<%BANNER%>

The Relationship between Middle Level Grade Configuration and Model Practices on Student Achievement in Urban School Dis...

Permanent Link: http://ufdc.ufl.edu/UFE0043561/00001

Material Information

Title: The Relationship between Middle Level Grade Configuration and Model Practices on Student Achievement in Urban School Districts
Physical Description: 1 online resource (202 p.)
Language: english
Creator: Kriznar, Dana L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: achievement -- advisory -- configuration -- education -- exploratory -- heterogeneous -- interdisciplinary -- k-8 -- looping -- middle -- planning -- school -- student -- teaming -- urban
Human Development and Organizational Studies in Education -- Dissertations, Academic -- UF
Genre: Educational Leadership thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Before investing a large amount of human and financial resources in grade configuration redesign, urban school administrators need to understand whether it is the large structural changes, like grade configuration redesign, or whether it is the practices implemented within those larger structures that make the biggest impact on student achievement. In this study, 32 K-8 configured schools in the southeast United States are paired with 32 6-8 configured schools with comparable SES and minority rates within the same urban school district. This research uses the results from 192 educator surveys to determine the implementation level of established model practices, including interdisciplinary teaming, common teacher planning, heterogeneous grouping, advisory periods, exploratory courses, and looping. In addition, the reading and math achievement gain scores from the Florida Comprehensive Test of 12,727 sixth grade students are used to determine if there is a relationship between any of the practices, the grade level configuration of the school and student achievement in an urban setting. There were differences in student achievement by grade configuration, with the students in K-8 schools making significantly greater gains in both reading and math; however, that increase may well be due to other factors such as grade level size or socioeconomic differences rather than the configuration of the school. Of the model practices examined, interdisciplinary teaming was the only model practice that had a significant difference in level of implementation, which was higher in the K-8 configured schools. When achievement gains were compared to the level of implementation of each of the six model practices, no significant relationships were found. Finally, there was no difference in combined effects of grade configuration and implementation of model level practices with gains in student achievement in reading or mathematics.
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 Dana L Kriznar.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Oliver, Bernard.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043561:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043561/00001

Material Information

Title: The Relationship between Middle Level Grade Configuration and Model Practices on Student Achievement in Urban School Districts
Physical Description: 1 online resource (202 p.)
Language: english
Creator: Kriznar, Dana L
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: achievement -- advisory -- configuration -- education -- exploratory -- heterogeneous -- interdisciplinary -- k-8 -- looping -- middle -- planning -- school -- student -- teaming -- urban
Human Development and Organizational Studies in Education -- Dissertations, Academic -- UF
Genre: Educational Leadership thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Before investing a large amount of human and financial resources in grade configuration redesign, urban school administrators need to understand whether it is the large structural changes, like grade configuration redesign, or whether it is the practices implemented within those larger structures that make the biggest impact on student achievement. In this study, 32 K-8 configured schools in the southeast United States are paired with 32 6-8 configured schools with comparable SES and minority rates within the same urban school district. This research uses the results from 192 educator surveys to determine the implementation level of established model practices, including interdisciplinary teaming, common teacher planning, heterogeneous grouping, advisory periods, exploratory courses, and looping. In addition, the reading and math achievement gain scores from the Florida Comprehensive Test of 12,727 sixth grade students are used to determine if there is a relationship between any of the practices, the grade level configuration of the school and student achievement in an urban setting. There were differences in student achievement by grade configuration, with the students in K-8 schools making significantly greater gains in both reading and math; however, that increase may well be due to other factors such as grade level size or socioeconomic differences rather than the configuration of the school. Of the model practices examined, interdisciplinary teaming was the only model practice that had a significant difference in level of implementation, which was higher in the K-8 configured schools. When achievement gains were compared to the level of implementation of each of the six model practices, no significant relationships were found. Finally, there was no difference in combined effects of grade configuration and implementation of model level practices with gains in student achievement in reading or mathematics.
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 Dana L Kriznar.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Oliver, Bernard.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043561:00001


This item has the following downloads:


Full Text

PAGE 1

1 THE RELATIONSHIP BETWEEN MIDDLE LEVEL GRADE CONFIGURATION AND MODEL PRACTICES ON STUDENT ACHIEVEMENT IN URBAN SCHOOL DISTRICTS By DANA L. KRIZNAR A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Dana Kriznar

PAGE 3

3 This d o cument is dedicated to my parents, George Schuchmann and Maureen Rodas, for providing me with a solid fou ndation t o my husband, Brad, for his 21 year s of steadfast love and support and t o my children, Austin and Tyler, for the fine young men they have become.

PAGE 4

4 ACKNOWLEDGMENTS I would like to offer my sincere thanks and appreciation to my committee chair Dr. Bernard Oliver for his advice and encouragement through this process. In addition, I would like to thank the other members of my committee, Dr. Jean Crockett, Dr. Eileen Oliver, and Dr. Linda Eldridge, for their time and guidance. A thank you is also extended to Dr. David Quinn and Dr. Jim Doud, who were a constant source of assistance throughout the program. I would also like to express my gratitude to the Instructional Research and Evaluation personnel within each of the districts s tudied for this p roject. Special thanks are owed to Tim Ballentine, Nina Briggs, Glenna Goings, Christina Knopf, Vickie Cartwright, Tarek Chebbi, and Ted Dwyer for their support. Thank you to my Duval and St. Johns cohort of peers who helped me through every step of this p rocess. Special thanks to Denise Hall, Myrna Allen, Lissa Dunn, Ed Pratt Dannals, Lori Turner, and Royce Turner. Your encouragement, assistance, and faith in me truly made the difference. To my coworkers and friends especially Dr. Terri Stahlman and Glori a Lockley, who understood when I was absent or preoccupied during the last few months, I appreciate you more than you know. Special thanks are owed to my parents Maureen and Lowell Rodas and Martha and the late George Schuchma nn, for their love and guida nce. They instilled in me at an early age the value and importance of education. I would also like to acknowledge my in laws, Mary Jean and the late Frank Kriznar, for all of their help and support throughout this program. Finally, words can not express my love and gratitude to my husband, Brad, and sons Austin and Tyler. The sacrifices they willingly made in order for me to accomplish this goal are too numerous to count.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF DEFINITIONS ................................ ................................ ................................ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 Background Information ................................ ................................ .......................... 15 Statement of the Problem ................................ ................................ ....................... 19 Purpose of the Study ................................ ................................ .............................. 21 Research Questions ................................ ................................ ......................... 22 Hypotheses ................................ ................................ ................................ ...... 23 Limitations ................................ ................................ ................................ ........ 27 Delimitations ................................ ................................ ................................ ..... 28 Assumptions ................................ ................................ ................................ ..... 28 2 A REVIEW OF THE LITERATURE ................................ ................................ ......... 30 A Historical Perspective of Grade Configuration for Adolescent Students .............. 32 Variations of Middle Level Grade Configurations ................................ .................... 35 Characteristics of Model Middle Schools ................................ ................................ 40 Interdisciplinary Teaming ................................ ................................ .................. 41 Common Teacher Planning Time ................................ ................................ ..... 45 Heterogeneous Grouping ................................ ................................ ................. 46 Student Advisory ................................ ................................ .............................. 48 Exploratory Courses ................................ ................................ ......................... 52 Looping ................................ ................................ ................................ ............. 54 Combined Effects of Developmentally Responsive Practices .......................... 56 Implementation of Model Practices by Grade Configuration ............................. 60 Eff ects of School to School Transitions ................................ ................................ .. 63 Characteristics of the K 8 Model for Middle Level Students ................................ .... 67 Student Achievement and Grade Co nfiguration ................................ ...................... 71 Grade Configuration Studies with No or Little Effects ................................ ....... 73 Grade Configuration Studies with Some Effects ................................ ............... 76 Summary of Research Findings ................................ ................................ ....... 80 Other Student Achievement Factors for Early Adolescents ................................ .... 81 Socioeconomic Factors ................................ ................................ .................... 81

PAGE 6

6 School Size ................................ ................................ ................................ ...... 82 Academic Rigor ................................ ................................ ................................ 84 Extracurri cular Activities ................................ ................................ ................... 86 Grade Configuration and Other School Factors ................................ ...................... 87 Student Behavior ................................ ................................ .............................. 87 Self Esteem ................................ ................................ ................................ ...... 89 Parent Involvement ................................ ................................ .......................... 92 Attendance ................................ ................................ ................................ ....... 93 Middle Level Teachers ................................ ................................ ..................... 93 Summary ................................ ................................ ................................ ................ 97 3 RESEARCH DESIGN AND METHODOLOGY ................................ ..................... 100 Resear ch Proposal ................................ ................................ ............................... 101 Research Questions ................................ ................................ ....................... 101 Hypotheses ................................ ................................ ................................ .... 102 Limitations ................................ ................................ ................................ ...... 106 Delimitations ................................ ................................ ................................ ... 107 Assumptions ................................ ................................ ................................ ... 107 Data Sources ................................ ................................ ................................ ........ 107 Population and Sample ................................ ................................ .................. 107 Data Collection and Instrumentation ................................ .............................. 112 Data Analysis ................................ ................................ ................................ 113 4 ANALYSIS OF DATA ................................ ................................ ............................ 117 Presentation of the Data ................................ ................................ ....................... 119 Descriptive Data ................................ ................................ ............................. 119 Statistical Compilation ................................ ................................ .................... 121 Analysis of Research Questions ................................ ................................ ........... 126 Research Question 1 ................................ ................................ ...................... 126 Research Question 2 ................................ ................................ ...................... 136 Research Question 3 ................................ ................................ ...................... 140 Research Question 4 ................................ ................................ ...................... 145 Summary ................................ ................................ ................................ .............. 151 5 CONCLUSIONS ................................ ................................ ................................ ... 153 Summary of Results ................................ ................................ .............................. 154 Discussion ................................ ................................ ................................ ............ 159 Grade Level Configuration ................................ ................................ ....... 160 Exemplary Middle Level Practices ................................ ........................... 162 Limitations of the Research ................................ ................................ .................. 164 Implications ................................ ................................ ................................ ........... 166 Implications for Students ................................ ................................ .......... 166 Implications for Schools ................................ ................................ ........... 167 Implications for School Districts ................................ ............................... 168

PAGE 7

7 Summary ................................ ................................ ................................ .............. 171 APPENDIX A UNIVERSITY OF FLORIDA IRB CONSENT ................................ ........................ 172 B DISTRICT CONSENTS TO CONDUCT RESEARCH ................................ ........... 173 C MIDDLE LEVEL EDUCATOR SURVEY PARTICIPANT CONSENT FORM ......... 178 D MIDDLE LEVEL EDUCATOR SURVEY ................................ ............................... 179 LIST OF REFERENCES ................................ ................................ ............................. 183 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 202

PAGE 8

8 LIST OF TABLES Table page 3 1 List of instructionally responsi ve practices for middle level learners ................. 11 6 4 1 Number of students excluded from study by demographic characteristics ....... 120 4 2 Aggregate demographic data for student population by grade configuration and descriptive demographic data ................................ ................................ .... 121 4 3 Number of student participants by grade configuration and gender, race/ ethnicity, limited English proficient, free or reduced price lunch rates and exceptionality ................................ ................................ ................................ .... 122 4 4 Mean gain scores and standard deviations for dependent variables by demographic factors ................................ ................................ ......................... 123 4 5 Educator survey participants by school configuration and demographic information ................................ ................................ ................................ ........ 124 4 6 Nonparametric correlations of reading and math gain scores wit h model practices ................................ ................................ ................................ ........... 127 4 7 Sig nificance of interdisciplinary teaming and student achievement gains with descriptive characteristic tested through linear regression ............................... 129 4 8 Sig nificance of common teacher planning and student achievement gains with descriptive characteristic tested through linear regression ........................ 130 4 9 Sig nificance of heterogeneous student grouping and student achievement gains with descriptive characteristic tested through General Linear Model procedure ................................ ................................ ................................ ......... 131 4 10 Sig nificance of advisory periods and student a chievement gains with descriptive characteristic tested through General Linear Model procedure ...... 132 4 11 Sig nificance of exploratory courses and student achievement gains with descriptive characte ristic tested through General Linear Model procedure ...... 134 4 12 Sig nificance of looping and student achievement gains with descriptive characteristic tested through General Linear Model procedure ........................ 135 4 13 Group statistics of exemplary middle practices by grade configuration ............ 137 4 14 Comparison of means of implementation levels of model practices by grade configuration ................................ ................................ ................................ ..... 137

PAGE 9

9 4 15 Group means and standard deviations of sixth grade FCAT reading gain scores from 2010 to 2011 by school grade configuration ................................ 141 4 16 Independent Samples test for sixth grade FCAT reading gain scores from 2010 to 2011 by school grade configuration ................................ ..................... 141 4 17 Linear regression of rea ding gains grouped by school configuration and using school factors as predictors ................................ ................................ .............. 142 4 18 Group means and standard deviations of sixth grade FCAT math gain scores from 2010 to 2011 by school gra de configuration ................................ ............. 143 4 19 Independent Samples test for sixth grade FCAT math gain scores from 2010 to 2011 by school grade configuration ................................ .............................. 143 4 20 Linear regression for math gains grouped by school configuration and using school factors as predictors ................................ ................................ .............. 144 4 21 Reading achievement gain scores and solutions for fixed effects .................... 148 4 22 Math achievement and solutions for fixed effects ................................ ............. 149 4 23 Type III Tests of Fixed Effects for reading achievement gain scores. ............... 150 4 24 Type III Tests of Fixed Effects for math achievement gain scores. ................... 151 5 1 Summary of findings for hypothesis 1 ................................ ............................... 156 5 2 Summa ry of findings for hypothesis 2 ................................ ............................... 157 5 3 Summa ry of findings for hypothesis 3 ................................ ............................... 158 5 4 Summa ry of findings for hypothesis 4 ................................ ............................... 159

PAGE 10

10 LIST OF DEFINITIONS Association of Middle Level Educators (AMLE) Formerly called the National Middle School Association, this organization is dedicated to the education of e arly adolescents. T he organization chang ed its name in 2011 to be more inclusive of the variety of grade spans in which the students are educated. Charter Schools Charter schools are independent public schools created to provide parents with additional choice. These schools are given more flexibility to innovate and offer specialized programs, while still maintaining accountability under the state guidelines. Co curricular Program Programs that provide opportunities outside of the regular graded cours ework that provide leadership roles, exploration in special interests, and socialization for students. Common Planning Teams of teachers with a common group of students are assigned a block of time to plan together on a daily basis. Developmental Scale Score (DSS) A score created from reported scale scores to help parents and educators understand the year to year progress of a student. Exploratory Courses Courses that provide required curricular opportunities for students to explore abilities, talents and interests through diverse elective opportunities. Florida Comprehensive Achievement Test (FCAT) A criterion referenced achievement test used as part of student, school, district, and state pro gress administered yearly to students in grades three through eleven. Free or Reduced Price Lunch Rate The rate of students in a school that qualify for a free or reduced price lunch, which is often used as an indicator of poverty. Grade Level Configur ations An organizational pattern of grade levels contained within a school. The schools used in this study were limited to sixth through eighth and kindergarten (or pre kindergarten) through eighth grade level configurations. K 8 School A school that con tains students enrolled in kindergarten (or pre kindergarten) through eighth grade. Heterogeneous Grouping Scheduling students of mixed abilities into courses together in order to create a relatively even distribution of students in each classroom.

PAGE 11

11 Int erdisciplinary Teaming Teams of 2 to 5 teachers that share a common group of students, common planning time, and are located in close proximity to each other. Looping Also called student teacher progression or multi year teaching, looping is the practice s of keeping a small group of teachers together with the same group of students for two or more years. Magnet Schools A public school that offers a specialized curriculum or program within district and state guidelines typically used to attract substanti al numbers of students from diverse backgrounds. Middle School A school between elementary school and high school, typically with students in grades six, seven, and eight. Minority Rate The term used in the state of Florida to describe the percent of students identified as non white within a school or school district. Sunshine State Standards (SSS) Developed as part of the Florida accountability system, these standards are broad statements regarding what students at each grade level should know and b e able to do. Transition The process of moving from one level of schooling to the next level at another school site.

PAGE 12

12 LIST OF ABBREVIATION S AMLE Association of Middle Level Educators Next word DSS Developmental Scale Score ESL English as a Second La nguage FCAT Florida Comprehensive Assessment Test FLDOE Florida Department of Education LEP Limited English Proficient NAESP National Association of Elementary School Principals NASSP National Association of Secondary School Principals NELS National Educat ional Longitudinal Study NMSA National Middle School Association OPPAGA Office of Program Policy Analysis and Government Accountability SES Socio Economic Status SSS Sunshine State Standards

PAGE 13

13 Abstract of Dissertation Presented to the Graduate School of t he University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE RELATIONSHIP BETWEEN MIDDLE LEVEL GRADE CONFIGURATION AND MODEL PRACTICES ON STUDENT ACHIEVEMENT IN URBAN SCHOOL DISTRICTS By D ana L. Kriznar December 2011 Chair: Bernard E. Oliver Major: Educational Leadership Before investing a large amount of human and financial resources in grade configuration redesign, urban school administrators need to understand whether it is the large structural changes, like grade configuration redesign, or whether it is the practices implemented within those larger structures that make the biggest impact on student achievement. In this study, 32 K 8 configured schools in the southeast United States a r e paired with 32 6 8 configured schools with comparable SES and minority rates within the same urban school district. This research uses the results from 192 educator surveys to determine the implementation level of established model practices, including i nterdisciplinary teaming, common teacher planning, heterogeneous grouping, advisory periods, exploratory courses, and looping. In addition, the reading and math achievement gain scores from the Florida Comprehensive Test of 12,727 sixth grade students a re used to determine if there is a relationship between any of the practices the grade level configuration of the school and student a chievement in an urban setting. There were differences in student achievement by grade configuration, with the students in K 8 schools making significantly greater gains in both reading and math; however, that increase may well be due to other factors such as grade level size or

PAGE 14

14 socioeconomic differences rather than the configuration of the school. Of the model practices examin ed, interdisciplinary teaming was the only model practice that had a significant difference in level of implementation, which was higher in the K 8 configured schools. When achievement gains were compared to the level of implementation of each of the six m odel practices, no significant relationships were found Finally, there was no difference in combined effects of grade configuration and implementation of model level practices with gains in student achievement in reading or mathematics.

