norming sample consisted of 18, 138 students in grades 1 through 6 from different regions of the country. These samples represented a variety of races. Cronbach alpha reliability coeffi cient for the survey ranged from .74 to .89 across grade levels (McKenna & Kear, 1990). Intervention The UFLI tutoring program is an early literac y intervention designed to help struggling readers develop beginning reading skills. Originally, UFLI was designed as

PAGE 107

107 an education tool to help prepare pre-servic e teachers in the area of reading. By conducting one-on-one tutoring wit h struggling readers, teachers are able to learn about the reading process and how it is acquired by children, the difficulties that many students face when they lear n to read, and the instructional methods that help struggling readers overcome their reading difficulties. The tutoring program was founded on current research regarding (a) early literacy development and (b) effective instructional strategies for struggling r eaders (Lane, Pullen, H udson, & Konold, 2009). Although UFLI was originally designed to be implemented in a one-on-one setting, several small-group modifications of the tutoring model have been successfully implemented (e.g., Pullen, 2000; Pullen, Lane, Lloyd, Nowak, & Ryals, 2005). The focus of this study was on sm all-group implementation. Early Literacy Components Addressed in UFLI Each UFLI tutoring session is designed to promote the develop ment of phonemic awareness, print awareness, decoding, reading fluency, comprehension, and strategy use (Lane et al., 2009). Some of these skills were addressed multiple times throughout the lesson. What follows is a description of how each reading component was addressed in each session. Phonemic awareness. In each tutoring session, phonemic awareness was developed in two ways. First, students we re required to count and identify phonemes using Elkoning boxes. Second, they l earned to blend and s egment phonemes using manipulative letters (Hayes, Lane, & Pullen, 2005; Lane et al., 2009). Print awareness. Print awareness was developed in every lesson through book reading and sentence writing activities. When students read new and familiar books, students were able to handle books, turn pages, and point to different parts of the text.

PAGE 108

108 By doing this, they learned about different parts of the book, reading directionality, and conventions of print (e.g., concept of wo rd, spacing, punctuation). In addition, when students were asked to write a sentence relat ed to the story read, they also practiced directionality, spacing, and punctuation (Hayes et al., 2005; Lane et al., 2009). Decoding. Decoding skills were developed during word work activities with manipulative letters, as well as reading and wr iting practice. In word work activities, students were required to manipulate letters in order to encode and decode real and nonsense words at the onset-rime and phonem e levels. During reading practice, students were required to decode new and unf amiliar words while repeatedly reading familiar books and while reading new and unfam iliar books for the fi rst time. During writing practice, students were asked to identify the phonemes in words, to encode the words using Elkonin boxes, and rewrit e words on a sentence page. With ample opportunities for practice, UFLI promoted decoding accuracy and automaticity (Hayes et al., 2005; Lane et al., 2009). Fluency. Fluency was primarily developed by repeated readings of connected text. Students reread familiar books at an independent reading level (more than a 95% accuracy). In the beginning, the focus of the tutoring was on developing reading accuracy. Once students became more accu rate while reading connected text, the focus shifted to reading automaticity. Finally, when students acquired appropriate reading automaticity the focus of the session s shifted to prosody (Hayes et al., 2005; Lane et al., 2009). Comprehension. Comprehension skills were develop ed before, during, and after reading connected text. Before the students read a new book, the tutor introduced the

PAGE 109

109 book and engaged the group in a picture walk with the goal of activating prior knowledge and creating context for the story. Then, students were encouraged to make connections and predictions related to the st ory. During reading, the tutor modeled and guided students to self-monitor for comprehension. After reading, the tutor and students engaged in a discussion of the st ory through the use of literal inferential, and evaluative questions. Children were also asked to su mmarize part of the story and then write a summary sentence (Hayes et al., 2005; Lane et al., 2005). Strategy use. Throughout the UFLI tutoring progr am, students acquired different reading strategies to identify words and che ck for accuracy. Strategies taught included use of grapho-phonemic information, semantic and syntactic cues to confirm decoding accuracy, monitoring, cross-checking, and se lf-correcting, among others. In addition, the tutor scaffolded the use of strategies by using the ABC mnemonic, which stands for Acquire, Build, Control. First, the tutor demons trated and modeled the use of a new reading strategy that student s needed to acquire. Then, as students built their strategy repertoire, the tutor prompted students to use specif ic strategies. Finally, once children had control over t he strategies, the tutor observed how students applied them to connected text and asked them to explain how they were able to figure out specific words. It was expected that students who we re able to select appropriate strategies without prompting and explain accurately how the strategies worked would be able to use them independently (Hayes et al., 2005; Lane et al., 2009). Effective Instructional Practices for ELLs The UFLI tutoring program included resear ch-based instructional practices that facilitate le arning among E nglish language learners. Thes e were (a) explicit and systematic instruction (Ger sten & Geva, 2003; Helman, 2009; Manyak & Bauer, 2008;

PAGE 110

110 Vaughn et al., 2006), (b) small-group instru ction (Gersten et al., 2007), (c) ample opportunities for practice (Gersten & Geva 2003), (d) assessment of students progress (Gersten & Geva, 2003; Helman, 2009), (e) in teractive teaching (Gersten & Geva, 2003), (f) modeling (Helman, 2009), (g) integration of vocabulary into reading instruction (Antunez, 2002; Helman, 2009), (h) use of vi suals (Gersten & Baker, 2000; Helman & Burns, 2008), (i) connection between oral and wr itten forms of words (Helman & Burns, 2008), (j) self-monitoring (Helman & Burn s, 2008), and (K) use of native language strategically (Ger sten & Baker, 2000). UFLI Sessions The tutoring sessions were designed bas ed on the small-group UF LI model by Lane, Pullen, and Hayes (2007) and modified for second grade English language learners. Each session had four steps: (a) gaining fluency and measuring progress, (b) word work with manipulative letters, (c ) introducing and reading a new book, and (d) writing for reading. The instructor follo wed a session guide t hat outlined each step (Appendix D). Each session had an accompanyi ng session notes sheet to be completed by the tutor (Appendix E). What follows is a brief description of each step, including modifications for ELLs made for the study. Step 1 Gaining fluency a nd measuring progress (10 minutes). The purpose of this step was twofold: (a) to develop reading fluency of connected text, and (b) to measure students reading progress. Fluen cy was developed by targeting word reading accuracy and automaticity, reading rate, and prosody. Students reread one to three books that had been read in previous sessions and that could be read by all group participants with 90% to 100% accuracy. Ea ch student read at least one book (choral reading, partner reading ), with minimal tutor assistance. Books selected for this step

PAGE 111

111 provided opportunities to practice decoding, sight word reading, as well as improve reading rate. Books that were too easy were not selected because they did not offer instructional value. As students reached appropr iate levels of accuracy and rate, the tutor shifted the focus to prosody. During progress monitoring, the tutor took a running record of one group member. The student was asked to read the new book that was introduced in the previous session, while the other group members we re reading familiar books. The running record served two purposes: (a) to determine the level for the new book to be read in step 3, and (b) to plan for instruction. To determine the new book level, the tutor calculated the rate of accuracy (total word s read correctly/total words read x 100). If the accuracy rate was below 90% (frustration leve l), the tutor decided whether to introduce a new book at the previous leve l or try a new one at the same level. If the accuracy rate was between 90% and 95% (instructional level) the tutor introduced a new book at the same level. If the accuracy rate wa s above 95% (independent level), the tutor introduced a new book at a higher level. To pl an for instruction, the tutor examined the students miscues (e.g., attempts, omissions, self-corrections, additions, repetitions, emerging patterns) and strat egy use. Immediately after, the tutor provided feedback on students use of reading strategies to help them develop awareness of how strategies were being used. When appropria te, the tutor asked the studen t to identify strategies that were used to figure out challengi ng words (Lane et al., 2001; Lane et al., 2005). Throughout the first step, the tutor supported the language needs of students by relying on visuals and by providing quick definitio ns of unfamiliar word s (Gersten & Baker, 2000).

PAGE 112

112 Step 2 Word work with ma nipulative letters (10 minutes). The purpose of this step was to develop an understanding of the alphab etic principle, as well as automatic word recognition skills. In this step, manipulat ive letters were used to conduct word work with new and familiar words. The tutor and each participant had a magnetic board and a set of magnetic letters. Students practi ced encoding and decoding real and nonsense words, at the onset-rime and phoneme leve l. When needed, the tutor pointed out the similarities and differences between Span ish and English sounds and word patterns (Linan-Thompson et al., 2003). The tutor pl anned ahead of time by selecting a known word from a familiar book, and then creating a list of real words and nonsense words that could be derived from m anipulating the known word. The list of words followed a sound sequence that moved from easier to diffi cult (e.g., continuous sounds in initial position before stop sounds). The process was as follows. First, the tuto r either pointed out the familiar word in the book or asked students to find it. Then, th e tutor asked students to either spell the word or read the word that was spelled with magnetic lette rs on their own boards. Next, the tutor prompted students to either encode or decode each of the words in the list by making changes to the onset-rime or phonemes in the word. To address the needs of ELLs, the tutor specified whether each word was a real word or a nonsense word. This was important because many ELLs were still building their vocabularies and were not able to distinguish between real and nonsense words (Linan-Thompson et al., 2003). In addition, vocabulary was supported through the use of picture cards, brief definitions, and in some instances, translation of key words. This was relevant given that vocabulary is one of t he biggest challenges ELLs face (Carlo et al., 2004). As

PAGE 113

113 students acquired word reading strategies, t he tutor introduced more complex spellings, including prefixes, suffixes, and other mo rphographic features (Lane et al., 2001; Lane et al., 2009). Step 3 Introducing and reading a new book (10 minutes). The purpose of this step was to learn and practice reading strategi es with progressively more challenging books. The tutor selected a book based on the rate of accuracy obtained during the running record. Then, the tutor introduced th e book to the students with a picture walk and a discussion that highlighted key vocabul ary words, unusual spellings, and/or repeated language patterns. Picture cards and quick definitions were used to support vocabulary development. In addition, students were encouraged to make connections to their personal lives, as well as predictions about the story, which were later confirmed or refuted. This provided participating ELLs wit h opportunities to practice oral skills and acquired vocabulary while learning comprehens ion strategies. After a brief discussion, students started to read the book while the tutor coached them in learning and applying reading strategies. Coaching was more expl icit at the beginni ng when students were learning new strategies and became more imp licit when the student s started acquiring and demonstrating control over them (Lane et al., 2007; Lane et al., 2009). In addition to reading connected text, the tutor engaged the students in word work with new words. Students were prompted to appl y different reading strategies to figure out unknown words. To do this, the tuto r selected one or two words that students struggled with when reading the new book. The tutor used manipulative letters or dryerase boards to (a) demonstrate how the new words were spelled, (b) point out similarities and differences between new words and familiar words, (c) use parts of the

PAGE 114

114 new words as base for practicing word fam ilies, (d) practice segmenting and blending using Elkonin boxes, or (e) practice writi ng multisyllabic words and/or sight words (Lane et al., 2007; Lane et al., 2009). As in previous word work activities, visuals and brief definitions were used to support the vocabulary needs of students. Step 4 Writing for reading (15 minutes). The purpose of this step was to use writing as a means to develop print awareness, phonemic awar eness, decoding, and encoding automaticity, and familiarity with sight words and word patterns. First, the tutor had a brief informal conversation with the st udents about the new book read. During the conversation, the tutor identified one or tw o meaningful sentences from each participant that provided optimal opportunities for inst ruction. The tutor jotted down one sentence for each student on the session notes sheet. Each sentence included at least one to three high frequency words and at least two to four decodable words. The tutor then asked the students to write t heir sentences in their writ ing books. Each writing book, formed of blank sheets of white paper, was turned sideways in order to have a top page for word work and a bottom page for sentence writing. Initially, the students wrote their sentence on the top page, whil e the tutor moved from one student to the other checking for spelling accuracy. As students finished writ ing their sentences, the tutor asked them to either write words that were spelled correctly in the sentence page, or do word work on the top page with words that were misspelled. Word work was conducted in different ways: (a) unfamiliar high frequency words with uncommon spellings (e.g., because, was) were spelled by the tutor first and then written multiple times by the students (i.e., write the word inside this box, write the word as large as you can, write the word as small as you can, write the word as fast as you

PAGE 115

115 can), and (b) unfamiliar words with regular s pelling patterns were practiced with Elkonin boxes first and then rewritten once again bef ore writing it on the sentence page. Uncommon words with unusual spellings were dealt with in one of three ways: (a) the tutor wrote the word on the sentence page, (b) the tutor fi rst wrote the word on the practice page and then the student wrote it on the sentence page, or (c) the tutor helped the student identify parts of the words he or she could spell and completed the word for the student before the student wrote it on the sentence page (Lane et al., 2007; Lane et al., 2009). Word work was conducted individually or as a group if every student was working on the same word. Every time a word wa s added to the sentence page, the child was required to reread what he or she had written to that point. This provided additional reading practice and reinforced grapho-phonemic connections. To support the needs of ELLs, the tutor helped students produce comple te and grammatically correct sentences by reinstating what the students had said and making t hem repeat the complete sentence before writing it. In some instanc es, when students used words in Spanish to express an idea, the tutor translated the wo rd for the student and asked them to repeat the word verbally, before writing it. Materials Based on the UFLI tutoring program, all participating groups had access to the same type of materials. These materials wer e provided by the investigator at no cost to the school or the students. Materials incl uded (a) leveled books, (b) magnetic letters and magnetic boards, (c) picture ca rds, and (d) other materials. Leveled books. The books selected for each lesson followed the Reading Recovery level criteria, which are designed to span from level 1 to level 20. Levels 1 to

PAGE 116

116 16 correspond to first grade, and levels 17 to 20 correspond to second grade. Leveled books match the text to the specific needs of readers (Lane et al., 2009), allowing them to practice decoding skills and sight word reading (Helman, 2009), and to apply reading strategies (Fountas & Pinnell, 1999). Wi thin each reading level, narrative and informational texts that covered a variety of topics were selected to support students interests and backgrounds. Magnetic letters and magnetic boards. Each student had a set of solid blue, lowercase magnetic letters and a magnetic board to manipu late during word work activities. Research shows that the use of magnetic letters supports the development of decoding skills. For example, Pullen (2000) studied the effect of alphabetic word work using manipulative letters on the reading skill s of first grade struggling readers. Results showed that students in the experimental gr oup (lessons with manipulative letters) had better decoding skills than students in the comparison condition (lessons without manipulative letters) and students in the control group (no treatment ). In a different study, Pullen and colleagues (2005) conduct ed a multiple-baseline design across groups study to evaluate the use of mani pulative letters during explicit decoding instruction. Students used magnetic letters to blend and segment, decode and recode target words from books previously read. The authors found that all participating students (N=9) increased the percentage of correct pseudowords read per minute, from baseline (average of 46.5%) to intervention (average of 86.5%). These findings offer additional evidence for the use of manipulative letters to during reading interventions that target word reading skills.

PAGE 117

117 Picture cards. With the purpose of facilitat ing comprehension throughout the tutoring lessons, picture cards of key vocabu lary words were used to help students gain understanding during reading and wo rd work activities. Sets of picture cards were either purchased or created by the re searcher. According to Gersten and Baker (2000), visual aids help students visualize the abstractions of language (p. 463). T herefore, they can effectively support English language learners as they deal with the language demands of each lesson. Other materials. In each tutoring session, th ree additional materials were required. First, session notes for each group to keep a detailed record of their performance. Second, a sentence writing book was given to each participant to complete step 4 of the less on. And third, pencils and a digital timer to ensure that adequate time was spent on each activity. Dependent Variable Two dependent variables were selected and measured throughout the different phases of this study. The variables were (a) rate of correct pseudowords read per minute (CPPM) and (b) rate of correct sight w ords read per minute (CSPM). Pseudoword reading, also known as nonsense word reading or nonword reading, has long been identified as a re liable predictor of readi ng achievement (Gough, 1983). Shankweiler and colleagues (1999) examined the relation between word reading, nonword reading, and reading comp rehension. Findings showed that there is a strong relation between word and nonword readi ng and between these skills and reading comprehension. According to Shankweiler and colleagues, nonwor d reading measures require the use of phonologically analytic decoding processes, which are critical to reading ability (p. 87) Pullen and colleagues ( 2005) also state that using a rate of

PAGE 118

118 pseudoword reading enables us to determine the level of automaticity with which readers decode words, that is, decoding fluency (Hudson et al., 2009). The rate of correct pseudowords read per minute (CPPM) was used in this study to make decisions about movement across phases (baseli ne, intervention, maintenance). In addition to pseudowords, the rate of correct sight words read per minute (CSPM) was monitored throughout the intervention to measure the development of word recognition skills. According to Ehri ( 2005b), a useful way in which readers might read a word is by sight or memory. A wo rd that is well known can be recognized automatically, as a single unit, allowing the readers attention to be focused on meaning while reading connected text (Ehri, 2005b; Eh ri & Snowling, 2004, LaBerge & Samuels, 1974). Sight words are learned by establis hing connections between the graphemes and phonemes in words. These connections are based on knowledge of the alphabetic principle and spelling patterns, which is dev eloping in beginning readers. Each time a word is read, the grapheme-phoneme c onnections are str engthened (Ehri, 2005b). According to Ehri and Snowling (2004), having an extensive sight word lexicon is central to reading. Therefore, measur ing students ability to read sight words accurately and fluently is key in evaluating the effect iveness of beginning reading interventions like UFLI. For the purposes of this study, si ght words are defined as high-frequency words with irregular spelling patterns. To measure the dependent variables, two types of probes were created by the researcher: (a) pseudoword probes and (b) sight word probes. Pseudoword probes were created from a pool of 100 word s that represent common letter-sound combinations and spelling patterns. From this pool of words, 15 different probes of 50

PAGE 119

119 words each were generated by random selecti on in order to prevent practice effects (see Appendix F for all pseudowor d probes). In a similar fash ion, 15 sight word probes were created from a pool of 100 high frequency words selected from Dolchs list of high frequency words for first and second grade. Ea ch probe consisted of 50 randomly selected sight words (see Appendix G for all sight word probes). To estimate the rate of CPPM and CSPM, the researcher conduct ed 1-minute timings with each participant every other day. No feedback was provided to the students at this time. The rate of CPPM and CSPM corresponded to the total number of words read correctly in one minute. Design This study implemented a multiple baseline across groups design (Kazdin, 1982; Kennedy, 2005) to assess the ef fects of the UFLI program on the reading skills of second grade, Spanis h-speaking, English la nguage learners. Single-subject designs are relevant to the field of literacy for various reasons: (a) they emphasize the individual as the unit of concern; (b) they systematically determine if a specific intervention is effective and for whom it is effective, allowing the analysis of responders and nonresponders to treatment; (c) they provide a practica l way to test educational procedures under typical educational conditions; (d) they incorporate ways to assess not only the outcomes of an intervention, but also the proce ss of change and the maintenance of change across time, and (e) they offer a costand time -efficient way to investigate critical instruct ional questions (Horner et al., 2005; Neuman & McCormick, 2000). In 2003, single-subject research was identif ied by the Division of Research of the Council for Exceptional Children as one of th e four research methodologies needed to

PAGE 120

120 identify evidence-based practices (Odom et al., 2005). Through the use of withinand betweensubject comparisons and systematic replication, single-subject design offers ways to control for majo r threats to external and internal validity, a major concern for research in general (Horner et al., 2005). The design for this study consists of th ree phases: (1) baseline, (2) intervention, and (3) maintenance. A detailed description of each phase and the tutor/researcher that conducted each phase is provided below. Follo wing that, procedural information is given on supplemental data collection. Baseline During bas eline, the dependent variables (CPPM and CSPM) were measured for each participant using resear cher created probes. Each data collection session lasted approximately three minutes per student. Once i ndividual rates were calculated, a mean group was estimated and plotted into a graph. All groups started baseline at the same time. When Group 1 showed a stable line on si x continuous data points (no significant increasing or decreasing trend in behavior), they moved to the intervention phase. In the meantime, the other two groups remained in baseline. Intervention During intervention, each group participated in a series of 45-minute, small-group tutoring lessons, two to five times a week Lessons for each group were scheduled in collaboration with the st udents teachers and the school pr incipal to avoid interfering with important instructional time. All less ons took place in a quiet and well illuminated room to avoid distractions. Each session fo llowed a structured format that involved the four steps described previously. In each tu toring lesson, students read familiar and new leveled books, did word work with magnetic letters, and wrote sentences related to the

PAGE 121

121 readings. During this phase, response to intervention was measured using the same set of pseudoword and sight word reading probe s and following the same procedure as in baseline. Data were collected for each parti cipant at the beginnin g of the tutoring session (6 to 10 minutes per group). Accordi ng to Pullen et al. (2005), administering probes at the beginning of a session is a mo re accurate measure since it reduces the chance of students applying what they had just practiced during the lesson. While one group member completed the probes, the other group members sat on the other side of the room and read familiar books. After each individual rate was established, a group mean was calculated and plotted into a graph. When Group 1 showed an increase in the rate of CPPM on four consecutive data points, Group 2 started the intervention phase. In the meantim e, Group 3 remained on baseline. When Group 2 showed an increase on four consecutive data points, Group 3 started intervention. Each group remained in the intervention phase until (a) every member of the group was readi ng at grade level (level 20 books) or (b) when every member of the group completed a minimum of 40 tutoring lessons. Maintenance The maintenance phase began for each group two weeks after the intervention phase ended. The same data colle ction procedures as in baseline were followed. Each data collect ion session lasted approximately th ree minutes per student. Once individual rates were established, a group mean was calculated and plott ed into a graph. The maintenance phase ended for each group when dat a points showed that a stable line was achieved, indicating that they were st ill reading at an appropriate reading level. The total number of data points varied across groups: Group 1 had 5 data points, while Group 2 and 3 had a total of 6 data points.

PAGE 122

122 Tutor/Researcher The researcher conducted all the tutori ng sessions. She was trained in the UFLI program by the programs developer and had experien ce im plementing it with Spanishspeaking English language learners, in one-on-one and small-group formats. The researcher was originally fr om Guatemala and had a Mexic an heritage as well. She was a native speaker of Spanish and spoke English fluently. The tutor had a B.A. in Psychology and Master of Education and Education Specialist degrees in Counselor Education. At the time of the study, the tutor was a doctoral candidate in Special Education. Her areas of expertise were early literacy, early reading difficulties, and English language learners. Data Analysis Data collec ted in each phase were analyzed using systematic visual comparison of responding within and across conditions (H orner et al., 2005, p. 169). In visual analysis, data points are graphed in order to ex plore and observe the types of patterns that arise over time (Kennedy, 2005). Visual comparison of data points allowed the researcher to identify changes in the dependent variable as a function of the independent variable (UFLI tutoring program). According to Horner et al. (2005), when the research participant is a group, and not an individual, the group gener ates a single score for eac h measurement period. In this study, mean rates of CPPM and CSPM we re calculated for each group based on each members individual score. The group mean was plotted in a noncumulative way using a simple line graph (Kazdin, 1982). Da ta were then analyzed by estimating an overall mean for each phase and conducting within group comparisons across phases.

PAGE 123

123 Supplemental Data Supplemental data include preand post-in tervention measures of early literacy skills an d attitudes toward reading, teac hers ratings of st udents reading abilities and classroom behaviors. It also addresses book le vels read by each group throughout the intervention phase. What follows is a brief description of the procedures followed for each of these. Preand PostIntervention Data At the beginning of the study, before baseline data were collected, all students were assessed indiv idually on measures of early literacy skills and attitudes toward reading. At the end of the study, when t he maintenance phase was completed for all groups, students early literacy skills and atti tudes toward reading were assessed again. Assessment times were scheduled with the hel p of the students teachers and the school principal. Pre-intervention assessm ent lasted approximately three hours per student, while post-intervention assessment lasted approximately 90 minutes. The time varied from preto post-intervention becaus e, at the beginning of the study, the Woodcock-Munoz Language Survey-Revised (WMLS-R) was administered for screening purposes. Assessment sessions took pl ace at the school and were completed in multiple sittings to accommodate the needs of young children and to avoid fatigue. Assessment at the beginning of the study was completed by the investigator and three graduate students from the University of Florida. The three graduate assistants were also fluent in English and Spani sh and had background knowledge of reading, second-language learning, and educational assessment. T hey were pursuing graduate studies in School Psychology, and were famili ar with a few of the assessments used in this study. The investigator provided one 2-hour training session where trainees were

PAGE 124

124 able to learn about the assessments procedur es and to practice implementing them. Assessment at the end of the study was completed only by the researcher, given that time requirements were less demanding since the WMLS-R was not administered at that time. In addition to directly measuring students early literacy skills, teachers were also asked to complete a reading behavior rati ng scale to measure students changes as a result of the intervention. The scale addres sed five areas of reading (i.e., phonemic awareness, phonics, fluency, comprehension, and vocabulary), as well as classroom behaviors (i.e., participation in class, mo tivation to read, and use of English to communicate in class). Each teacher rece ived a rating scale before the study started and one at the culminati on of the study. The Book Levels during Intervention During intervention, data were collected for each group regarding the changes in book levels read throughout the tutoring less ons. These data were then plotted into a graph for visual analysis. The graph depicts the starting reading le vel for each group, the number of lessons that each group rema ined at each reading level, and the level reached at the end of the intervention. This data provided additional information on the effectiveness of the intervention. Interobserver Agreement To ensure that the dependent variables (CPPM and CSPM) were measured and recorded with integrity, interobserver agr eements (IOA) were es tablished between the tutor and an observer. The tutor and observer listened to students reading of pseudowords and sight words probes duri ng one-minute timings. Independently, the tutor and the observer recorded the number of words read correctly and calcula te the

PAGE 125

125 rate of accuracy. To calculate the leve l of agreement, a frequen cy-ratio approach was used. In this approach, the researcher calc ulated the total number of responses that each observer recorded. Then, the smaller to tal of responses (S) was divided by the larger total of responses (L) and then was mu ltiplied by 100 (S/L x 100). An acceptable level of agreement is 80% (Kazdin, 1982; Kennedy, 2005). The level of agreement for all pseudoword observations was 93.8% and for sight word observations was 98.40%. These levels of agreement were based on a to tal of five observations for pseudoword and sight words conducted for each group: one during baseline, three during intervention, and one during maintenance. The only exception was an additional observation conducted with one student in Gr oup 1. On that occasion, Amelia was absent and Pedro was not able to speak out loud due to illness. An IOA was conducted based on only one student. Treatment Integrity Several measures were taken to ens ure that the tuto ring program was implement ed in a reliable manner. First, duri ng each lesson the tutor had a copy of the session guide that outlined eac h of the four steps. The session notes sheets also served as an additional guide for the tutor sinc e they outlined each of the steps in the lesson. Second, the tutor conducted self-evaluations using a treatment fidelity checklist (Appendix H). One checklist was completed every week for each of the participating groups. For Group 1 a total of 11 checklists were completed. The mean percentage of adherence was 97.45%. For Gr oup 2, 14 checklists were completed with a mean percentage of adherence of 96. 6%. For group 3, a total of 13 checklists were completed. The mean percent age of adherence was 96.2%.

PAGE 126

126 Third, the tutors adherence to the program was evaluated by one of two observers, who completed a treatment fide lity checklist on each visit. The tutor was observed two times with each group. Observer 1, a professor from the University of Florida who developed the UFLI tutori ng program completed two observations. Observer 2, a doctoral student in Special Educati on at the University of Florida and experienced UFLI tutor, co mpleted four observations. Af ter each session, the observer and the tutor met to discuss the integrity of the intervention. The researcher then calculated the percentage of adherence by dividing the total number of items completed by the total number of lesson components an d multiplying it by 100. The mean adherence percentage for all six observations was 96%. Adherence percentage for each treatment integrity observation is reported in Table 3-8. Table 3-8. Treatment integrity Group number Observation number Number of items completed Percentage of adherence 1 1 39/42 93% 2 40/42 95% 2 1 41/42 98% 2 41/42 98% 3 1 40/42 95% 2 40/42 95% Social Validity Social validity is defined by Kennedy (2005) as the esti mation of the importance, effectiveness, appropriateness, and/or satisf action various people ex perience in relation to a particular intervention (p 219). This is especially im portant in educational settings because it gives the researcher an indication of the level of acceptability of the intervention among its consumers (p. 218). To estimate social valid ity, a subjective evaluation approach (subjective evaluation) was conducted to obtain feedback from

PAGE 127

127 students and teachers about the effectiveness and the acceptability of the intervention (Kennedy, 2005). First, each student participated in a brief in terview at the end of the maintenance phase. Questions focused on students opinions about the UFLI tutoring lesson and its components. Each interview took approximatel y five minutes to be completed. Second, all the teachers of the par ticipating students were invi ted to observe a videotaped tutoring lesson and complete a social validity questionnaire. The goal of the questionnaire was to evaluate the accept ability of the tutoring program and its components and to determine whether teachers w ould see themselves using this lesson or any of its components in their cla ssrooms. The questionnaire was adapted from Pullen (2000) to address the UF LI program components as it re lates to ELLs. A copy of the students interview questions and the teachers questionnaire can be found in Appendix I. The researcher sent home consent forms asking parents for permission to videotape their children during a tutoring le sson (see Appendix A). A total of five students received parental consent to participate in the video, three from Group 2 (Maggie, Maria, and Viviana) and two from Group 3 (Aldo and Jennifer). Because of scheduling difficulties, one studen t from Group 2 (Viviana) and two students from Group 3 (Jennifer and Aldo) were selected to parti cipate. Once the videotape was completed, teachers watched the tape and were asked to complete the questionnaires. Teachers were provided a copy of the lesson guide to help them identif y the different steps of the program as they observed the video. Due to time constraints, teachers opted to complete the questionnaire at a late r time. The response rate was 100%.

PAGE 128

128 Delimitations and Limitations There were several delimitations and limitat ions to the study. First, delimitations and their impact on external validity are addre ssed. Second, the studys limitations are described as well as how each may have influenced the findings and their generalizability. Delimitations The first delimitation pertains to the gr oup of students selected to participate. All students were in second-grade and all spok e Spanish as their first language. This may limit the generalization of findings to other grade levels and other groups of ELLs that might speak another language at home. A second delimitation is related to the par ticular setting in which the program was implemented. This study was conducted in only one school located in a rural area of North Florida. This particular setting has a se t of variables that may differ from other schools (e.g., student population, human resources, instructional resources, scheduling) and these variables may have influenced intervention effects in some way (Kazdin, 1982). For example, the schools schedule allowed for daily 45-minute sessions that facilitated the provision of t he UFLI program to all groups for an extended period of time. While the number of weekly sessions varied across groups due to school activities that conflicted with the establis hed schedule, the length of t he intervention was extended until students reached grade leve l reading or until they completed a minimum of 40 tutoring sessions. Also, since the school serves all second and third grade students in the county, there was a suffici ent number of Spanis h-speaking ELLs in second-grade to group second graders appropriately by reading level. This might not be the case in other

PAGE 129

129 schools, where there may be a smaller number of ELLs Changing group composition could influence the efficacy of the UFLI program and could alter the results. A third delimitation is related to the tutor implementing the intervention. In this study, the intervention was conducted only by the researcher. According to Kazdin (1982), there is a possibility that the characte ristics of the individual implementing the intervention may help attain the intervention e ffects. In this case, the researcher shared a similar heritage with the majority of the participants, was proficient in English and Spanish, had extensive kno wledge about the reading proc ess and literacy development among ELLs, and had experience implementi ng the UFLI progr am with Spanishspeaking ELLs. It is unknown at this point if similar intervention effects can be attained if the UFLI program is implemented with ELLs by other tutors with different sets of characteristics, knowledge, and skills. Limitations The first limitation is inherent to single-subject design and pertain s to sample size. One of the main criticisms against this met hodology is that the results of a particular study may not be generalized to larger groups of subjects due to the limited sample size (Kazdin, 1982; Neuman & McCormick, 1995). To increase the external validity of results, direct and systematic replication is necessary. The second limitation is related to treatm ent interference. In this study, students were participating in the SRA Early Reading Intervention (ERI) program and/or Harcourt Interventions while the study took place. While the use of a multiple baseline design was used to establish control across groups it is unknown to what degree their participation in other programs mi ght have influenced the results.

PAGE 130

130 Description of One-on-One Tutoring Methodology For the student that received one-onone tutoring (Miriam pseudonym), data were collec ted on rates of CPPM and CSPM during baseline, intervention, and maintenance. Preand post-intervention data were collected using the same assessment instruments implem ented with the other par ticipants. Running record trials at the beginning of the study showed that she was reading level 6 books at an instructional level. Table 3-9 shows the rate of accuracy in each running record. What follows is a brief description Miriams demographic data, instructional information, reading scores used for selection, language levels, and classroom behaviors. Table 3-9. Running records results used for Miriam Student Trial 1 Trial 2 Trial 3 Trial 4 Trial 5 Trial 6 Miriam 10 (85%) 9 (88%) 8 (89%) 7 (83%) 6 (92%) 6 (93%) Miriam was a Hispanic female, aged 7 years-11 months, enrolled in second grade for the first time. Her parent s were from Mexico and they only spoke Spanish at home. Her mother reported not finishing high school. Her mother stated t hat Miriam had had four years of instruction in the US. As part of her daily r eading instruction, she received the SRA/EIR program in small groups, five times a week, for 45 minutes. School records showed that Miriam had an LEP stat us, but did not receive LEP classes. Reading scores showed that at the end of first grade, Miriam scored in the 46th percentile on the SAT-10. Her DIBELS scores for the end of first grade were 50 on PSF (low risk), 36 on NWF (moderate risk), and 27 on ORF (moderate risk). At the beginning of second grade, she scored between the 10th and 20th percentil e on the Invented Spelling Assessment. Language scores on the WMLS-R demonstrated that she had limited to fluent English oral language ability (Level 3.5) and limited Spanish oral

PAGE 131

131 language ability (Level 3). On the C-ATRS, her teacher r eported that Miriam had no attitude or behavior problems in class. She affirmed that Miriam was a very easy-going student and that she was almost always focused on her work. The student received the UFLI individual tutoring program, which follows the same principles and addresses the same reading skills as the group program, but follows a different lesson format. In each lesson, a fifth step is included, which focuses on extending literacy practices by exposing students to different types of reading genre. An outline of the steps for one-on-one UFLI tu toring can be found in Appendix J. Pilot Study A pilot study was conducted prior to this research study with the purpose of evaluating the effectiveness of the UFLI tutori ng program on the reading skills of Spanish-speaking English language learners, who we re experiencing reading difficulties. In the pilot study, a mult iple baseline across participants design was implemented with a total of three first-grade students. The par ticipants were selected based on their English language learner status and their atrisk reading status as determined by DIBELS scores. The experi mental procedures (movement across phases) were similar to the ones described in this study. The dependent variable was the rate of correct pseudowords read per minute (CPPM), using progress monitoring probes from DIBELS. The tutoring lessons had all the components addressed in this study plus the addition of a fifth step that fo cused on extending literacy practices. In that step, students were exposed to different types of reading genre with the goal of increasing awareness of different text structures (Lane et al., 2009). Students participated in an average of 22 tu toring lessons, 4 to 5 days a week. Each lesson lasted approximately 45 minutes. Analysis of data showed that all students

PAGE 132

132 made a marked improvement by the end of the intervention. Student A went from a total of 4 CPPM during baseline to 24 CPPM at the end of the in tervention. Student A also improved on book levels read, going from level 3 to level 10 in 27 sessions. Student B went from a total of 10 CPPM during baseline to 34 CPPM at the end of the intervention. In book reading, she went from level 3 to level 17 in only 16 sessions. Student C went from 1 CPPM to 9 CPPM after intervention, and from a level 1 books to a level 5 in 25 sessions. Students in the pilot study varied in their in itial level of oral language proficiency based the WMLS-R CALP levels. Student A and B demonstrated limited English oral proficiency (Level 3), while student C demonstrated very limit ed English oral proficiency (Level 2). Regardless of the level of English oral profic iency, all students responded to the intervention, a finding that was also r eported in studies conducted by Gunn et al. (2000; 2002; 2005), and LinanThompson et al. (2003). As a result of the pilot study, and in agr eement with other res earch studies (LinanThompson et al., 2003; Mohr & Mohr, 2007) a fe w modifications were made to the UFLI tutoring program to address the language needs of English language learners. First, picture cards and quick definitions were us ed to clarify unfamiliar words during reading and word work activities. Second, real words and nonsense words were identified during decoding and encoding activities to help students make connections between written words and their meaning. Finally, complete sentences and standard grammar were modeled by the tutor to help students express complete ideas.

PAGE 133

133 CHAPTER 4 RESULTS The purpos e of this study was to evaluate the effectiveness of a modified version of the University of Florida Literacy In itiative small group tutoring program on the reading skills of second grade Spanish-spea king English language learners who are struggling to read. For this purpose, a multiple baseline across groups design was implemented to measure students response to intervention. The goal of this chapter is to present the results obtained throughout the study. The chapter is divided in several sections: (a) dependent variables, (b) total number of tutoring sessions (c) supplemental data, (d) social validity, (e ) summary of one-on-one tutori ng data, and (f) summary of findings. Dependent Variables Throughout the three phases of the st udy, CPPM and CSPM data were collected using researcher-created probes. After each probe was implemented, the rate of CPPM and CSPM was first calculated for each gr oup member. A few students (i.e., Amelia, Pedro, Jessica, and Maggie) completed some of the CSPM probes within one minute. On those occasions, the time remaining was recorded and t hen used to estimate what the actual number of correct sight word s per minute would hav e been if students had continued reading for the full minute. Once individual rates of CPPM and CSPM were calculated, a group mean was estimated and plotted into a graph. At t he end of each phase, overall mean rates of CPPM and CSPM were calculated for each group. Decisions about movement from one phase to the next were made based on each groups rate of CPPM.

PAGE 134

134 Results for the two dependent variables are organized by phases: (a) baseline, (b) intervention, and (c) maintenan ce. In each phase, group data ar e reported, including the range of mean scores and the overall mean fo r the entire phase. In addition, the difference in overall means between baselin e and intervention, as well as between intervention and maintenance are reported. Individual data can be found in Tables 4-1 to 4-3. Baseline Phase During bas eline, the rate of CPPM and CSPM were calculated until a stable line of response was established for each group. Each data session lasted approximately 6 to 10 minutes per group. Individual da ta is presented in Table 4-1. Group 1 data. The mean rate of correct pseudow ords read per minute for Group 1 ranged from 6 to 7 CPPM with an overall phase mean of 6.33 CPPM. The mean rate of sight words ranged from 48.67 to 52.39 CSPM with an overall phase mean of 50.82 CSPM. Data were collected during a total of six baseline sessions. Group 2 data. The mean rate of pseudoword reading for Group 2 ranged from 4.25 to 5 CPPM, with an overall phase mean of 4.75 CPPM. The mean rate of sight words ranged from 22.5 to 24.3 CPPM, with an overall mean of 23.82 CSPM. Data were collected over a total of seven sessions. Group 3 data. The mean rate of correct pseudowor ds read per minute for Group 3 ranged from 1.66 to 3 CPPM, with a phase m ean score of 2.42 CPPM. For sight word reading, the mean ranged from 1.7 to 3.7 C SPM, with an overall p hase mean of 2.63 CSPM. Data were collected over 11 sessions. Summary. During baseline, all groups displa yed no significant increase or decrease in the trend of response for CPPM or CSPM. Baseline levels differed across

PAGE 135

135 groups. Group 1 had the highest baseline rates for both variables, followed by Group 2 and Group 3. The difference across groups for CPPM was smaller than for CSPM. Between Group 1 and Group 2, there was a di fference of 1.58 CPPM, compared to a difference of 27 CSPM. Similarly, betw een Group 2 and Group 3 there was a difference of 2.33 CPPM, compared to 21.19 CSPM. Table 4-1. Individual rates of CPPM and CSPM during baseline CPPM CSPM Participants Range Mean Range Mean Amelia 12-15 13.5 62.72-68.1866.35 Pedro 1-3 1 49.41-5552.80 Jessica 3-5 3.83 32-34 33.33 Viviana 1-2 1.57 12-15 14.3 Ernesto 2-5 4.14 28-31 29.9 Maggie 7-9 8.28 30-33 31.7 Maria 4-7 5 18-21 42 Jennifer 3-5 3.90 2-5 4 Aldo 0 0 0-1 0.36 Jorge 2-4 3.36 2-5 3.54 Intervention Phase During intervention, probes were given to each participant at the beginning of the tutoring session (6 to 10 minutes per group). T he purpose of this phase was to establish a change in the rate of CPPM and CSPM from baseline, as well as the progress in rate as a result of the interv ention. Individual dat a is presented in Table 4-2. Group 1 data. Once a stable trend line was es tablished in baseline, Group 1 started the intervention phas e. During this phase, the group received a total of 24 pseudoword and sight word probes. The mean rate of pseudoword ranged from 11 to 34 CPPM, with an overall phase mean of 23. 88 CPPM. This represents a positive change of 17.55 CPPM from baseline to intervention. For sight wo rds, the mean rate

PAGE 136

136 ranged from 51.53 to 73.99 CSPM with a phase mean of 62.96 CSPM. This shows an increase of 12.14 CSPM from baseline to intervention. Group 2 data. When Group 1 showed an increase in the rate of pseudoword reading over four consecutive sessions, Gro up 2 started the intervention phase. In the meantime, Group 3 cont inued on baseline. During the in tervention phase, Group 2 received a total of 26 pseudoword and si ght word probes. The mean rate of pseudoword ranged from 7.25 to 18.3 CPPM, with a phase mean of 13.41CPPM. This shows a positive change of 8.66 CPPM from baseline to inte rvention. For sight words, the mean rate ranged from 24.25 to 41.43 CSPM, with a phase mean of 33.42 CSPM. This shows an increase of 9.6 CSPM from baseline to intervention. Group 3 data. When Group 2 showed an increase in the rate of pseudoword reading over four consecutive sessions, Gro up 3 started the intervention phase. During the intervention phase, Group 3 received a total of 25 pseudoword and sight word probes. The mean rate of pseudoword ranged fr om 5 to 13 CPPM, with an overall mean of 8.53 CPPM. This shows a positive c hange of 6.11 CPPM from baseline to intervention. For sight words, the mean rate ranged from 3 to 12.3 CSPM, with an overall mean of 7.14 CSPM. This shows an increase of 4.51 CSPM from baseline to intervention. Summary. During intervention, all groups s howed an increase in the rate of pseudowords and sight words correctly read per minute. The rate of improvement during intervention varied across groups on bot h variables. Based on the overall phase mean for each group, it was observed that Group 1 had the highest rate of improvement on CPPM and CSPM, followed by Group 2 and Group 3. From baseline to intervention,

PAGE 137

137 Group 1 had an increase of 17.55 CPPM and 12.14 CSPM. Group 2 had an increase of 8.66 CPPM and 9.6 CSPM, while Group 3 had an increase of 6.11 CPPM and 4.51 CSPM. Table 4-2. Individual rate of CPPM and CSPM during intervention CPPM CSPM Range Mean Change from baseline Range Mean Change from baseline Amelia 17-42 30.29 16. 7966.66-93.7576.25 9.9 Pedro 11-39 27.04 27.38 51.6-72.6365.23 12.42 Jessica 6-24 14.20 10.3733-57.64 47.45 14.12 Viviana 4-7 11.11 9.5414-31 23.88 9.6 Ernesto 0-16 11.80 7.6630-47 40.38 10.53 Maggie 9-24 17.5 9. 2223-52.63 41.17 9.46 Maria 7-18 13.23 8.2319-38 28.26 8.84 Jennifer 9-19 14.12 10.225-19 10.52 6.52 Aldo 0-6 1.96 1.960-8 3.2 2.84 Jorge 5-14 9.52 6.163-12 7.72 4.18 During this phase, Aldos rate of pseudoword showed a unique trend. He moved up and down between 0 and 2 CPPM for 19 pseudoword probes. On the 20th probe, he moved up to 5 CPPM and by the 25th probe he was reading 6 CPPM. Maintenance Phase The maintenance phase started for each group two weeks after the intervention phase ended, with the purpose of determining whether the rate of CPPM and CSPM were sustained over time. The maintenanc e phase ended for each group when data points showed a stable line at an appropriate reading level. Individual data can be found in Table 4-3. Group 1 data. During the maintenance phase, data were collected over five sessions. The mean rate of pseudowords reading for Group 1 ranged from 36 to 40 CPPM with an overall mean of 37.87 CPPM, an increase of 13.99 CPPM. In addition, the mean rate of sight word reading ranged from 72.71 to 75.56 CS PM, with an overall

PAGE 138

138 phase mean of 74.03 CSPM. This shows a pos itive improvement fr om intervention to maintenance of 11.07 CSPM. Group 2 data. Maintenance data were collected over six sessions. Data showed the rate of pseudowords r anged from 15.25 to 21.8 CPPM, with an overall phase mean of 18.88 CPPM. In comparison to the intervention phase, the maintenance phase showed a positive improvement of 5.47 CPPM. Data also showed that the rate of sight words ranged from 38.75 to 44.17 CSPM, with a phase mean of 41.51 CSPM. This shows in increase of 8.09 C SPM from the previous phase. Group 3 data. During the maintenance phase, dat a were collected over six sessions. Group 3 showed a rate of pseudowor ds that ranged from 13 to 14.7 CPPM, with a phase mean of 13.8 CPPM When compared to the intervention phase, Group 3 showed an increase of 5.27 CPPM. Data gat hered on rate of sight words ranged from 10 to 13 CSPM, with a phase mean of 12.22 C SPM. This shows an increase of 5.08 CSPM from intervention to maintenance. Summary. During maintenance, all groups sustai ned and improved on their rates of CPPM and CSPM. For CPPM, Group 2 showed a larger difference from the previous phase (18.88 CPPM), followed by Group 1 (11.71 CPPM), and Group 3 (5.27 CPPM). For CSPM, Group 1 showed the largest diffe rence from the previous phase (14.56 CSPM), followed by Group 2 (8.09 CSPM), and Group 3 (5.08 CSPM). Summary of Dependent Variables Visual inspection of the data (Figures 4-1 and 4-2) reveals that all groups achieved an immediate improvement in both dependent variables once the intervention started. Group 1, for example, moved from a mean rate of 6 CPPM in baseline to 11 CPPM after only one tutoring session. Group 2 went from 4.8 CPPM in baseline to 7.25 CPPM and

PAGE 139

139 Table 4-3. Individual rate of CPPM and CSPM during maintenance CPPM CSPM Range Mean Change from intervention Range Mean Change from intervention Amelia 40-4342 11.71 88.23-94.8390.81 14.56 Pedro 38-4240 12.96 73. 16-76.9275.03 9.8 Jessica 29-3431.6 17.4 54.23-58.856.25 8.8 Viviana 14-1816.16 5. 05 29.34 31.83 7.95 Ernesto 13-1916.16 4.36 42-47 44.16 3.78 Maggie 19-2924.16 6.66 47-53.5750.39 9.22 Maria 15-2119 5.77 35-45 39.66 11.40 Jennifer 17-1918.33 4.21 16.18 17 6.48 Aldo 9-1110.16 8.2 6-9 7.83 4.63 Jorge 12-1513.5 3.98 8-14 11.83 4.11 Group 3 moved from 2.66 CPPM in baselin e to 5.33. The trend of improvement continued for all groups as the tutoring sessions progressed. For example, the rate of CPPM during intervention ranged from 11 to 34 for Group 1, 7.25 to 18.25 for Group 2, and 5.33 to 13 for Group 3. A comparison of overall group means between baseline and intervention showed a dramatic change ac ross groups. Specific ally, Group 1 moved from a mean rate of 6.33 CPPM to 23.88 CPPM. Similarly, Group 2 mo ved from a mean rate of 4.75 CPPM to 13.41 CPPM, while Group 3 moved from 2.42 CPPM to 8.53 CPPM. Improvement was also noticed for the rate of correct sight word s read per minute. The immediacy of change observed after one tutoring lesson was less dramatic for CSPM. Group 1, for example, moved from a mean rate of 51.27 C SPM in baseline to 51.33 CSPM. Group 2 went from 24 CSPM to 24.25 CSPM. Similarly, Group 3 moved from 2.33 CSPM to 3 CSPM. The trend of impr ovement continued for all groups as the

PAGE 140

140 Figure 4-1. Rate of correct pseud owords per minute (CPPM) across groups.

PAGE 141

141 Figure 4-2. Rate of correct sight words read per minute (CSPM) across groups.

PAGE 142

142 tutoring sessions progressed. For example, the rate of CSPM during intervention ranged from 51.53 to 73.99 CSPM for Group 1, 24.25 to 41.45 CSPM for Group 2, and 3 to 12.33 CSPM for Group 3. A comparison of overall group means between baseline and intervention showed a noticeable change across groups. Specifically, Group 1 moved from a mean rate of 50.82 CSPM to 62.96 CSPM. Similarly, Group 2 moved from a mean rate of 23.82 CSPM to 33.42 CSPM, while Group 3 moved from 2.63 CSPM to 7.14 CSPM. The effect of the UFLI tutoring program on the rate of correct pseudowords and sight words read per minute was sustained tw o weeks after the intervention ceased. Moreover, the overall mean rate of CPPM and CSPM fo r all groups continued to increase during the maintenance phase, dem onstrating that students had acquired the reading skills taught and continued to use them independently. For Group 1, the overall mean rate of CPPM went from 23.88 to 37.87 CPPM, while the rate of CSPM went from 62.96 to 74.03 CSPM. Similarly, for Group 2, the rate of CPPM we nt from 13.41 to 18.88 CPPM, while the rate of CSPM went fr om 33.42 to 41.51. For Group 3, the rate of CPPM went from 8.53 to 13. 8 CPPM, and the rate of CSPM went from 7.14 to 12.22. Total Number of Sessions The number of the tutoring sessions va ried across groups. Two elements were considered in determining the total number of sessions for each group. First, tutoring sessions would be terminated once all mem bers were reading level 20 books with at least 90% accuracy (instructional level). Se cond, additional tutoring sessions would be added to allow all participants wh o had been absent to participate in a minimum of 40 sessions. A summary of the total number of sessions fo r each group is presented in Table 4-4.