PAGE 15

15 CHAPTER 1 INTROD UCTION Background Information With the nationwide downturn in the economy, every sector of society seems to be learning to do more with less. Because of their dependence on state and federal revenue, public schools are not exempt from tightening budgets (M cNeil, 2009). Even though school coffers have been greatly diminished, federal and state mandates for student performance have not. Perhaps now more than ever, public school administrators are scrutinizing every aspect of their organizations to determine t hose practices that provide the best results for the least amount of money. One trend in educational reform that has gained attention recently is how to best configure grade spans in schools to maximize both efficiency and student performance (Craig, 2006; Epstein, 1990; Howley, 2002). Ten years ago, a one year dip in student academic achievement for middle level students was not perceived as a monumental loss for most school administrators. The decline could be attributed to a variety of factors, including the transition to a new school, a change in socialization, or the onset of puberty (Alspaugh, 1995; Alspaugh, school leaders are scrambling to maximize student learni ng gains at every level (United States Department of Education, 2008). In addition, most states have developed accountability systems that result in status and monetary rewards for gains in achievement and sanctions for achievement losses (United States De partment of Education, 2009). Any dip in achievement at any grade level can mean a loss in state

PAGE 16

16 and federal funds, a decline in student enrollment, the stigma of being labeled as a failing school, and even threats of reconstitution or school closure. Scho ol districts across America have used a variety of grade configurations throughout our history. When free public education was first introduced in this country, one room schoolhouses servicing all students in grades one through eight were the norm (Manning 2000; Sailor, 1986). The junior high school movement began in the 8 and 9 12 configuration model wasted the middle years that could be better spent preparing youth for postsecondary school. Th e middle school movement with a sixth through eighth grad e configuration began in the 197 0s, growing out of a concern from educators and researchers that junior high schools were not meeting the developmental and educational needs of early adolescents (Geo rge & Alexander, 1993). By the 1990s, most students in grade six were attending a middle school with a 5 8 or 6 8 grade configuration. Most of the remaining K 8 configured schools were either private or parochial institutions, or located in more rural area s of the country (Craig, 2006; Eccles et al., 1991; Howley, 2002; Manning, 2000). Founded in 1973, the National Middle School Association (NMSA) was developed to bring together educators who were dedicated to the implementation of research based, developme ntally appropriate practices designed to improve the outcomes of middle level learners (NMSA, 1982). The middle school design was developed out of a concern that middle level students needed an instructional program that was more developmentally responsive to the unique needs of early adolescents than was offered in our elementary and junior high schools (Alexander & McEwin,

PAGE 17

17 1989b; George & Alexander, 1993; Manning, 2000; NMSA, 1982). At the time, the design had more to do with the instructional program and design structure than grade span, although there was considerable debate about the most developmentally appropriate setting for fifth and sixth grade students, in particular. Through research, NMSA identified a number of exemplary practices designed to m eet the needs of these students (1982). They included the following: ( a) heterogeneous grouping, ( b) common planning for teachers, ( c) student teacher progression or looping, ( d) advisory programs, ( e) exploratory courses, and ( f) interdisciplinary teaming with students arranged in smaller learning communities within the school. Historically, research has shown that middle schools are more likely to implement these exemplary practices than other grade configurations, leading to more frequent progress monit oring and assistance, better instructional practices, and more exploratory opportunities for students (Alexander & McEwin, 1989b; Epstein & MacIver, 1990; Irvin, Valentine, & Clark, 1994; Jenkins & McEwin, 1992; MacIver & Epstein, 1993, McEwin & Alexander, 1990) More recently, however, larger school districts around the country are choosing to convert a portion of their large middle schools to neighborhood K 8 configured schools (Look, 2002a; Office of Program Policy Analysis and Government Accountability [ OPPAGA], 2005; Pardini, 2002; Patton, 2005). School districts have high hopes that the conversion will result in improved student discipline, higher attendance rates, more parental involvement, and greater student achievement. Since public K 8 schools are to support the changeover. In part because of the rising number of schools making the

PAGE 18

18 shift to a K 8 configuration NMSA changed its name in 2011 to the Association of Middle Le vel Educators (AMLE) to better reflect the common commitment to educating early adolescents, regardless of the configuration of the school (AMLE, 2011). Most of the earlier academic research centering on grade configuration focuses on rural areas, which m ay not translate to the desired results in urban areas of the country. Studies have been conducted with mixed results in many rural areas of the country, including Tennessee, Louisiana, Missouri, Connecticut and Maine (Alspaugh, 1995; Blair, 2007; Coladarc i & Hancock, 2002b; Franklin & Glascock, 1998; Johnson, D. J., 2002; Tucker & Andrada, 1997; Wihry, Coladarci, & Meadow, 1992). Several of these studies have provided evidence that academic achievement levels in reading, math, and science for students in g rades 6 through 8 attending a K 8 school may be higher than their middle school peers, but the researchers also cite other factors that may contribute to the increased scores, such as school size and smaller numbers of students in each grade level within t he school. Some of the most current research testing the K 8 model in urban areas has been conducted by school districts themselves. Cleveland, Cincinnati, Philadelphia, Miami, and Baltimore public school systems have all undertaken studies of the ir respe ctive K 8 redesign s (Abella, 2005; Balfanz, Spiridakis, & Neild, 2002; Baltimore City Public School System Division of Research, Evaluation and Accountability, 2001; Pardini, 2002). These districts cite improved discipline, attendance, and student achievem ent for middle level learners in the K 8 design. Much of this internal research was conducted by district staff in their respective school systems or by contracted researchers in an effort to find supporting evidence for costly school reconfigurations.

PAGE 19

19 Bec ause much of the research was conducted internally, external validation would add to the credibility of the findings. There is a growing body of research that indicates that transitions from one level of schooling to the next causes stress in children, wh ich could lead to a fragmentation in their learning experiences (Blyth, Simmons, & Bush, 1978; Paglin & Fager, 1997; Toepfer, 1990a). Both Wise (2000) and Alspaugh and Harting (1995) found student achievement losses the year following a transition regardle ss of the grade in which the transition occurred. Researchers have also found a correlation between the number of transitions a student is required to make and dropout rates (Alspaugh, 1995). Simmons and Blyth (1987) found that an increased number of trans itions has a negative effect on self esteem, which is most pronounced for adolescent girls. Toepfer (1990b) suggested that the effects of transitioning can be lessened by responsive practices within the school, such as orientations, summer programs, and te aming. Statement of the Problem Before investing a large amount of human and financial resources in grade configuration redesign, urban school administrators need to understand whether it is the large structural changes, such as grade configuration redesig n, or the practices implemented within those larger structures that make the biggest impact on student achievement. Middle school advocates argue that the developmentally appropriate practices implemented as part of the middle school design are most import ant to the success of early adolescents (Ecker, 2002; George & Shewey, 1994; Irvin et al., 1994). On the other hand, some researchers are proposing that the elimination of a transition from one school to the next using a K 8 grade configuration will contri bute more to the

PAGE 20

20 success of middle level learners (Alspaugh, 1995, 1998; Eccles et al., 1991; Offenberg, 2001). This research compares the level of implementation of middle level exemplary practices established by the NMSA/AMLE (NMSA, 1982; NMSA, 1995) to determine if there is a relationship between any of the practices and student achievement for students in an urban setting. In addition, the research will compare the level of implementation of the same best practices to determine whether the degree of im plementation varies by grade configuration of the school. Next, the study will compare academic achievement for students in a K 8 configured school with those in a 6 8 design. Finally, the research will study the combined effects of grade configuration and model practice implementation on academic achievement gains for sixth grade students in urban districts. The study will use both educator questionnaires to determine the level of implementation of best practices, as well as existing student achievement da ta in both reading and mathematics from the state accountability test s. The Florida Comprehensive Assessment Test (FCAT) is a criterion referenced test that measures student academic achievement in reading and mathematics on a yearly basis for students in grades three through ten ( Florida Department of Education, 20 11b). Although student scores are expressed as proficiency levels from one to five for general reporting, the students also receive a Developmental Scale Score (DSS), which is used to track stud ent progress from year to year. In addition to determining individual student performance, the data are also used as high stakes measures of school and district progress under the state and federal accountability programs. Publicly funded schools within th e state of Florida can also receive sanctions or financial rewards

PAGE 21

21 depending on the collective achievement levels and progress of the students within the school. The study also contains a survey component developed by the National Association of Secondary School Principals (NASSP) to determine the level of implementation of those exemplary practices identified by the NMSA/AMLE, including interdisciplinary teaming, common teacher planning, advisory programs, heterogeneous grouping, exploratory opportunities, and looping. The NASSP has sponsored two extensive national studies regarding successful middle level leadership and effective practices for middle level learners (Valentine, Clark, Hackmann, & Petzko, 2002; Valentine, Clark, Hackmann, & Petzko, 2004). In preparation for the 2002 study, the research team developed a lengthy survey instrument containing 170 items. The survey is divided into sections, including school and principal demographics, school leadership, school programs, school reform and professio nal development, instructional practices, curricular and co curricular programs, school organization, and leadership issues. Over 14,000 middle level principals participated in the 2002 study. Since this research focuses mainly on the implementation of eff ective middle level practices, only the portions of the survey corresponding to those practices were used. Purpose of the Study The purpose of this study is to compare, when controlling for school and individual student demographic factors the degree to w hich the implementation of exemplary practices within the two grade configurations were related to student achievement measures. In ad dition, the study also attempts to compare the level of implementation of the exemplary practices to determine which setti ng is most likely to fully implement the practices. The study further compares student achievement in

PAGE 22

22 reading and mathematics for students in a K 8 grade configuration with that of students in a 6 8 middle school model. Finally, the study examines the comb ine d effects of model practices and grade configuration on academic achievement levels in reading and mathematics for sixth grade students enrolled in K 8 urban schools t o those of students who attend a traditional 6 8 middle school in urban school distric ts. Research Questions Question 1 : Is there a relationship between any of the exemplary practices identified by NMSA /AMLE, including ( a) interdisciplinary teaming, ( b) common teacher planning time, ( c) heterogeneous grouping, ( d) student advisory, ( e) expl oratory courses, and ( f) looping, and student achievement as measured by student gains on the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students in K 8 and 6 8 middle schools in lar ge urban school districts when controlling for school and student demographic factors ? Question 2: Are there differences in levels of implementation of NMSA/AMLE identified exemplary practices in urban K 8 configured schools when compared to 6 8 configure d middle schools? Question 3: Are there differences in academic achievement in reading and mathematics for sixth grade public school students in K 8 configured schools when compared to sixth grade students in 6 8 configured middle schools, as measured by t he Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system? Question 4: Is there a relationship between the level of implementation of exemplary middle level practices and academic achievement as measured by the

PAGE 23

23 Flo rida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students enrolled in urban K 8 configured schools compared to similar students in 6 8 middle schools? Hypotheses The following null hypotheses gui ded the investigation: Ho 1 : There is no relationship between the implementation of exemplary middle level practices and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in urban school district s when controlling for school demographic factors of school size, grade level size, minority rate, SES rate, LEP rate, and ESE rate Ho 1 1: Interdisciplinary Teaming Ho 1 1 a : Reading Achievement Ho 1 1 b : Mathematics Achievement Ho 1 2: Common Teacher Planning Time H o 1 2 a : Reading Achievement Ho 1 2 b : Mathematics Achievement Ho 1 3: Heterogeneous Grouping Ho 1 3 a : Reading Achievement Ho 1 3 b : Mathematics Achievement Ho 1 4: Student Advisory Ho 1 4 a : Reading Achievement Ho 1 4 b : Mathematics Achievement Ho 1 5: Exploratory Cour ses Ho 1 5 a : Reading Achievement

PAGE 24

24 Ho 1 5 b : Mathematics Achievement Ho 1 6: Looping Ho 1 6 a : Reading Achievement Ho 1 6 b : Mathematics Achievement Ho 2 : There are no significant differences in the implementation of exemplary middle level practices in schools confi gured as K 8 compared to traditional middle schools with a 6 8 grade configuration Ho 2 1: Interdisciplinary Teaming Ho 2 1 a : 6 8 Grade Configuration Ho 2 1 b : K 8 Grade Configuration Ho 2 2: Common Teacher Planning Time Ho 2 2 a : 6 8 Grade Configuration Ho 2 2 b : K 8 Grade Configuration Ho 2 3: Heterogeneous Grouping Ho 2 3 a : 6 8 Grade Configuration Ho 2 3 b : K 8 Grade Configuration Ho 2 4: Student Advisory Ho 2 4 a : 6 8 Grade Configuration Ho 2 4 b : K 8 Grade Configuration Ho 2 5: Exploratory Courses Ho 2 5 a : 6 8 Grade Configu ration Ho 2 5 b : K 8 Grade Configuration Ho 2 6: Looping

PAGE 25

25 Ho 2 6 a : 6 8 Grade Configuration Ho 2 6 b : K 8 Grade Configuration Ho 3 : There are no significant differences in student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grad e students in a K 8 school compared to the same students in a traditional 6 8 middle school. Ho 3 1: Reading Achievement Ho 3 1 a : 6 8 Grade Configuration Ho 3 1 b : K 8 Grade Configuration Ho 3 2: Mathematics Achievement Ho 3 2 a : 6 8 Grade Configuration Ho 3 2 b : K 8 Grade Configuration Ho 4 : There is no relationship between exemplary middle school practices and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students who attend K 8 schools implementing exemplary middle l evel practices compared to the same students attend traditional middle schools implementing the practices at the same level Ho 4 1: Interdisciplinary Teaming Ho 4 1 a : 6 8 Grade Configuration Ho 4 1 a 1 : Reading Achievement Ho 4 1 a2 : Mathematics Achievement Ho 4 1 b : K 8 Grade Configuration

PAGE 26

26 Ho 4 1 b1 : Reading Achievement Ho 4 1 b2 : Mathematics Achievement Ho 4 2: Common Teacher Planning Time Ho 4 2 a : 6 8 Grade Configuration Ho 4 2 a 1 : Reading Achievement Ho 4 2 a2 : Mathematics Achievement Ho 4 2 b : K 8 Grade Configuratio n Ho 4 2 b1 : Reading Achievement Ho 4 2 b2 : Mathematics Achievement Ho 4 3: Heterogeneous Grouping Ho 4 3 a : 6 8 Grade Configuration Ho 4 3 a 1 : Reading Achievement Ho 4 3 a2 : Mathematics Achievement Ho 4 3 b : K 8 Grade Configuration Ho 4 3 b1 : Reading Achievement Ho 4 3 b2 : Mathematics Achievement Ho 4 4: Student Advisory Ho 4 4 a : 6 8 Grade Configuration Ho 4 4 a 1 : Reading Achievement Ho 4 4 a2 : Mathematics Achievement Ho 4 4 b : K 8 Grade Configuration Ho 4 4 b1 : Reading Achievement Ho 4 4 b2 : Mathematics Achievement

PAGE 27

27 Ho 4 5: Exploratory Courses Ho 4 5 a : 6 8 Grade Configuration Ho 4 5 a 1 : Reading Achievement Ho 4 5 a2 : Mathematics Achievement Ho 4 5 b : K 8 Grade Configuration Ho 4 5 b1 : Reading Achievement Ho 4 5 b2 : Mathematics Achievement Ho 4 6: Looping Ho 4 6 a : 6 8 Grade Configu ration Ho 4 6 a 1 : Reading Achievement Ho 4 6 a2 : Mathematics Achievement Ho 4 6 b : K 8 Grade Configuration Ho 4 6 b1 : Reading Achievement Ho 4 6 b2 : Mathematics Achievement Limitations There were a number of limitations to this study. These limitations are lis ted below. 1. The study was limited to the Florida Comprehensive Achievement SSS data in reading and mathematics reported to the state of Florida for the 2009 2010 and 2010 2011 school years. 2. Although the researcher recognizes that sixth graders can be educa ted in a variety of grade level configurations, this study was limited to two configurations: ( a) sixth grade students in schools with sixth through eighth grade, and ( b) pre kindergarten or kindergarten through eighth grade configurations. 3. The study was limited to a comparison of public K 8 and middle schools in urban districts larger than 100,000 students in the state of Florida

PAGE 28

28 participating in the state accountability system and may or may not be representative of other K 8 and middle schools in the Un ited States. 4. The extent to which the survey responses by educators in the respective schools represent the actual practices within the schools may limit the quality of the findings. 5. The study was limited to only those students who had achievement scores for both the 2009 2010 and the 2010 2011 school years. In addition, the study only included students in the K 8 design who had achievement data for the previous school year within the same K 8 school. Therefore, this study fails to account for the achievem ent of students who transfer to or from different school systems. Schools with high mobility rates would tend to confound the number of school to school transitions. Delimitations In addition to the limitations already presented, there were several delimi tations associated with this research study. Those delimitations are presented below. 1. The study is delimited to the data offered by the research and evaluation departments provided by the identified districts of the participating schools. 2. Only middle scho ols and K 8 schools identified by the Florida Department of Education ( FLDOE, 2011) were included in this study because an ample number of K 8 and comparable middle schools were available in that region, and all students took identical achievement tests. 3. K 8 and middle schools identified as charter schools or dedicated magnet or theme based schools were not included in this study. 4. Achievement scores were limited to gains from fifth to sixth grade only. Gain scores from students retained in sixth grade dur ing the 2009 2010 school year were not included in this study. 5. Information from educators within the school about implementation of practices was delimited to a single set of questions related to degree of implementation provided in an online or paper sur vey. Assumptions The study also makes two assumptions regarding this research. Those assumptions are the following:

PAGE 29

29 1. The study was based on the assumption that the student achievement data were appropriate measures for comparing academic achievement of mid dle schools and K 8 schools. 2. A further assumption was that comparable kind and quality of curriculum and instruction were present in both configurations of schools.