PAGE 143

143 Group 1 reached level 20 books on session 35. On that session, Amelia read a level 19 book with 96% accuracy, allowing t he group to advance to level 20. To ensure that all members of the gr oup were reading level 20 books at an instructional level, three more sessions were conducted. By t hat time, everyone in Group 1 was able to read level 20 books at an independent level (98% to 100%). Af ter considering the total number of absences each student had, Amelia participated in a total of 34 sessions. She had six absences but was able to make up two sessions. The two make up sessions were done individually and covered all the steps of t he small-group lesson format. Jessica participated in a total of 37 sessions, while Pedro participated in a total of 38 sessions. Group 2 had a total of 47 tutoring sessions. This group did not reach level 20 at the end of the intervention. Therefore, to ensure that all member s participated in a minimum of 40 sessions, seven extra session s were added to compensate for students absences. Vivian had four absences, so she participated in a total of 43 sessions. Ernesto missed five sessions, so he participat ed in a total of 42 sessions. Maria missed seven sessions, participating in a total of 40 tutoring sessions. Finally, Maggie received the most sessions (47 total) since she did not miss one day of tutoring. Group 3 had a total of 45 sessions to compensate for students absences. This group did not reach level 20 at the end of the intervention. Jennifer did not miss one day of tutoring, so she participated in all 45 sessions. Aldo had two absences, so he participated in a total of 43 sessions, while Jorge, with five absences, participated in 40 sessions.

PAGE 144

144 Table 4-4. Total number of tuto ring sessions received by each student Participants Total absences Total number of sessions attended/group total Amelia 6 34/38 Pedro 0 38/38 Jessica 1 37/38 Viviana 4 43/47 Ernesto 5 42/47 Maggie 7 40/47 Maria 0 47/47 Jennifer 0 45/45 Aldo 2 43/45 Jorge 5 40/45 Supplemental Data In addition to the dependent variable, preand post-intervention data were collected o n students early literacy skills and attitudes toward reading, as well as teachers ratings of students reading abilitie s and classroom behaviors. Also, data were gathered on book levels read by each group during intervention. What follows is detailed description of findings. Results can be found in Tables 4-5 to 4-13. Measures of Early Literacy Skills and Attitudes to wards Reading Several measures of reading ability were measured before and after intervention. These included: (a) decoding accuracy and fl uency, (b) word recognition accuracy and fluency, (c) reading fluency, and (d) reading co mprehension. Students attitudes towards both recreational and academic reading were measured as well. Decoding accuracy and fluency. Decoding accuracy was assessed by the Nonsense Word Decoding subtest (NWD KTEA-II), while decoding fluency was measured by the (a) Phonemic Decoding Efficiency subtes t (PDE TOWRE) and the (b) Nonsense Word Fluency test (NWF DIBELS). Students made a marked improvement in both skills after interv ention. The NWD showed that all students

PAGE 145

145 improved on decoding accuracy, with a mi nimum gain of 6 nonsense words read correctly and a maximum of 21 words read correctly. The PDE test showed improvement in reading fluency with gains that ranged from 4 correct nonsense words to 16 nonsense words. A dramatic change was also observed on the NWF (DIBELS), where students gains ranged from 11 to 68 correct letter-sound correspondences per minute. According to risk levels, before inte rvention there were tw o students at high risk, six at moderate risk, two at low risk, and none at above average. In contrast, after intervention, there were no st udents at high or moderate risk, while six were at low risk and 4 at above average. Individual raw sco res and percentile ranks can be found in Tables 4-5 to 4-7. Word recognition accuracy and fluency. Word recognition fluency was measured by the Sight Word Efficiency te st (SWE TOWRE), while accuracy was measured by the Letter Word Recognition subtest (LWR KTEA-II). Results showed that all students made positive gains in both skills. Students gains in word reading accuracy ranged from 5 to 15 words read corre ctly, while gains in word reading fluency ranged from 0 to 15 words per minute. In dividual data, including raw scores and percentile ranks can be found in Tables 4-5 and 4-6. Table 4-5. Preand post-interventi on TOWRE raw scores and percentiles ranks Participants Pre-SWE Post-SWEPre-PD E Post-PDE Pre-total Post-total Amelia 47 (73%) 62 (84%) 20 (64%) 35 (89%) 70% 91% Pedro 46 (70%) 57 (74%) 14 (45%) 24 (66%) 61% 74% Jessica 41 (61%) 41 (39%) 13 (42%) 22 (61%) 52% 50% Viviana 25 (27%) 33 (25%) 2 (12%) 15 (39%) 14% 29% Ernesto 34 (45%) 45 (48%) 4 (16%) 17 (45%) 25% 42% Maggie 36 (50%) 46 (48%) 7 (21%) 18 (48%) 32% 48% Maria 27 (32%) 38 (32%) 11 (32%) 15 (39%) 29% 32% Jennifer 11 (8%) 29 (19%) 8 (25%) 24 (66%) 10% 39% Aldo 5 (3%) 18 (6%) 0 (10%) 12 (29%) 3% 10% Jorge 13 (10%) 34 (25%) 10 (29%) 22 (61%) 14% 39% SWE= sight word efficiency. PDE = phonemic decoding efficiency. Total = total word reading efficiency.

PAGE 146

146 Table 4-6. Preand post-intervention KT EA-II raw scores and percentiles ranks Participants Pre-LWR Post -LWR Pre-NWDPost-NWD Amelia 44 (50%) 49 (53%) 4 (25%) 23 (75%) Pedro 35 (30%) 50 (58%) 8 (45%) 23 (75%) Jessica 40 (42%) 49 (53%) 9 (50%) 15 (58%) Viviana 24 (5%) 36 (18%) 2 (16%) 16 (58%) Ernesto 29 (16%) 40(27%) 2 (16%) 13 (53%) Maggie 39 (30%) 44 (37%) 3 (21%) 21 (70%) Maria 31 (21%) 41 (30%) 0 (3%) 21 (70%) Jennifer 23 (4%) 30 (10%) 3 (21%) 14 (55%) Aldo 15 (0.5%) 29 (8%) 0 (3%) 11 (45%) Jorge 20 (2%) 31 (30%) 4 (5%) 14 (55%) LWR stands for letter-word recognition. NW D stands for Nonsense Word Decoding. Table 4-7. Preand post-intervention DIBELS Nonsense Word Fluency scores Participants Preintervention Postintervention Amelia 67(LR) 116 (AA) Pedro 45 (MR) 113 (AA) Jessica 47 (MR) 112 (AA) Viviana 46 (MR) 57 (LR) Ernesto 46 (MR) 63 (LR) Maggie 66 (LR) 64 (LR) Maria 35 (MR) 54 (LR) Jennifer 42 (MR) 99 (AA) Aldo 3 (HR) 51 (LR) Jorge 38 (HR) 59 (LR) Scores refer to the number of correct lette r-sound correspondences read per minute. AA = above average, LR = low risk, MR = moderate risk, and HR = high risk. Reading fluency. The Oral Reading Fluency test (ORF DIBELS) was used to measure students fluency of connected tex t. Results showed that all students made gains from the beginning of second grade passage to the end of second grade passage, with gains ranging from 9 to 54 correct word s per minute. On grade-level passages, at the beginning of the interventi on, six students were at high risk and 4 were at low risk. In contrast, at the end of the intervention, five students were at high risk, three at moderate risk, and two at low risk. Students who were at moderate risk or high risk on any of the passages were administered easier passages until a low risk or ab ove average status

PAGE 147

147 were achieved, or until passages corres ponding to the middle of first grade were reached. Individual data can be found in Tables 4-8 and 4-9. Table 4-8. Pre-intervention DI BELS Oral Reading Fluency scores Participants Beginning-second grade ORF passage End-first grade ORF passage Mid-first grade ORF passage Amelia 55 (LR) Pedro 63 (LR) Jessica 61 (LR) Viviana 14 (HR) 11 (HR) 19 (MR) Ernesto 25 (HR) 34 (MR) 47 (AA) Maggie 45 (LR) Maria 16 (HR) 21 (MR) 18 (MR) Jennifer 7 (HR) 10 (HR) 6 (HR) Aldo 3 (HR) 1 (HR) 3 (HR) Jorge 7 (HR) 7 (HR) 3 (HR) Scores correspond to the total number of words read correctly per minute. Passages with no scores were not administered to those particular students. AA = ab ove average, LR = low risk, MR = moderate risk, and HR = high risk. Table 4-9. Post-intervention DI BELS Oral Reading Fluency scores ORF Passages Participants End-2nd grade Mid-2nd grade Beginning 2nd grade End-1st grade Mid-1st grade Amelia 102 (LR) Pedro 94 (LR) Jessica 75 (MR) 73 (LR) Viviana 49 (HR) 48 (HR)46 (LR) Ernesto 79 (MR) 65 (MR)74 (AA) Maggie 79 (MR) 66 (MR)62 (LR) Maria 56 (HR) 52 (MR)48 (LR) Jennifer 33 (HR) 34 (MR)28 (MR) 33 (LR) Aldo 12 (HR) 11 (HR)11 (HR) 17 (MR) Jorge 32 (HR) 35 (MR)34 (MR) 31 (LR) Scores correspond to the total number of correct word s read per minute. Passages with no scores were not administered to those particular students. AA = ab ove average, LR = low risk, MR = moderate risk, and HR = high risk. Reading comprehension. The Qualitative Reading Inventory-4 (QRI-4) was administered to measure reading comprehension in two ways: (a) ideas recalled, and (b) questions answered. The passage se lected to read was contingent on ORF (DIBELS) risk status. Before intervention, four students read level two passages, one

PAGE 148

148 student read a primer pass age, and four students did not read any passages. In contrast, after intervention, seven students read level two passages, two read primer passages, and only one student did not read any passages. Results showed that all students increased the number of ideas recalled. For the total number of questions answered correctly (explicit and implicit), scores remained the same for one student and increased for the other nine students. Individual data c an be found in Tables 4-10. Attitudes toward reading. The Elementary Reading Attitude Survey (ERAS) was used to assess students attitudes toward reading. This survey measures attitudes toward academic reading and recreational readi ng. It also provides a total attitude toward reading score. Results showed that all students made a marked improvement in attitudes toward reading in general, with gai ns in raw scores ranging from 15 to 38. Changes in attitudes toward recreational reading ranged from 4 to 19, and toward academic reading ranged from 5 to 23. A su mmary of raw scores and percentile ranks is presented in Table 4-11. Reading Ability Rating Scale. Before and after the intervention, teachers evaluated the reading ability of their students using the Read ing Ability Rating Scale. This scale addressed five key areas of r eading: (i.e., phonological awareness, phonics, fluency, comprehension, and vocabulary), and one area related to classroom behaviors (i.e., participation in class, use of English to communicate in cla ss, and motivation to read). Mean scores were calculated and report ed for each of the five reading areas and raw scores were reported for classroom behav iors. Preand post-intervention scores showed equal or higher ratings for all students on phonological awareness, phonics, fluency, and comprehension. In contrast, in the area of vocabul ary, six students

PAGE 149

149 Table 4-10. Pre/Post-intervent ion QRI-4 comprehension raw scores Passage Level Ideas Recalled Explicit Questions Implicit Questions Total Questions Reading Level Student Pre Post Pre Post Pre Post Pre Post Pre Po st Pre Post Amelia 2 2 5 12 3 4 3 2 6 6 Inst Inst Pedro 2 2 5 19 4 4 2 4 6 8 Inst Ind Jessica 2 2 4 8 1 3 1 1 2 4 Frus Frus Viviana N/A 2 5 2 0 2 Frus Ernesto P 2 0 9 1 3 0 2 1 5 Frus Frus Maggie 2 2 8 9 3 4 0 2 3 6 Frus Inst Maria N/A 2 10 4 3 7 Inst Jennifer N/A P 5 2 1 3 Frus Aldo N/A N/A Jorge N/A P 4 2 1 3 Frus Students read one of three passages: P = Primer, 1 = Level 1, 2 = Level 2. Ideas recalled, explicit questions, implicit questio ns, and total questions answered correctly correspond to the passage read by each student. Reading levels were identified based on the number of total correct answers based on QRI-4 norms. Inst = Instructional, Ind = Independent, Frus = Frustration. Students with no scores did not take the QR I-4 due to risk status on DIBELS ORF.

PAGE 150

150 Table 4-11. ERAS Pre-and post-interv ention raw scores and percentile ranks Recreational Reading Academ ic Reading Total Reading Participants Pre Post Pre Post Pre Post Amelia 27 (31%) 33 (68%)25 (32%)36 (85%)52 (29%) 69 (79%) Pedro 23 (11%) 29 (44%)28 (46%)37 (89%)51 (26%) 66 (71%) Jessica 21 (6%) 30 (50%)25 (32%)34 (78%)46 (1%) 64 (66%) Viviana 20 (4%) 39 (94%)21 (15%)36 (85%)41 (6%) 75 (92%) Ernesto 30 (50%) 34 (74%)19 (8%)32 (67%)49 (20%) 66 (71%) Maggie 21 (6%) 36 (84%)23 (23%)37 (89%)44 (9%) 73 (88%) Maria 24 (15%) 38 (92%)26 (37%)35 (81%)50 (23%) 73 (88%) Jennifer 23 (11%) 30 (50%)21 (15%)37 (89%)44 (9%) 67 (74%) Aldo 16 (1%) 31 (56%)14 (1%)37 (89%)30 (0%) 68 (77%) Jorge 13 (0%) 21 (8%)19 (8%)27 (41%)32 (1%) 48 (18%) obtained higher ratings, three obtained equal ratings, and one obtained a lower rating. For classroom behaviors, the ma jority of students received equal or higher ratings on participation in class, use of English in cl ass, and motivation to read. For each of these behaviors, only one or two students obtained lower rates from their teachers. Data can be found in Tables 4-12 and 4-13. Table 4-12. Preand post-intervention mean rate scores for reading skills on the RARS Phonemic Awareness Phonics Fluency Comprehension Vocabulary Name Pre Post Pre Post Pr e Post Pre Post Pre Post Amelia 2.3 4 1.8 3.3 1.8 3 1.6 3 4 4 Pedro 3 3 3.2 3 2.2 2.2 2.8 3.2 3.66 4 Jessica 2 4 2.5 3.5 2 3.2 1.6 2.6 2 3 Viviana 1 2 1 1.8 1 1.2 1 1.4 1 1 Ernesto 1 1 1.5 1.6 1 1.2 1 1.6 1.6 1.6 Maggie 2.6 3 2.5 3.331.3 2.831.2 2.6 3.3 3.66 Maria 3 4 2.5 3.662.2 2.5 1.6 2 2.6 2.66 Jennifer 1 2 1 1.6 1 1 1 1.2 1 1.3 Aldo 1 3.33 1 2.831 1.5 1 1.4 1 2.66 Jorge 2.3 3 1.3 2.661 1.831 2.2 3.66 2.66 Scores represent mean rates that range from 1 (very weak) to 4 (very strong). Book Levels Read during Intervention As a result of the tutori ng programs, all groups showed a marked improvement in book levels Group 1 started at a higher level than the other groups. At the beginning of

PAGE 151

151 Table 4-13. Preand post-intervention ra te scores for classroom behaviors on the RARS Participation in Class Use of English in Class Motivation to Read Pre Post Pre Post Pre Post Amelia 2 2 4 4 3 3 Pedro 4 4 4 4 3 3 Jessica 3 4 3 4 3 4 Viviana 2 2 4 3 2 3 Ernesto 2 3 2 3 4 4 Maggie 3 4 4 4 4 4 Maria 3 2 3 4 3 2 Jennifer 3 3 3 3 3 2 Aldo 1 3 2 3 1 4 Jorge 3 3 4 4 1 2 Scores for classroom behaviors repres ent the raw score for each of the three items included in this area. Scores go from 1(very weak) to 4 (very strong). the intervention (early November), students were reading books at level 9, corresponding to the middle of first grade. After 38 tutoring sessions (mid-January), all members of the group were reading books at level 20 with 98 to 100% accuracy. According to Reading Recovery book levels, students in second grade should reach level 20 by the end of the school year. This means that students in Group 1 were able to catch up to and surpass the expected reading levels for the specific time of year. Group 2 also showed notable im provement in book reading. At the beginning of the intervention (beginning of November), the group start ed reading books at level 2, which corresponded to the beginning of first grade. After 47 sessions the group reached level 16, corresponding to the end of firs t grade. While these students were still behind their peers, it is important to note that in approximately three and a half months, they had made the equivalent of one years progress according to Reading Recovery book levels. For Group 3, the improvement was less drastic, but still relevant. All members of this group started the intervention phase reading level 1 books, corresponding to the

PAGE 152

152 beginning of first grade. Level 1 books contain the most basic level of text, and the text is heavily supported by the pictures. Children ar e expected to read level 1 books in late kindergarten or at the beginning of first grade. At the end of 45 sessions (three and a half months), the group progre ssed to level 7. These students remained in each book level far longer than those in the other groups showing marked difficulties acquiring the decoding strategies being taught. Yet, with systematic and explicit instruction, they were able to make more improvement in three and a half months of intervention than they had made in all of their schooling before t he intervention began. It is important to consider that all three mem bers were repeating second gr ade, so their progress during intervention can be considered their first ac tual reading growth. A graph depicting the progression of book levels by ea ch group can be found in Figure 4-3. Figure 4-3. Book levels read by each group during intervention.

PAGE 153

153 Summary of Supplemental Data Supplemental data showed that st udents made marked improvements on all measures of early literacy skills, on book le v els read, and on attitudes toward reading. Some of the largest improvements were on decoding accuracy (NWD from KTEA-II) and decoding fluency (PDE from TOWRE and NWF from DIBELS). Word recognition accuracy (LWR from KTEA-II) and fluency (SWE from TOWRE) were also impacted but to a lesser degree. The effect of the inte rvention was also observed on students oral reading fluency (ORF DIBELS) where students were able to read more words per minute in connected text. Furthermore, at th e end of the intervention, QRI-4 scores demonstrated that students were able to comprehend texts, by recalling more ideas from the stories read and by answering more explicit and implicit questions. Improvements in students reading skills we re also observed by teachers who rated students higher on phonemic awareness, phoni cs, fluency, comprehension, and vocabulary. Results also showed that students in ev ery group made considerable advances on the levels of books they read during intervention making at least one halfs year progress. In addition, a much more positive outlook on reading was observed as students exhibited dramatic improvements in their attitudes toward academic and recreational reading. Social Validity Data collec ted through the students interviews and the teachers questionnaire showed that the program was regarded as im portant, effective, and feasible. What follows is a summary of students and teac hers responses. A visual depiction of teachers answers can be found in Figures 4-4 and 4-5.

PAGE 154

154 Student Interview At the end of the maintenanc e phase, students participat ed in a social validity interview that consisted of seven questions. The interviews took approximately five minutes and were conducted individually. Results showed that all ten participants had positive feelings about the tutoring lessons. They all thought the lessons helped them become better readers and that other students should participate in the program. Nine students affirmed that they will us e w hat they learned when they read. In relation to students favorite part of the lesson, six out of ten said they preferred doing word work with magnetic letters, while four said they liked reading books. When asked about their least favorite part of t he lesson, only four st udents stated disliking a part of the lesson. They all r eported that writing was their least favorite. Finally, nine out of ten students said that they enjoyed working in small groups. Teacher Questionnaire At the end of the intervent ion, teachers completed a social validity questionnaire after observing a videotaped tutoring session The questionnaire was divided in six different sections. The first five sections focused on t he different components of the UFLI program (e.g., repeated reading of familiar books, progr ess monitoring, word work with manipulative letters, reading a new book wr iting for reading). In each of the five sections, teachers had to answer yes or no to four questions: (1) Do you feel that this part of the lesson will help EL Ls develop reading skills? (2) Do you believe that this component is easy to implement with ELLs? (3) Currently, do you use a similar strategy when you work with ELLs? (4) Would you be likely to use a strategy like this one with ELLs during your reading instruction? Besides answering yes or no, teachers had the option of explaining the reason behind their answers.

PAGE 155

155 The sixth section focused on the overall tutoring lesson. Teachers had to answer six questions: (1) Do you feel that the tutoring lesson can help ELLs develop reading skills? (2) Do you believe that this lesson is easy to implement? (3) Would you be willing to implement the lesson in your classr oom? (4) Would you be willing to implement components of the lesson in y our classroom? (5) Would you be interested in learning how to implement this tutoring program? (6) Do you think other teachers of ELLs might be interested in learning to implement this program? Teachers also had the option of explaining the reas on for their answers. Results showed that all teachers believe d that each of the components and the UFLI lesson as a whole can help ELLs dev elop their reading skills. They also agreed that the UFLI lesson and all its component s would be easy to implement in the classroom, except for one teac her who stated that progre ss monitoring and writing for reading would be difficult to execute with ELLs. According to her, it gets difficult with a full class. When asked if they were using similar stra tegies with ELLs, they all reported using repeated reading in the classroom. Four out of six teachers said they were conducting progress monitoring, and using a strategy simila r to reading for writing, while five out of six reported doing word work and reading ne w books. All teachers stated that they would be likely to incorpor ate the UFLI lesson or any of its components in their classrooms, but only four out of six were interested in learning how to implement it. Also, five out of six agreed that other teacher s of ELLs might be interested in learning how to implement the UFLI program. Teachers answers for each question related to

PAGE 156

156 UFLI tutoring components are depicted in Figure 4-4, while answers for questions related to the UFLI tutoring program as a while are represented in Figure 4-5. Figure 4-4. Teachers responses to the social validity questi onnaire about the UFLI components Figure 4-5. Teachers responses to the social validity questi onnaire about the UFLI program Summary of One-on-One Tutoring Data In this secti on, a summary of findings is provided for the student receiving one-onone tutoring. First, changes in the tw o dependent variables (CPPM and CSPM) are

PAGE 157

157 reported for each of the phases. Then, supplemental data is reported, including preand post-intervention measures of early readi ng skills, attitudes toward reading and teachers rating scales, as well as changes in leveled book reading. Dependent Variables The rates of CPPM and CSPM were calculat ed for each probe and plotted into a graph for visual analysis (see Figure 4-6). Du ring baseline, Miriam received a total of five pseudoword and sight word probes. The rate of pseudowords ranged from 3 to 5 CPPM with a mean score of 4.66 CPPM, while the rate of sight word reading ranged from 13 to 16 CSPM, with a mean rate of 14.66 CSPM. Once a stable trend line was established during baseline, Miriam started the intervention phase. During this phase, she received a total of 26 pseudoword and sight word probes, one for each tutoring session. The rate of pseudoword reading ranged from 10 to 28 CPPM, with a mean score of 17.65 CPPM, which shows an in crease of 12.99 CPPM from baseline to intervention. Furthermore, the rate of si ght word reading ranged from 16 to 46 CSPM, with a mean score of 30.1 CSPM. This show s a positive change of 15.44 CSPM from baseline to intervention. Intervention ended after 26 tutoring lessons when she was able to read books at level 20 with more than 95% accuracy. Two weeks after the intervention ceased, maintenance data were co llected using the same set of probes. At that time, Miriams rate of pseudoword reading ranged from 24 to 27 CPPM, with a mean score of 25.4 CPPM, which shows an in crease of 7.75 from t he previous phase. Furthermore, the rate of sight word reading ranged from 43 to 46 CSPM, with a mean score of 44.2 CSPM. This shows a positiv e change of 14.1 CSPM from intervention to maintenance.

PAGE 158

158 Figure 4-6. Rate of correct pseudoword and sight word read per minute for Miriam. Supplemental Data Before intervention, Miriams raw score on Sight Word E fficiency (SWE) was 28, which placed her at the 32nd percentile, while her score on Phonemic Decoding Efficiency (PDF) was 6, which placed her on the 19th percentile. Miriams percentile score on the Total Word Readi ng Efficiency was 21%. By the end of the intervention, she obtained a score of 54 on the SWE (67th percentile) and 20 on the PDF (84th percentile). Her Total Word Readi ng Efficiency placed her at the 64th percentile. The KTEA-II scores improved after intervention as well. On Letter Word Recognition (LWR) she went from the 36 correct words read (32nd percentile) to 43 words (34th percentile). On Nonsense Word Decoding (NWD), she went from 6 nonsense words (37th percentile) to 20 nonsense words read correctly (68th percentile). On DIBELS, Miriams NWF scores went from 51 (low risk) to 94 (above average), showing an improvement of

PAGE 159

159 43 correct letter-sound correspondences per minute. ORF scores showed an increase of 48 words, going from 18 (high risk) to 66 words per minute (high risk). Given her high risk status, easier passages were administer ed at both times. At the beginning of the study, she read 18 words on t he end of first grade passage (h igh risk), and 20 words on the middle of first grade passage (low risk) At the end of the study, Miriam read 68 words on the middle of second grade passage (low risk). Comprehension scores on the QRI-4 befor e intervention were based on a primer level passage. Miriam was able to recall two ideas, and answer one explicit question correctly. This showed that she was reading prim er level passages at a frustration level. After intervention, she was given a level two passage. Miriam was able to recall 14 ideas from the story and answe red four explicit and three im plicit questions. A score of seven points indicated that she was reading level two passages at an instructional level. Attitudes toward reading showed a mark ed improvement after intervention. Initially, Miriam obtained 20 points (4th percentile) on recreational reading and 22 points (18th percentile) on academic reading. Overa ll, she scored a total of 42 points (7th percentile). After intervention, recreation reading score went up to 32 points (62th percentile) and academic went up to 35 points (81st percentile). Her total score was 67, which corresponded to the 74th percentile. The teachers ratings on the RARS increas ed after intervention for most areas of reading ability. The mean score for phonological awareness remained at 3 (strong). In contrast, phonics mean rates went from 2.66 (weak) to 3.3 (strong) fluency rates from 1.83 (very weak) to 2.3 (weak), comprehens ion from 2.4 (weak) to 3.2 (strong), and vocabulary from 3 (strong) to 4 (very strong). Classroom behaviors remained the same

PAGE 160

160 for class participation with a score of 3 (str ong) and use of English to communicate in class with a score of 4 (very strong). Moti vation to read increased from a score of 3 (strong) to 4 (very strong). Miriam also showed rapid progress in leveled books throughout the intervention. At the beginning of the study, she was reading books at level 6, which correspond to early first grade. On session 25th, she moved up to level 20, which corresponds to the end of second grade. On session 26th, she read a level 20 book with more than 95% accuracy. Figure 4-7 depicts Miriams progr ess compared to the changes in groups book reading. Figure 4-7. Book levels read by Miriam during intervention. This figure compares the changes in book levels during one-onone tutoring compared to the progress made by students in small-group tutoring and illustrates the faster rate of growth for the student rece iving one-on-one tutoring. Summary of Findings The findings described in this chapter rev eal that the UFLI tu toring program was an effective small-group reading intervention that produced drasti c changes in students

PAGE 161

161 reading abilities. Improved rates of CPPM and CSPM were reached during the intervention phase and sustained two wee ks after it ended. Changes in pre-and postintervention measures, as well as changes in book levels read throughout the lessons provide additional evidence of the programs effectiveness. Fu rthermore, social validity data demonstrated that the UFLI tutoring progr am was considered effective, efficient, and important by students and t eachers. Chapter 5 provides interpretation of these findings, including their implicatio ns for practice and research.

PAGE 162

162 CHAPTER 5 DISCUSSION The purpos e of this study was to evaluate the effectiveness of a modified version of the University of Florida Literacy Initiati ve (UFLI) small-group tutoring program on the reading skills of second-grade Spanish-speaking, English language l earners, who were struggling to read in English. The aim of th is chapter is to summarize and interpret the results obtained in this study. The chapter is organized in several sections: (a) overview of the study, (b) summary of findings, (c) interpretation of findings, (d) interpretation of supplemental data, (e) social validity, (f) e ffectiveness of the UFLI tutoring program, (g) implications for future res earch, (h) implications for practice, and (i) conclusions. Overview of the Study A single-subject, multiple baseline across groups des ign was implemented to examine the effects of the UFLI tutoring program on the reading s kills of 10 Spanishspeaking, English language le arners who were struggling to read in second grade. Participants ranged in age from 7 years 8 mont hs to 10 years 1 month at the beginning of the study. School personnel identified st ruggling second-grade ELLs, whose native language was Spanish, and who were performing below level on their latest end-of-year SAT-10 and DIBELS scores. Ba sed on these two scores, as well as beginning-of-year Invented Spelling Assessment scores, student s were selected to participate in the small-group tutoring program designed to address word reading skills and modified to support their language needs. The level of English and Spanish oral language proficiency of the participants va ried from negligible to fluent. Students were grouped based on the level of book they were able to read with 90% to 95% accuracy, as determined by runni ng records and using Reading Recovery

PAGE 163

163 book levels. A total of three groups were formed. Group 1 was composed of three students (1 male and 2 female) reading level 9 books. Group 2 was composed of four students (1 male and 3 female) reading leve l 2 books. And Group 3 was composed of three students (2 male and 1 fema le) reading level 1 books. The focus of this study was an examination of the effects of the UFLI program on two reading skills: decoding and sight word reading. Decoding was measured by assessing the rate of correct pseudowords read per minute, while sight word reading was measured by assessing the rate of corr ect sight words read per minute. For this purpose, data were collected during three phases : (a) baseline, (b) intervention, and (c) maintenance. During baseline, students were given pseudoword and sight word probes with the goal of establishing the rate of co rrect words each was able to read per minute. When Group 1 demonstrated a stable line over six data points, the group moved to the intervention phase. Meanwhile, the other groups remained in baseline. When Group 1 showed an increase in four consecutive data points, Group 2 moved from baseline to intervention. The cycle was repeated for Group 3. During intervention, 10 student s participated in smallgroup tutoring sessions for 45 minutes, 2 to 5 times a week. The tota l number of sessions varied across groups based on students attendance and book level reached by the group. The maintenance phase of the study was designed to examine w hether the effects of the UFLI program were sustained two weeks after the intervention concluded. One additional student participated in one-on-one tutoring sessions using the same instructional methods. In addition to rate of pseudowords and sight words read per minute, supplemental data were collected, including preand postmeasures of reading skills (e.g., decoding,

PAGE 164

164 word recognition, fluency, comprehension, attitudes toward reading) as well as teachers ratings of students reading ab ilities and classroom behaviors. Also, changes in book reading throughout the intervention we re recorded for each group. Finally, social validity information was collected thr ough student and teacher social validity questionnaires. Summary of Findings An examination of the data gathered showed several key findings. First, the UFLI small-group tutoring program was effectiv e in producing substantial improvement on decoding and word recognition skills of student s, as measured by the rate of correct pseudowords read per minute and the rate of correct sight words read per minute. Furthermore, students sustained these improved rates two weeks after the intervention ceased, demonstrating the effectiveness of the UFLI program in producing a lasting effect on students reading skills. Second, in addition to the two main variables, all groups showed a marked improvement when reading leveled books. Group 1 made the equivalent of one and a half years progress in 38 sessions, going from level 9 to level 20 books. Group 2 made a years progress in 47 sessions, going from le vel 2 to level 16 books. Finally, Group 3 showed approximately one-half y ears progress in 45 sessions, moving from level 1 to level 7 books. In other words, all groups made more rapid progress during the intervention than would be expected duri ng typical classroom instruction. Third, supplemental data showed t hat participants improved on pre-and postintervention measures of decoding, wo rd recognition, or al reading fluency, comprehension, and attitude toward r eading. In particular, students showed improvement in decoding and word recognition accuracy and fluency. Changes in

PAGE 165

165 students abilities were also observed by their teachers, who ra ted them higher on phonemic awareness, phonics, fluency, vocabul ary, comprehension, and classroom behaviors. Fourth, all participants benefitt ed from the intervention, r egardless of their initial level of English oral language proficiency. It is important to note that not all students in the higher reading performing group had the highes t level of oral proficiency. On the contrary, there were some students in lowe r groups who had equal or higher levels of oral proficiency. Thus, the rate of impr ovement differed across groups with different reading levels and not different proficiency levels. Fifth, the student who received one-on-one intervention using the same instructional methods also demonstrated considerable reading growth on the two dependent variables. In fact, her progress was more rapid than that of the students receiving small-group instruction. Supp lemental data also showed a marked improvement in reading ability after intervention. Finally, social validity data showed that the UFLI program was well received by students and teachers who noticed its effectiveness in improving students reading skills. The program was not only deemed effect ive, but also feasible. Students believed that other ELLs might benefit from the program, and teac hers reported their willingness to implement it in their classrooms. Interpretation of Findings Findings from this study provide evidenc e to support the use of early reading interventions in Englis h for ELLs who are st ruggling to learn to read in their second language. Specifically, the UF LI tutoring program was effe ctive in promoting word reading skills among second-grade, Spanish -speaking ELLs who were receiving

PAGE 166

166 reading instruction and who had been performing below grade level in measures of reading achievement. As a result of daily, small-group tutoring sessions focusing on word reading skills, partici pating students showed a mark ed improvement in their decoding skills, as measured by the rate of correct pseudowords read per minute (CPPM), and word recognition, as measur ed by correct sight words read per minute (CSPM). These findings are consistent with previous studies reporting the effectiveness of word-level reading interventions on t he reading skills of Spanish-speaking ELLs, particularly on word attack (Gunn et al., 2000, 2002, 2005; Linan-Thompson, Vaughn et al., 2003; Vaughn, Cirino et al., 2006; Vaughn, Linan-Thompson, & Hickman, 2003; Vaughn, Mathes et al., 2006) and word identification skills (Al Otaiba, 2005; Gunn et al., 2000, 2002, 2005; Quiroga et al., 2002; Vaughn, Cirino, et al., 2006). Decoding and automatic word recognition are critical skills in the early stages of reading development. Beginning readers need to develop the ability to analyze and manipulate the sounds in speech, as well as to recognize the letter-sound correspondences so that they can effectiv ely decode unfamiliar words in print (Ehri, 2004). Furthermore, students must develop the ability to read words accurately and automatically to achieve fluent reading (Stahl 2004). In order to promote these reading skills, researchers have underscored the impor tance of providing instruction in phonemic awareness and phonics not only for native speakers of English (Adams, 1990; Ehri, 2004), but also for English langu age learners (Ehri & Roberts, 2006; Helman & Burns, 2008). The present study provides further evidence to support the importance of such instruction.

PAGE 167

167 Progression through Phases of Word Recognition Development Based on recent findings suggesting that native Englis h speakers and ELLs develop word recognition skills in similar ways (Chiappe & Siegel, 2006; Chiappe, Siegel, & Gottardo, 2002; Chiappe, Siegel & Wade-Wolley, 2002; Geva et al., 2000; Jackson & Lu, 1992; LeSaux, Koda, Siegel, & Shanahan, 2006; Limbos & Geva, 2001; Verhoeven, 1990, 2000; Wade-Woolley & Siegel 1997), students response to the UFLI intervention can be described following Ehr is phases of word reading development. According to Ehri and McCormick (1998), reader s progress through five phases of word reading development: (a) pre-alphabetic, (b) partial-alphabetic, (c) full-alphabetic, (d) consolidated-alphabetic, and (e) automatic (Ehri & McCormick, 1998). Each phase is characterized by the learners underst anding and use of the alphabetic system (p. 140). Before the UFLI intervention, some of the second-grade par ticipants displayed reading behaviors typical of the partialalphabetic phase, while others displayed behaviors of an incipient full-alphabetic phase. According to Ehri (2005b), the partial alphabetic phase is commonly seen among kinder garten and first grade students, while the full-alphabetic phase is co mmonly seen among first-graders. Nevertheless, it is not unusual to find older struggli ng readers at any of these phases. Students who have not developed strong skills in the full-alphabetic phase, particula rly struggling readers, need systematic phonics and phonemic awareness instruction to help them develop the ability to read words accurately and automatic ally (Ehri & McCormick, 1998). Consistent with this premise, after participating in t he UFLI tutoring program, all ten participants exhibited a more advanced word reading abi lity, signaling a movement towards more proficient reading.

PAGE 168

168 For example, evidence from the reading behaviors exhibited by students in Group 1 indicated that they had begun transitioning into the fullalphabetic phase prior to the intervention. They had started developing a full working knowledge of the alphabetic system as well as more advanced segmentat ion ability, but they were still lagging substantially behind their average performi ng peers. Their reading was slow and arduous, but they had started to represent full sight words in memory, to use decoding and analogy to read unfamiliar words, and to s pell words more accurately. According to Ehri and McCormick (1998), to help students move from slow deliberate decoding to faster decoding it is impor tant to provide ample opportuni ties to practice (p. 152). Specifically, instruction should give opport unities to practice analyzing within-word grapho-phonemic associations, dividing words into smaller units (e.g., syllable, onsetrime, phonogram), working with word fa milies, applying decoding and analogy strategies when reading connected text, and using prediction to confirm reading accuracy (Ehri & McCormick). Given that the UFLI tutoring program is adapted to the indivi dual needs of the students, instruction was geared towards providing this type of instruction and practice. As a result, students in Group 1 were able to move into the consolidated-alphabetic phase by the end of the intervention, demonstrated by the chunking strategies they were using during lessons. By the end of t he intervention, student s had not only showed a marked improvement in the rate of CPPM and CSPM, but also the level of reading accuracy during level book reading. At that time, students were reading books at level 20, which corresponds to the end of second grade. This showed that they had caught up to and surpassed their second-grade peers based on the book levels expected for

PAGE 169

169 that grade. By this time, students were able to read more words accurately and automatically, using a more complete k nowledge of the alphabet ic system, recognizing and using consolidated units (i.e., rimes, syllables, root words, morphemes) to read multisyllabic words. They continued using decoding and analogy to read unfamiliar words, while their sight vocabulary continued to grow (Ehri & McCormick, 1998). Before the intervention, students in Group 2 performed considerably below students in Group 1 on measures of decoding and word recognition yet their reading behaviors indicated that each had started the transition into the full-alphabetic phase. One reading behavior characteristic of this trans ition relates to the type of reading errors made when students decode unfamiliar words. While younger students tend to produce errors that do not resemble target word s, students advancing in the development of word reading ability tend to produce errors that resemble the target words, signaling the use of a decoding strategy (Ehri, 2005a). In order to complete the transition in to the fullalphabetic phase, students need instruction in decoding and sight word learning. Specifically, instruction s hould guide students in processing all the letters in words, instead of just initial or final letters (Ehri, 2005a; Ehri & McCormick, 1998). With an intervention tailored to their needs, students in Group 2 showed a marked improvement in their rate of CPPM and CSPM as well as the level of accuracy during book reading activities. At t he end of the intervention, students were reading books at level 16, which corresponds to the end of first grade. This shows that the group made the equivalent of one years progress after 47 sessions ( approximately 3 months). By this time, it seemed that students in Gr oup 2 had completely moved into the fullalphabetic phase and had begun dem onstrating reading behaviors indicative of a

PAGE 170

170 transition into the conso lidated-alphabetic phase. They had a better knowledge of grapho-phonemic correspondences and were consis tently using decoding skills to read and spell unfamiliar words (Ehri, 2005a). Thei r growing sight vocabulary was apparent during reading and writing activities as we ll. For example, during reading, students began to cover word parts and read other parts as chunks. Nevertheless, it is important to note that their reading was still slow and laborious. Thus, continued instruction in decoding and sight word readi ng (Ehri, 2005a) is recommended for them to complete the transition into the cons olidated-alphabetic phase. Prior to the intervention, students in Group 3, the lower performing group, exhibited reading behaviors characteristic of the partial-alphabetic phase. Given their incipient knowledge of letters and sounds an d their developing phonemic segmentation ability, they were unable to use basic reading strategies, like decoding, analogy, or sight word reading. This is exemplified by their initial low rates of CPPM and CSPM, as well as their low level of reading accuracy during book reading activities. Ehri and McCormick (1998) identified common mistakes exhibited by students in the partialalphabetic phase: (a) confusing words that had similar spelli ngs (e.g., think and thank), (b) misreading words by using letter names inst ead of sounds (e.g., jo instead of go), (c) guessing words based on partial letter-sound relations and context cues (e.g., house instead of home), (d) reading words backwards due directionality problems (e.g., was instead of saw), (e) misreading words with graphem es (i.e., /c/ /h/ /e / /s/ /t/ instead of /ch/ /e/ /s/ /t/), and (f) writ ing words using only salient sounds (e.g., mret instead of market). Students in Group 3 exhibited all th ese behaviors before the intervention. To

PAGE 171

171 develop their word reading skills, these students needed to transition into the fullalphabetic phase. Juel, Griffith, and Gough (1986) identified f our elements needed for this transition: (a) phonemic awareness (segmenting, blending, s ubstitution), (b) exposu re to print (text level being read), (c) cipher knowledge ( nonword decoding), and (d) sight word knowledge (recognition of misspellings). Given that the UFLI tutoring program addresses these four components, students in Group 3 started making the transition to the full alphabetic phase as the intervention progressed. At first, Jennifer and Jorge were advancing at a faster rate than Aldo. Jennifer and Jorge were acquiring a more complete knowledge of the alphabetic system and were using this knowledge to decode unfamiliar words and to acquire sight words. Initially, their reading was extremely slow and laborious, but with practice it started to become more fluent. Their rate of CPPM and CSPM started increasing at a similar pace and their level of text reading accuracy started to improve. In contrast, Aldo continued having difficulty making letter-sound connections and identifying all the sounds in words. Many of hi s attempts to read unfamiliar words and high frequency words tended to be unsuccessful. His rate of CPPM and CSPM remained close to zero. Torgesen (2000) stated that many struggling readers or treatment resisters need more intensive intervention. Torgesen further states that many struggling students who are facing severe risk factors may require more than 1 or 2 years of intervention. In a study conducted by Vaughn, Linan-Thomps on and Hickman (2003), researchers found that when more intervention was provided a nd instruction was modified to meet the specific needs of each individual, students responded positively. In the present study,

PAGE 172

172 intervention was modified for Aldo. Star ting on session 20, Aldo received 15 extra minutes of instruction, in the form of one-on-one tutoring th at took place before his small-group session, focu sing on the letter-sound correspondences to be practiced during the small group session. Immediately, he showed a small improvement on CPPM and CSPM. Instead of reading a 0 to 1 C PPM, he started reading between 0 and 2 CPPM. Similarly, he went from reading 0 to 3 CSPM to reading between 2 and 4 CSPM. After 19 sessions of additional tutoring, Aldo made a marked improvement. His rate of pseudoword reading went from a range of 0 to 2 CPPM to a rate of 4 to 6 CPPM. His rate of sight word reading w ent from a range of 2 to 4 CSPM to a range of 4 to 8 CSPM. This change was also visible during book r eading and writing activities. He started segmenting and blending sounds more accura tely, and he started showing independent use of self-monitoring and self-correction. Th ese changes were also noticed outside the tutoring sessions. Aldos teachers reported that not only was he using reading strategies in the classroom, but for the fi rst time, he was also enthusiast ic about reading in class. By the end of the intervention, Aldo started showing reading behaviors ty pical of the fullalphabetic phase, just like J ennifer and Jorge, which incr eased his self-confidence. Despite their substantial improvements, it is important to recognize that students in Group 3 were still performing significantly below their second-grade peers. All three demonstrated urgent need for continued reading intervention, given that they were repeating second grade. Also, given Aldos overall performance, admi nistration of other measures could help uncover additional characteristics or the presence of a disability, which might be influencing his progress. Some suggested areas commonly associated

PAGE 173

173 to learning disabilities include phonological pr ocessing, working memory, word retrieval, and naming speed (Altmann, Lom bardino, & Puranik, 2008). Prior research makes it clear that students who continue struggling require systematic, explicit instruction in phonem ic awareness and phonetic decoding (Ehri & McCormick, 1998; Torgesen, 2000). Further more, teachers should provide daily opportunities to read connected text using easy materials that can be read with at least 96% accuracy (Allington, 1983, p. 555). Unfortunately, many struggling readers do not have these opportunities in their regular classrooms, as their peers move onto more advanced skills. Without continued intervention that is tailored to the individual needs of these students, it is unlikel y that struggling readers wil l catch up to their higher performing peers. In contrast to the groups response to intervention, Miriam moved to a more advanced phase of word recognition development in considerably less time. At the start of the intervention, she showed reading behav iors typical of an incipient full-alphabetic phase, where she started using decoding strategies to read unfamiliar words, as well as recognizing high frequency words by sight. Still, her reading was slow and laborious. This was demonstrated not only by low rates of CPPM and CSPM, but also by connected text reading during running record s. In response to the intense one-on-one intervention, Miriam rapidly moved into the consolidated-alphabetic phase, where she was able to decode larger word units and us e those to decode mult isyllabic words with accuracy and speed. Her full knowledge of letter-sound correspondences translated into an enhanced sight word reading ability and a more accurate and fluent reading of connected text.