PAGE 30

30 CHAPTER 2 A REVIEW OF THE LITE RATURE One of the longest running debates in middle educ ation research concerns the ideal grade span for early adolescents (MacIver & Epstein, 1993). Conrad Toepfer (1990b), one of the earliest advocates of the middle school movement, states that ce to recapture millions of youth adrift. Yet all too often these schools exacerbate the problems of young high school education or less will not be able to make a liv able income off blue collar jobs (United States Department of Education, National Center for Education Statistics, 2010). Therefore, it is more important today than ever before that children receive an education that will help them become competitive in th e global marketplace of the twenty first century. Over the past four decades, student achievement has shown steady increase for middle level learners. Haycock and Ames (2000) reported that in the 1970s and 1980s, not only did middle school achievement sc ores rise in reading, mathematics, and science for eighth grade students, the racial student achievement gap narrowed significantly. However, that gap began to widen again in the early 1990s, with white students making steady progress, while the academic g rowth of their minority race peers remained stagnant. In 1995, achievement scores for all subgroups in eighth grade began improving; however, the racial disparity gap between subgroups remained the same. In addition, while eighth grade scores have improved over the past two decades, researchers have suggested that the value added scores have actually declined, and that the improvement in scores may have more to do with progress in fourth grade and

PAGE 31

31 below than with anything that is happening in the middle gra des (Haycock & Ames, 2003; Heller, Calderon, & Medrich, 2003). The implementation of the No Child Left Behind Act in 2001 caused public schools and school districts to reexamine all aspects of their school organization, including grade configurations, sche duling, grouping, and the length of the school day and school year (Lashway, 2000). There are many factors surrounding grade configuration that may contribute to conditions that support the performance of middle level students. These include the actual gro uping of grade levels within a school, the implementation of recommended practices for early adolescents, the number of transitions between schools that students are required to make, the size of the school and the number of students in each of the grade l evels within the school (Howley, 2002; Lucas & Valentine, 2005). Middle school advocates argue that many schools configured with a sixth through eighth grade span are not doing well because they have simply changed grade configurations, without altering th eir program to include exemplary middle level practices (Hough, 2005; Pardini, 2002). Little empirical research has been done on grade span and the ir effects on student achievement for middle level learners (Renchler, 2000). Most of the data available are local or regional and qualitative and anecdotal in nature (Pardini, 2002). Of the studies available, most compare and contrast achievement of students in schools with varied grade arrangements and do not consider fidelity to the implementation of recommen ded practices for middle level students (Offenberg, 2001). The evidence from existing studies suggests there is little clear middle level organizational configuration

PAGE 32

32 that is better than the others in terms of student achievement gains and implementation o f developmentally appropriate practices (Blair, 2007; Klump, 2006: Viadero, 2008). A Historical Perspective of Grade Configuration for Adolescent Students The debate surrounding grade configuration dates back to 1888, when Harvard President Charles Eliot stated that the current configuration of grades 1 8 and 9 12 was wasted time that could be better spent on preparing students for college (Manning, 2000; Sailor, 1986). In the early 1900s, colleges and universities proposed that schools develop a 1 6 and 7 12 model, moving seventh and eighth grade students to secondary schools in an effort to provide larger numbers of students with specialized training to enter college or a chosen vocation. The junior high school movement began in 1909 with the ope ning of Indianola Junior High School in Columbus, Ohio (Lounsbury & Vars, 2003, Manning, 2000). In 1913, the Committee on Economy of Time in Education presented a report proposing that the grade configuration be changed from a 6 6 organization to a 6 3 3 e lementary/junior high/high school model (Pate & Muth, 2003). Five years later, the Commission on the Reorganization of Secondary Education released a supporting document titled Cardinal Principles of Secondary Education which also advocated the 6 3 3 juni or high model (Lounsbury & Vars, 2003). Sailor (1986) reported that in 1920, only 10% of students starting first grade finished high school, and nearly 30% dropped out of school before completing the ninth grade. It was during this era that Koos (1920) wr ote the seminal book, The Junior High School which advocated the adoption of the junior high school model, spanning grades seven through nine, for many of the same reasons that the middle school reform was adopted decades later (Pate & Muth, 2003). There were three main imperatives that

PAGE 33

33 spawned the junior high movement: (a) more rigorous curricular offerings for college bound students, (b) increased vocational courses for those students who would be entering the workforce, and (c) more developmentally appr opriate instruction for early adolescents (Manning, 2000). Despite the recommendation of experts at the university level 80% of early adolescents were still attending K through 8 schools during this time period (Alexander & McEwin, 1989a). When roads imp roved after World War I, schools began to consolidate and increase in size, encompassing larger geographic areas (Howley, 2002). By 1945, the junior high model had become the majority model for use (Lounsbury & Vars, 2003; Strauss, 2005). By 1960, 80% of a ll graduates attended an elementary school, followed by a junior high and a three year high school (Alexander & McEwin, 1989a). Although early documents indicate more altruistic reasons for the change in grade configuration, several authors suggest that th e initial junior highs were created to relieve overcrowding in elementary schools caused by the high birth rates following World War I and the implementation of early childhood programs in the 1950s and 60s (Craig, 2006; George & Alexander, 1993; Sailor, 1 986). By the 1960s, the baby boomers who had overcrowded the elementary schools in the 1950s were now in upper elementary schools. Rather than build new schools, fifth and/or sixth grade students were moved to the junior high schools to create more room in the elementary schools for the growing populations (Beane & Lipka, 2006). For example, in a study from 1979, researchers found that of 41 middle schools in New York, the majority were created for building space and financial reasons (Moss, Jackson, & Jack son, 1979).

PAGE 34

34 According to Banks (2004), the first middle school was created in Bay City, Michigan in 1950; however it was not until 1963 that William Alexander, who is considered by most the father of the middle school movement, first used the term Muth, 2003). Heller (2003) reported that re considered more than just symbolic. First, it rejected the notion that education for early Additionally, the new name signified a renewed professionalism and distinction for middle level educators. In sum, it marked the beginning of a new movement of reform in the education of early adolescents. Ironically, many of the reasons used to support the movement to middle schools were the same used by educators decades earlier to support the reorganization to junior high schools (Valentine et al., 2002). The middle school movement initially began as a reaction to the failure of junior high schools to respond to the unique needs of early adolescents, to develop an atmosphere of more rigorous learning, and to create a place for modern teaching techniques to occur, such as op en classrooms and team teaching (Manning, 2000; Sailor, 1986). By the late 1960s, many 6 8 configured schools began to open in the Midwest (George & Alexander, 1993; Pate & Muth, 2003). A coalition of Midwestern schools formed a professional organization in 1971 called the Midwest Middle School Association, which was reorganized in 1973 as the National Middle School Association, and then renamed the Association for Middle Level Education in 2011 (Association for Middle Level Education [AMLE], 2011; Lounsbu ry & Vars, 2003; Pate & Muth, 2003).

PAGE 35

35 Throughout the 1970s and 80s, grade school configurations across the country began to shift to a K 5, 6 8, 9 12 configuration (Craig, 2006). By 1985, the middle school configuration of 6 3 4 had become the most common f orm of school organization for early adolescents (Lounsbury & Vars, 2003). From 1971 to 2000, the percentage of schools using the middle school configuration increased by 404%. During the same time period, the percentage of junior high schools with a 7 9 g rade configuration decreased by 85% (Valentine, 2000). By 2000, three out of four middle schools enrolled sixth grade students (Cook, MacCoun, Muschkin, & Vigdor, 2007). Although a variety of grade configurations still exist throughout the United States, t he vast majority of students ages 10 through 14 are educated in middle schools that begin in either fifth or sixt h grade and end at eighth grade (Valentine, 2000). As many researchers continue to point out, however, the title of a school or the cluster of grades contained within does not make it an effective school. Variations of Middle Level Grade Configurations Differences in grade configuration are important because they drive decisions regarding scheduling, staffing, funding, and grouping practices th at effect student learning (Lashway, 2000; McPartland, Coldiron, & Braddock, 1987). There are five common grade configurations for students in the 10 14 year old age range. The first is a combination of elementary and middle school grades, usually encomp assing kindergarten through the eighth grade. Middle/high schools contain grades 6 or 7 through grade 12 K through 12 schools includes elementary, middle and high school students. Junior high schools typically serve students in grades seven through nine. Finally, middle schools begin either in fifth or sixth grade and end with the seventh or eighth grade (Epstein, 1990).

PAGE 36

36 Over the past few decades, the number of middle schools has increased dramatically, while the number of junior high schools has decline d. For example, from 1970 to 2000, the number of middle schools grew over 650%, from 1,500 to 11,500 (Rockoff & Lockwood, 2010). In 1987, roughly half of sixth grade students in public schools attended a K 6 school. By 2007, only one out of five students i n sixth grade attended school in an elementary K 6 setting. In recent years, middle schools have been considered the most dominant organizational structure for early adolescents (DeJong & Craig, 2002; Herman, 2004; Mizell, 2005). In studying middle grade o rganization over a 30 year period, Valentine (2000) calculated that by the year 2000, 86% of early adolescents attended a school with some variation of the middle school organization (5 8, 6 8, or 7 8), with the remaining students in either a junior high 7 9 model (5%) or some other unspecified grade configuration (9%). The K 8 model has always been a popular choice with private and parochial schools (Herman, 2004). In 2000, Renchler reported that 10% of all schools for early adolescents were elementary/m iddle combinations, or K 8 schools. Historically, most K 8 schools have been in either private institutions or rural areas (Eccles et al., 1991; Howley, 2002). In 1990, only about 9% of all sevent h graders attended a K 8 school; however nearly 32% of all school buildings were K 8 combination schools, which suggests that K 8 schools were typically smaller in size since they constitute such a smal l proportion of overall seventh grade students (Epstein, 1990). Approximately 1% of schools house students in gra des K 12 (Howley, 2002). In 2000, principals were asked to identify the ideal grade configuration for middle level students. Sixty five percent (65%) indicated 6 8, 9% preferred the 5 8 model, 16%

PAGE 37

37 listed 7 8, and only 3% listed the old 7 9 junior high sc hool model. The remaining 7% listed other configurations, such as K 8, 3 8, K 12, or 6 12 (Lucas & Valentine, 2005). Researchers believed this difference in opinion had more to do with philosophy than with hard data. For instance, most administrators belie ved that schools housing early elementary grades along with middle level students have a student centered orientation, while secondary configurations have more of a subject area orientation (McPartland et al., 1987 ). In addition, the junior high model has been viewed as more bureaucratic and administrative in structure than the K 8 model (Eccles et al., 1991 ). McPartland et al. (1987 ) found that elementary configurations were more likely to have self contained classrooms and ability grouping within the clas srooms than their secondary counterparts. In addition, teachers in schools containing elementary grades are less likely to have certification in their subject areas (Sparks, 2004). Conversely, secondary schools are usually departmentalized, with separate t racks grouped by academic performance. Middle schools have nearly equal representations of both types of groupings. There are many reasons why the schools change their grade configurations. Some schools choose configuration based on developmental similari ty or to adopt a new program (Epstein & MacIver, 1990). More frequently, grade span is determined for other reasons, such as enrollment numbers, building space, parent preference and distance to the school, rather than on what is best for early adolescents (Craig, 2006; Paglin & Fager, 1997; White, 1993). For example, using data from the Schools and Staff Survey (SASS) from 1993 94, Alt and Hammer (2000) reported that the middle school grade configuration is less likely to be located in very small districts because they have

PAGE 38

38 less flexibility to configure their schools in a variety of ways. Other factors for changing grade configurations could include changing demographics of the area, transportation costs, school size, number of transitions made by students, and overall school goals (Craig, 2006; Howley, 2002; Renchler, 2000). Howley (2002) contended that the issue of grade span causes a great deal of controversy in rural areas, where reconfiguring grade structures can mean closing local schools. This notion is reinforced by Paglin and Fager (1997), who argued that changing grade configurations to isolate middle level students can be detrimental in small rural communities. Removing and separating middle level students can result in schools that are so small, they must consolidate in order to survive. Franklin & Glascock (1998) warned that schools must be careful not to create grade segregated schools for purely financial or administrative purposes. Ecker (2002) stated that proponents of the middle school desig n have created a has led to dissatisfaction with middle schools, resulting in consideration of other grade configurations. In 2002, Pardini predicted that the newest t rend in grade span will be a return to the K 8 organization. In fact, the K 8 model has become an increasingly popular design in urban settings. A number of large cities, including New York, Minneapolis, Philadelphia, Baltimore, Cleveland, Cinncinati, Miam i, Boston, New Orleans, San Francisco, Milwaukee, and Memphis, have converted or are in the process of converting a number of their schools to a K 8 design (Beane & Lipka, 2006; Hough, 2005; OPPAGA, 2005; Pardini, 2002; Patton, 2005; Reising, 2002; Turque, 2008; Viadero, 2008; Wallis, Miranda, & Rubiner, 2005; Waugh, 2004). Likewise,

PAGE 39

39 Massachusetts, Pennsylvania, Ohio, Tennessee, Oklahoma, Maryland, and New York have statewide reforms which encourage the K 8 grade configuration (Hough, 2005; Reising, 2002). Reasons given by advocates for the conversion to the K 8 design include greater articulation in grades K through 8, fewer behavioral issues, improved safety, reduced transportation costs, and elimination of a transition at a critical developmental point fo r early adolescents (Beane & Lipka, 2006; DeJong & Craig, 2002). Other reasons given for this trend have included parent request, small scale local research on academic achievement, and a growing discontent with the current structure. As MacIver and Epste in (1993) have pointed out, grade organization alone is not a major determinant of whether a school is implementing practices responsive to the middle level student. Paglin and Fager (1997) offer several factors when considering the grouping of early adole scents, including the following: ( a) consistency with community needs, ( b) staff training and attitudes, ( c) student needs, and ( d) effects on other students in the school. The effectiveness of grade configuration is not simply a matter of changing grade s pans, but rather of implementing programs responsive to the unique needs of middle school students. Therefore, when studying the effects of grade configuration, the researcher must also address the level of implementation of exemplary practices, not just g rade configuration alone (Hough, 2005; Sailor, 1986). As The extent of grade configuration effect depends on the grade spans, but they all have the option of adopting best practices that support middle

PAGE 40

40 level students (Epstein & MacIver, 1990). In other words, it may not be the grade span of the school, but what happens within those configurations that matters most. Characteristics of Model M iddle Schools In 1982, the National Middle School Association (NMSA), which changed its name to the Association for Middle Level Educators (AMLE) in July 2011, published the education of young adolescents. NMSA/AMLE maintained that early adolescents need an educational program that is different than both elementary and high school. The true goal of the middle school movement was not about restructuring grade configurations but about the implementation of programmatic elements that would best help students transition into adolescence (Alexander & McEwin, 1989b; Lounsbury & Vars, 2003). The p rovide students with care and support, as well as challenging programs that will There are several practices that have been deemed as effective in working with early adolescents (Alexander & McEwin, 1989b; Ecker, 2002; Johnson, A., 2002; Lipsitz, Mizell, Jackson, & Austin, 1997; Lucas & Valentine, 2005; MacIver & Epstein, 1993, NMSA, 1995; Neill, 1999). Key components of the middle school design include the following: ( a) interdisciplinary teaming with flex ible scheduling, ( b) common planning time for teams of teachers, ( c) heterogeneously grouped students, ( d) advisory periods, ( e) exploratory courses, and ( f) looping (Valentine et al., 2004). Middle school advocates argue that s tructural changes such as ho uses, interdisciplinary teams, and advisories create a more personalized learning environment (Lipsitz et al 1997). These special middle level structures allow for increased exploration, better interaction with

PAGE 41

41 adults and peers, greater competence, clear structure and limits, meaningful participation, and more diversity in scheduling, teaching and curriculum (Dorman, Lipsitz, & Verner, 1985). Other exemplary practices for middle level learners include varied instructional strategies and orientation activi ties for both students and parents (Alexander & McEwin, 1989b; Ecker, 2002; Lucas & Valentine, 2005; NMSA, 1995). Just because schools have changed grade configurations does not necessarily mean that they have changed practices and implemented these recom mended changes. In fact, MacIver and Epstein (1993) presented research from the 1980s, at the height of the middle school reform movement, that suggested new middle schools and old junior high schools were surprisingly alike. Neill (1999) surveyed a group of principals and teachers in 32 accredited middle schools in Montana to determine both the perceived degree of importance of a number of practices for middle schools, and then the level of implementation in their own middle schools. The essential elements in the survey included practices such as common planning time for teachers, interdisciplinary teaming, exploratory experiences, heterogeneous grouping and advisory programs. The researcher found that although the educators rated all of the moderate level of implementation in their own schools. The establishment of a common set of developmentally appropriate practices for early adolescents has allowed researchers to focus on t he effectiveness of true middle school programs (Neighbors, 1998). Interdisciplinary Teaming An interdisciplinary team is a group of two to five teachers who share the same students, have common planning time, and are located in the same area of the buil ding.