PAGE 174

174 Overall, all students made subs tantial progress as a resu lt of the intervention. Students in Group 1 are likely to be able to succeed in reading without additional intervention. Students in Group 2 improved considerably but still will likely require additional support to continue making progress. Students in Gr oup 3, despite significant gains, remained well below expected performanc e for their grade level and will require continued intensive intervention. Similarly to Group 1, Miriam is likely to continue thriving in the classroom without additional intervention. Relation between CPPM and CSPM Two important findings concerning the relation between CPPM and CSPM merit further discussion. First, there was a marked difference between the CPPM rate and the CSPM rate for Groups 1 and 2. Throughout t he study, both groups consistently had a higher rate of CSPM than CPPM. This finding is consistent with t hat of Ehri and Wilce (1983), who found that less skilled readers were less able to decode nonsense words than real words. Ehri and Wilce stated that these students are inaccurate and excessively slow in decoding nonsense words, and that while practice improved their pseudoword reading speed, it remained slower than real words (p. 15). One possible explanation is that students who lack decoding skills are relying on visual features of words (Gough & Juel, 1991), or they are still relying on visual cue reading strategies characteristic of the pre-al phabetic phase (Ehri, 2002) to compensate for their lack of decoding skills. Second, it was noticeable in a ll groups that the rate of CSPM increased as the rate of CPPM improved, even t hough the emphasis of instruction was on decoding. According to Ehri (2005a), this re lation is possible because decoding helps build students sight vocabulary. When readers develop a working knowledge of the alphabetic system, they use th is knowledge to acquire and retain words in memory

PAGE 175

175 (Ehri, 2005b). This was certainly the case for Group 3, who initially decoded high frequency words and later started recognizing words by sight. Interpretation of Supplemental Data In addition to data collected during each phase of the si ngle-subject design, supplemental data were collected to provi de additional information about the effects of the intervention. Data were collected on student s early reading skills, attitudes toward reading, and teachers ratings of students reading ability. What follows is detailed analysis of these data. Early Literacy Skills Preand post-intervention data provide addi tional support for t he effectiveness of the UFLI program on students decoding an d wo rd recognition skills. These two skills were assessed under untimed conditions (KT EA-II) to evaluate reading accuracy and under timed conditions (TOWRE and DIBELS-NWF) to evaluat e reading fluency. The largest improvement in all participants wa s on decoding accuracy and fluency. This is most likely the result of t he programs prime focus on decodi ng instruction. Results also showed that students ability to recognize words by sight in creased after intervention. Instruction also targeted sight word reading, but to a lesser degree. It is possible that the newly acquired decoding ski lls impacted students ability to read words by sight. According to Ehri (2005a), decoding is a sel f-teaching mechanism that supports the acquisition of sight words (p. 149). W hen students decode unfamilia r words, those words are retained in memory, adding to their growing sight vocabulary. Another important finding from this study relates to reading fluency. Preand postintervention data on the ORF (DIBELS) showed that all the participants made drastic improvements on their rate of words read per minute. Aldo (Group 3) for example, who

PAGE 176

176 was reading three words per minute at t he beginning of the school year and who had difficult decoding basic VC words, was able to read 12 words per minute after intervention. Although this was still very weak reading, this change represents a substantial improvement considering that he was repeating second grade and the previous year he had read 4 words per minute on the end of year ORF passage. Dramatic improvements were also observed in other students. Ernesto (Group 2), for example, went from 25 words per minute to 79 words per minute, while Amelia (Group 1) went from 55 to 102 words per minute. Reading fluency is a complex construct that encompasses multiple processes and skills, of which decoding fluency is a signifi cant one (Hudson, Pullen, Lane, & Torgesen, 2009). According to Hudson and colleagues, an i nefficient decoding process interferes with the readers ability to read text fluently, given that a c onsiderable amount of energy is expended trying to decipher unfamiliar word s. This was certainly the case for the participants in this study. Their initial lack of decoding fluency interfered with their ability to read connected text with accuracy and autom aticity. However, once students started acquiring more efficient decoding processe s their ability to read connected text improved. This was not only reflected in t heir post-intervention fluency scores, but also throughout the lessons as they increased thei r level of reading accuracy of leveled books. The benefit of enhanced word reading skills does not stop with improved reading fluency. It is well known that a close relation exists between fluency and comprehension. When a reader fails to dec ode and recognize words automatically, reading fluency is compromised (Hudson et al., 2009). When reading fluency is

PAGE 177

177 compromised, comprehension suffers. Acco rding to LaBerge and Samuels (1979), this is because students are not able to focus their energy on higher order skills, so they are unable to derive meaning from text. Students comprehension scores in this study are a perfect example of this connection. Before t he intervention, half of the students did not take the QRI-4 because they were not abl e to read second-grade or even first-grade passages fluently. One student was given a primer level passage and due to his extreme difficulty with decoding and word recognition, he was not able to understand what he was reading. His comprehension sco res showed that he was reading at a frustration level. Furthermore, three of t he four students whose ORF scores showed low risk for second grade passages, and who were given QRI-4 level two passages, also had difficulties reading words accurately and fluently. His comprehension scores also showed that they were reading at a frustrat ion level according to QRI-4 norms. After intervention, when word level skills improved and fluency rates increased, a development in comprehension ability was readily observed. For example, this time, seven out of ten students were able to r ead level two passages. Of these students, three were reading at an instructional leve l and one at an independent level. Moreover, four of the five students who we re not given the QRI-4 before intervention, were able to take it after. Two of t hese students read level two pass ages and two read primer level passages. This shows that the transition toward a more effective and efficient reading ability had initiated. Students ability to derive meaning from te xt was also influenced by the programs high emphasis on reading comprehension deve lopment. Each tutori ng lesson included activities designed to help students acqui re effective comprehension skills. For

PAGE 178

178 example, before the students read a new book, the tutor introduced the book and engaged the group in a discussion of the pict ures, with the goal of activating prior knowledge and creating context for the story. Then, as students read the books, they were encouraged to make connections and predictions related to the story. The tutor also modeled and guided students to self-monito r for comprehension. After reading, the tutor and students engaged in a discussion of the story through t he use of literal, inferential, and evaluative questions. Children were also asked to summarize part of the story and then write a summary sentence (H ayes et al., 2005; Lane et al., 2005). Finally, the English vocabulary support pr ovided by the tutor throughout each lesson contributed to students under standing of what they re ad (Carlo et al, 2004). Attitudes toward Reading Another change obser ved after intervention was an increased positive attitude toward reading. Results from the ERAS showed that at the beginning of the study students did not have strong positive attitudes toward reading, particularly recreational reading. This meant that st udents preferred reading activi ties associated with school more than reading for fun outside of school. These findings are most likely associated with students history of r eading difficulties. Students wh o have not acquired basic reading skills and who have experienced repeat ed failure during reading activities are less likely to become motivated and self-r egulated readers who are willing to read outside of school (Quirk & Schwanenflugel, 2005). It is also important to note that despit e showing more positive attitudes toward academic than recreational readi ng at the beginning of t he study, their scores placed them considerably below second-grade norms. Quirk and Schwanenflugel (2005) stated that, typically, students who are in rem edial reading interventions have already

PAGE 179

179 experienced at least a year of repeated failu re associated with reading. Thus, students low academic reading attitude scores might have been a reflection of this occurrence. For example, six of the ten students who were repeating second grade when the study started were still having trouble reading grade-level books. Unless these students acquired the necessary reading skills to be su ccessful readers, they would continue to experience negative attitudes toward reading. Fortunately, not only did reading skills improved after intervention, but their attitudes toward academic and recreational r eading improved, as well. It is likely that with successful reading came a heightened sense of self-efficacy, which in turn translated into a positive disposition for reading (Quirk & Schwanenflugel, 2005). This connection between reading abilit y, self-efficacy, and attitudes toward reading was reflected not only on students scores, but al so on students performance in the tutoring program and in the classroom. For example, the tutor noted that, initially, there was a tendency among students to complain and avoi d work that was too challenging for them. As their skills started to improv e, students began to participate more and complain less. For example, at the beginn ing, Aldo complained when asked to read a page that had more than one sentence. Toward the end of the intervention, he would ask to read the whole book by himself. This trend was observed in the other groups, as well. Students would ask to read more than one page at a time or to read more books. Teachers also reported that students were vo lunteering to read in class and were eager to read more books. Students improved attitudes toward reading may contribute to the lessening of the Matthew Effect. Stanovich ( 1986) states that good readers continue to improve more

PAGE 180

180 rapidly than poor readers at least in part because they read more, thereby getting more practice in the skills necessary for improvement. It was evident from their preintervention reading achievement and attitude scores that the students in this study were not practicing reading nearly enough. With improved skills and accompanying improved attitudes toward reading, perhaps these students will continue to read more and get the practice they need to close the gap between their reading and that of their peers. Teacher Reading Ability Rating Scale (RARS) Chang es in reading ability were also not iced by students classroom teachers. Overall, teachers rated students reading abilitie s higher after intervention. This is very important because it indi cated that students newly acquired skills were being generalized to the classroom setting. Teacher s observations of students progress were also corroborated through their informal co mmunications with the researcher. A few times, teachers expressed their contentment with students increased reading ability and increased levels of participation in class. According to Perry and Meisels (1996), teachers are in a good position to evaluat e their students achievements because they interact with them on a regular basis. Still, a closer look at specific tasks withi n each of the five reading skills showed that, for some students, teachers failed to recognize a change despite students marked improvement on the measurements used in the study. Interestingly, some of these teachers expressed on different occasions that their students were still struggling or that they still could not read. This raises the ve ry important issue of how teachers measure progress. In these classrooms, for example, struggling r eaders were required to read grade-level texts. Students who were performing significantly below grade level before

PAGE 181

181 intervention (e.g., reading primer level passages) and who made considerable growth as a result of the intervention (e.g., readi ng first grade passages), may not be able to demonstrate this progress when teachers provide reading materials that are still above their reading ability. Furt hermore, it is likely that for some teachers it might be difficult to recognize students progress because, as Allington (1983) acknowledged, many struggling readers tend to work arduously thr ough any text, even the ones they can read accurately. Allington recommended that str uggling readers be given opportunities to read easy materials so that they get to read more like good readers (p. 555). Social Validity Single-subject research helps identify inte rventions that target socially relevant outcomes (Horner et al., 2005). D emonstrati ng the social validity or appropriateness (Wolf, 1978) of an intervention is import ant when working in educational settings (Kennedy, 2005). According to Horner and colle agues, the social validity of a singlesubject study is enhanced when teachers and other intervention agents report that procedures are acceptable, feasible, and effectiv e. In this study, the goal was to receive feedback from students and teachers. Students Responses The acceptability of the interventi on by direct consumers (Kennedy 2005), students in this case, is very important becaus e it can influence how they respond to the intervention. The fact that students had positive feelings towa rd the program might have contributed to their positiv e disposition throughout the lessons. Most students were eager to work and worked hard to complete all the tasks. Their positive feelings were also reflected in the excitement they showed when the tutor would pick them up in their classroom. Interestingly, Jorge, who refu sed to work on many occasions and created

PAGE 182

182 distractions for other students, reported enjoying tutoring. One possible explanation for his positive response is that tutoring prov ided an escape from the regular classroom where he frequently got in trouble. Neverthel ess, when the work from the tutoring started to get difficult for him, he would get frustrated a nd immediately engage in covert and overt misbehavior (e.g., making fun of ot hers, ripping books, throwing his pencil or magnetic letters, doing minimal work, not answ ering questions). For students like Jorge, who are in critical need of s upport it is important to devise appropriate ways to also help him take control of his emotions and find better ways to express his frustration. An important finding from the intervie ws is that students recognized that participating in the tutoring program helped them become better readers. Some students stated that they were reading mo re, reading better, and reading faster. Jennifer, in particular, said she actually l earned to read as a resu lt of the tutoring program. Aldo recognized that he was reading on his own. He said: Now I read books. You no help much. The fact that the Aldo was able to sound out words on his own and read faster than before was very motivating for him. Toward the end of the intervention, he repeatedly said that he was reading without t he tutors help. This finding is relevant given the tendency of struggli ng readers to have low confidence in their ability to read (Guthrie & Davis, 2003). In relation to the different components of a UFLI lesson, students had different preferences. Some students liked doing word work with magnetic letters and others preferred reading books. Yet, when asked about their least favorite part, the only activity that was identified by students was writing sentences (Writing for Reading). Some of the reasons given were that it was too hard, too much work, and too tiring. Writing for

PAGE 183

183 Reading involves a lot of work for thes e students: (a) encoding familiar and unfamiliar words by segmenting and blendi ng, (b) writing sight words by practicing correct spellings, (c) figuring out the appropriate me chanics of writing like capitalization and punctuation, and (d) rereadi ng the sentence created after each work is added. Unfortunately, many of these students were not accustomed to persisting until they could produce a correct sentence, and this probabl y is why it felt like too much work for them. Fortunately, despite being a challe nging program, all par ticipants agreed that other struggling readers should parti cipate in the tutoring lessons. Teachers Responses Given that the intervention was implemented by the rese archer, it was important to know what indirect consumers (Kennedy, 200 5), in this case the teachers, thought about the intervention. Teachers opinions about the acceptability feasibility, and effectiveness of the intervention (Horner et al., 2005) can influence th eir decisions about adopting the program and implementing it with ELLs. All teachers agreed that the UFLI program and each of its components can hel p ELLs become better readers, and that they would be likely to incorporate the UFLI lesson or any of its components in their classrooms. This is relevant given that teachers beliefs about the importance of an intervention influence their disposition to adopt and implement new practices (Sparks, 1988). One teacher recognized that one of her students who couldnt put two sounds together at the beginning of the school year started reading as a result of his participation in the program. Similar comme nts were expressed by other teachers throughout the study, who st arted noticing an improv ement not long after the intervention started.

PAGE 184

184 Despite all stating that t hey were willing to implem ent the program in their classrooms, two teachers reporte d not being interested in l earning how to implement it. One stated that she had a good understanding of it, while th e other said that she had done that type of program befor e. Unfortunately, just being familiar with it or having a good understanding is not enough to ensure appropriate implem entation. Participating in professional development is critical fo r teachers to acquire the necessary knowledge and skills to properly implement the interventi on, especially as it addresses the needs of ELLs. It is well documented in the literature that professional development is associated not only with improved instructional practices, but also with students learning (Borko, 2004). Finally, it is important to acknowledge that all teachers reported using similar strategies in their classrooms, particul arly repeated reading. Some teachers did progress monitoring, while others did word wo rk, or writing activities. Many of them stated that the SRA/EI R program that they use in their classrooms has some similar activities. Yet, they stated that word work ac tivities did not inclu de manipulative letters and writing activities did not go as in -depth as the UFLI program does. These distinctions could explain the differences in relative effectiveness of the two intervention programs. Overall, the UFLI program was we ll received among st udents and teachers. Moreover, other school personnel (e.g., school principal, r eading coach) reported their satisfaction with the program and their desire to use it in the future. They stated that other teachers had heard about students progress and were interested in learning

PAGE 185

185 about it. Professional development will be provi ded by the researcher to all interested teachers. Effectiveness of the UFLI Tutoring Program There are three plausible explanations or the effectiveness of the UFLI tutoring program. These include: (a) ear ly intervention and language prof ic iency, (b) provision of instructional practices for ELLs, and (c) consistency with the RTI approach. What follows is a brief description of each explanat ion and the theoretical foundation behind it. Early Intervention and Language Proficiency The role of early reading interventions in the prevention of later reading difficulties has been well documented in the literature (E hri, Dreyer, Flugman, & Gross, 2007; Gersten & Dimino, 2006; Tor gesen et al., 2001; Vaughn, Cir ino et al., 2006). Early interventions are parti cularly important giv en the tendency of struggl ing readers to fall further and further behind their higher performing peers. According to Gunn and colleagues (2000), students who have not learned to read by the end of third grade are less likely to successfully develop their literacy skills. While Chall (1983) stated that up until third grade the focus of the curriculum is on learning to read and that the focus shifts from learning to read to reading to learn in fourth grade, in practice and with increased accountability, it seems that this shi ft occurs much earlier. A close look at the Florida Sunshine State Standards, for exampl e, shows that students in third grade require reading proficiency to fully benefit from the curriculum (FLDOE, 2010d). Therefore, it is critical that struggling readers develop the necessary reading skills in order to successfully complete third grade tasks. Unfortunately, there is a tendency for educ ators to delay interventions until nonnative speakers develop their English skill s or to exclude students with limited English

PAGE 186

186 proficiency because they have lower probab ilities of benefitting from interventions (Ashdown & Simic, 2000). These notions have been supported by available literature that underscores how ELLs struggle to unders tand the words they read (Helman, 2009), and consequently, struggle to comprehend texts (Peregoy & Boyle, 2005). Unfortunately, these approaches lim it the opportunities for ELLs to receive the help they need. Furthermore, when educators delay inte rvention, children get farther behind, making it more difficult for t hem to catch up to their peers. Consequently, the initial gap that existed between Spanish-speaking ELLs and their English-speaking peers continues to widen, forcing educators to fi nd and implement more intense interventions. Fortunately, there is sound evidence t hat ELLs benefit from English reading interventions regardless of t heir level of English oral l anguage proficiency (Gunn et al., 2000, 2002, 2005; LeSaux & Siegel, 2003; Quir oga, Lemos-Britton, Mostafapour, Abbott, & Berninger, 2002; Roberts, 2003; Stuart, 1999). The present study provided additional support for t he use of early reading interventions for ELLs with different levels of oral language proficie ncy. It also provided evidence that delaying intervention while language proficien cy develops is unnecessary. Data showed that all partici pating students made a marked im provement by the end of the intervention, including students who scored negligible in oral language proficiency on the WMLS-R. Out of the ten participants, six had a CALP score of 3, equivalent to limited oral proficiency. Of the six students, three were in Group 1 (higher performing), two were in Group 2 (lower performing), and one was in Group 3 (lowest performing). Furthermore, the two lower performing group s (2 and 3) had at least one member who outscored the higher performing group in level of oral proficiency. Maria, in Group 2,

PAGE 187

187 had a CALP score of 4 (fluent) and Jorge, in Group 3, had a CALP sco re of 3.5 (limited to fluent). This finding shows that reading ability is not entirel y contingent on oral language proficiency. Therefor e, there is no compelling reason for delaying literacy instruction and interventions for struggling readers who are learning to read in their second language (Ehri & Roberts, 2006). Perhaps the UFLI program would have been even more effective if it had been implemented earlier. Yet, providing it in second grade was still early enough to remediate and a ccelerate their learning process. It should be noted, however, that the fact that students with limited oral language proficiency improved as a result of this reading intervention does not preclude the need for English language development. Oral language proficiency is critical for students overall achievement. Developing proficiency in English oral skills should remain a key component of ELL instruction. Effective Instructional Practices for ELLs Approximately 60% of ELLs in this count ry are receiving reading instruction in English (August, 2006; Goldenber g, 2008). Rec ently, the NLP repor ted that students who have no access to native language instruction can succeed if high quality instruction is provided in their second l anguage (Snow, 2006). According to Gersten and Geva (2003) there are six inst ructional strategies that ar e highly effective when working with ELLs: (a) explicit teaching (e.g., model skills and strategies, provides prompts, adjust own use of English during lesson, ma kes relationships overt); (b) systematic instruction in phonemic awareness, lette r-sound correspondence, and decoding; (c) English learning (e.g., uses visuals or m anipulatives to teach content, encourages elaborate responses, uses gestures or facial expressions to clarify meaning); (d) vocabulary development (e.g., teaching vo cabulary prior to and during a lesson,

PAGE 188

188 provides opportunities to speak English, engages students in meaningful interactions about text); (e) interactive teaching (e .g., maintains students engaged during lessons, incorporates students ideas and experiences in to lessons, provides sufficient time for students to respond); and (f) instruction geared toward low performers (e.g., promotes response accuracy, opportunities to practice independent practice, ongoing monitoring of understanding and performance, modifies instruction according to students needs). Furthermore, Helman and Burns (2008) recommend giving students ample opportunities to hear and discuss novel words in context, to connect oral and written forms of words, and to use novel words in students own sentences. The effectiveness of the UFLI tutoring program in promoting the reading skills of ELLs can also be attributed to its compli ance with the aforement ioned practices. The UFLI tutoring program, as it was origin ally designed, addressed many of the instructional practices identified by Gerst en and Geva (2003) as well as Helman and Burns (2008), given that many of these st rategies have also been identified in the literature as effective practices for struggling readers in general (Torgesen, 2002; Vaughn, Gersten, & Chard, 2000). In additi on, the UFLI program was modified to support the language needs of Spanish-speaking ELLs. Based on the findings from the pilot study and recommendations from Linan-Thompson and colleagues (2003), the researcher made sure to distinguish bet ween real and nonsense words, use picture cards, and provide quick definitions of unfamiliar words when needed. In addition, based on Mohr and Mohrs (2007) recommendation, the researcher extended students interactions by accepting phrases or parti al answers and modeling complete sentences with standard grammar.

PAGE 189

189 These practices were consistently impl emented throughout the tutoring session with all groups, and while all students benefitted from these practices, it was observed that students with lower levels of oral language proficiency relied more heavily on them. Anecdotal data showed that Jennifer (CAL P 1 negligible) and Viviana (CALP 2 very limited) had limited E nglish vocabulary and relied heavily on the use of picture cards and quick definitions. These strategies were consistently used with them during word work, reading, and writing activities. Further more, during writing activities, Jennifer, Viviana, and Aldo tended to provide incomplete or grammatically incorrect sentences. In response, the tutor supported them by acc epting their answers, rephrasing their ideas into complete and grammatically correct sentences, and asking them to repeat the rephrased sentence before attempting to write it. Consistency with the Respons e to Intervention Approach Respons e to Intervention (RTI) is a multi-tiered educational approach that promotes early identification and early inte rvention for students who may be at risk for reading difficulties (NRCLD, 2007). This approach provides increasingly intense levels of service (Mellard, 2004) for students who do not respond well to the instruction provided (Vaughn, Wanzek, Woodruff, & Linan-Thompson, 2007). At each level, students benefit from research-based interventions, continuous progress monitoring, and data-driven instruction (NJCLD, 2005). Tier 1 consists of research-based core reading programs that address t he needs of the majority of students. Tier 2 offers more strategic and intense interventions in addi tion to the core reading instruction (approximately 30 minutes more). Tier 3 provides sustained, intensive and scientifically based interventions that are tailored to the s pecific needs of each individual (McMaster, Fuchs, Fuchs, & Campton, 2005; Vaughn et al., 2007).

PAGE 190

190 In this study, the UFLI program was impl emented in addition to the core reading program; thus, it can be considered a Tier 2 intervention. Cons istent with the RTI approach, the UFLI program incorporated continuous progr ess monitoring and datadriven instruction. Based on students running records, the tutor determined the area of focus for word work and the level of book reading for that particular lesson. Furthermore, running records and the use of leveled books assisted the tutor in establishing students current reading level to make decisions about the need for further intervention. For example, students in Gro up 1 were able to catch up to grade level (level 20 books) after 38 sessi ons. At that time, students exited the intervention program and moved back to receiving ju st Tier 1 instruction. The RTI model also allows students who do not respond to Tier 2 intervention to receive a more intensive Tier 3 program. It is estimated that between 5% and 10% of students will require Tier 3 interventions (V aughn et al., 2007). According to Torgesen (2002) available research suggests that mu ltilayered interventions can help reduce the number of students who fail to show adequate reading growth. In this study, for example, one participant (Aldo) remained nonresponsive for the majority of the intervention. As a result of this lack of responsiveness, the researcher decided to provide additional one -on-one instruction for that st udent. These extra 15 minutes, which could be considered Tier 3 instructi on, provided the necessary support to start producing a change in Aldos re ading ability. This change was observed in an increased CPPM and CSPM, as well as an increased accuracy level in book reading. Furthermore, this improvement was sustai ned two weeks after the intervention had ended.

PAGE 191

191 Data obtained through Miriams one-on-one tutoring also supported the use of UFLI as more intensive Tier 3 interventi on for students who are performing below grade level. While groups made substantial progre ss, Miriams advancement was considerably faster than that of t he small-group participant s, including that of students who started at a higher book level. Specifically, Miriam made the equivalent of a year and a halfs progress in only 26 sessions. This finding is crit ical for educators that face the challenge of helping students who are significantly below grade level make quick progress in order catch up to their higher performing peers. It should be noted that, wit hout replication of the one-on-one approach with simila r students, it is impossible to know for certain the extent of the role that UFLI played in Miriams rapid im provement. However, given that other aspects of her schooling were the same as students in the small-group intervention and given the result s of the pilot study and other prior UFLI studies that employed one-on-one instructi on, the assumption that one-on-one tutoring produces more rapid gains than small-group tutoring is a reasonable one. It should be noted that in a true RTI model the Tier 2 and Tier 3 interventions would have been initiated with these strugglin g students much earlier in their school career, as soon as progress-monitoring data indicated that they were not keeping up with their peers. Also, it is unc lear whether the core reading instruction provided at the school would meet the criteria for research-based instru ction. Still, even under less than ideal circumstances for implementation, the UFLI intervention proved to be an effective tool that would be appropriate for Tier 2 or Tier 3 in an RTI model. Implications for Future Research The results of this study showed that t he UFLI tutoring program is an effective English reading interv ention for sec ond-grade Spanish-speaking ELLs, who are

PAGE 192

192 struggling to read. Furthermore, these findings are consistent with previous research demonstrating the efficacy of English inte rventions for struggling English language learners. This study holds several import ant implications for future research. One of the main issues with any single-s ubject design has to do with sample size. Given the limited number of participants in this study, it is important to ensure its external validity; that is, the extent to which the results can be generalized to other participants, places, and conditions (Kennedy 2005). For this purpose, direct and systematic replications of the study are necessary. Dir ect replications should be conducted to make sure that these findings generalize to other students who share similar traits with the participants in this study Also, it is critical to conduct systematic replications by purposely sele cting new but dissimilar subjects, settings, and contexts. In particular, systematic replication will help determine whether the findings of this study extend across different school settings, parti cipants, and tutors. Finally, examining the intervention in an experimental control-group design with a larger sample size would be a logical next step to enhance th e generalizability of findings. Since the study was conducted in only one sc hool located in a rural area of North Florida, it is recommended that the UFLI program be examined in different school settings. The effectiveness of the program may be impacted in some way by the characteristics of each particular school. For example, schools may have different schedules that may not allow for daily 45-mi nute sessions or for extended interventions. Reducing the duration of each session, the frequency of sessions, or the length of the intervention may greatly impact the results.

PAGE 193

193 Similarly, given that the participants in this study were in one school and shared similar traits, it is critical that other studies be conducted to examine how the UFLI program can impact the reading skills of ot her Spanish-speaking ELLs. There might be within-group differences that ca n impact students response to intervention. Therefore, it is important to know how other groups wit h different heritages, socioeconomic status, immigration status (i.e., migrants, U.S. residents, U.S. nat ionals), etc., respond to the UFLI tutoring program. Future research should also examine t he effectiveness of the UFLI program as different types of educators implement it (i .e., mainstream teachers, special education teachers, ESOL teachers, reading specialist, paraprofessionals). According to Kazdin (1982), the characteristics of those implementing a program may influence the intervention effects. For example, in a st udy conducted by Ehri, Dreyer, Flugman, and Gross (2007) about the effectiveness of a tutoring intervention for language minority students, the researchers compared result s across reading specialists, credentialed teachers, and paraprofessionals One of the findings showed that paraprofessionals did not affect students ability to decode pseudowords as well as reading specialists. Ehri and colleagues hypothesize that lack of backg round knowledge in phonics instruction might have been the reason behind their lower per formance. At this point, it is unknown if students response to the intervention would differ if the UFLI program was implemented by other professionals, with different levels of knowledge about reading. Besides replicating the study using single-subject designs, between-group comparisons using quantitative analysis should be conducted to further investigate the effectiveness of the UFLI pr ogram in promoting the readi ng skills of beginning readers

PAGE 194

194 who are learning to read in English. For ex ample, given that students were able to maintain the decoding and word recognition rates two weeks after the intervention ceased, it is also important to conduct fo llow-up studies to determine if gains can be sustained for longer periods of time. In addition, it is important to determine whether supplemental data continues to show an improved reading ability. It is also critical to examine how effect ive the program is wit h other ELLs whose native language is not Spanish. While Spanish is the most common language spoken by ELLs, other languages are also represented in the classroom. Klindler (2002) reports that for the year 2000-2001, approximately 10% of students were ELLs and that the most common languages after Spanish we re Vietnamese (2%), Hmong (1.6%), Cantonese (1%), and Korean (1%). Therefore, a logical next step would be to test the UFLI program with speakers of other languages. Furthermore, since many ELLs are being instructed in English alongside their native-English speaking peers (Goldenberg, 2008), it would be important to examine the impact of the UFLI program when ELLs and native-English speaker s are grouped together. In this study, data showed that all participants improved in decoding and word recognition skills as a result of the interv ention, independent of their initial level of English oral language profici ency. While these results are consistent with previous findings (Gunn et al., 2000, 2002, 2005; LeSaux & Siegel, 2003; Quiroga, LemosBritton, Mostafapour, Abbott, & Berninger 2002; Roberts, 2003; Stuart, 1999; Vaughn et al., 2003), examining the UFLI program with a larger samp le would allow researchers to further examine the role of oral proficiency in students response to intervention. One

PAGE 195

195 option would be to compare across groups that are not only grouped by reading level, but also by oral language proficiency. The recent emphasis on the RTI model highlights the import ance of identifying effective interventions that can address the growing needs of struggling readers, especially treatment resist ers or nonresponders (Torge sen, 2000). Given that the UFLI program was originally designed for one-on-one tutoring and later adapted for small-group intervention, exam ining its effectiveness as part of an RTI model has great implications for the fields of reading and special education. The program should be examined as a multilayered approac h, first as a Tier 2 inte rvention (i.e., small group) and then as a Tier 3 intervention (one-on-one; Torgesen, 2002). Consistent with RTI and the use of more int ensive interventions, it is important to evaluate the impact that group size has on the effectiveness of the UFLI program. Given the rapid response of the student receivi ng one-on-one tutoring compared to students receiving small-group tutoring (almost half the number of sessions as Group 2), it is important to do a cost-benefit analysis. In a cost benefit analysis, researchers can directly compare student achievement in diff erent size groups and the cost of providing interventions in smaller groups or one-on-one. RTI also highlights the importance of preventing severe reading difficulties by providing early interventions; thus, the UF LI program should also be examined with younger ELLs, starting in first grade. This is pivotal given that first grade struggling students tend to remain poor readers in later grades (Juel, 1988). Furthermore, it is well known that reading problems in older students are more difficult to remediate even when intensive interventions are provided (Torgesen et al., 2001).

PAGE 196

196 Since the UFLI program was originally designed with the intent of training preservice teachers to become better reading teachers, it would be valuable to investigate the effect that UFLI training and UFLI implementation would have on teachers knowledge about reading. Also, it would be helpful to determine if it has an impact on teachers instructional practi ces in the general classroom. In particular, an examination of the effects of the modified version of UF LI implemented in this study on teachers knowledge of effective reading interv ention for ELLs would be warranted. It is also crucial that studies be conduct ed to examine the most efficient ways to expand the implementation of the UFLI program at the school, district, state, and eventually national level. According to Klingner, Ahwee, Pilonieta, and Menendez (2003), educators face many barriers that can keep them from consistently implementing research-based interventions. So me of these barriers include lack of time, lack of support from administrators, mismat ch between the interventions, other methods mandated by the district, among ot hers. Therefore, future research should also focus on identifying what works best, for whom, and under what conditions (McDonald, Keesler, Kauffman, & Schneider, 2006) so that any scaling-up efforts become feasible. Implications for Practice The findings from this study have signific ant implications for educators working with Spanish-speaking ELLs. The UFLI tutori ng program is an effective English reading intervention that helps develop the skills of beginning readers who are struggling to read in their second language. Teachers can impl ement this small-group intervention to supplement their core reading program and address the needs of many students. It is also important that teachers of E LLs provide interventions for beginning readers that emphasize word reading ski lls, particularly phonemic awareness and

PAGE 197

197 decoding. Providing explicit and systematic instruction in t hese two skills are necessary for the development of students reading fluency. Interventions should also provide daily opportunities for ample practice reading con nected text at an appropriate reading level (Allington, 1983) in order to develop reading automaticity. The fact that all participants benefitted fr om the intervention regardless of their initial level of oral language proficiency indicates t hat educators can provide early interventions to those students who are experiencing difficulties lear ning to read. This has important implications for practice given the tendency for educators to postpone reading interventions until students develop su fficient oral language proficiency. In accordance with Ehri and Roberts (2006) sugges tion, there is no compelling reason for delaying reading interventions for ELLs. Students can learn to read while they continue developing English oral proficiency. Another implication relates to gro uping practices and language support. Since students reading improvement was closely related to their init ial reading level and not to oral proficiency level, t eachers can group students with diffe rent levels of English proficiency to receive intervention. In doing so, it is also impor tant to incorporate instructional practices that support the varying language needs of ELLs so that everyone attains the maximum benefit. Findings from this study have particular relevance for schools, given the recent emphasis on RTI. Results show that UFLI is an effective small-group tutoring program that can be used as a Tier 2 interventi on for Spanish-speaking ELLs who are not responding adequately to Tier 1 instruction. Its emphasis on progress monitoring and data-driven instruction makes this interv ention optimal for RTI. Furthermore, given

PAGE 198

198 Aldos response to additional instructional ti me following the UFLI protocol, teachers can consider using the UFLI program as a Tier 3 intervention, as well. This suggestion is also supported by previous data from the pilo t study that examined the UFLI program as a one-on-one tutoring program for Spanish-speak ing ELLs. In that study, three firstgrade students showed marked improvement on decoding and leveled book reading as a response to intervention. Conclusion Children in the United States whose native language is not Englis h are at particular risk for reading diffi culties (Snow et al., 1998). While the number of language minority students in the U.S. is growing (Klingner, Artiles, & Barletta, 2006), access to effective bilingual programs (August, 2006; Gomez-Bellenge, Chen, & Schulz, 2008) is limited. Up to 60% of ELLs in the nation ar e receiving reading instruction in English (August, 2006; Goldenberg, 2008). Because the level of achievement for ELLs is drastically below their m onolingual peers, there is a need for effective reading interventions at an early stage (Cartl edge et al., 2009; Denton & Mathes, 2003). Unfortunately, many schools do not provide early interventions because they believe that literacy instruction should be delayed until ELLs develop sufficient oral language proficiency (Ehri & Roberts, 2006; Tabors & Snow, 2002). Fort unately, there is evidence that ELLs who participate in English readi ng interventions originally designed for monolingual students respond positively despite having limited levels of oral proficiency (Gunn et al., 2000, 2002, 2005; LeSaux & Siegel, 2003; Quiroga, Lemos-Britton, Mostafapour, Abbott, & Berninger, 2002; Roberts, 2003; Stuart, 1999). These initial findings highli ght the importance of identifying English interventions that have proven to be effective with str uggling monolingual students and evaluate how

PAGE 199

199 effective they are in promoting the reading skills of developing ELLs. For this purpose, a multiple-baseline design across three groups was used to evaluate the effectiveness of the University of Florida Literacy Initiati ve (UFLI) small-group tutoring program for Spanish-speaking ELLs who were strugg ling with reading. The results of this investigation demonstrated that the UFLI tu toring program is an effective reading intervention for second-grade, Spanish-speaking ELLs who are struggling to read in English. Providing an intervent ion that promotes word re ading skills of young learners effectively improves the reading skills of EL Ls. Furthermore, the program proved to be effective despite students varying levels of English oral language proficiency. These findings contribute to the growing literatur e on the effectiveness of English reading intervention for struggling ELLs of varying levels of language proficiency. As Ehri and Roberts (2006) state, there is no compelling reason for postponing reading interventions for struggling ELLs. This study demonstrates that well-des igned intervention can support the reading development of struggling students who are also ELLs. Given the evidence generated by this study and others like it, it is essentia l that schools use what is known to provide appropriate intervention for students with r eading and language difficulties. Making this happen may require reallocation of resources, rethinking of intervention models, or additional professional developm ent for teachers, but the ne ed is too great to ignore.

PAGE 200

200 APPENDIX A IRB DOCUMENTATION

PAGE 201

201

PAGE 202

202

PAGE 203

203

PAGE 204

204

PAGE 205

205

PAGE 206

206

PAGE 207

207

PAGE 208

208

PAGE 209

209

PAGE 210

210 Childs Intervention Assent Form (English and Spanish versions) Hi, my name is Stephanie Arriaza, and I'm from the University. I would like to do some reading activities with you and work with sounds and letters. We will be working together for a few weeks. You can earn sti ckers for participating and you can stop working at any time. I also want you to k now that whatever you decide, this will not affect your grades. Would you like to work with me? Hola, mi nombre es Stephanie Arriaza y vengo de la Universidad. Me gustara hacer algunas actividades de lectura contigo y tr abajar con letras y sus sonidos. Estaremos trabajando juntos por algunas semanas. Tendrs la oportunidad de ganar calcomanas y podrs parar de trabajar cuando tu lo desees. Tambin quiero que sepas que si decides dejar de participar esto no afecta ra tus calificaciones. Quieres trabajar conmigo? Childs Assessment Assent Form (English and Spanish versions) Hi, my name is Stephanie Arriaza, and I am from the University. We will work together for about an hour today and maybe an hour some other day. We will do some reading activities and respond to some questions in English and Spanish. You can earn stickers for participating and you can stop working at any time. I also want you to know that you can stop working at any time. I also want you to know that whatever you decide, this will not affect your grades. Would you like to work with me? Hola, mi nombre es Stephanie Arriaza y vengo de la Universidad. Vamos a trabajar juntos alrededor de una hora al da de hoy y posiblemente una hora otra da. Vamos a hacer algunas actividades de lectura y contestaras algunas preguntas en Ingles y Espaol. Tendrs la opor tunidad de ganar calcomanas y podrs parar de trabajar cuando tu lo desees. Tambin quiero que seas que si decides dejar de trabajar, esto no afectara tus calificaciones Quieres trabajar conmigo?

PAGE 211

211 APPENDIX B HOME LANGUAGE SURVEY AND REMINDER LETTERS Hom e Language Survey Childs name: _________________________ Childs age: ___________________________ Your name: ___________________________ Todays date: __________________________ Household Demographics What is your relationship to the child? mother father other _________________ What is your ethnicity/nationality? ________________________________________________________ How many people live in your household? What are their ages? ___________________ How long have you lived in your current home? ______________________________________________ How many years has your child been in school in this country? __________________________________ What is your highest level of education? Did not complete high school Completed high school or equivalent Community college Obtained undergraduate degree Other Home Language Practices How many languages are spoken at home? ___________ Specify which language are spoken at home ________________________________________________ Which language do you use to interact with your child? _______________________________________ Do you use different languages for different purpo ses? Yes ____ No ____. Specify ________________ ____________________________________________________________________________________ Do you consider yourself a fluent speaker of English? Yes ___ No ___ Is there anything else about your childs language or literacy development that youd like to add?

PAGE 212

212 Lenguaje en el Hogar Encuesta Nombre del nio(a):_____________________ Edad del nio(a):________________________ Nombre del Padre/Madre:_________________ Fecha de hoy: __________________________ Informacin General Cul es su parentesco con su hijo/hija? madre padre otro__________________ Cul es su nacionalidad? ________________________________________________________________ Cuantas personas viven en su hogar? Que edades tienen? ____________________ Cuanto tiempo ha vivido en su actual casa? _________________________________________________ Cuanto aos ha estado su hijo en la escuel a en este pas? _______ ________________________________ Cul es el nivel ms alto de educacin que usted tiene? No complete el bachillerato/preparatoria/high school Complete el bachillerato/preparatoria/high school Estudio Tcnico Licenciatura Otro Actividades de Lenguaje en el Hogar Cuantos idiomas o lenguajes se habla en su hogar? ___________ Especifique que idiomas o lenguajes se hablan en su hogar______________________________________ Con que idioma o lenguaje usted se comunica con su hijo/hija?__________________________________ Usa usted diferentes lenguajes para dife rentes propsitos? Si ____ No ____. Especifique___________________________________________________________________________ Considera usted que habla el idioma Ingles? Si ___ No ___ Hay algo ms que usted quisiera agregar acerca del desarrollo del lenguaje y escritura de su hijo o hija?

PAGE 213

213

PAGE 214

214

PAGE 215

215 APPENDIX C TEACHERS READING ABILITY RATING SCALE (RARS) READING ABILITY RATING SCALE Student Teacher Date Please rate your students ability to perform th e following reading tasks by circling one of the numbers on the scale (1 = very weak 4 = very strong) for each item. Phonological Awareness very weak.....very strong Detect individual sounds in words 1 2 3 4 Segment sounds in words 1 2 3 4 Blend sounds to form words 1 2 3 4 Phonics very weak.....very strong Decode simple words (e.g., cat, boat) 1 2 3 4 Decode more complex words (e.g., multisyllable) 1 2 3 4 Spell simple words 1 2 3 4 Spell more complex words 1 2 3 4 Read high frequency words by sight 1 2 3 4 Use context clues to figure out words 1 2 3 4 Fluency very weak.....very strong Read words accurately in connected text 1 2 3 4 Read words quickly in connected text. 1 2 3 4 Read text with inflection and expression. 1 2 3 4 Read connected text on grade level. 1 2 3 4 Self-monitoring to check for accuracy 1 2 3 4 Self-correcting reading errors 1 2 3 4 Comprehension very weak.....very strong Make predictions about a story. 1 2 3 4 Make personal connections to the story. 1 2 3 4 Summarize text. 1 2 3 4 Engage in discussions of a story or text read. 1 2 3 4 Comprehend grade level texts. 1 2 3 4 Vocabulary very weak.....very strong Comprehend English words while reading 1 2 3 4 Use English vocabulary in speech 1 2 3 4 Use English vocabulary in writing 1 2 3 4 Classroom behaviors very weak.....very strong Participate in class 1 2 3 4 Use English to communicate in class 1 2 3 4 Motivation to read 1 2 3 4 Additional comments about your studen ts reading or language abilities:

PAGE 216

216 APPENDIX D UFLI SMALL GROUP SESSION GUIDE UFLI Small-Group Session Guide STEP 1: GAINING FLUENCY & MEASURING PROGRESS Select and read 1-3 familiar books Have students pair-read a famili ar book while taking running record Present book from prev ious session to student Take running record while child reads Provide feedback regar ding self-corrections you observed STEP 2: WORD WORK WITH MANIPULATIVE LETTERS Hand out magnetic boards and magn etic letters to each student Do word work with familiar words STEP 3: INTRODUCING & READING A NEW BOOK Introduce the new book Coach students through the text Discuss the story Work with new words Record strategies used and focus of word work STEP 4: WRITING FOR READING Elicit and record sentences Ask each student to write their own sentences (check for spelling) Work on each word, while coaching students

PAGE 217

217 APPENDIX E UFLI SMALL-GROUP SESSION NOTES Date: __________________ Session time ________ to _______New book level:_______ Absentees: _____________ Step 1: Gaining Fluency & Measuring Progress Familiar Book Level/Title ________________________________________ ___________________________________ Running Record book title/level ___________________________________________________ Running Record Student ___________________________________ Accuracy _________% Observations Step 2: Word Work Areas of focus for word work: words decoded words encoded ELL accommodations: Step 3: Introducing and Reading a New Book Book Level/Title Strategies introduced/practiced: Working with new words: ELL accommodations:

PAGE 218

218 S tep 4: Writing for Reading Student Sentence Student Sentence Student Sentence Student Sentence ELL accommodations: S ession Observations:

PAGE 219

219 APPENDIX F PSEUDOWORD PROBES jorp dasp tig croob faf gake sape tham taz stup doil vop fike korsh dod nuz lurt jath rast keef zeck beash zid joad shan sev keb snill yabe tusp huz bipe buk bim doon heesh yud meab vift jit waib mip zow fump rofe weff trin vake pef naz

PAGE 220

220 marp nep parb zeck naz zow huz waib jorp hup rofe meab moz rast slig puzz heesh zid shog tusp crin snop mox meez lav rox trin lurt sev tep fump sarb fent tham wut cham yabe sape doil beash keef tig stup dod belp yin jath lish keb cack

PAGE 221

221 rast waf cham nuz marp rell sarb pash keef pef flaf bipe doon puzz dod sev zow keb wut vame doil jorp joad faf yin lav yabe delk dush buk shog zeck tusp mip tep plup tig fump vup parb tane voy sape fent baf shan vake dosh gerg fike

PAGE 222

222 mox doon vake belp doil groob korsh drup meab cack sarb jath tane tusp shog delk keef tep fump zeck fent beash vame joad gerg waib dasp pash jorp mip snop cag mox leat trin zeck plup beal lurt fike huz tham vope yabe parb snill bipe nep wope dush

PAGE 223

223 cag jorp wope flaf korsh zeck tham boz bipe jath croob lish snop vift beal cham tusp beash belp sarb dasp vake yud voy doil plup vame parb grup waib snill bipe lurt stup fike gerg shog keb dasp belp wope dreck rast yabe goft trin rofe bipe tane yud

PAGE 224

224 dreck keef boz fent rofe lav wope sarb zid trin plup shog weff jorp pef wut cham snop flaf buk croob zot meab tane heesh beal gerg jit tep vope beash tusp vop zow tham marp nep tig sev waib rox korsh slig baf voy taz stup parb dosh puzz

PAGE 225

225 b af fike jath meez waf d reck joad cack nuz crin l ish yin cag stup tane b uk moz huz dasp flaf r ast rofe mox parb snill p lup tham tig vame zot c roob shog nep hup rox c ham gerg weff korsh doon v up sev fay gake sape m ip beal beash snop marp

PAGE 226

226 cham sarb lish vake mox delk veek rofe fent boz keb jit slig sape doil buk plup jath dasp cag croob hup nep baf zow puzz doon vame bim pef beal yud stup meez flaf weff dush voy dreck snop parb goft shan joad snill tusp rox vift zeck nuz

PAGE 227

227 jath jorp drup vift lish sape rell mox nep gake snill flaf lurt meez waib bim bipe vame marp heesh moz taz zid sarb baf boz tig zeck voy wope dasp joad huz pef rofe croob delk yud mip fike cag korsh goft belp tane beal keef plup naz yabe

PAGE 228

228 pef dush tane yin snop jit trin weff vup grup doon gerg zeck slig marp jath cack rast vame beal fent mox croob parb korsh lurt dreck stup flaf huz doil fay rox gake dod drup faf yabe goft vope nuz waf plup heesh lav moz sev veek snill tig

PAGE 229

229 goft puzz parb cack delk doon trin fay mox vope vift lish joad dosh waf rofe lurt vame yin korsh shog taz drup bim voy keef tusp yabe tep gerg vop bipe plup keb baf rast lav sarb mip shan sape vake stup waib leat nep rell tig gake grup

PAGE 230

230 sarb heesh lurt grup dreck zid shan tep doil faf rox pash pef tig voy lav goft rell baf beash tane parb zow beal leat sape plup shog moz vame nuz slig nep korsh drup flaf yabe wope yin dod vake waib wut belp keb jath dosh doon trin sev

PAGE 231

231 buk yin tane croob korsh doon parb dosh nep fump plup vup weff marp vope dush yud dod fike leat sev dasp crin grup keb moz gake delk snop slig goft veek bim rast drup waib fent gerg rox beash voy beal tep tusp vop pash lish doil fay zid

PAGE 232

232 snill puzz vake vop dosh plup naz meab snop rox keef jit huz taz gake hup boz pef leat dush korsh flaf shan tig pash waib cham mox stup zot slig joad beash tusp yabe gerg doil yud sev bipe zid nuz zeck grup rell shog bim fay wut dreck

PAGE 233

233 weff sarb lurt snop boz vift bim trin pef jit tig doon nep vake vop marp jath meez stup shog voy joad dod cack zow wut tep fike delk gerg dasp hup slig snill huz drup fay lish vup puzz cham mip vope zeck naz mox goft doil faf lav

PAGE 234

234 APPENDIX G SIGHT WORD PROBES are big first know some their our because any going had saw where right again may did take upon very why this does could ask by yes always your read stop under please must did came all good she there under help gave many her ate here new saw they

PAGE 235

235 live had these like four wash going both after give would their which please your come one off first myself about some done around white many never keep thank her how before use them over because pull right found best would think have funny once what its came just under

PAGE 236

236 look funny live its could work these green bring after many read because open gave would give very found had let once over always one does together wish four about upon them own under done stop cold wash use ask goes white were again pretty take round what her would

PAGE 237

237 upon goes would those live give pull your both had off some sleep why just one dont jump work come very funny let buy think again first have once going green pretty them got wish sit could brown found much cold would ask best been keep what like know write

PAGE 238

238 round their much because like fast were have had upon wash any those walk off which again buy open myself give over take sing under let wish use before been why got funny these work five stop how pull where right pretty both together about does done best sleep keep

PAGE 239

239 her could because pull where best look goes first funny cold made take green dont have always much their again from what right let pretty please done gave there like around live think open thank been upon were which found walk any many had call after some once them jump

PAGE 240

240 does your around write those together look found wash ate every keep any why brown had five cold their much think before own away which upon round going gave work white bring where sing what jump after many made ask pull its got were buy four them very read know

PAGE 241

241 its open buy made gave four again which from around goes look work would your sleep before off what many never over best myself five very their funny right cold white give one walk first please going bring pretty where dont them got how pull always have sing once these

PAGE 242

242 round had work what look use wish those its call sing these five together gave green many off walk goes wash pretty going bring open found before ate why first her best some myself were under thank came cold does upon every there ask know just got buy brown always

PAGE 243

243 look came always going their open once best keep never funny please live why gave from them those first wash bring these call sleep which jump ate does let come read stop thank walk your her pretty much green brown got cold just where its would been work because dont

PAGE 244

244 green live them buy after work four done wash their wish round from five about stop please own have made does goes one once always around cold these going jump were never both what know ask ate bring upon sit come came sing open off gave could pull why because

PAGE 245

245 know their best stop away been together does never thank like read after open both once dont let think around before her why found your going buy pretty write these four under brown look call any give about please upon ate work over from much right very always pull myself

PAGE 246

246 stop five over could open have sing goes around just read from which sleep keep would write let her much give think round wash gave bring call own funny about upon these them know there pretty its had sit every does any myself got before wish both very brown never

PAGE 247

247 could much both just jump white funny how got what come these buy any why were ate always write away after take sleep stop came before ask read goes going work think pull look four every once off cold its use where please done your wish their upon thank first

PAGE 248

248 some about these open why going green let found could bring round just off upon right after again goes which would had never myself always many cold write funny together how please gave there jump every those from live its five got very once dont your their work before read

PAGE 249

249 APPENDIX H TREATMENT INTEGRITY CHECKLIST Treatment Integrity Checklist Group No._____________________________ Date: _________________________________ Observer ______________________________ School: _______________________________ STEP 1: GAINING FLUENCY & MEASURING PROGRESS YES NO NA The tutor provided appropriate books (within a 2-level range from the current new book level; read with 90% accuracy or better). Students read familiar books. Students were paired up for read ing while tutor took a running record Tutor completed a running record with one student using yesterdays new book. Tutor analyzed students strategies/errors. Tutor determined book level based on accuracy. Tutor provided feedback about strategies used by the child. STEP 2: WORD WORK WI TH MANIPULATIVE LETTERS YES NO NA Tutor handed out a magnetic board and ma gnetic letters to each student Tutor led students in word work with familiar words/spelling patterns Decoded words at the onset/rime level Encoded words at the onset/rime level Decoded words at the phoneme level Encoded words at the phoneme level Used real words Used nonsense words Tutor helped students locate word(s) from their word work in the text. Tutor identified which words were real words and which words were nonsense words. Tutor made comparisons between English and Spanish words and sounds. Tutor used picture cards to illustrate unfamiliar word meanings.