PAGE 42

42 In terms of organizational practices, students should be heterogeneously grouped on interdisciplinary teams, with teachers who share a common daily planning period (Alexander & McEwin, 1989b; Ecker, 2002; Lucas & Valentine, 2005; MacIver & Epstein, 1 create a feeling of closeness within a l arger, more impersonal school. In interdisciplinary t eaming, t he academic block of instructional time with students is flexible, allowing teachers the freedom to adjust schedules within the time allotted to the team (Lucas & Valentine, 2005; Paglin & Fager, 1997). Extended time during the school day for acad emic enrichment or remediation is also included in the organizational structure of the school (Ecker, 2002). There are many benefits to interdisciplinary teaming. For example, teaming is linked to a greater student sense of belonging, improved relationsh ips with the families of students, and a more integrated curriculum (Beane & Lipka, 2006; Washington, 2000). From a survey of teachers, students, and administrators in 155 schools, researchers found that schools with interdisciplinary teaming had a more po sitive school climate, increased contact with parents, greater teacher job satisfaction, and higher student achievement (Flowers, Mertens, & Mulhall, 1999). Teaming is also associated with more meaningful relationships between adults and students (Hackman & Valentine, 1998; Phillips, 2003). In a study by Arhar (1994), students in teamed schools rated schools than students in schools that were not organized into teams. In another study, students from th ree middle schools in a major urban school district were interviewed

PAGE 43

43 regarding their experiences in middle school. The researcher found that interdisciplinary teaming was related to group closeness, school family identity, preparation for the future, and b etter student teacher relationships (Gray, 2004). Advocates of teaming indicate that the practice allows better communication, may lead to greater student achievement (H ackman & Valentine, 1998). Aycock (2005) studied the effect of interdisciplinary teaming on student achievement in 49 middle schools in Mississippi. The researcher used results from the Mississippi Curriculum Test based state assessment, as the measure of student achievement in schools with and without interdisciplinary teaming, and found that eighth grade academic achievement was higher in schools with interdisciplinary teaming. Similarly, in a study of 10 middle schools in a large urban district in the Midwest, educators were assessed regarding the level of implementation of several middle school practices. Interdisciplinary teaming was positively related to student achievement scores, especially in mathematics (Russell, 1997 ). Balfanz and MacIver (2000) studied five high poverty Talent Development Middle Schools in Philadelphia, and found similar student achievement gains As part of the as mo re personalized learning environments and increased student teacher interactions within a small learning community. When these practices were combined with a variety of other components, such as professional development, student safety nets, and ongoing ev aluation, students made significant gains in learning during the first year of

PAGE 44

44 the implementation, although it is not stated in the research the degree to which communal organizational practices contributed to the learning gains. Because the benefits of i nterdisciplinary teaming appear to be great, DeJong and Craig (2002) have recommended a teamed approach, regardless of grade level or grade configuration. Despite the advantages, interdisciplinary teaming occurs most frequently in middle schools than in ot her organizational patterns (Ritzenthaler, 1993). In a random sample of 300 schools, Alexander and McEwin (1989a) found that 6 8 configured middle schools more frequently used interdisciplinary teaming and flexible student grouping. Five years later, Irvin et al. (1994) found that teaming was most often used with students in sixth grade, and almost never used with students in the ninth grade. The trend is beginning to turn, however, as more high schools have started to debate departmentalization versus tea ming through smaller learning communities (DeJong & Craig, 2002). Lee and Smith (1995) researched the effects of high school restructuring and school size on student achievement and engagement. Using data from nearly 12,000 tenth graders from 830 schools p articipating in the National Educational Longitudinal Study (NELS), the researchers found that students who attended schools that implemented three specific practices had greater learning gains, were more academically engaged, and had more equitable gains in student achievement. Those practices included keeping students in the same homeroom throughout high school, implementing interdisciplinary teaching, and developing smaller learning communities of teachers and students within the school, much the same as the developmentally responsive practices suggested for middle level learners.

PAGE 45

45 Common Teacher Planning Time Common planning time is an essential element of interdisciplinary teaming. In an effort to reduce the isolated learning of academic content by subj ect area, middle school teachers are grouped into interdisciplinary teams in which they are expected to collaborate and plan together (Toepfer, 1990a). To better coordinate activities and instruction, team teachers must be provided time together on a daily basis. The National Center for Public Education and Social Policy has recommended that teachers be provided with 3 to 4 hours of common planning a week (Erb & Stevenson, 1999). This time is to be used to plan for curriculum integration across subject area s, review student work, coordinate student assignments, analyze assessment data, communicate with parents, and collaborate on the effectiveness of instructional approaches (NMSA, 2003). White (1993) maintained that with common planning, teachers are able t o provide a curriculum centered on thematic based units that are directly related to student interests and needs. In a study of the Michigan Middle Start Initiative, the researchers found that with increased amounts of common planning time, the quality of team interactions also increased. In addition, teachers with increased common planning were more likely to implement desired instructional practices. They also found that the highest gains in average reading and mathematics scores for seventh grade studen ts over a two year period were achieved by schools that provided increased levels of common planning time for their teachers (Mertens, Flowers, & Mulhall, 1998). In a related study of 155 middle schools in Michigan, researchers studied the impact of teamin g on five research based outcomes (Flowers, Mertens, & Mulhall, 1999). Using survey data from teachers, administrators, and students, as well as

PAGE 46

46 student achievement data from the Michigan Ed ucational Assessment Program (MEAP ), the researchers determined th at common planning was essential to the effectiveness of interdisciplinary teaming. In addition, schools with the highest levels of common planning had the highest gains in student achievement on the MEA P. Middle school advocates have stressed that common planning is a key factor in student success (Heller et al., 2003). Hoy and Sabo found that student achievement increased where teachers felt they had strong support among themselves (1998). Likewise, researchers conducted a case study of three nationally r ecognized Blue Ribbon middle schools serving students from grades six through eight in an Oklahoma suburban school district. Through interviews and observations of teachers, students, counselors, and administrators, they found there were two common practic es associated with all three schools: a wide array of exploratory courses and common planning for teachers (Weibling, 1997). Heterogeneous Grouping Turning Points: Preparing American Youth for the 21 st Century (Jackson & Davis, 2000), a report from the Ta sk Force on Education of Young Adolescents, cited a number of recommendations for middle level educators, including the elimination of tracking. Although there are arguments for ability grouping, such as the need for teachers to adapt instruction to specif ic needs of students, there are far more reasons to abandon the practice. Researchers cite damage to low achievers, less qualified teachers in lower ability level classrooms, and the perpetuation of class and racial inequalities as a result of tracking (He ller et al., 2003; Slavin, 1993; Stroud, 2002). Mills (1998) found that tracking has a negative effect on lower tracked student motivation for more challenging courses. In the study, the researcher found that 82% of

PAGE 47

47 middle schools used some form of abili ty grouping, with math and reading being the subjects most commonly grouped by ability level. Mills also found that the practice of tracking makes school schedules less flexible. Using data from the Longitudinal Study of American Youth (LSAY), Hoffer (199 2) conducted a four year study of middle and high school mathematics and science teachers, as well as students drawn in pairs from middle and high schools to determine the long term effects of tracking. Schools were chosen by community type and region of t he country, and the results were controlled for social background and initial achievement effects. The researcher found that placing low achieving student s in low level math courses had no positive long term effects. In fact, Heller et al. (2003) suggested that low to average level students benefit from being intentionally placed in a track higher than they would ordinarily be assigned. In a similar study by Mason, Schroeter, Combs, and Washington (1992), 24 students were placed in advanced mathematics cla sses. Those students scored higher on a math assessment than the cohort group. Perhaps more importantly, those same students went on to take more advanced classes in mathemati cs than their peers who were placed in the less challenging mathematics courses In a review of 27 research studies regarding middle grades tracking, Slavin (1993) found that the effects of ability grouping are zero. Similarly, Betts and Shkolnik (2000) have stated that grouping has little effect on achievement regardless of whether st udents are high, average, or low achieving. Since tracking appears to have little or no positive effects, national experts advise that model middle school teams should consist of a group of diverse learners with flexible grouping practices within the team of

PAGE 48

48 students In this way, the needs and abilities of all the students can be better served ( N ational Association of Elementary School Principals [NAESP] 1984 ). Student Advisory One of the most notable practices in a school that serves middle level studen ts is is that every child should be known well by at least one caring adult (Carnegie Council on Adolescent Development, 1989). The National Middle School Association (1995) has identified advisory groups as one of the six general characteristics of a developmentally responsive middle school. In advisory, the role of the teacher is expande d to that of a supporter, role model, and critical adult friend for a specified group of students (MacLaury & Gratz, 2002). In an effort to keep the groups small, most schools use all certified staff within the school, including guidance personnel, adminis trators, and elective and physical education teachers. The purpose of the advisory groups is not counseling, because the time is not spent discussing individual difficulties. Instead, advisories are comprised of activities that provide an opportunity for the adult to get to know the students on a more personal may need more intensive help, both academically and socially (NMSA & NAESP, 2000). There are generally six purposes of advisory programs including the following: ( a) building relationships between student and teacher; ( b) fostering a sense of team spirit among the students; ( c) building life skills; ( d) relaxing and recharging while engaged in fun activities; ( e) devel oping and monitoring study habits; and ( f) housekeeping and organizational activities, such as daily announcements or organizing for a field trip

PAGE 49

49 (Galassi, Gulledge, & Cox, 1997). There is some disagreement among educators about what kinds of activities sh ould be included in advisories, but virtually all iterations include a caring adult who regularly communicates with students in a positive way (Caccamo, 2000). Although there was little empirical evidence of its effectiveness at the time, in 1984 Florida p assed legislation that funded school advisory programs, providing $4 million for middle schools (Galassi et al., 1997). Since that time, a body of research has emerged that suggests that advisory programs have a significant impact on students at all grade levels (Caswell, 2003). Schools with teacher led advisory groups are viewed by students as having a more personalized learning environment, in which students view their teachers as a primary source for academic or personal problems (Fibkins, 1999). For ex ample, MacLaury and Gratz (2002 ) surveyed 71 low income minority fifth and sixth grade students about their academic performance, the school environment, and social supports within the school. 44 of the students participated in an advisory group and 27 did not. Students who participated in the advisory groups were more likely to feel safe confiding in a teacher and indicated that they were more likely to seek out help. In a case study of students from an urban middle school in Philadelphia, Brown (2001) re ported that students participating in advisory groups were more proficient in goal setting, participated in more community service activities, and had established more personal trusting relationships within the school. Goodwin (2003) reported similar advan tages in a qualitative study of one urban middle school in New York. Study responses indicated that advantages of advisory groups include a less chaotic learning

PAGE 50

50 environment, relationships between teachers and students that are characterized by care, and t he presence of a positive role model to act as a mentor and guide when students experience problems. Establishing a supportive school environment through advisories can also lead to increased academic achievement (Caswell, 2003). Roeser, Midgley, and Urdan (1996) surveyed nearly 300 eighth grade students in two suburban middle schools and found that positive perceptions of teacher student relationships were predictive of end of year student grades. In a study of over 200 middle schools in 16 states, Caswel l (2003) researched the effects of advisory programs on student adjustment and academic outcomes. The schools in the study were all participants in the Carnegie Corporation Middle Grade School State Policy Initiative (MGSSPI). Using the High Performance Learning Community Assessments, surveys were collected from 3,837 teachers and 113,598 students. The researcher studied the length of the advisory structure, advisory practices, teacher attitudes, and student experiences and their effect on student grades and classroom behavior. The research revealed that advisory structures were significantly positively related to five advisory practices: ( a) conversations about career and college issues, ( b) implementation of student initiated projects, ( c) individual and small group meetings, ( d) parent communication, and ( e) total advisory practices combined. In addition, students reported that advisory practices influenced their feelings of safety. Advisory programs are also linked to other research that indicates its s ignificant positive impact. For instance, students in schools with advisory programs experience

PAGE 51

51 less absenteeism (Lipsitz et al., 1997), have fewer disciplinary problems (Connors, 1992; Lounsbury, 1996), and are less likely to drop out of school (MacIver & Epstein, 1993; Ziegler & Mulhall, 1994). Advisory groups also appear to have positive effects for teachers. These teachers report a greater sense of professional efficacy, (Felner, Seitsinger, Brand, Burns, & Bolton, 2007; Lounsbury, 1996), communicate wi th parents more frequently (Ziegler & Mulhall, 1994), and collaborate more closely with their peers (Caswell, 2003; Lipsitz et al, 1997). Despite the reported merits of using advisory groups, Caswell (2003) found that only one third of the 200 middle schoo ls in her study had advisory programs. While middle school implementation of advisory programs may have hit a plateau, the practice ed as one of the seven guiding principles of restructured high schools by the Institute for Student Achievement, a non profit educational organization in New York that works with high school students with academic or social challenges (Hochman, Tocci, & Al len, 2005). Similar to the middle school advisory, responsibility for counseling with students is distributed across the entire high school faculty so all students can be known by at least one caring adult within the learning community. Educators for Socia l Responsibility also advocate the use of advisory practices in an effort to personalize the learning environment for high school students (Poliner & Lieber, 2004). There are several factors that are critical to the success of an advisory program. Teachers need tangible support from their school administrators in the form of time, materials, and supplies. Secondly, activities that students enjoy spanning a variety of

PAGE 52

52 topics need to be developed. Finally, the best advisory programs have a school wide coordin ator who facilitates the planning of student activities and assists the teachers with planning (Galassi, Gulledge, & Cox, 1997). Exploratory Courses In addition to the core academic courses, middle level students need exposure to a wide range of high inter est activities called exploratories (NMSA, 1995). Exploratory activities were developed around the notion that early adolescents need high interest, wholesome opportunities for socialization. This outlet can be provided in a number of ways, including mini courses, intrascholastic athletics, performing groups, clubs, and electives (George & Alexander, 1993; George & Shewey, 1994; Lounsbury & Vars, 2003). Exploratory activities include, but are not limited to, student participation in music, the visual and pe rforming arts, world languages, technology, physical education, and vocational arts (Raymond, 2005). Because the goal is to expose students to a wide array of interest areas, exploratory courses are typically six to nine weeks long for 40 minutes to an hou r daily (Manning, 2000; Aycock, 2005). According to Heller et al. (2003), academic success is significantly related to the degree to which student interests are incorporated into the curriculum. Exploratory activities provide students with an opportunity t o develop interests in activities that support the core curriculum. The effects of exploratory courses were researched in a study that included over 300 middle schools in Colorado (Raymond, 2005). Using student achievement data on the Colorado State Achiev ement Test, and school wide compared the number of exploratory courses offered with school achievement, and

PAGE 53

53 found a positive correlation between school achievement and the numb er of exploratory course offerings. The positive effects of exploratory courses on student learning are well documented. For example, there are a number of research studies that cite the benefits of the arts in education. Butzlaff (2000) researched 24 corr elational studies about the effects of music on reading and found that students studying music have significantly higher scores on standardized tests; however, the author is careful to note that this correlation does not necessarily imply causality. In a r eview of 16 research studies, skills. Burton, Horowitz, and Abeles (2000) conducted a research study of over 2,000 children in 28 schools from 4 different regions of the country to determine if cognitive skills were developed through the arts. The research, which combined both qualitative and quantitative data, determined that a flexible art curriculum, including arts integration, is associated with creative thinking, fluency, focused perception, and persistence. In addition, students exposed to the arts have a better perception of accomplishment in reading and mathematics. In a review of 25 studies, including both experimental and correlational designs, Vaughn (2000) f ound a small positive effect on student mathematics performance during math tests. Outside of arts, Schreiber and Chambers (2002) conducted a research study including over 8,000 students in grades eight and ten, and found that academic extracurricular acti vities outside core instruction were consistent predictors of student achievement. McEwin and Alexander (1990) found that since many middle school facilities were newer than K 8 schools, they tend to have more updated facilities designed

PAGE 54

54 specifically for e xploratory activities. This may be changing, as more new schools are being designed specifically for K 8 students. For example, Miami Dade County Public Schools have designed a prototype school designed specifically for students in grades K through 8 that include specifically designed space for computer labs, collaborative activities, arts suites, and outdoor space suitable for both elementary and middle level students (Mi ami Dade County Public Schools [M DCPS] 2011). 2011 marked the end of the federal sti mulus funding for education, resulting in a $100 billion shortfall on top of deficits already experienced in states and local Washington Post, 2011). While in the past, many districts have tried to spare programs like elective courses, art, music, and physical education, these new cuts will invariably have an effect on the ability of schools to maintain these programs (Krugman, 2011). Looping Looping is the practice of keeping a small group of teachers together with the same group of students for two or more years (Aycock, 2005). Looping is also referred to as teacher student progression, multi year teaching, multi year grouping or multi year assignment (McCown & Sherman, 2002). The pr emise of looping is that teachers will invest more of themselves and be less likely to give up on marginal students, because they know they will be working with the students for a number of years (George & Alexander, 1993). In addition, relationships with families are enhanced by years of interaction with the same groups of students. Another benefit is that teachers save valuable instructional time at the beginning of the school year, which is typically lost with introductions to classroom rituals and routi nes that have already been established the previous year (Darling Hammond, 1997).