PAGE 250

250 STEP 3: INTRODUCING & READING A NEW BOOK YES NO NA The tutor introduced the book and discussed the story. Tutor led students in discussion about illustrations. Tutor promoted student involvement in discussion. Tutor pointed out repetitive language. Tutor used vocabulary from story. Tutor used picture cards or gave quick definitions of unfamiliar words. Tutor helped students make predictions about the story. The tutor coached studen ts through the book. Tutor prompted strategy use. Tutor encouraged students to re -read sentences after decoding a word. Tutor used word work (with manipulat ive letters or dry-erase boards) to introduce new words or spelling patterns. STEP 4: WRITING FOR READING YES NO NA Tutor elicited a sentence from each student based on the new book. Tutor recorded each sentence. Tutor asked each student to write their own sentence on the writing book Student coached each student Sight words were practiced appr opriately (1-3 words) Elkonin boxes were used appropriately (2-4 words) Students read and reread their sentence as each word was added. PROCEDURES YES NO NA Tutor worked with every student in the group. Tutor started the session on time Session did not exceed 45 minutes. Tutor had all the materials ready ahead of time. Tutor handled materials with ease. Observations:

PAGE 251

251 APPENDIX I SOCIAL VALIDITY INSTRUMENTS Student Interview 1. How do you feel about participa ting in the tutoring lessons? 2. Do you think that these lessons helped to become a better reader? How? 3. What part of the lesson was your favorite? Why? 4. What part of the lesson was your least favorite? Why? 5. How do you feel about worki ng in small groups? 6. Will you use what you learned from the lesson in your reading? 7. Do you think other students should pa rticipate in the tutoring lessons?

PAGE 252

252 Teachers Social Validity Questionnaire Thank you for participating in th e evaluation of a UFLI tutoring lesson. Your opinion about the tutoring program and procedures will help us improve the lessons and procedures followed with English language learners (ELLs). Your ident ity will be kept confident ial. If you have any questions, please cont act Stephanie Arriaza at (352) 870-7056 or send an email to satienza@ufl.edu As you observe the tutoring lesson, please use the lesson hand out as a guide and respond to the following questions. Repeated Reading of Familiar Books 1. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 2. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 3. Currently, do you use a similar strategy when you work with English language learners? Yes No 4. Would you be likely to use a strategy like this one with ELLs during your reading instruction? Please explain. Yes No Progress monitoring 5. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 6. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 7. Currently, do you use a similar strategy when you work with English language learners? Yes No 8. Would you be likely to use a strategy like this on e with ELLs during your read ing instruction? Please explain. Yes No

PAGE 253

253 Word work with magnetic letters 9. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 10. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 11. Currently, do you use a similar strategy when you work with English language learners? Yes No 12. Would you be likely to use a strategy like this on e with ELLs during your read ing instruction? Please explain. Yes No Reading a new book 9. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 10. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 11. Currently, do you use a similar strategy when you work with English language learners? Yes No 12. Would you be likely to use a strategy like this on e with ELLs during your read ing instruction? Please explain. Yes No

PAGE 254

254 Writing for Reading 13. Do you feel that this part of the lesson w ill help ELLs develop reading skills? Why or why not? Yes No 14. Do you believe that this component is easy to implement with ELLs? Why or why not? Yes No 15. Currently, do you use a similar strategy when you work with English language learners? Yes No 16. Would you be likely to use a strategy like this on e with ELLs during your read ing instruction? Please explain. Yes No Overall tutoring lesson 17. Do you feel that the tutoring lesson can help ELLs develop reading sk ills? Why or why not? Yes No 18. Do you believe that this lesson is easy to implement? Why or why not? Yes No 19. Would you be willing to implemen t the lessons in your classroom? Yes No 20. Would you be willing to implement com ponents of the lesson in your classroom? Yes No 21. Would you be interested in learning how to implement this tutoring program? Yes No 22. Do you think other teachers of ELLs might be in terested in learning to implement this program? Yes No

PAGE 255

255 APPENDIX J UFLI INDIVIDUAL TUTO RING SESSION GUIDE

PAGE 256

256 LIST OF REFERENCES Adams, M. J. (1990). Learning to read: Thinking and lear ning about print. Cambridge, MA: MIT Press. Adams, M. J., Foorman, B. R., Lundberg, I., & Beeler, T. (1998). Phonemic awareness in young children: A classroom curriculum. Baltimore, MD: Brookes Publishing. Al Otaiba, S. (2005). How effective is code-bas ed reading tutoring in English for English language learners and preser vice teacher-tutors? Remedial and Special Education, 26 (4), 245-254. Allington, R. L. (1983). The reading instru ction provided readers of differing reading abilities. The Elementary School Journal, 83(5), 548-559. Altmann, L. J. P., Lombardino, L. J., & Pu ranik, C. (2008). Sentence production in students with dyslexia. International Journal of Language and Communication Disorders, 43 (1), 55-76. doi :10.1080/13682820701284522 Alvarado, C. G., Ruef, M. L., & Schrank, F. A. (2005). Woodcock-Muoz Language SurveyRevised: Comprehensive Manual. Itasca, IL: Riverside Publishing. Anderson, R. C., Hiebert, E. H., Scott, J. A., & Wilkinson, I.A. (1985). Becoming a nation of readers: The report of the commission on reading. Washington, DC: The National Institute of Education. Antunez, B. (2002). Implementing Reading First with English language learners. Directions in Language and Education, 15, 1-12. Ashdown, J., & Simic, O. (2000). Is early intervention effective for English language learners? Evidence from Reading Recovery. Literacy Teaching and Learning, 5 (1), 27-42. August, D. (2006). Demographic overview. In D. August & T. Shanahan (Eds.), Developing literacy in second-language lear ners: Report of the national literacy panel on language-minority children and youth (pp. 43-49). Mahwah, NJ: Lawrence Erlbaum Associates. August, D., Carlo, M., Dressler C., & Snow, C. (2005). The cr itical role of vocabulary development for English language learners. Learning Disabilities Research & Practice, 20 (1), 50-57. August, D., & Hakuta, K., (1997). Improving schooling for language minority children: A research agenda. Washington, DC: National Research Council. August, D., & Shanahan, T. ( 2006). Developing literacy in second-language learners: Report of the national literacy panel on language-minority children and youth. Mahwah, NJ: Lawrence Erlbaum Associates.

PAGE 257

257 Beck, I. L., Farr, R. C., St rickland, D. S., Amor, A. F., Hudson, R. F., McKeown, M. G.,Washington, J. A. (2007). StoryTown (K to 6th grade Core Reading Program). Orlando, FL: Harcourt School Publishers. Bear, D. R., Helman, L., Templeton, S. Invernizzi, M., & Johnston, F. (2007). Words their way with English learners: Word study for phonics, vocabulary, and spelling instruction. Upper Saddle River, NJ: Pearson. Berninger, V. W., Abbott, R. D., Vermeulen, K., Ogier, S ., Brooksher, R., Zook, D., & Lemos, Z. (2002). Comparison of fast er and slower responders to early intervention in reading: differentiating features of thei r language profiles. Learning Disabilities Quarterly, 25 (1), 59-76. Borko, H. (2004). Professiona l development and teacher l earning: Mapping the terrain. Educational Researcher, 33 (8), 3-15. Bravo, M. A., Hiebert, E. H., & Pearson, P. D. (2006). Tapping the linguistic resources of Spanish-English bilinguals: The role of cognates in science. In R. K. Wagner, A. E. Muse, & K. R. Tannenbaum (Eds.), Vocabulary acquisition: Implications for reading comprehension (pp. 140-154). New York, NY: The Guilford Press. Campbell, M. L. (1984). Corrective Reading program evaluated with secondary students in San Diego. Direct Instruction News, 7 (4), 15-17. Cartledge, G., Gardner, R., & Ford, D. Y. (2009). Diverse learners with exce ptionalities: Culturally responsive teaching in the inclusive classroom. Upper Saddle River, NJ: Pearson. Cardoso-Martins, C., Rodrigues, L. A., & Ehri, L. C. (2003) Place of environmental print in reading development: Evi dence from nonliterate adults. Scientific Study of Reading, 7 (4), 335-355. doi: 10.1207/S1532799XSSR0704_2 Carlo, M. S., August, D., Maclaughlin, B., Snow, C. E., Dressler, C., Lippman, D. N.,White, C. (2004). Closing the gap: Addressing the vocabulary needs of English-language learners in bilingual and mainstream classrooms. Reading Research Quarterly, 39 (2), 188-215. doi:10.1598/RRQ.39.2.3 Carnine, D., Silbert, J., & Kameenui, E. (1990). Direct instruction reading. Columbus, OH: Merrill Publishing Company. Center,Y., & Wheldall, K. ( 1992). Evaluating the effectivene ss of Reading Recovery: A critique. Educational Psychology, 12 (3/4), 263275. Center, Y., Wheldall, K., Freeman, L., Outhred, L., & McNaught, M. (1995). An evaluation of Reading Recovery. Reading Research Quarterly, 39 (2), 240-263. Chall, J. (1983). Stages of reading development. New York, NY: McGraw-Hill.

PAGE 258

258 Chhabra, V., & McCardle, P. (2004). Contributions to evidence-based research. In P. McCardle & V. Chhabra (Eds.), The voice of evidence (pp. 3-11) Baltimore, MA: Paul H. Brookes Publishing. Chiappe, P., Siegel, L. S., & Gottardo, A. (2002). Reading related skills in kindergartners from diverse linguistic backgrounds. Applied Psycholinguistics, 23, 95-116. doi:10.101 7.S014271640200005X Chiappe, P., Siegel, L. S., & Wade-Woolley, L. (2002). Linguistic diversity and the development of reading skills. Scientific Studies of Reading, 6 (4), 369-400. doi:10.1207/S1532799XSSR0604_04 Clay, M. M. (1972). Sands: The concepts about print tests. Auckland, New Zealand: Heinemann. Clay, M. M. (1985). The early detection of reading difficulties. Auckland: Heinemann. Clay, M. M. (1987). Reading Recovery: S ystemic adaptations to an educational innovation. New Zealand Journal of Educational Studies, 22, 35-58. Clay, M. M. (1993a). An observation survey of early literacy achievement. Auckland, New Zealand: Heinemann. Clay, M. M. (1993b). Reading Recovery: A gui de for teachers in training. Portsmouth, NH: Heinemann. Collier, V. P. (1989) How long? A synthesis of resear ch on academic achievement in a second language. TESOL Quarterly, 23 (3), 509-531. Conners, C. K., Sitarenios, G., Parker, J. D., & Epstein, J. N. (1998). Revision and restandardization of the Conners Teacher Rating Scale (CTRS-R): Factor structure, reliability, and criterion validity. Journal of Abnormal Child Psychology, 26(4), 279-291. doi:10.1023/A:1022606501530 Cummins, J. (1981). Age on arrival and immigr ant second language learning in Canada: A reassessment. Applied Linguistics, 2 (2), 131-149. doi:10.1093/applin/II.2.132 DeFord, D. E. (1995). Early wr iting: Teachers and children in Reading Recovery. In S. Swartz & A. Klein (Eds.), Research in Reading Recovery (pp. 148-172). Portsmouth, NH: Heinemann. Denton, C. A., Anthony, J. L., Parker, R., & Hasbrouck, J. E. (2004). Effects of two tutoring programs on the English readi ng development of Spanish-English bilingual students. The Elementary School Journal, 104 (4), 289-305. Denton, C. A., Ciancio, D. J., & Fletcher, J. M. (2006). Validity, reli ability, and utility of the Observation Survey of Early Literacy Achievement. Reading Research Quarterly, 41 (1), 8-34.

PAGE 259

259 Denton, C. A., & Mathes, P. G. (2003). Interventions for struggling readers: Possibilities and challenges. In B. R. Foorman (Ed.), Preventing and remediating reading difficulties: Bringing science to scale (pp. 229-252). Timonium, MD: York Press. Ehri, L. C. (1989). The development of s pelling knowledge and its role in reading acquisition and reading disability. Journal of Learning Disabilities, 22 (6), 356-365. Ehri, L. C. (1991). Development of the ability to read words. In R. Barr, M. Kamil, P. Mosenthal, & P. Pearson (Eds.), Handbook of reading research (Vol. II, pp. 383417). New York, NY: Longman. Ehri, L. C. (1997). Learning to read and learning to spell are one and the same, almost. In C. A. Perfetti, L. Rieben, & M. Fayol (Eds.), Learning to spell: Research, theory, and practice across languages (pp. 237-269). Mahwah, NJ: Erlbaum. Ehri, L. C. (1998). Grapheme-phoneme knowledge is essential for learning to read words in English. In J. Me tsala & L. C. Ehri (Eds.), Word recognition in beginning literacy (pp. 3-40). Mahwah, NJ: Erlbaum. Ehri, L. C. (2002). Phases of acquisition in learning to read words and implications for teaching. In R. Stainthor p & P. Tomlinson (Eds.), Learning and teaching reading (pp. 7-28). London, England: British Journal of Educational Psychology Monograph Series II. Ehri, L. C.(2004). Teaching phonemic awar eness and phonics: An explanation of the national reading panel metaanalysis. In P. McCardle & V. Chhabra (Eds.), The voice of evidence in reading research (pp. 153-186). Baltimore, MA: Brookes Publishing Ehri, L.C., (2005a). Development of sight word reading: Phases and findings. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 135-154). Malden, MA: Blackwell Publishing. Ehri, L. C. (2005b). Learning to read words: Theory, findings, and issues. Scientific Studies of Reading, 9 (2), 167-188. doi: 10.1207/s1532799xssr0902_4 Ehri, L. C., Dreyer, L. G., Flugman, B., & Gross, A. (2007). Reading Rescue: An effective tutoring intervention model for language minority students who are struggling readers in first grade. American Educational Research Journal, 44 (2), 414-448. doi:10. 3102/0002831207302175 Ehri, L. C., & McCormick, S. (1998). Phases of word learning: Implications for instruction with delayed and disabled readers. Reading & Writing Quarterly, 14(2), 135-163. doi :10.1080/1057356980140202

PAGE 260

260 Ehri, L. C., & Roberts, T. ( 2006). The roots of learning to read and write: Acquisition of letters and phonemic awareness. In D. K. Dickinson & S. B. Neuman (Eds.), Handbook of early literacy research (Vol. II, pp. 113-131). New York, NY: The Guilford Press. Ehri, L. C., & Snowling, M. (2004). Developmental variation in word recognition. In A. Stone, E. Silliman, B. Ehren, & K. Apel (Eds.), Handbook of language and literacy: Development and disorders (pp. 433-460). New York: The Guilford Press. Ehri, L. C., & Wilce, L., ( 1983). Development of word identification speed in skilled and less skilled beginning readers. Journal of Educational Psychology, 75 (1), 3-18. doi:10.1037/00220663.75.1.3 Ehri, L. C., & Wilce, L. (1985). Movement into reading: Is the first stage of printed word learning visual or phonetic? Reading Research Quarterly, 20 (2), 163-179. Engelmann, S., & Bruner, E. (1988). Reading Mastery. Chicago, IL: Science Research Associates. Engelmann, S., Carnine, L., & Johnson, G. (1988). Corrective Reading: Word-attack basics. Chicago: Science Research Associates. Engelmann, S., & Engelmann, K. (2004). Impediments to scaling up effective comprehensive school reform models. In T. Glennan, S. Bodilly, J. Galegher, & K. Kerr (Eds.), Expanding the reach of education reforms: Perspectives from leaders in the scale-up of educational interventions (pp. 107-133). Santa Monica, CA: RAND. Escamilla, K. (1994). Descubriendo La Lectu ra: Un programa en Espaol de lectoescritura para las primeras etapas. Literacy, Teaching and Learning, 1(1), 72-85. Fitzgerald, J. (1995). English-as-a-s econd-language learners cognitive reading processes: A review of the research in the United States. Review of Educational Research, 65 (2), 145-190. Florida Center for R eading Research (2006). DIBELS risk levels chart. Retrieved May 28th, 2010 from http://www.fcrr.org/pmrn/docs/ dibles_risklevels_k3_0 607_0809.pdf Florida Department of Education (2009). Student performance results: School reading demographic report. Retrieved May 28th, 2010 from https://app1.fldoe.org/FCATDemographics/Selections.aspx?lev el=School&subj= Reading Florida Department of Education (2010a). Florida school indicators report 2008-2009: Membership by category, school data. Retrieved May 29th, 2010 from http://www.fldoe.org/eias/eiaspubs/0809fsir.asp

PAGE 261

261 Florida Department of Education (2010b). Florida school indicators report 2008-2009: Membership by category, district and state data. Retrieved May 29th, 2010 from http://www.fldoe.org/eias/eiaspubs/0809fsir.asp Florida Department of Education (2010c). Florida school indicators report 2008-2009: Membership by grade, school data. Retrieved May 29th, 2010 from http://www.fldoe.org/eias/eiaspubs/0809fsir.asp Florida Department of Education (2010d). Next generation Suns hine State Standards Retrieved July 15th, 2010 from http://www.floridastandards.org /Standards/FLStandar dSearch.aspx Fountas, I. C., & Pinnell, G. S. (1999). Matching books to readers: Using leveled book in graded reading, K-3. West port, CT: Heinemann. Francis, D., Lesaux, N., & August, D. (2006). Language of instruction. In D. August & T. Shanahan (Eds.), Developing literacy in second-l anguage learners: Report of the national literacy panel on language-minority children and youth (pp. 365-413). Mahwah, NJ: Lawrence Erlbaum Associates. Fuchs, L. S., & Deno, S. L. (1992). Effects of curricu lum within curriculum-based measurements. Exceptional Children, 58, 232-243. Fuchs, L. S., & Fuchs, D. (2007). A model for implementing responsiveness to intervention. Teaching Exceptional Children,39 (5), 14-20. Genesee, F., Lindholm-Leary, K., Saunders, W ., & Christian, D. (2005). English language learners in U.S. Schools: An overview of research findings. Journal of Education for Students Placed at Risk, 10 (4), 363-385. doi:10.1207/s 15327671espr1004_2 Gerber, M., & Eng lish, J. (2003). Project La Pater: Final report. Santa Barbara, CA: Center for Advanced Studies of Individual Differences. Gerber, M., Jimenez, T., Leafsted t, J., Villaruz, J., Richards, C., & English, J. (2004). English reading effects of small-group intensive intervention in Spanish for k-1 English learners. Learning Disabilities Research & Practice, 19 (4), 239-251. doi:10.1111/j.1540-5826.2004.00109.x Gersten, R., & Baker, S. ( 2000). What we know about effect ive instructional practices for English language learners. Exceptional Children, 66 (4), 454-470. Gersten, R., Baker, S., Shanahan, T., Linan-Thompson, S., Collins, P., & Scarcella, R. (2007). Effective literacy and Eng lish language instruction for English learners in the elementary grades (NCEE 2007-4011). Washington DC: National Center for Education Evaluation and Regional Assistance. Retrieved October 14th, 2009 from http://ies.ed.gov/ncee/wwc/publications/practiceguides

PAGE 262

262 Gersten, R., & Geva, E. (2003). Teachi ng reading to early language learners. Educational Leadership, 60 (7), 44-49. Geva, E., Yaghoub-Zadeh, Z., & Sc huster, B. (2000). Understan ding differences in word recognition skills of ESL children. Annals of Dyslexia, 50 123-154. doi:10.1007/s 11881-000-0020-8 Glynn, T., Crooks, T., Bethune, N., Ballard, K., & Smith, J. (1989). Reading Recovery in context. Wellington: New Zealand Department of Education. Goldenberg, C. (2008). Teaching English la nguage learners: What the research does and does notsay. American Educator, Summer. Good, R. H., & Kaminski, R. A. (2002). Dynamic indicators of basic early literacy skills (6th edition). Eugene, OR: Institute for the Development of Educational Achievement. Retrieved August 17th, 2009 from https://dibels.uoregon.edu/measur es/files/admin_and_s coring_6th_ed.pdf Gough, P. (1983). Context, form, and inte raction. In K. Rayner (Ed.), Eye movement in reading (pp.331-358). Cambridge MA: MIT Press. Gough, P. (1996). How children le arn to read and why they fail. Annals of Dyslexia, 46 1-20. Gough, P., & Juel, C. (1991). The first stages of word recognition. In L. Reiben, &. C. A. Perfetti (Eds.), Learning to read: Basic re search and its implic ations (pp. 47-56). Hillsdale, N.J: Erlbaum. Greene, J. P. (1997). A meta -analysis of the Rossell and Baker review of bilingual education research. Bilingual Research Journal, 212 (2/3), 1-22. Gregory, R., Hackney, C., & Gregory, N. (1982). Corrective Reading programme: An evaluation. British Journal of Educational Psychology, 52, 33-50. Gunn, B., Biglan, A., Smokow ski, K., & Ary, D. (2000). The efficacy of supplemental instruction in decoding skills for His panic and non-Hispanic students in early elementary school. The Journal of Special Education, 34 (2), 90-103. Gunn, B., Smolkowski, K., Bigl an, A., & Black, C. (2002). Supplemental instruction in decoding skills for Hispanic and non-Hispa nic students in early elementary school: A follow-up. The Journal of Spec ial Education, 36(2), 67-79. Gunn, B., Smolkowski, K., Bi glan, A., Black, C., & Blai r, J. (2005). Fostering the development of reading skill through s upplemental instruction: Results for Hispanic and non-His panic students. The Journal of Special Education, 39 (2), 66-85.