PAGE 55

55 Although looping has attracted increased attention in the past two decades, it is not a new concept. It was first suggested in 1913 by the United States Department of Educat ion for students in grade leveled city schools, so that the teachers could better build on the work of the earlier years of instruction. In Germany, students routinely stay with the same teacher for first through fourth grades (McCown & Sherman, 2002). Le e, Smith, Perry and Smylie (1999) found that the practice of looping with students for two or more years resulted in student reports of increased feelings of support. McCown and Sherman (2002) conducted a case study of looping in one Kansas middle school i n which 1 team of students and teachers looped from seventh to eighth grades, while the other two teams of students did not. The researchers found a number of advantages associated with looping. For instance, the teachers reported that they already knew th e academic needs of students and were able to accomplish more instructional goals during the first two weeks of school. In addition, teachers understood the emotional needs of their students, particularly those students with special needs. Both parents and teachers reported a stronger rapport between home and school. Students reported less apprehensive about starting a new school year. Teachers reported that the greatest benefit may have been for students with disabilities, and said that they found many of their ESE students were performing on grade level by year 2 and required less specialized support, as a result. Recent research supports the use of looping to increase student achievement. In a study of 240 middle school students in Georgia, Gregory (2009) studied the impact of looping on academic achievement and social experiences. Using standardized test data and surveys of student perception, the researcher found that looping had a positive

PAGE 56

56 effect on student perceptions of social experiences. Also signif icant was the positive correlation between reading, writing, and mathematics scores and the degree of participation in looping. In a similar study, Fuller (2006) conducted research to determine if looping resulted in academic achievement in reading, langua ge and writing. The researcher used student achievement data from 69 students who remained with the same group of teachers for seventh and eighth grade, and 142 students who were assigned two different groups of teachers for each year. Results from the stu dy indicated that the students in the group that looped experienced greater learning gains than their non looping peers. Combined Effects of Developmentally Responsive Practices In the earlier years of the reform, the middle school movement prospered more on faith than research (Lounsbury & Vars, 2003). In 1990, Epstein reported that most schools that enrolled seventh grade students had not yet adopted the majority of the recommended responsive middle school practices. More recently, other authors have also suggested that middle schools are moving away from developmentally appropriate adopted the middle school concept in grade configuration alone, without adopting the affective and organizational components of the design (Beane & Lipka, 2006, Hough, 2005; Sailor, 1986; Valentine et al., 2002). Lipsitz et al. (1997) have stated that many middle schools are poised for reform because of structural or organizational changes, but have yet to make the next step to academic excellence

PAGE 57

57 The body of research regardi ng the implementation of multiple developmentally appropriate practices has resulted in a variety of findings. Anfara and Lipka have stated, achievement have yielded ambiguous example, in a study of 75 middle schools in Alabama, Neighbors (1998) studied the relationship between the implementation of middle school practices and student achievement. Using the New England League of Middle Schools (NELMS) recommendations for appropriate middle level practices, Neighbors found no significant relationship between student achievement and high levels of implementation of the middle school model. The NELMS recommendations seem to be more about e ducational practices that would be applicable to any level of schooling, such as teaching a core of common knowledge, engaging families, and empowering teachers, rather than practices that have been specifically designed for middle school students, such as interdisciplinary teaming, common teacher planning, and student teacher advisory groups. In 2003, Spurgeon found more encouraging results in a study that compared the implementation of middle school practices with student achievement. In this study, data were collected about the attendance, disciplinary referrals, and achievement in science and math. Participants in the study included 600 students in three Midwestern middle schools. The researchers used the following practices to determine implementation of the middle school model: ( a) interdisciplinary teaming, ( b) advisory programs, ( c) varied instruction, ( d) exploratory programs, ( e) transition programs, and ( f) grade level configuration change. The researcher found that the implementation of the pract ices

PAGE 58

58 had a significant effect on sixth grade attendance rate, seventh grade disciplinary referral rate, and eighth grade student achievement, indicating that the combined effects of the practices may have some effect on the overall performance of students. A team of researchers conducted a study of the Illinois Middle Grades Network (IMGN), consisting of 97 middle schools participating in a restructuring reform using 1997) As part of the study, the researchers developed what they referred to as the Carnegie Index of Middle School Transformation structures for effective middle level implementation, including teaming, common planning, low te acher to student ratio, and regular advisory periods. Their research indicated that middle schools with the highest level of implementation of these best practices had the highest scores in mathematics, language and reading on student achievement tests. Me rtens and Flowers (2003a) had similar result s in a research study using data from the School Improvement Self Study conducted by the Center for Prevention Research and Development at the University of Illinois. The researchers collected surveys from 3,500 participants, including teachers, administrators, parents and students at 121 schools in rural communities in Arkansas, Louisiana, and Mississippi that were considered high poverty and ethnically diverse. Student achievement was based on standardized test scores from the SAT9 and the Iowa Tests of Basic Skills. Findings included a strong association between school level curriculum coordinator and interdisciplinary instructional practices. In addition, common planning was found to be related to higher readin g scores, and the sustained implementation of teaming

PAGE 59

59 combined with common planning showing the most positive results in terms of student achievement. Blue Ribbon Schools is a national program that recognizes effective, exemplary schools based on a variety of eligibility requirements including the following: ( a) visionary leadership, ( b) shared purpose, ( c) positive school climate, ( d) challenging curriculum, ( e) student academic achievement, ( f) stakeholder support and ( g) commitment to improvement. Ranta (2001) studied seven Blue Ribbon Schools in New Jersey to determine that implementation of the essential characteristics of effective middle schools. He found that there was a positive relationship between the implementation of middle school exemplary prac tices and middle schools identified as Blue Ribbon Schools of Excellence, with Blue Ribbon Schools in New Jersey having a higher level of implementation of essential middle school characteristics than the comparison group of schools. Bush (1999) conducted a case study of one middle school that had been fully implementing the middle school concept for 12 years. Through surveys, interviews, and a review of school based documents, she found that the most influential factors to full program implementation inclu ded: ( a) a dedicated faculty, ( b) administrative support, ( c) district support for the identified organizational structures and practices, and ( d) student attitudes. She also listed a number of factors that were identified as impeding the implementation, i ncluding: ( a) teacher resistance, ( b) lack of professional development, ( c) limited time for collaboration, ( d) state mandates and accountability issues, ( e) student behavior issues, ( f) class size, and ( g) changes in school based leadership.

PAGE 60

60 Effective mid dle level programs are not solely defined by grade configuration. The concepts appropriate to middle level students can be carried out in any kind of school grade configuration (Hough, 2005; Offenberg, 2001). For example, Manning and Saddlemire (1996) have suggested the implementation of a variety of middle school concepts in high schools, including advisory groups, exploratory programs, and interdisciplinary teams. The goals of middle school reform are apparently not unique to early adolescents. The purpos performing schools that are developmentally responsive, academically excellent, and Implementation of Model Practices b y Grade Configuration Much of the research regarding the implementation of middle grades model practices by grade configuration is limited and older and may not reflect the current implementation of the new K 8 model. Most of the studies focused on the ol d junior high structure as compared to what was considered the modern middle school, at the time. K 8 data regarding implementation of the practices was considered almost anecdotal, since the majority of the students were being educated in junior highs or middle schools. However, there were a few studies that shed light on where these practices were being implemented to the fullest. Using data from the Hopkins Education in Middle Grades Survey (Becker, 1987) and the National Education Longitudinal Study of 1988 (Ingels, Abraham, Rasinski, Karr, Spencer, & Frankel, 1990), researchers conducted a series of analyses to determine differences in grade organization for middle level students. The researchers found that middle schools with 6 8 or 7 8 grade configur ations were slightly more likely to adopt responsive practices recommended for adolescents (MacIver & Epstein, 1993).

PAGE 61

61 In a national random sample of 300 schools, Alexander and McEwin (1989 a ) found that 6 8 configured middle schools more frequently use inte rdisciplinary teaming, flexible grouping, and random assignment in basic classes. Middle school teachers were also found to visit other middle schools more frequently than teachers from schools of other grade configurations. The K 8 configuration had almos t as much departmentalization as the middle and junior high models, but with much less team planning. Common planning periods for interdisciplinary teams were most often found in 6 8 and 7 8 schools (Epstein, 1990). Interdisciplinary teams were most frequ ently found in 5 8 or 6 8 teams, and least likely to be found in 7 9 junior high schools. Teaming was offered to sixth grade students most often and almost never to stud ents in the ninth grade (Irvin et al. 1994). With 98% offering re medial courses, middl e schools we re most likely to have additional help for students during the school day (MacIver & Epstein, 1993) although it is not clear whether remedial course offerings effect the ability of schools to heterogeneous group students In addition, middle s chools use d informal progress reports more often than other grade configurations (Alexander & McEwin, 1989 a ). Instructionally, middle school teachers were found to be more likely to use the following instructional practices: ( a) discovery methods in sci enc e, ( b) problem solving in mathematics, ( c) increased writing across the curriculum, and ( d) student debate. At the same time, they we re less likely to use drill and practices then their K 8 and junior high school counterparts (Epstein & MacIver, 1990). Mi ddle schools we are K 8 schools (Alexander & McEwin, 1989 a ; McEwin & Alexander, 1990). Lounsbury and Johnston (1988) found that middle school programs deal t more directly with current

PAGE 62

62 issues, suc h as sex, alcohol, drugs, social and emotional development, and personal adjustment than schools in the K 8 configuration. In a survey of principals who serve d students in the middle grades, Epstein (1990) found that middle school principals in a 6 8 confi guration ranked personal growth and development higher than principals in any other grade configuration. McEwin and Alexander (1990) found that s ince the middle school concept wa s the newest of the designs, middle schools tend ed to have newer facilities t hat are needed for the exploratory courses. In f act, studies at the time showed that students in middle schools have more exploratory choices than middle level students in other grade configurations (Jenkins & McEwin, 1992; Lounsbury & Johnston, 1988). Sch ools with 6 8 grade configurations tend to offer health, computer educa tion, and sex education, which we re not commonly offered in the K 8 configuration (Alexander & McEwin, 1989 b ). In addition, all club types, including photography, newspaper, dances, int ramur al sports, and student council we re found to be more prevalent in the middle and junior high configuration than in the K 8 design (McEwin & Alexander, 1990). The advantages of the middle school concept include more interdisciplinary teaming and teach er advisory programs, schedules that are more flexible, and more comprehensive exploratory and intramural programs (McEwin & Alexander, 1990 Valentine et al 2002). However, the research regarding middle school practices is not all positive. One of the m ain criticisms seems to be that middle schools have placed too much emphasis on creating a nurturing environment and too little focus on academic rigor and achievement (Pardini, 2002). In addition, Eccles et al. (1991) found that many schools with 6 8 grad e configurations do not follow the middle school philosophy.

PAGE 63

63 Instead, they look and operate just like the old departmentalized ju nior high schools they replaced, with very little implementation of the developmentally responsive practices that are a foundat ion of the middle school redesign. Effects of School to School Transitions Some researchers suggest that transitions, rather than grade span, may be the primary area of concern for middle level students (Alspaugh, 1995; Lucas & Valentine, 2005; Schiller, 1 999; Simmons & Blyth, 1987). The smaller the grade span within a school, the more transitions students must make between kindergarten and twelfth grade, which can be a stressor for some children (Paglin & Fager, 1997). Toepfer (1990a) found that one or tw o grade level centers offer a fragmented educational experience for students. In early adolescence, children are dealing with the changes in their body as well as transitioning to a new school. Students in this age range may place a higher priority on soci aliz ation than on their studies in s chool. In addition, students may begin to experiment without new groups of f riends and with their own imag e, from the way they dress to the color of their hair (Daniels, 2005). ial, psychological, and developmental needs is essential before beginning to work on their academic needs (Daniels, 2005; Lipsitz, 1984). Rapid physical, social and emotional changes mean students need more flexible learning modalities which balance struct ure and choice (Ecker, 2002). Jenkins and McEwin (1992) indicated that programs and practices in elementary schools are focused on childhood needs, rather than the needs of young adolescents. The goal of middle school is to focus on the unique needs of you ng adolescents (George & Oldaker, 1985/1986). The challenge of schools serving early adolescents is to develop a program

PAGE 64

64 based on needs and characteristics of an incredibly diverse and varied population (NMSA, 1995). Some theorists have suggested that decl ines in student motivation may be a result of adolescents coping with two major transitions: pubertal change and school change (Eccles et al., 1991, p. 522). Simmons and Blyth (1987) argued that the effects of simultaneous multiple life changes are negativ e and are more prolonged, especially for females. Their research shows that girls suffer losses in self esteem when the numbers of life change events to which they are exposed increases. Both males and females experience declines in grade point average and extracurricular participation with increased numbers of transitions. Alspaugh (1995) found that the year the student transitioned to another school, regardless of the year, achievement for that student dropped. However, he reported that students tend to r ecover from transition losses the year after the transition. In a study of sixth grade students in North Carolina, researchers found that behavior problems increased and academic achievement declined the first year after a transition from one school to ano ther, even after adjusting for economic and demographic factors (Cook et al., 2007). In a similar study of 303 students in rural South Dakota, the researcher compared academic achievement of students in their last year of elementary school with their first year of middle school (Johnson, D., 2002). Using scores from the SAT9 and report card grades, the study found that SAT9 scores declined in both math and reading after the transition from fifth to sixth grade. Transitions may have other negative effects on student success. Rudolph, Lambert, Clark, and Kurlakowsky (2001) studied the effects of school to school

PAGE 65

65 transitions on two self regulatory beliefs in early adolescents. In the study, 329 students and their teachers from Midwestern school districts were s urveyed in the spring of their fifth grade year, then again in the fall of sixth grade. 187 of the students transitioned from elementary to middle school after fifth grade, with the remaining 142 students staying in the same school for sixth grade. Feeling s of lack of influence over their own success in school were more prevalent in the group of students that transitioned to middle school for sixth grade. In a study of 45 schools in Missouri, Howley (2002) compared the grade level in which students transi t ioned to high school with drop out rates and found that schools that housed students from grades K 12 that had no school to school transitions experienced the lowest dropout rates. In addition, increased numbers of transition were correlated with higher dro pout rates. Brown (2004) studied school districts in Ohio in which all students within the district experience the same number of transitions. The study compared the number of school to school transitions with dropout rate, graduation rate, and aggregate p ass rates on the Ohio Proficiency Test (OPT). The research found that the number of transitions is a significant predictor for graduation rates and ninth grade aggregate achievement on the OPT. Simmons and Blyth (1987) offer a hypothesis regarding the effe cts of transitions on early adolescents that appears to have some substance. Their Developmental Readiness Hypothesis states that a change in environment will have negative effects if it occurs when children lack the maturity to cope with the change. Simmo ns and Blyth found that the negative effects can be prolonged, with the greatest loss to the self esteem of females.

PAGE 66

66 Toepfer (1990b) suggested that total educational effectiveness is enhanced when there is a bridge between elementary, middle and high schoo l. The stress of transitioning between schools can be lessened by articulation practices such as field trips to the new school, presentations and orientations for parents and students, and faculty communication between feeder patterns (Grolnick, Kurowski, Dunlap & Hevey, 2000; MacIver & Epstein, 1993; Paglin & Fager, 1997). Research seems to bear out this theory. In a study of 56 high schools in Florida and Georgia, Hertzog and Morgan (1999) surveyed students about what types of transition programs they wer e offered. Schools with two or fewer transitional practices had significantly high ninth grade dropout and attrition rates than those that implemented three or more practices. In a study of urban students transitioning to the ninth grade, one factor that distinguished students that made a successful transition to high school from students who were not as successful was the ability the student to develop a peer group that supported their academic goals. High ability, low income, urban minority youth partici pating in the Youth Scholars Program through Ohio State University identified three coping strategies for transitioning, which included individual strategies such as strateg Myers, & Smith, 2000). Some of the responsive practices recommended by middle level experts, such as teaming, common planning, and advisory, are designed to help ease the transition i nto middle school (Toepfer, 1990b). Interestingly, some of the practices advocated in the high school reform used to ease the transition of students from middle school to high

PAGE 67

67 or teams for ninth grade students and student teacher advisory (Hertzog & Morgan, 1999). Shoffner and Williamson (2000) have suggested a number of activities that smooth the transition to middle school, including summer bridge activities, interactions between staff in feeder patterns, buddy systems between older and younger students, and classroom guidance activities. Characteristics of the K 8 Model for Middle Level Students Although not a new idea, many school districts are beginning to believe that middle level students can be better served in a K 8 school design (Howley, 2002). The K 8 model is considered the norm in European countries that consistently outperform nation where elementary school students spend time in a middle or junior high school before 8 model has always been the norm for private schools, but now this trend is beginning to catch on in the public school sector (P aglin & Fager, 1997). Another variation of this configuration is a modified K 8 which offers multiple schools within a single building, each with their own course offerings, bell schedules, and unique programs (Toepfer, 1990b). Hough (1995) called this de school, which includes both elementary and middle grades. However, students in the middle grades have special programs, courses, and teachers with specialized training in working with early adolescents. In a study of over 100 middle le vel school administrators, McEwin, Dickinson, and Jacobson (2004) found that middle grade students in K 8 schools are more likely to have a comprehensive core curriculum than middle schools or junior high schools. For

PAGE 68

68 example, K 8 schools are more likely t o offer reading courses and instruction in basic skills. Of the schools surveyed, students in the K 8 schools were scheduled into core academic instruction, such as reading and math, an average of 30 minutes more per day than their peers in 6 8 middle scho ols. In addition, K 8 schools are more likely to offer independent study opportunities and use more flexible grouping practices. Schools with the K 8 grade configuration are also more likely to have a variety of organizational plans within the school, such as self contained classes and departmentalization of classes according to student need. The most common organizational design includes self contained or partially self contained sixth grade students and more departmentalized seventh and eighth grade class es. Advocates for the K 8 school design have theorized that there may be additional benefits to the K 8 grade configuration. The K 8 structure may foster more supportive interpersonal relationships (Offenberg, 2001). Some school district administrators be lieve that by shifting to K 8 schools, fewer students will be lost in the shuffle of the larger middle schools (Pardini, 2002). Others have reported that K 8 schools are perceived as safer than schools that house middle school students (Waugh, 2004; Weiss & Kipnes, 2006). Some researchers have suggested that the smaller school size of K 8 may actually be an advantage to implementing several developmentally appropriate practices, such as teaming and heterogeneous grouping (Coladarci & Hancock, 2002a; Hough, 2005; Offenberg, 2001; Yakimowski & Connolly, 2001). Another stated advantage of K 8 schools is that students have opportunities for leadership roles for longer periods of time, since they stay in the same school longer (Simmons & Blyth, 1987). In addition the increased number of years in one school

PAGE 69

69 provides a longer period of time for relationships between teachers, parents, and students to form (Offenberg, 2001). Alspaugh (1998) believed that the social structure and bonding that occurs when students are grouped together for many years may be more important than originally understood. Simmons and Blyth (1987) reported that students in K 8 schools have the added benefit of not facing the transition of moving to a new school until they are more developmenta lly suited for the change. In an analysis of students remaining in K 8 schools compared to students shifting to a middle school, the researchers found that students in K 8 schools had higher levels of social engagement, better attitudes toward school, and higher levels of self esteem. On the other hand, K 8 schools have been criticized for operating just like elementary schools, with no special attention given to the unique needs of students in grades 6 through 8 (Jenkins & McEwin, 1992). Ecker (2002) sugg ested that there may be a perception that K 8 schools lack the rigor needed for middle aged children. For example, algebra is taught less frequently in K 8 schools, because elementary teachers are rarely trained or certified to teach more advanced subjects Additionally, K 8 schools are usually smaller in size than consolidated middle schools which typically draw students from a number of elementary feeder schools. With fewer students, K 8 schools often have proportionally fewer resources which can result i n reduced offerings in exploratory classes, gifted and accelerated courses. Reduced resources can also mean limited support for students with disabilities, and fewer extracurricular activities (Beane & Lipka, 2006). Despite the critics, advocates for the K 8 school design firmly believe it is the next big idea in school reform. Communities that are making the shift from middle

PAGE 70

70 schools to the K 8 configuration offer a variety of reasons for the change. William t of Education, is convinced that a shift to K 8 can dramatically improve schools for a relatively low cost (Pardini, 2002). In Milwaukee, where the public school district has expanded the number of K 8 schools by 19 over a two year period, district person nel stated they were moving to a K 8 design to put schools back in the communities where the students live (Cook & Bloom, 2007; Gewertz, 2004). They argue that the new grade configuration will save on transportation costs and better anchor the schools in t he communities. Herman (2004) reported that K 8 schools can be more cost efficient because of reduced transportation expenses and from overhead saved from operating one building rather than two. Cost savings was not the impetus for the switch to K 8 schoo ls in Cleveland and Cincinnati, where the Facilities Master Plan includes over $1 billion for the conversion (DeJong & Craig, 2002). Reasons for the shift in these two large Ohio school districts included frustration with the high rates of absenteeism and suspension, overcrowded classes in the middle schools, and parent dissatisfaction (OPPAGA, 2005). Although the K 8 movement seems to be gaining steam in many urban areas, some researchers are concerned about the locations of the K 8s within the districts. For example, Gewertz (2004) found that the wealthier neighborhoods in Baltimore have converted more schools to the K 8 design than in more impoverished areas of the city. Balfanz et al. (2002) noted a similar trend in Philadelphia. In Cleveland, where it t akes a 75% majority of parent votes to convert to K 8, parent participation may be easier to garner in more affluent areas.

PAGE 71

71 In Miami Dade, the public school district has converted 26 schools to the K 8 configuration, with six conversions occurring the pas t two school years. When comparing school free and reduced price lunch rates, there appears to be a marked disparity between K 8 schools and traditional middle schools. For example, in comparing the percentage of schools with more than half of the students on free or reduced price lunch, 92% of the middle schools meet the qualification, in contrast to only 54% of the K 8 schools. To further exacerbate the disparity, five of the six new K 8 schools that converted or opened from 2008 to 2010 had less than hal f of the students on free or reduced priced lunch (Florida Department of Education (FLDOE), 2011a). Several researchers have warned that this continued trend could lead to greater segregation in schools (Beane & Lipka, 2006; Gewertz, 2004). In a case study of five failed K 8 school conversions in Philadelphia, Look (2002b) concluded that several factors must be present when attempting to successfully move to a new grade configuration as part of a district wide reform agenda. First, the district and affected schools must have consistent leadership who are committed to the new design. Secondly, there must be substantial stakeholder investment in the effected community. Finally, the current policies and practices must be revised to reflect the newly designed gr ade configurations. Student Achievement and Grade Configuration Because a number of urban districts are adopting the K 8 model, expending great financial and human resources in the process, it is important to understand what the research says regarding the relationship between grade configuration and student achievement. There are a number of studies that attempt to determine this relationship, particularly for schools that educate middle level learners.

PAGE 72

72 In the 1980s, the biggest debate regarding grade con figuration was whether sixth grade students belonged in the elementary school or the middle/junior high school. As the middle school movement gained momentum, much of the research focused on middle schools in comparison to junior high schools. Because K 8 schools were so few in number and largely consisted of rural, private, or parochial settings, the student achievement differences that existed within those schools could be largely dismissed because the student bodies were dissimilar to most other schools serving early adolescents. In 1995, the Third International Mathematics and Science Study revealed that fourth graders from the United States were doing well in mathematics against their counterparts from other countries, but the performance of our eighth grade students was well below the international standard (Peak, 1996). Researchers gained a new interest in finding out why student achievement in the elementary schools was significantly better than student achievement in middle and high school (Wren, 200 3). there has been a renewed interest in determining the relationship between grade configuration and student achievement, as more districts began to invest in reconfiguration to the K 8 model. This is a result of heightened accountability for student achievement and a scarcity of financial resources for schools. Administrators at both the district and school level are grasping at solutions to increase student performance in th e face of drastic budget cuts. The majority of studies that attempt to determine the relationship between grade configuration and student achievement have found no effect or little effect, particularly

PAGE 73

73 after controlling for demographic or school level fact ors. The following sections summarize the findings of those studies. Grade Configuration Studies with No or Little Effects In 1986, Sailor wrote a review of the research concerning the effects of grade configuration on student achievement. At that time, he reported that the research in the configuration had any effect on the student achievement of sixth grade students (p. 7). Instead, Sailor suggested that any effects of grade co nfiguration were small compared with other factors, such as school climate and after school supports, and stated that 1). In a national study of eighth grade students, researchers studied the effects of grade configuration on student achievement, controlling for a number of factors (Eccles et al., 1991). Overall, students in K 8 schools performed better academically than their 6 8 counterparts. When controlling for socioeconomic status, however, the researchers found that grade configuration had no effect on student academic performance. School size had some effect on academic achievement, with students in smaller schools outperforming students in larger schools. Vaccaro (2000) compared 119 schools in eastern Tennessee in a study of academic achievement in K 8 schools and middle schools over a three year period. Using student achievement data in reading, language, math, sc ience, and social studies from 57 middle schools and 62 K 8 schools, the researchers determined that there was no one best grade configuration for middle level learners. For example, middle school students outperformed K 8s in seventh grade math and scienc e, while K 8 schools

PAGE 74

74 showed greater achievement in sixth and eighth grade reading. In all areas, there appeared to be no significant differences between the two configurations. The researcher indicated that there may factors other than grade configuration that have a greater impact on student achievement, such poverty levels and teacher experience. In response to what they deemed the abysmal performance of their middle schools, Philadelphia embarked on a full scale conversion to a K 8 design as part of the Philadelphia Education Longitudinal Study, Weiss and Kipnes (2006) studied both academic and nonacademic outcomes from a random sample of 1,483 students from 45 Philadelphia public schools after their eighth grade year. The sample included students who had attended both K 8 and middle schools. The researchers found little difference in student performance based on school type. The only significant differences found were a greater se nse of anonymity and threat for students who attended middle schools. A year later, Byrnes and Ruby (2007) also compared eighth grade student achievement in Philadelphia Public Schools in a study for Johns Hopkins Center for Social Organization of Schools In this study, the researchers divided the K 8 schools into cohorts based upon the number of years they had been reconfigured. Schools were schools that had been con verted in the last one to five years. Using data from 40,883 eighth grade students in 95 schools, and controlling for population demographics and school characteristics, the researchers compared math and reading achievement for students in the old K 8s, th e new K 8s, and students in the middle school configuration.

PAGE 75

75 Overall, students in the old K 8 schools had significantly higher performance in both reading and math than students in the middle schools, but the researchers also found that the same setting ha d significantly fewer minority and high poverty students. In addition, the old K 8 schools had much small average grade sizes, lower rates of student mobility, less teacher mobility, and a greater percentage of certified teachers. Therefore, researchers co ncluded that the older K 8 schools served a significantly different population of students than those in the middle schools. In comparing the newer K 8s with middle school performance, the researchers found fewer differences. The new K 8 had lower levels o f student achievement and teachers who were less experienced than the middle school cohort. The new K 8s had smaller grade sizes than middle schools, but a greater percentage of minority students. After controlling for population demographics, the research ers found no significant differences in math achievement between the new K 8s and middle schools, and only slightly higher levels of reading achievement in the new K 8s. Old K 8s had the highest overall academic achievement in both reading and math. The mo st interesting finding is that after controlling for grade level size, the researchers found no differences between old K 8s, new K 8s and middle schools in terms of academic achievement, which suggests that academic achievement has more to do with grade l evel size than grade level configuration. analysis of 160 middle and K 8 schools in 4 counties in middle Tennessee. Using Terra Nova scores for students in grades six through e ight, Blair concluded that although there were differences in achievement, no one grade configuration had a clear

PAGE 76

76 advantage regarding student achievement overall. For example, in language arts achievement, sixth grade students had better performance in K 8 schools, while middle school students had the edge by eighth grade, and overall, there were no significant differences in the total group in language arts achievement. Similar findings were reported in mathematics, science, and social studies, with incons istent advantages at certain grade levels, with no significant overall effect. Dove, Pearson, and Hooper (2010) studied the relationship between grade span configuration and academic achievement using data from 281 Arkansas schools with sixth grade studen ts that had maintained their current grade configuration for the past three years. Using mathematics and reading scores from the Arkansas Benchmark 6 Examination, the researchers compared aggregate school level scores and found no relationship between grad e span configuration and academic achievement for sixth grade students. Grade Configuration Studies with Some Effects Most studies comparing student academic achievement during the 1980s showed a mixed bag of results between grade configurations for early adolescents. For example, in 1983, 62% of schools configured to a 6 8 grade span reported consistent academic improvement (George & Oldaker, 1985/1986). In a study of 18 schools in New York City, Moore (1984) compared reading achievement in K 8 schools to middle schools with similar racial and socioeconomic status demographics. He found that seventh and eighth grade students in K 8 schools had higher reading achievement scores when compared to students in middle schools. Similarly, using the Pennsylvania Ed ucational Quality Assessment for students in 330 schools in Pennsylvania, Becker (1987) found that the elementary setting is most advantageous for middle level students

PAGE 77

77 from low socioeconomic backgrounds, but had no effect on other groups, even when contro lling for other factors such as ability grouping and enrollment per grade level. In 1992, researchers compared academic achievement of students in schools with K 8, K 9 or 3 8 grade configurations with those that were configured 4 8, 5 8, or 6 8 (Wihry et al., 1992). In the study, the performance of eighth graders in 163 Maine schools was compared by grade configuration. After controlling for poverty and parent educational attainment, the researchers found that student achievement on the Maine Educational Assessment was from one third to a full standard deviation higher in the schools with the larger grade spans than those configured around the middle grades. A comparison of four enrollment groups in K 8 and K 12 rural Missouri school districts revealed si xth and eighth grade achievement levels higher than the state average (Alspaugh, 1995; Alspaugh & Harting, 1995). In these studies, it is difficult to discern whether the achievement of K 8 schools should be attributed to the rural nature of the district, rather than the configuration of the school (Alspaugh, 1995). Tucker and Andrada (1997) found in a survey of schools in Connecticut, schools that know they will be accountable for sixth grade results, such as K 6 and K 8 schools, show consistently better sixth grade results than 6 8 configured schools on the Connecticut Mastery Test The researchers attributed the higher performance to heightened feelings of accountability of teachers to their colleagues within the school and better fifth grade preparation In a study of student academic performance in rural schools in Louisiana, Franklin and Glascock (1998) studied achievement scores of sixth and seventh grade students in schools with elementary (grades K 6/7), middle/junior high (grades 6/7 8/9),

PAGE 78

78 seconda ry (grades 7/8/9 12), and unit (K 12) grade configurations. Sixth and seventh grade students performed equally well in an elementary configuration as in a K 12 unit configuration, and actually performed better than the sixth and seventh graders in a middle school setting. The researchers suggested that the lower achievement levels in middle school are due to a more fragmented school experience due to the additional transition. The researchers further stated that in many cases, the middle school design was c reated for fiscal and administrative purposes, rather than to meet the social and academic needs of early adolescents. Prior to the last decade, most K 8 schools were primarily located in rural areas (Epstein, 1990). In an attempt to study grade configurat ion on a more urban area, Offenberg (2001) conducted a comparison of student achievement in Philadelphia public schools, which had more recently converted some of its urban schools to a K 8 model. He found that students in the K 8 schools had better grade point averages and scored significantly higher on standardized tests than students in the middle school configuration, even when the data were controlled for poverty and race. Wren (2003) studied the effects of student transition on student achievement in a larger, urban inner city school district in the Midwest, with a 91% minority rate. The researcher sorted student achievement scores from the Michigan Education Assessment Program (MEAP) test from 232 schools by the number of transitions each student had experienced. Students in elementary, K 8, or K 12 schools were coded as having only one transition (home to school), middle school students were coded with two transitions (home to elementary to middle school), and high school students were coded with thre e transitions. The researcher found that the number of transitions was

PAGE 79

79 negatively associated with student achievement, and that achievement scores for elementary students were highest, regardless of the grade configuration. In her discussion, the researche r hypothesized that when students come to one large middle school from multiple elementary schools, there is more loss of achievement than when one elementary feeds directly into one middle school. Freeman (2005) used aggregated school achievement and att endance data, along with teacher and principal interviews to study the effect of grade configuration in North Carolina schools. Using all K 8, 6 8, and 7 9 configured schools in the state, Freeman found that K 8 schools had better attendance and student ac hievement on the North Carolina end of grade tests in reading and mathematics. The researcher credited fewer school to school transitions, greater continuity, and higher levels of personal support of the K 8 structure for the differences in attendance and achievement. In a follow up study of the conversion of middle schools to K 8 schools in Miami Dade County Public Schools, Abella (2005) was interested in the long term effects of grade configuration on student achievement. In the initial research for the f irst five K 8 conversion schools after the first year of implementation, researchers found that sixth graders in the K 8 schools outperformed nearby middle schools in student achievement had better attendance, and fewer out of school suspensions. In study ing the same students three years later, however, Abella found that by the eighth or ninth grade, those differences in student achievement levels had disappeared. There is a growing body of research that suggests that the transition to middle school may be particularly harmful to minority and high poverty students. Black (2009) researched the difference in student achievement by the configuration of grades for

PAGE 80

80 Alabama sixth grade students. In addition, he studied whether the differences varied by gender, ethnicity, and socioeconomic status. After researching data from 242 schools, the researcher concluded there were no differences in student achievement based on grade level configuration or by gender and grade configuration combined. One interesting findin g within the research was that black students and high poverty students in K 8 schools had higher achievement scores in math than their middle school peers. In New York City, Rockoff and Lockwood (2010) followed students from grades three to eight to dete rmine the effects of grade configuration on early adolescents. Some of the students transitioned to middle school, while others remained in the same K 8 school. The researchers found that the year the students transitioned to middle school, whether in the sixth or seventh grade, they experienced declines in both reading and mathematics achievement. Students who were already performing below grade level fared the worst in the transition, experiencing drops that were 50% greater than transitioning high achiev ing students during the first year alone, with a cumulative deficit of 200% in a three year period over high achieving students making the same transition. The researchers found no differences in funding or class size between K 8 and middle schools; howeve r, there was a significant difference in school size, with middle schools much larger than K 8s. Summary of Research Findings There are no consistent findings regarding the relationship between grade configuration and student achievement. Many of the studi es that do report differences also report other factors that may influence the outcomes, such as school size, class size, or demographic differences between the schools. Nearly every researcher indicated that any differences that did exist may have more to do with the practices

PAGE 81

81 within the building, which may or may not be a by product of the grade configuration within the school. Urban districts that are converting to a K 8 design are doing so without much long term evidence to suggest the conversions will produce the desired effects on student achievement. Other Student Achievement Factors for Early Adolescents In addition to grade configuration, there are a variety of other factors that may have an effect on student achievement for early adolescents. Some of the factors that have been pointed to by researchers include the following: ( a) socioeconomic factors, ( b) school size, ( c) academic rigor, and ( d) extracurricular activities. Socioeconomic Factors The largest body of research linked to student perfor mance at all grade levels is poverty index is the strongest predictor of achievement test scores (Beaton, Mullis, Martin, Gonzalez, Kelly, & Smith,1996; Bracey, 1997; Schmidt, Mc Knight, Jakwerth, Cogan, & Houang 1999). Barth (2001) reported that significant differences in middle level student achievement were linked to socioeconomic status, not grade span configuration. In a study of fourteen southern states, researchers found tha t low income and minority students are less likely to attend schools with highly qualified teachers, and teachers of high poverty students tended to score lower on indicators of instructional effectiveness (Cooney & Bottoms, 2003). Goldschmidt and Wang (19 99) have found that dropout rates are higher in schools with a greater number of students on free or reduce priced lunches. Using data from the National Education Longitudinal Study of 1988, Hough and Sills Briegel (1997) studied 40 consolidated schools m atched with 40 community based

PAGE 82

82 schools throughout the United States. In their research, they found that socioeconomic status, not school type, was the most significant factor in determining student achievement. Similarly, in a study of reading and mathemat ics achievement in 3,856 Illinois public schools, researchers found four relationships predictive of student achievement, with the poverty and minority rate having the strongest predictive value (Sutton & Soderstrom, 1999). In a study of 73 high schools i n New Hampshire, researchers examined math and reading achievement scores, post secondary aspirations, attendance, and percent of students taking college entrance exams in relation to district socioeconomic factors, including district unemployment rate, pe degree, and percent of students on free or reduced price lunch. The researchers found that socioeconomic factors account for much of the variation in school level outcomes of student achievement (Toukoushian & Ta ylor, 2005). School Size Combination schools such as K 8, K 12, and 7 12 tend to be smaller in size than junior highs and middle schools (Ecker, 2002; Renchler, 2002). Of all the configurations for middle level students, junior high schools tend to have th e largest enrollment, followed by middle schools (Eccles et al., 1991; Viadero, 1995). Howley (2002) has found that schools with the largest enrollments tend to have the fewest number of grade levels within the school. Bickel and Howley (2000) studied the joint influence of school size and district size with student performance of eighth and eleventh grade students in Georgia. Using data from 367 middle schools and 298 high schools, the researchers found that both school size and district size was predicti ve of student achievement. The researchers