PAGE 263

263 Guthrie, T. J., & Davis, M. H. (2003). Motivating struggling readers in middle school through an engagement model of classroom practice. Reading & Writing Quarterly, 19 (1), 59-85. doi :10.1080/10573560308203 Hasbrouck, J. E., Ihnot, C., & Rogers, G. H. (1999). Read Naturally: A strategy to increase oral reading fluency. Literacy Research and Instruction, 39 (1), 27-37. Hayes, L. F., Lane, H. B., & Pullen, P. C. (2005). University of Florida Literacy Initiative: Tutoring for beginning readers. Unpublished manual, Univer sity of Florida, Gainesville, Florida. Helman, L. (2009). Opening door s to texts: Planning effect ive phonics instruction with English learners. In L. Helman (Ed.), Literacy development with English learners (pp. 138-155). New York, NY: The Guilford Press. Helman, L., & Burns, M. ( 2008). What does oral language have to do with it? Helping young English-language learners acqui re a sight word vocabulary. The Reading Teacher, 62 (1), 14-19. Hickman, P., Pollard-Durodola, S., & Vaughn, S. (2004). Storybook reading: Improving vocabulary and comprehension for English language learners. The Reading Teacher, 57 (8), 720-730. Hobsbaum, A. (1995). R eading Recovery in England. In S. Swartz & A. Klein (Eds.), Research in Reading Recovery (pp. 132-147). Portsm outh, NH: Heinemann. Hoover, W. A., & Gough, P. B. (1 990). The simple view of reading. Reading and Writing, 2 127-160. Horner, R. B., Carr, E. G., Halle, J., McGee, G., Odom, S., & Wolery, M. (2005). The use of single-subject research to ident ify evidence-based practice in special education. Exceptional Children, 71 (2), 165-179. Hudson, R. F., Pullen, P. C., Lane, H. B., & Torgesen, J. K. (2009). The complex nature of reading: A multidimensional view. Reading & Writing, 25 (1), 4-32. doi:10.1080/10573560802491208 Individuals with Disabilit ies Education Act of 2004, Pub. L. No.108-446. Jitendra, A., Edwards, L., St arosta, K., Sacks, G., Jacobs on, L., & Choutka, C. (2004). Early reading instruction for children with reading difficulties: Meeting the needs of diverse learners. Journal of Learning Disabilities, 37 (5), 421-439. Jordan, G. (2006). What is SRA Early Interventions in Reading Level 1? Florida Center for Reading Research, Retrieved November 24th, 2008 from http://www.fcrr.org/FCRRReports/PDF/SRAEIR.pdf

PAGE 264

264 Juel, C. (1988). Learning to read and write: A longitudinal study of 54 children from first through fourth grades. Journal of Educational Psychology, 80 (4), 437-447. doi:10.1037/00220663.80.4.437 Juel, C., Griffith, P. L., & Gough, P. B. (1986). Acquisition of literacy: A longitudinal study of children in first and second grade. Journal of Educational Psychology, 78(4), 243-255. doi:1 0.1037/0022-0663.78.4.243 Kamps, D., Abbott, M., Greenwood, C., A rreaga-Mayer, C., Wills, H., Longstaff, J.,Walton, C. (2007). Use of evidence-based, small group reading instruction for English language learners in element ary grades: Secondary-tier intervention. Learning Disabilities Quarterly, 30 (3), 153 168. Kaufman, A. S., & Ka ufman, N. L. (2004). Kaufman Test of Educational Achievement-II, Comprehensive form manual. Minneapolis, MN: Pearson Assessments. Kazdin, A. E. (1982). Single-case research designs: Methods for clinical and applied settings. New York, NY: Oxford University Press. Kazelskis, R., Thames, D., & Reeves, C. (2004). The Elementary Reading Attitude Survey: Factor invariance across gender and race. Reading Psychology, 25 (2), 111-120. doi:10. 1080/02702710490435682 Kelly, P. R., Gomez-Bellenge, F. X., Chen, J., & Schulz, M. M. (2008). Learner outcomes for English language learner lo w readers in an early intervention. TESOL Quarterly, 42 (2), 235-260. Kennedy, C. H. (2005). Single-case designs for educational research. Boston, MA: Allyn and Bacon. Kindler, A. (2002). Survey of the states limited English pr oficient students and available educational programs and services 2000-2001 summary report. Klingner, J. K., Ahwee, S., Pilonieta, P., & Menendez, R. (2003). Barriers and facilitators in scaling up research-based practices. Exceptional Children, 69 (4), 411-430. Klingner, J. K., Artiles, A., & Barletta, L. (2006). English language learners who struggle with reading: Language acquisition or LD? Journal of Learning Disabilities, 39 (2), 108-128. Krashen, S. D. (1991). Bilingual education: Arguments for and (bogus) arguments against. In J. E. Alat is & A. Tan (Eds.), Round table on languages and linguistics (pp. 111-127). Washington, D. C.: Georgetown University Press. LaBerge, D., & Samuels, J. (1974). Toward a theory of automatic information processing in reading. Cognitive Psychology, 6 (2), 293-323. doi:10.1016/00100285(74)90015-2

PAGE 265

265 Lane, H. B., & Pullen, P. C. (2004). Phonological awareness assessment and instruction: A sound beginning. Boston, MA: Allyn and Bacon. Lane, H. B., Pullen, P. C ., & Hayes, L. F. (2007). UFLI: Small group intervention handbook: Kindergarten & first grade. Unpublished manual, University of Florida, Gainesville, Florida. Lane, H. B., Pullen, P. C., Hudson, R. F., & Konold, T. R. (2009) Identifying essential instructional components of literacy tutoring for st ruggling beginning readers. Literacy Research and Instruction, 48 (4), 1-20. doi: 10.1080/19388070902875173 Leafstedt, J. M., Richards, C. R., & Gerber, M. M. (2004). Effectiveness of explicit phonological-awareness instruction fo r at-risk English language learners. Learning Disabilities Research & Practice, 19(4), 252-261. Lee, J., Grigg, W., & Donahue, P. (2007). The nations report card: Reading 2007. Washington, DC: National Center for Educational Statistics. Lemos-Britton, Z., & Mostafapour, E. ( 1997). Spanish phonological awareness assessment and training for ESL students (SPAAT). Concord, CA: Author. Lesaux, N. K., & Siegel, L. S. (2003). The development of reading in children who speak English as a second language. Developmental Psychology, 39 (6), 1005-1019. doi:10.1037/00121649.39.6.1005 Lesaux, N. K., Koda, K., Siegel, L. S., & Shanahan, T. (2006). Development of literacy. In D. August & T. Shanahan (Eds.), Developing literacy in second-language learners: Report of the national litera cy panel on language-minority children and youth (pp. 75-122). Mahwah, NJ: Law rence Erlbaum Associates. Leslie, L., & Caldwell, J. (2006). Qualitative Readi ng Inventory-4. Boston, MA: Allyn and Bacon. Limbos, L. M., & Geva, E. (2001). Accuracy of teac her assessments of secondlanguage students at risk for reading disability. Journal of Learning Disabilities, 34(2), 136-151. Linan-Thompson, S., Vaughn, S., Hickman-Davis, P., & Kouzekanani, K. (2003). Effectiveness of supplemental reading in struction for second-grade English language learners with r eading difficulties. The Elementary School Journal, 103 (3), 221-238. Linan-Thompson, S., Vaughn, S., Prater, K., & Cirino, P. (2006). The response to intervention of English language lear ners at risk for reading problems. Journal of Learning Disabilities, 39(5), 390-398.

PAGE 266

266 Lindsey, K. A., Manis, F. R., & Bailey, V. E. (2003). Predicti on of first-grade reading in Spanish-speaking Eng lish-language learners. Journal of Educational Psychology, 95(3), 482-492. doi:1 0.1037/0022-0663.95.3.482 Lonigan, C. J. (2006). Devel opment, assessment, and promot ion of preliteracy skills. Early Education & Development, 17 (1), 91-114. doi:10.1207/s15566935eed1701_5 Manyak, P. C., & Bauer, E. B. (2008). Expl icit code and comprehension instruction for English learners. The Reading Teacher, 61(5), 432-434. Marchand-Martella, N., Slocum, T., & Martella, R. (2004). Introduction to direct instruction. Boston, MA: Pearson Education. Martson, D. (1989). Curriculum-based measurem ent: What is it and why do it? In M. Shinn (Ed.), Curriculum-based measurement: Assessing special children (pp. 1878). New York, NY: Guilford Press. Mathes, P. G., Denton, C. A., Fletcher, J. M., Anthony, J. L., Francis, D. J., & Schatschneider, C. (2005). The effects of theoretically differ ent instruction and student characteristics on the skills of struggling readers. Reading Research Quarterly, 40 (2), 148-182. doi:10.1598/RRQ.40.2.2 Mathes, P. G., & Torg esen, J. K. (2005). SRA Early interventions in reading: Level 1. Columbus, OH: SRA/McGraw-Hill. McCardle, P., Mele-McCarthy, J., Cutting, L., Leos, K., & DEmilio T. (2005). Learning disabilities in English language learners: Identifying the issues. Learning Disabilities Research & Practice, 20(1), 1-5. doi:10.1111/j.15405826.2005.00114.x McDonald, S. K., Keesler, V. A., Kauffm an, N. J., & Schneider, B. (2006). Scaling-up exemplary interventions. Educational Researcher, 35 (3), 15-24. McKenna, M. C., & Kear, D. J. (1990). Measuring attitude to ward reading: A new tool for teachers. The Reading Teacher, 43 (9), 626-639. McMaster, K. L., Fuchs, D., Fuchs, L. S., & Compton, D. L. ( 2005). Responding to nonresponders: An experimental field tr ial of identification and intervention methods. Exceptional Children, 71 (4), 445-463. Mohr, K. A., & Mohr, E. S. (2007). Ext ending English-language learners classroom interactions using the Response Protocol. The Reading Teacher, 60 (5), 440-450. doi:10.1598/RT.60.5.4 Mortensen, E. (1984). Reading achievement of native S panish-speaking elementary students in bilingual vs. monolingual programs. Bilingual Review, 11 (3), 31-36.

PAGE 267

267 National Center for Educat ion Statistics. (2001). Entering kindergarten: Findings from the condition of education 2000. Retrieved November 24th, 2008 from http://nces.ed.gov/pubs2001/2001035.pdf National Commission on Exc ellence in Education. (1983). A nation at risk: The imperative for educational reform. Washington, DC: U.S. Department of Education. National Data Evaluation Center. (2008). Technical report: Reading Recovery and Descubriendo La Lectura national report 2006-2007. Retrieved November 15th, 2008 from http://www.ndec.us/Documentation.asp National Reading Panel. (2000). Teaching children to read: An evidence-bas ed assessment of the scientific research literature on reading and its implications for reading instruction. Washington, DC: Na tional Institute of Child Health and Human Development. Neal, J. C., & Kelly, P. R. (1999). The success of Reading Recovery for English language learners and Descubriendo La Lect ura for bilingual students in California. Literacy Teaching and Learning, 4 (2), 81-108. Neuman, S. B., & McCormick, S. (2000). A case for single-subject experiments in literacy research. In M. L. Kamil, P. D. Pearson, R. Barr, & P. B. Mosenthal (Eds.), Methods of literacy research (pp. 181-194). Mahwah, NJ: Erlbaum. Nicholson, T. (1989). A comm ent on Reading Recovery. New Zealand Journal of Educational Studies, 24 95-97. No Child Left Behind Act of 2001, PL 107-110, 20 U.S.C. 6301 et seq. OConnor, R. E., Notari-Syverson, A., & Vadasy, P. F. (1998). Ladders to literacy: A kindergarten activity book. Baltimore, MA: Brookes Publishing. Odom, S. L., Brantlinger, E., Ge rsten, R., Horner, R. H., T hompson, B., & Harris, K. R. (2005). Research in special education: Scientific methods and evidence-based practices. Exceptional Children, 71 (2), 137-148. Peregoy, S. F., & Boyle, O. F. (2005). Eng lish learners reading Eng lish: What we know, what we need to know. In Z. Fang (Ed.), Literacy teaching and learning: Current issues and trends (pp. 18-27). Upper Saddle Riv er, NJ: Pearson Education. Perry, N. E., & Meisels, S. J. (1996). How accurate are teachers judgments of students academic performance? (Working Paper No. 9608,National Center for Educational Statistics).W ashington, DC: U.S. Department of Education. Pinnell, G. S., Lyons, C. A., De Ford, D. E., Bruk, A. S., & Se ltzer, M. (1994). Comparing instructional models for the literacy education of high-risk first graders. Reading Research Quarterly 29(1), 8-39.

PAGE 268

268 Planty, M., Husar, W., Sny der, T., Kena, G., KewalRamani A., Kemp, J., Bianco, K., Dinkes, R. (2009). The condition of education 2009 (NCES 2009-081) Washington, DC: National Center for Educational Statistics. Retrieved on April 19th, 2010 from the http://nces.ed.gov/pubs2009/2009081.pdf Pollar-Durodola, S. D., Mathes, P G. Vaughn, S., Cardenas-Hagan, E., & LinanThompson, S. (2006). The role of or acy in developing comprehension in Spanish-speaking Eng lish language learners. Top Language Disorders, 26(4), 365-384. Pullen, P. C. (2000). The effe cts of alphabetic word work with manipulative letters on the reading acquisition of struggling fi rst-grade students (Unpublished doctoral dissertation). University of Florida, Gainesville. Pullen, P. C., Lane, H. B., Lloyd, J. W., Nowak. R., & Ryals, J. (2005). Effects of explicit instruction on decoding of struggling fi rst grade students: A data-based case study. Education & Treatment of Children, 28 (2), 63-76. Quirk, M. P., & Schwanenflugel, P. J. (2005). Do supplemental remedial reading programs address the motiva tional issues of struggling readers? An analysis of five popular programs. Literacy Research & Instruction, 43 (3), 1-19. Quiroga, T., Lemos-Britton, Z ., Mostafapour, E., A bbott, R., & Berni nger, V. (2002). Phonological awareness and beginning reading in Span ish-speaking ESL first graders: Research into practice. Journal of School Psychology, 40 (1), 85-111. doi:10.1016/S00224405(01)00095-4 Richards, C., Leafstedt, J. M., & Gerber, M. M. (2006). Qua litative and quantitative examination of four low-performi ng kindergarten English learners. Remedial and Special Education, 27(4), 218-234. Richards, C., Leafstedt, J. M., Gerber, M. M., Jimenez, T., & Filippini, A. (2003). La Patera phonological awareness test (LP-PAT). Santa Barbara: University of California, Center for the Advanc ed Study of Individual Differences. Roberts, T. A. (2003). Effects of alphabet-letter instruction on young childrens word recognition. Journal of Educational Psychology, 95 (1), 41-51. doi:10.1037/00220663.95.1.41 Samuels, J. (1979). The method of repeated readings. Reading Teacher, 32, 403-308. Santoro, L., E., Jitendra, A. K., Staros ta, K., & Sacks, G. (2006). Reading well with Read Well: Enhancing the reading performanc e of English language learners. Remedial and Spec ial Education, 2(2), 105-115.

PAGE 269

269 Scarborough, H. S. (2 003). Connecting early language and literacy to later reading (dis) abilities: Evidence, theory, and practice. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (pp. 97110). New York, NY: The Guilford Press. Schwartz, R. M. (2005). Literacy learni ng of at-risk first-grade students in Reading Recovery early intervention. Journal of Educational Psychology, 97 (2), 257-267. doi:10.1037/00220663.97.2.257 Schirmer, B. R., Casbon, J., & Twiss, L. L. (1996). Innovative literacy practices for ESL learners. The Reading Teacher, 49 (5), 412-414. Seymour, P., & Duncan, L. (2001) Learning to read in English. Psychology: The Journal of the Hellenic Psyc hological Society, 8 281-299. Seymour, P., & Elder, L. (1986). Beginning reading without phonology. Cognitive Neuropsychology, 3 (1), 1-36. doi :10.1080/02643298608252668 Shanahan, T., & Barr, R. (1995) Reading Recovery: An i ndependent evaluation of the effects of an early instructional intervention for at-risk learners. Reading Research Quarterly, 30 (4), 958-996. Shanahan, T., & Beck, I. (2006). Effectiv e literacy teaching for English-language learners. In D. Augus t & T. Shanahan (Eds.), Developing literacy in secondlanguage learners: Report of the national literacy panel on language minority children and youth (pp. 415-488) Mahwah, NJ: Lawrence Erlbaum Associates. Shankweiler, D., Lundquist, E., Katz, L., Stu ebing, K. K., Fletc her, J. M., Brady, S.,Shaywitz, B. A. (1999). Comp rehension and decoding: Patterns of association in children wit h reading difficulties. Scientific Studies of Reading, 31(1), 69-94. doi:10. 1207/s1532799xssr0301_4 Siegler, R. S., & Crowley, K. (1991). A microgenetic me thod: A direct means for studying cognitive development. American Psychology, 46 (6), 606-620. doi:10.1037/0003066X.46.6.606 Silliman, E., Wilkinson, L., & Brea-Spahn, M. (2004). Policy and practice imperatives for language and literacy learning: Who will be left behind? In C.A. Stone, E.R. Silliman, B.J. Ehren, & K. Apel (Eds.), Handbook of language and literacy (pp. 97-129). New York, NY: The Guilford Press. Slaving, R. & Cheung, A. (2005). A synthesis of res earch on language of reading instruction for English language learners. Review of Educational Research, 75(2), 247-284. Smith, S. (2004). What is Corrective Reading? Florida Center for Reading Research. Retrieved on November 23rd, 2004 from http://www.fcrr.org/FCRRReports/PDF /corrective_reading_fina l.pdf

PAGE 270

270 Snow, C. (2006). Cross-cutting themes and future research directions. In D. August & T. Shanahan (Eds.), Developing literacy in second-l anguage learners: Report of the National Literacy Panel on langua ge-minority children and youth (pp. 631-651). Mahwah, NJ: Lawrence Erlbaum Associates. Snow, C., Burns, M., & Griffin, P. (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press. Sparks, G. M. (1988). Teachers a ttitudes toward change and subsequent improvements in classroom teaching. Journal of Educational Psychology, 80(1), 111-117. doi:10.1037/0022-0663.80.1.111 Sprick, M. M., Howard, L. M., & Fidanque, A. (1998). Read Well: Critical foundations in primary reading. Longmont, CO: Sopris West. Stahl, S. (2004). What do we know about fluen cy? Findings from the National Reading Panel. In. P. McCardle & V. Chhabra (Eds.), The voice of evidence in reading research (pp. 187-211). Baltimore, MA: Brookes Publishing. Stanovich, K. E. (1980). Toward an inte ractive-compensatory model of individual differences in the development of reading fluency. Reading Research Quarterly, 16(1), 32-71. Stanovich, K. E. (1986). Matt hew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21 (4), 360407. Stanovich, K. E. (1990). Concepts in developm ental theories of reading skill: Cognitive resources, automaticity, and modularity. Developmental Review, 10 (1), 72-100. doi:10.1016/02732297(90)90005-O Stuart, M. (1999). Getting ready for r eading: Early phoneme awareness and phonics teaching improves reading and spelling in inner-city second language learners. British Journal of Educational Psychology, 69 587-605. Swartz, S. L., & Kl ein, A. F. (1995). Research in Reading Recovery. Portsmouth, NH: Heinemann. Tabors, P. O., & Snow, C. E. (2002) Young bilingual childr en and early literacy development. In S. B. Neuman & D. K. Dickinson (Eds.), Handbook of early literacy research (pp. 159-178). New York, NY: The Guilford Press. Tindal, G., Marston, D ., & Deno, S. L. (1983). The reliability of direct and repeated measurement (Research Report 109). Minneapolis, MN: University of Minnesota Institute for Research on Learning Disabilities.

PAGE 271

271 Torgesen, J. K. (1998). Catch them before they fall: Identification and assessment to prevent reading failure in young children. American Educator, 22, 32-39. Torgesen, J. K. (2000). Indivi dual differences in response to early interventions in reading: The lingering problem of treatment resisters. Learning Disabilities Research & Practice, 15 (1), 55-64. Torgesen, J. K. (2002). The prev ention of reading difficulties. Journal of School Psychology, 40(1), 7-26. doi:10.1016/S0022-4405(01)00092-9 Torgesen, J. K., Alexander, A. W., Wagner, R. K., Rashotte, C. A., Voeller, K. K., & Conway, T. (2001). Intensive remedial in struction for children with severe reading disabilities: Immediate and long-term outcomes from two instructional approaches. Journal of Learning Disabilities, 34 (1), 33-58. Torgesen, J. K., Wagner, R. K. & Rashotte, C. A. (1999). Test of word reading fluency. Austin, TX: Pro-Ed. Vaughn, S., Cirino, P. T., Li nan-Thompson, S., Mathes, P. G. Carlson, C. D., Hagan, E. C.,Francis, D. J. (2006). Effectiv eness of a Spanish intervention and an English intervention for English-languag e learners at risk for reading problem. American Educational Research Journal, 43 (3), 449-487. Vaughn, S., Gersten, R., & Chard, D. J. (2000). The underlying message in LD intervention research: Findings from research syntheses. Exceptional Children, 67(1), 99-114. Vaughn, S., Linan-Thompson, S. & Hickman, P. (2003). Res ponse to instruction as a means of identifying students with reading/learning disabilities. Exceptional Children, 69 (4), 391-409. Vaughn, S., Linan-Thompson, S., Kouzekanani, K., Bryant, D. P., Dickson, S., & Blozis, S. A. (2003). Reading instruction grouping for students with reading difficulties. Remedial and Spec ial Education, 24 (5), 301-315. Vaughn, S., Linan-Thompson, S. Pollard-Durodola, S., Ma thes, P. G., & Hagan, E. (2006). Effective interventions for Eng lish language learners (Span ish-English) at risk for reading difficulties. In D. Dickinson & S. Neuman (Eds.), Handbook of early literacy research (pp. 185197). New York, NY: The Guilford Press. Vaughn, S., Mathes, P. G., Linan-Thompson, S., Cirino, P., Carlson, C., PollardDurodola, S.,Francis, D. (2006). Effectiveness of an English intervention for first grade English lan guage learners at risk fo r reading problems. The Elementary School Journal, 107 (2), 153-180. doi:10.1086/510653

PAGE 272

272 Vellutino, F. R., Scanlon, D. M., Sipay, E. R., Small, S., Pr att, A., Chen, R., & Denckla, M. B. (1996). Cognitive profiles of diffi cult-to-remediate and readily remediated poor readers: Early intervention as a vehi cle for distinguishing between cognitive and experiential deficits as basic causes of specific reading disabilities. Journal of Educational Psychology, 88 (4), 601-638. doi:10. 1037/0022-0663.88.4.601 Velluntino, F. R., Scanlon, D. M., & Tanzman, M. (1994). Components of r eading ability: Issues and problems in operationalizing wo rd identification, phonological coding, and orthographic coding. In G. R. Lyon (Ed.), Frames of reference for the assessment of learning disabilitie s: New views on measurement issues (pp. 279324). Baltimore, MA: Brookes Publishing. Verhoeven, L. T. (1990). Acquisition of reading in a second language. Reading Research Quarterly, 25 (2), 90-114. Verhoeven, L. T. (2000). Components in ear ly second language reading and spelling. Scientific Studies of Reading, 4 (4), 313-330. doi:10.120 7/S1532799XSSR0404_4 Wade-Woolley, L., & Siegel, L. S. (1997). The spelling pe rformance of ESL and native speakers of English as a function of reading skill. Reading & Writing, 9 (5), 387406. doi:10.1023/A:1007994928645 Wagner, R. K., Torgesen, J. K., & Rasho tte, C. A. (1994). De velopment of reading related phonological processing abilities: New evidence of bidirectional causality from a latent variable longitudinal study. Developmental Psychology, 30 (1), 7387. doi:10.1037/0012-1649.30.1.73 Wanzek, J., & Vaughn, S. (2008). Response to varying amounts of time in reading interventions for students with low response to intervention. Journal of Learning Disabilities, 41(2), 126-142. doi :10.1177/0022219407313426 Watts-Taffe, S., & Truscott, D. M. (2000). Using what we know about language and literacy development for ESL students in the mainstream classroom. Language Arts, 77 (3), 258-265. Whal, M. (2007). What is Read Well? Florida Center for Reading Research. Retrieved November 23rd, 2008. from http://www.fcrr.org/fcrrrepor ts/PDF/Read_W ell_Report.pdf What Works Clearinghouse. (2007). Intervention: Reading Recovery. Retrieved November 20th, 2008 from http://ies.ed.gov/ncee/wwc/reports/ beginning_reading/reading_recovery/ Whitehurst, G. & Lonigan, C. (2002). Emergent literacy: Development from prereaders to readers. In S. Neuman & D. Dickson (Eds.), Handbook of early literacy research (pp. 11-29). New York, NY : The Guilford Press.

PAGE 273

273 Wolf, M. M. (1978). Social validity: The case for subjective measurement on how applied behavior analysis is finding its heart. Journal of Applied Behavior Analysis, 11 (2), 203-214. doi: 10.1901/jaba.1978.11-203 Woodock, R. W., Muoz-Sandoval, A. F., R uef, M. L., & Alvar ado, C. G. (2005). WoodcockMuoz Language Survey Revised, English. Itasca, IL: Riverside Publishing. Woodock, R. W., Muoz-Sandoval, A. F., R uef, M. L., & Alvar ado, C. G. (2005). Woodcock-Muoz Language Survey Revised, Spanish. Itasca, IL: Riverside Publishing.

PAGE 274

274 BIOGRAPHICAL SKETCH Stephanie Arriaza de Allen was born and ra ised in Guatemala city, Guatemala in 1975. She attended a bilingual k-12 school a nd graduated with an English/Spanish High School degree in 1992. She earned her bachel ors degree in psychology from the Universidad del Valle de Guatemala in 2000. She came to the United States in 2000 as an international student and completed one year of English as a second language courses at the University of Florida and Santa Fe College in order to prepare for her graduate studies. In 2001, she was accepted into the Ma ster of Education and Educat ional Specialist degree in counselor ed ucation at the University of Florida. She completed the program in 2004 with a focus on school counse ling. Towards the end of the program, she completed three field experiences in elementary schools where she worked with students who were experienci ng emotional and behavioral diffi culties. Many of these students were English language learners. As she conducted individual and small-group counseling, she realized that many of their issues stemm ed from academic difficulties, particularly in the area of reading. An inte rest in learning about struggling readers and effective ways to serve them developed from these observations. She decided to pursue a masters degree in Special Education at the University of Florida to learn how to help students become successful readers. At the time, her plan was to return to Guatemala and serve the specia l education community in that country. As the masters program progressed and she learned about effe ctive reading intervention for struggling readers and the particular needs of English l anguage learners, her passion for helping grew stronger and stronger. Toward the end of the masters progr am, she decided to

PAGE 275

275 apply to the Special Education do ctoral program at the University of Florida. In 2006, she was accepted and awarded an alumni fe llowship to complete her studies. During the four year program, she focused on literacy development, reading disabilities, early interventions, and Englis h language learners. While completing her course work, she had the opportunity to co-teach and teach undergraduate and graduate level courses in liter acy, reading disabilities, bi lingual special education, and family involvement. She worked in research projects related to literacy and teacher education. She has published an article in a peer reviewed journal. She has been a reviewer and a presenter in several state and national conferences. In 2007, she was awarded second place for best paper present ed at the Language and Globalization: Policy, Education, and Media Conference at Georgetown Universi ty. Other awards received include the Outstanding Internati onal Student Award for the College of Education, as well as certific ates for excellence in academi c achievement offered by the International Student Center at the University of Florida. In 2010, she graduated with a Ph.D. in Specia l Education. Her professional goals include teaching at the university level and becoming a national and international consultant for the prevention of reading di fficulties and the promotion of effective reading interventions for all struggling readers. She also plans to open a tutoring center for underprivileged children who are in need of reading support.