PAGE 83

83 concluded that both small school size and small district size is more advantageous for achievement of low socioeconomic students. In Texas, smaller schools with broad grade spans have better academic achievement s cores and are more cost effective than larger schools (Bickel, Howley, Williams, & Glascock, 2001). In a study of 140 middle schools participating in the Michigan Middle Start Initiative, Mertens, Flowers, and Mullhall (2001) collected data from 2,400 tea chers and 30,000 students. Schools in the study were divided into three size categories: ( a) less than 500 students, ( b) 500 749 students, and ( c) more than 750 students. The researchers found that both categories of schools with fewer than 750 students ha d greater parental involvement, more common planning, and better instructional practices, which may lead to greater student achievement. The small grade level enrollment in combination schools such as K 8 and K 12 configurations make it impractical to impl ement some effective middle level practices, such as common planning and interdisciplinary teams of teachers (Offenberg, 2001; Epstein, 1990). As enrollments by grade level increase, schools experience more opportunities for grouping within grade levels (E pstein, 1990; Epstein & MacIver, 1990; Paglin & Fager, 1997). On the other hand, schools with a larger span of grades have greater opportunities for cross age grouping and parent involvement for a greater length of time. However, fewer classrooms per grade make it harder to match students to a teaching style that best suits their particular learning needs or to separate kids from each other when needed (Paglin & Fager, 1997). In rural areas, many schools districts are finding that consolidation may not be t he answer to their financial worries. Students in consolidated schools find a loss of

PAGE 84

84 student achievement, longer bus rides to school, reduced parental involvement, declining extracurricular participation, and less sense of ownership (Bickel, et al., 2001) Researchers are not advocating the abolishment of large schools, but rather the regrouping of teachers and students into smaller learning communities within those schools (Dunne, 2001; MacIver & Epstein, 1993). Researchers suggest that large middle schoo ls need to be made smaller to create better student adult relationships and to build longer lasting relationships with parents (Hough, 2005; Offenberg, 2001). The number of students included at each grade level alone does not determine whether instructiona l approaches are responsive to early adolescents (MacIver & Epstein, 1993; Dunne, 2001). Academic Rigor Middle schools have been accused of emphasizing social skills at the expense of academic rigor. For example, Lounsbury and Johnston (1988) stated that s ixth grade students are rarely challenged to think critically. Others have said that students in middle school are not offered enough challenging academic courses (MacIver & Epstein, 1993). Toepfer (1990b) advocates a curriculum rich in problem solving for middle level students. Middle level students should be made aware of the complex issues that will affect our society, and the middle school curriculum should include topics of future concerns. With the high stakes testing that accompanies No Child Left Be hind legislation, teachers are placing increased emphasis on high accountability areas, in particular reading and mathematics (Lounsbury & Vars, 2003). Even before the new law, however, MacIver and Epstein (1993) found that most middle schools emphasized p assive drill and practice over other more complex and varied cognitive tasks. The

PAGE 85

85 researchers also found that average to upper socioeconomic status students were more frequently offered advanced classes and richer instructional approaches than students fr om lower socioeconomic status backgrounds. There is a growing body of research that suggests placing students in rigorous courses has a positive effect on student achievement, particularly in mathematics. For example, in a study of over 3,100 eighth grad e students, Cooney and Bottoms (2002) found that students placed in a higher level math class than their ability like peers had a lower failure rate in mathematics. In addition, those same students are more likely to take more advanced mathematics courses in high school. Furthermore, students who take algebra in eighth or ninth grade are far more likely to take calculus in high school and enroll in postsecondary education. Gamoran and Hannigan (2000) also found that exposure to algebra in middle school prod uces as much achievement gains in low performing students as it does their high achieving peers. Conversely, Hoffer (1992) found that low performing math students who are enrolled in remedial or lower level math classes do not make significant gains, and t he remedial coursework is not beneficial. The implementation of rigorous performance standards may also have a positive effect on student achievement. For example, in a study of students enrolled in an earth science class in New York, students were told t hey needed to master 100% of the course competencies on the New York Regents Exam in order to receive credit for the course. Students were given specific information about each of the competencies. As a result of the stated expectations for learning, 71 ou t of 72 students met the goal (Nave, Miech, & Mosteller, 2000).

PAGE 86

86 Lee et al. (1999) studied the effects of social supports and academic press on the achievement levels of sixth and eighth grade students in the Chicago Public Schools. Academic press is the s tress placed on rigor and accountability, while social supports is the degree that relationships are formed among adults and students within the schools. The study examined Iowa Tests of Basic Skills scores of over 28,000 students and survey data from over 5,000 teachers. The researchers found that high levels of academic press alone resulted in greater achievement gains, and that academic press combined with social support resulted in the greatest achievement gains. The authors stressed that academic stres s provides pressure on both teachers and students to conform to specific standards. In addition, it provides specific direction for student work, letting both teachers and students know what they need to accomplish. Finally, academic press creates built in incentives to motivate teachers and students. Extracurricular Activities There is a modest body of research that suggests that students who are involved in academic, athletic, or recreational extracurricular activities have higher academic achievement. F or example, Schrieber and Chambers (2002) studied over 8,000 students in eighth and tenth grade as part of the National Education Longitudinal Study (NELS) of 1988. Extracurricular academic pursuits were more highly correlated with academic achievement. It is not clear whether extracurricular activities are a causal factor for increased academic achievement, or if academically successful students are more likely to participate in extracurricular activities.

PAGE 87

87 Grade Configuration and Other School Factors Rese archers have found that the grade configuration of a school may have an effect on factors other than just student achievement. The affected areas include: ( a) student behavior, ( b) self esteem, ( c) parental involvement, ( d) attendance, and ( e) teacher plac ement. Student Behavior Early adolescence is typically when a trend of poor behavior and diminishing academic achievement in school begins (Eccles et al., 1991). In a survey of 170 students majoring in education, only 9% indicated they were interested in middle grades education because of perceived disciplinary problems and perceptions about the attitudes of early adolescents (Carter & Carter, 2000). Some research suggests that peer influence for middle level learners may not be as strong as previously th ought. For example, Berndt, Laychak, and Park conducted an experimental study to determine the effects of peer influence on academic achievement motivation (1990). In the study, 118 students were matched with a close friend. Half of the pairs were assigned to an experimental group, and the remaining pairs were assigned to a control group. All students were independently administered a questionnaire that required them to choose between two courses of action when posed with a scenario related to academic moti vation. For example, students were asked to decide if they would attend a rock concert after they found out that an important exam had been scheduled for the following morning. After the first round of questionnaires, the matched pairs were brought togethe r for a discussion. The experimental group discussed their responses to the questionnaire, while the control group had a discussion unrelated to school topics. After the matched pair discussions, the students were once

PAGE 88

88 again administered the questionnaire. Although the experimental group did shift their responses to become more alike, there was not a dramatic change. More importantly, of the changes in decisions that did occur, friends tended to influence the responses in a more positive direction. In earli er research, George and Oldaker (1985/1986) found the middle school reorganization had a positive result on student behavior. In the study, the researchers identified a preliminary list of prospective exemplary middle schools using the 1982 Study of Well D isciplined Schools sponsored by Phi Delta Kappa and the 1983 Department of Education National Secondary School Recognition Program. The list was narrowed by a panel of ten experts in middle level education who researched each school to determine the implem entation of model middle school practices. In all, 130 middle schools that were deemed exemplary by fidelity to program implementation were included in the study. Participants, including teachers and administrators, from each school were administered surve ys. Of the respondents, 80% reported a significant reduction in referrals and suspensions after implementing the middle school model. In addition, 90% of the teachers reported a greater confidence in maintaining discipline in the middle schools. Teaming an d common planning were also linked with most consistent procedures for handling student class disruptions. As cited in a previous section, Abella (2005) conducted a longitudinal study in Miami Dade, where some of the public schools had reconfigured to a K 8 design. Students in the K 8 design received significantly fewer out of school suspensions during the middle years, with the most pronounced difference in the sixth grade. However, by the time both groups reached the ninth grade, there were no significan t differences in

PAGE 89

89 out of school suspension rates between the two groups. Similarly, Weiss and Kipnes (2006) found no significant difference in suspension rates of eighth grade students attending urban K 8 schools compared to their middle school peers. Cook et al. (2007) found different results in a study of sixth grade students in North Carolina. In a sub sample of 342 public schools that served sixth grade students, the researchers found that sixth graders in middle schools were more likely to be cited for disciplinary infractions that sixth grade students in elementary schools. Franklin and Glascock (1998) also reported decreased suspension rates for Louisiana sixth and seventh grade students in K 8 configured schools as compared to their middle school pee rs. Whether perceived or real, reducing behavioral disruptions is one of the frequently cited reasons school districts give for converting to the K 8 model. In Cleveland, Ohio, Superintendent Barbara Byrd Bennett stated that the shift from middle schools to the K cited fewer disciplinary problems as one of the major reasons for her controversial overhaul of the D.C. public school system, which included school closures and K 8 consolidations (Turque, 2008). Deputy Superintendent Robert Alfaro said that, while San Antonio schools were converted to K 8s for academic reasons, one of the positive outcomes has b een reduced disciplinary infractions (Reeves, 2005). Self Esteem needs is essential before beginning to work on their academic needs (Daniels, 2005; Lipsitz, 1984). Rapid physic al, social and emotional changes mean students need more

PAGE 90

90 flexible learning modalities which balance structure and choice (Ecker, 2002). Hough (1989, 1995) maintained that it is during this time period that giant gaps begin to emerge between children in ter ms of maturity. The developmental diversity of students in this age group makes it difficult to meet the needs of all students (NMSA, 1982, 1995). Early adolescence is a very important stage because it is during this time that children begin to determine their behavior as adults. Self concept, learning, interests, skills and values are all forming during this stage of development (National Middle School Association, 1995; NMSA & NAESP, 2000). For example, early adolescence is characterized by more abstract thinking, concern over sexual relationships, a desire for autonomy from adult control, and a greater concern for peer relationships (Eccles et al., 1991). It is also during this time that young adolescents begin to transition from an egocentric point of v iew to a more group centered way of thinking (Daniels, 2005). In terms of emotional development, early adolescence is a time when children begin to have feelings of diminished security (Lounsbury & Johnston, 1988). At this age, students will work hard to demonstrate their autonomy, but are still very dependent on significant adults. Early adolescents are very emotional and their social interactions greatly affect their school behavior. Most students in middle schools are very self conscious, sensitive, and vulnerable. Conflicting messages they receive from family, friends, and even the media can be very confusing (George & Alexander, 1993). Students moving into middle school have a great number of stressors, including concerns about navigating the school, adjustments to the organization of the school, and developing new relationships with peers (Wren, 2003). In earlier research,

PAGE 91

91 Simmons and Blyth (1987) found some evidence that smaller K 8 schools are better for the self esteem of early adolescents than lar ger junior high schools. Others believe that early adolescents have more opportunities for leadership roles, such as bus monitors and safety patrols, in K 8 schools (Pardini, 2002). Eccles et al. (1991) found that seventh grade students in junior high scho ols are actually given fewer opportunities for decision making than when they were sixth grade students in the elementary school setting. value decisions (Toepfer, 1990b). Sinc e students in K 8 schools stay in the same school for a longer period of time, the likelihood is greater of forming stronger adult student sense of belonging. In a r esearch study of over 90,000 students in 132 schools included in the National Longitudinal Study of Adolescent Health, Anderman (2002) found that students in urban schools had a lower sense of belonging and students in suburban schools. When correlated wit h student achievement data, the researcher found that a greater sense of belonging is related to higher grade point averages. Changing from a small protected elementary school to a large impersonal middle school can be detrimental to self esteem, particul arly for girls (Simmons & Blyth, 1987; Weiss & Kipnes, 2006). Sixth grade students in elementary settings have higher self ratings of appearance, sports ability, intelligence, and popularity. Students report that they like their grade level better and they perceive themselves to be less anonymous. Girls suffer losses in self esteem when the number of life events to which they are exposed increases. For example, when Simmons and Blyth compared the self esteem

PAGE 92

92 scores of girls transitioning to seventh grade, t he girls in the K 8 group scored higher than those transitioning to junior high school. When both groups transitioned to high school, the results were even greater, with the K 8 girls significantly higher. The researchers conclude that the effects of multi ple life changes simultaneously are negative and have prolonged effects, especially for females. MacIver and Epstein (1993) found that the differences in self esteem can be ameliorated by interdisciplinary teaming and teacher student advisories. They foun d that these two practices had positive effects on student bonding to teachers and resulted in positive feelings toward the school. Similarly, there were fewer reports of depression, anxiety, and aggressive behavior in schools with the highest level of imp lementation of middle school structures (Felner et al., 1997). Parent Involvement Compared to the junior high school model, the middle school configuration seems to encourage more parental involvement. In earlier research, George and Oldaker (1985/1986) measured attendance at open house and parent conferences, numbers of volunteer hours, and participation in school wide events and found increased levels of involvement in the middle school configuration. For the most part, however, parent participation fal ls off significantly in the middle grades (Viadero, 1995). Offenberg (2001) reports that middle school parents in 6 8 schools are less likely to come to the school because it is usually not as close to their home as the elementary school and because their children attend middle schools for only a few years. In Cincinnati, school officials reconfigured to K 8 schools after they found that leaving for private schools once they were of middle school age (Pardini, 2002). They

PAGE 93

93 found that the parents in Cincinnati were more comfortable keeping their children closer to home in their neighborhood schools, but were uneasy sending their children to the larger middle school. As a si de benefit, school officials found that parents were more inclined to remain involved when their children attended a K 8 school. Attendance Although the results are mixed, research seems to suggest that attendance is better in K 8 schools. In a study of Cleveland public schools, sixth graders in K 8 schools posted better attendance than the students still in the 6 8 grade configuration (Pardini, 2002). Across a four year period, Abella (2005) found that students in the K 8 grade configuration increased t heir absenteeism at a slower pace than did their peers in the middle school setting. Even when the groups of students entered high school, the students who had attended the K 8 schools had significantly lower absentee rates. Similar results of improved att endance for students attended K 8 schools were found in Louisiana, North Carolina, Milwaukee, and Miami (Abella, 2005; Cook & Bloom, 2005; Franklin & Glascock, 1998; Freeman, 2005). However, Weiss and Kipnes (2006) studied eighth grade attendance rates in a large urban school district and found no significant difference in attendance rates between the K 8 cohort and their middle school peers. Middle Level Teachers One of the most important elements of a successful school, regardless of the grade levels or s chool type, is the quality of the instructional staff. In fact, many researchers theorize that the qualifications of the staff within each type of school have more to do with any differences that exist than the grade configurations of the schools themselve s (Gomez, 2010; McPartland et al., 1987; Mertens & Flowers, 2003a). In a

PAGE 94

94 national random sample of 300 schools, McEwin & Alexander (1990) found that the most important difference in the educational value received by students is the quality of the teacher. Unfortunately, for many middle level students, the quality of teaching may be lacking. K 8 schools may have the advantage in this area, since middle schools have a harder time recruiting and retaining quality professionals (Pardini, 2002). On the other ha nd, students in K 8 schools are unlikely to have qualified teachers in algebra, geometry, and world languages, because those schools tend to be smaller, requiring teachers with more general teaching certifications (Gomez, 2010, Juvonen, Le, Kaganoff, Augus tine, & Constant, 2004, Yakimowski & Connolly, 2001). Before changing grade configurations and school structures, school personnel must be properly prepared to work with early adolescents (NMSA, 1982, 1995). Reed, Dickinson and Cuozzo (2000) noted that the re is very little explicit training for middle school teachers, despite the specific developmental needs of early adolescents. Some colleges and universities have instituted programs or departments specifically designed to train middle school teachers and many states now have licensure requirements to teach middle level students (Lounsbury & Vars, 2003). In a survey of 209 members of the Tennessee Assoc i ation of Middle Schools, Boren (1997) found that as a whole, the members favored a special endorsement f or middle level educators. Professional organizations, such as the Association for Middle Level Educators have been established to provide support for middle level educators, and both the National Association of Secondary School Principals and National

PAGE 95

95 As sociation of Elementary School Principals include some emphasis on middle level education (Lounsbury & Vars, 2003). Regarding ongoing professional development, Toepfer (1990b) found that staff development in middle schools was more likely to be provided b ased on teacher needs than in other grade configuration. K endorse staff development in early adolescence and in specific strategies for the middle 443). In a more recent st udy, Schmitt (2004) found that middle school teachers were more likely to be highly engaged in professional development than teachers in K 8 or K 12 schools. Using classroom teacher observations matched with student scores, Miller (2008) found that math te achers with secondary certification had significantly higher levels of stage environment fit in classroom management, climate and emotional support than those with elementary certification. There is mounting evidence that subject area expertise may be a f actor in student achievement for middle level learners. For example, in researching the sharp decline of eighth grade math scores on an internationally benchmarked exam, the TIMSS 1999 study reported that eighth grade students in the United States are less likely to have a teacher certified in math or math education than countries outperforming the U.S. (Mullis et al., 2000). In a study of 271 middle schools in Pennsylvania, Lutz (2007) found that eighth grade students who had teachers with secondary mathem atics certification scored higher on the math portion of the Pennsylvania System of School Assessment (PSSA) than students who were instructed by teachers with elementary certification or had an alternative certification. In a meta analysis of 5 studies, S parks

PAGE 96

9 6 (2004) found that student achievement gains in both math and reading were significantly higher when students were taught by fully certified teachers. Darling Hammond (2000) similarly found a positive correlation between student math scores on the Nat ional Assessment of Educational Progress and instructors with both a subject area major and full state certification. This is important because sixth grade students in middle school are more likely to be instructed by subject area experts than their peers in K 8 configured schools (Bedard & Do, 2005). While many middle schools use the team approach, middle school teachers are still overwhelmingly considered subject area purists (White, 1993). Pardini (2002) reports that secondary school trained teachers t ypically have a focus on content discipline and sometimes do not even know which teachers their students go to for other subjects. Teachers working in K 8 schools have usually been trained to teach elementary students and have a more student centered appro ach to teaching. Additionally, K 8 teachers are more accustomed to teaming, planning, and working with the same groups of children. Mertens and Flowers (2003b) found that combining the effects of teacher certification with the practice of interdisciplinary teaming increased student achievement scores. Teacher experience may also play a critical role in student achievement results. In a study of 2,000 students and over 200 middle school teachers, teacher experience was found to be the strongest predictor of overall student achievement scores (Johnson, T., 2005). Richardson (2008) studied teacher qualifications and mathematics achievement in Montgomery, and found that students who were instructed by teachers with five or more years experience had better gains than their less experienced

PAGE 97

97 counterparts. In a study of 282 middle school mathematics teachers, Swan (2006) discovered that one third were in their first year of teaching and only half had more than three years of experience. The researcher also found tha t teachers with higher seniority, advanced degrees, or math certification had better student performance. This is important because Pardini (2002) reported that K 8 schools have an easier time recruiting and retaining teachers. Thompson (2007) conducted a case study in one New York City school to research why some teachers left and others stayed. The researcher found in this case, teacher retention was related to perceived support from colleagues and sense of motivation teachers got from working with their students. Summary In 1995, the National Middle School Association issued the following charge in its landmark work, This We Believe : (p. 16). This was not a move away from the social support structures emphasized in their earlier work; the education of early adolescents has always been the primary mission of this organization. In July of 2011, NMSA changed its name to the Association fo r Middle Level Education (AMLE) to more accurately reflect the composition of its membership. On their updated website, the organization explains, who work with kids in grades 5 9, regardless of the grade configurations of the schools that house them, which acknowledges that middle schools are not uniquely capable of educating early adolescents. The United States Department of Education (2003) listed four principles of education reform, including: ( a) stronger accountabil ity, ( b) expanded flexibility and

PAGE 98

98 more localized control, ( c) more choices for parents, and ( d) research based instructional methods. The research based practices that are associated with higher student achievement for middle level learners include a rigor ous curriculum, challenging lessons and assignments, highly qualified teachers, and adequate resources for instruction (Haycock & Ames, 2000). Early adolescents need both academic rigor and social supports. This combination is predictive of student achiev ement regardless of student demographics (Lee & Smith, 1995). This is supported by a recent study comparing the relationship between grade configuration and student achievement. Through classroom observation, focus groups, and data collection, the research er found several common elements in schools with stable or increased student achievement, including consistent application of the following practices: 1. Rigorous and rele vant curriculum and instruction; 2. Focused and directed prof essional development; 3. A learni ng community structure comprised of collaborative teams meet ing together on a regular basis; 4. A student advisory program and strong tr ansition plans for all students; and 5. A student centered focus (Cooksley, 2010). This research is consistent with that of He ller et al. (2003), who reported a list of research based best practices to enhance academic achievement for students in the middle grades, which included: ( a) an accelerated, rich curriculum, ( b) high academic standards, ( c) student engagement in new skil ls that are relevant, ( d) strong family connections, ( e) grouping all students with a caring adult, ( f) coordinating curriculum and sharing data, and ( g) highly qualified teachers.

PAGE 99

99 There is not an overabundance of research on the effects of grade configura tion on middle level students, particularly taking into account the implementation of developmentally responsive practices advocated by AMLE (Howley, 2002; Pardini, 2002). Of the research studies conducted, there is no clear relationship between student ac hievement and grade configuration alone. Therefore, it may be a mistake to look at big structures, such as grade configuration, when it may be the practices within those structures that have the bigger effect. One common thread through all of the literatur e concerning grade span is that the excellence of a school depends on more than just the configuration of its grades. Beane and Lipka (2006) recognize that high quality instruction and learning for early adolescents can exist in a variety of grade spans. Lipsitz et al stated that the issue is Significant progress has been made in the journey to provide young adolescents with a developmentally responsive education. However, we are only halfway up the mountain, with the most important and perhaps most difficu lt part of the climb remaining (p. 5).

PAGE 100

100 CHAPTER 3 RESEARCH DESIGN AND METHODOLOGY To date, research has offered a mixed bag of results concerning the best organizational structure for early adolescents. Middle school proponents believe that adolescents flourish in age appropriate settings that implement the developmentally appropriate practices of the middle school design. Other researchers, however, argue that the elimination of a transition from one level to the next is more beneficial to middle level learners. Little research exists to explain the effects of K 8 grade configurations on stude nt achievement in large urban districts, because until recently most of the K 8 schools were either private/parochial or located in more rural areas of the country. Despite the lack of research, many urban school districts are eliminating or reducing middl e schools in favor of the K 8 configuration. Since pockets of K 8 schools are emerging in urban areas, it is now possible to study the effects of the new design with its middle school counterpart. The purpose of this study is to compare, when controlling f or minority rate and SES, the academic achievement levels in reading and mathematics of sixth grade students enrolled in K 8 urban schools to those of students who attended a traditional 6 8 urban middle school for students. In addition, the study will mea sure the degree of implementation of exemplary practices determined by the National Middle School Association (NMSA, 1982; NMSA, 1995) for the two grade configurations in relation to sixth grade student achievement. Finally, this study determines if any of those practices are predictive of student achievement in relation to the two grade configurations.

PAGE 101

101 Research Proposal Research Questions This study will explore the following questions: Question 1: Is there a relationship between any of the exemplary pract ices identified by NMSA /AMLE, including ( a) interdisciplinary teaming, ( b) common teacher planning time, ( c) heterogeneous grouping, ( d) student advisory, ( e) exploratory courses, and ( f) looping, and student achievement as measured by student gains on t he Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students in K 8 and 6 8 middle schools in large urban school districts when controlling for SES and minority rate ? Question 2: Are there differences in levels of implementation of NMSA/AMLE identified exemplary practices in urban K 8 configured schools when compared to 6 8 configured middle schools? Question 3: Are there differences in academic achievement in reading and mathematics for six th grade public school students in K 8 configured schools when compared to sixth grade students in 6 8 configured middle schools, as measured by the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system? Question 4: Is there a relationship between the level of implementation of exemplary middle level practices and academic achievement as measured by the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students enrolled in urban K 8 configured schools compared to similar students in 6 8 middle schools?

PAGE 102

102 Hypotheses The following null hypotheses guided the investigation: Ho 1 : There is no relationship between the implementation of interdisciplinary teams a nd student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in an urban school district when controlling for school demographic factors of school size, grade level size, minority rate, SES rate, LEP rate, and ESE rate Ho 1 1: Interdisciplinary Teaming Ho 1 1 a : Reading Achievement Ho 1 1 b : Mathematics Achievement Ho 1 2: Common Teacher Planning Time Ho 1 2 a : Reading Achievement Ho 1 2 b : Mathematics Achievement Ho 1 3: Heterogeneous Grouping Ho 1 3 a : Reading Achievement Ho 1 3 b : Mathematics Achievement Ho 1 4: Student Advisory Ho 1 4 a : Reading Achievement Ho 1 4 b : Mathematics Achievement Ho 1 5: Exploratory Courses Ho 1 5 a : Reading Achievement Ho 1 5 b : Mathematics Achievement Ho 1 6: Looping

PAGE 103

103 Ho 1 6 a : Reading Achievement Ho 1 6 b : Mathematics Achievement Ho 2 : There are no significant differences in the implementation of exemplary middle level practices in schools configured as K 8 compared to traditional middle schools with a 6 8 grade configuration. Ho 2 1: Interdisciplinary Teaming Ho 2 1 a : 6 8 Grade Configuration Ho 2 1 b : K 8 Grade Configuration Ho 2 2: Common Teacher Planning Time Ho 2 2 a : 6 8 Grade Configuration Ho 2 2 b : K 8 Grade Configuration Ho 2 3: Heterogeneous Grouping Ho 2 3 a : 6 8 Grade Configuration Ho 2 3 b : K 8 Grade Configuratio n Ho 2 4: Student Advisory Ho 2 4 a : 6 8 Grade Configuration Ho 2 4 b : K 8 Grade Configuration Ho 2 5: Exploratory Courses Ho 2 5 a : 6 8 Grade Configuration Ho 2 5 b : K 8 Grade Configuration Ho 2 6: Looping Ho 2 6 a : 6 8 Grade Configuration Ho 2 6 b : K 8 Grade Configuratio n

PAGE 104

104 Ho 3 : There are no significant differences in student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in a K 8 school compared to the same students in a traditional 6 8 middle school. Ho 3 1: Reading Achievem ent Ho 3 1 a : 6 8 Grade Configuration Ho 3 1 b : K 8 Grade Configuration Ho 3 2: Mathematics Achievement Ho 3 2 a : 6 8 Grade Configuration Ho 3 2 b : K 8 Grade Configuration Ho 4 : There is no relationship between exemplary middle school practices and student achievemen t gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students who attend K 8 schools implementing exemplary middle level practices compared to the same students who attend traditional middle schools implementing the practices at th e same level Ho 4 1: Interdisciplinary Teaming Ho 4 1 a : 6 8 Grade Configuration Ho 4 1 a 1 : Reading Achievement Ho 4 1 a2 : Mathematics Achievement Ho 4 1 b : K 8 Grade Configuration Ho 4 1 b1 : Reading Achievement Ho 4 1 b2 : Mathematics Achievement

PAGE 105

105 Ho 4 2: Common Teac her Planning Time Ho 4 2 a : 6 8 Grade Configuration Ho 4 2 a 1 : Reading Achievement Ho 4 2 a2 : Mathematics Achievement Ho 4 2 b : K 8 Grade Configuration Ho 4 2 b1 : Reading Achievement Ho 4 2 b2 : Mathematics Achievement Ho 4 3: Heterogeneous Grouping Ho 4 3 a : 6 8 Gr ade Configuration Ho 4 3 a 1 : Reading Achievement Ho 4 3 a2 : Mathematics Achievement Ho 4 3 b : K 8 Grade Configuration Ho 4 3 b1 : Reading Achievement Ho 4 3 b2 : Mathematics Achievement Ho 4 4: Student Advisory Ho 4 4 a : 6 8 Grade Configuration Ho 4 4 a 1 : Reading Ac hievement Ho 4 4 a2 : Mathematics Achievement Ho 4 4 b : K 8 Grade Configuration Ho 4 4 b1 : Reading Achievement Ho 4 4 b2 : Mathematics Achievement Ho 4 5: Exploratory Courses Ho 4 5 a : 6 8 Grade Configuration

PAGE 106

106 Ho 4 5 a 1 : Reading Achievement Ho 4 5 a2 : Mathematics Ach ievement Ho 4 5 b : K 8 Grade Configuration Ho 4 5 b1 : Reading Achievement Ho 4 5 b2 : Mathematics Achievement Ho 4 6: Looping Ho 4 6 a : 6 8 Grade Configuration Ho 4 6 a 1 : Reading Achievement Ho 4 6 a2 : Mathematics Achievement Ho 4 6 b : K 8 Grade Configuration Ho 4 6 b 1 : Reading Achievement Ho 4 6 b2 : Mathematics Achievement Limitations The study was limited to the following conditions: 1. The study was limited to the Florida Comprehensive Achievement SSS data in reading and mathematics reported to the state of Florida for the 2009 2010 and 2010 2011 school years. 2. Although the researcher recognizes that sixth graders can be educated in a variety of grade level configurations, this study was limited to two configurations: ( a) sixth grade students in schools with sixth throug h eighth grade, and ( b) pre kindergarten or kindergarten through eighth grade configurations. 3. The study was limited to a comparison of public K 8 and middle schools in urban districts larger than 100,000 students in the state of Florida participating in th e state accountability system and may or may not be representative of other K 8 and middle schools in the United States. 4. The extent to which the survey responses by individuals in the respective schools represent the actual practices within the schools may limit the quality of the findings.

PAGE 107

107 5. The study was limited to only those students who had achievement scores for both the 2009 2010 and the 2010 2011 school years. In addition, the study only included students in the K 8 design who had achievement data for the previous school year within the same K 8 school. Therefore, this study fails to account for the achievement of students who transfer to or from different school systems. Schools with high mobility rates would tend to confound the number of school to sc hool transitions. Delimitations The following were delimitations to the study: 1. The study is delimited to the data offered by the research and evaluation departments provided by the identified districts of the participating schools. 2. Only middle schools and K 8 schools identified by the Florida Department of Education ( FLDOE, 2011) were included in this study because an ample number of K 8 and comparable middle schools were available in that region, and all students took identical achievement tests. 3. K 8 and m iddle schools identified as charter schools, magnet schools, or schools of choice were not included in this study. 4. Achievement scores were limited to gains from fifth to sixth grade only. 5. Information from educators within the school about implementation o f practices was delimited to a single set of questions related to degree of implementation provided in an online or paper survey. Assumptions The study also makes the following assumptions: 1. The study was based on the assumption that the student achievement data were appropriate measures for comparing academic achievement of middle schools and K 8 schools. 2. A further assumption was that comparable kind and quality of curriculum and instruction were present in both configurations of schools. Data Sources Popul ation and Sample This research focuses on the academic achie vement of sixth grade students by differences in the implementation of model practices, by grade configuration, and by the

PAGE 108

108 combination of both fac tors. Comparison groups include students in a K 8 grade configuration in large urban sc hool districts and sixth grade students in traditional 6 8 middle schools with similar student demographics. Schools a re also classified by their implementation of the following model practices: ( a) interdisciplinary te aming, ( b) common teacher planning, ( c) advisory groups, ( d) heterogeneous grouping, ( e) exploratory courses, and ( f) looping. For accountability purposes, all kindergarten through twelfth grade public schools within the state of Florida are identified one of four types: 1 Elementary School 2 Middle School 3 High School 4 Combination School combination schools, middle/high combination schools, virtual schools, an d other configurations and settings that can not be classified in the first three categories. K 8 designation. A list was generated from the Florida Department of Education S chool 2011a). From the list, schools with middle/high or junior/senior in the school name were eliminated from consideration for this research. For the purpose of defi ning a large urban district, the researcher chose school districts with student populations over 100,000. Of the 67 public school districts in Florida, only seven had student enrollments over 100,000 for the 2010 2011 school

PAGE 109

109 year. All of the seven large ur ban districts had schools id entified as combination schools; Florida, charter and magnet schools are considered schools of choice, and parents must actively select to enroll t heir children in these special programs. Since charter and magnet schools may constitute a special population by virtue of selection factors, either on the part of the parent or the school program, these types of schools were eliminated in this research st udy. Charter schools were specifically identified on the list generated from the School Accountability webpage. Magnet schools were identified via a search of district s had K 8 schools that fit the initial research criteria. After the preliminary list was pared down to the non magnet, non charter schools in districts over 100,000, each school remaining on the list was researched individually. The exact grade configurat ion within each school was determined via school and district websites. Only schools that housed grades K 8 or PreK 8 were included in the K 8 category for this study. After all of the schools not meeting specified criteria were removed, there were 32 rema ining K 8 schools during the 2010 11 school year that met the research conditions for this study. The accountability data provided on the FLDOE website also included data 2011 school y ear. The researcher used these data to match each of the identified K 8 schools with a traditional middle school within the same district. At the advice of Dr. Jerry Valentine from the Middle Level Leadership Center at the University of Missouri Columbia, schools were matched first by the SES indicator, and then by minority rate, since SES

PAGE 110

110 seems to have a more profound effect on achievement level than minority status. This was confirmed by the work of Offenberg (2001), who found that socioeconomic variables had a much more significant effect on student achievement than did the ethnic /racial background variable. The mean difference in Free or Reduced Priced Lunch rate between the total enrollment of the K 8 and matched 6 8 school was 9.24 percentage points, a nd the mean difference in Minority Rate was 0.52 percentage points. Overall, the mean Free or Reduced Priced Lunch rate was 57.61% for the K 8 cohort and 66.85% for the 6 8 group. The mean minority rate was 83.64% for the K 8 schools and 83.15% for the 6 8 schools. The population of this study consisted of 32 K 8 schools and 32 middle schools with a 6 8 grade configuration within urban school districts with a student population over 100,000 located in Florida. All public, non charter K 8 schools within the state were included in the research. Each middle school was selected by matching minority rates and SES rates as reported by the Florida Department of Education with each of the K 8 schools. For the purpose of this study, the student sample consisted of 4 ,230 sixth grade students in K 8 schools, and 10,523 sixth grade students in the matched middle schools. The 32 K 8 schools had a total enrollment of 37,782 in 2010 for an average of 1,145 students per school. Enrollment in the 32 6 8 configured schools wa s 33,846, for an average of 1,026 students per school. Although the average enrollment at K 8 schools was slightly higher than the traditional 6 8 middle schools, the sixth grade cohort was much smaller due to the increased number of grades in K 8 schools. The K 8 schools had an average of 132 students in the sixth grade, while the 6 8 schools had

PAGE 111

111 an average of 322 sixth graders per school. Only students with fifth grade FCAT scores during the 2009 2010 school year and sixth grade FCAT scores during the 201 0 2011 school year were used in the study. Additionally, students who moved to or from one of the sample schools or were retained in sixth grade from the previous school year were removed from the sample. The FCAT data was requested directly from the Resea rch and Evaluation departments of each of the four Florida school districts used in this study. When submitting the data, districts were asked to include FCAT developmental scale scores for all sixth grade students from the 2010 and 2011 FCAT administratio ns. In addition, districts were asked to provide demographic information including free or reduced lunch status, ethnicity, gender, LEP status, Special Education status, and school attended by the student the previous year. Permission to conduct research in the four large Florida school districts included in this study was obtained from the University of Florida Institutional Review Board. Since all student identification information was deleted, and this study used existing data, there was no direct conta ct with students. Therefore, informed consent forms for students were not required. However, informed consent forms were required for the learning leader survey participants. To determine the extent in which exemplary practices were implemented in each of the schools, the building principal was asked to identify five learning leaders within the school. Learning leaders were also identified by school websites by title, such as Each of the part icipants was sent via letter or email a link to an electronic survey. Participants could

PAGE 112

112 also opt to complete a paper copy of the survey, by request. The survey questions were developed by the National Association of Secondary School Principals (Valentine, et al 2002) as a part of a national study of effective leadership in middle level schools. Only the survey questions pertaining to the level of implementation of exemplary middle level practices and participant demographic information were used in this study. Data Collection and Instrumentation The Florida Comprehensive Achievement Test (FCAT) was used as the measure for student academic achievement in reading and mathematics for this study The FCAT is a criterion referenced test administered annually t o Florida students in grades three through ten in reading and mathematics (Flori da Department of Education, 2011 a). Additionally, students in publicly funded schools are administered a state writing assessment in grades four, eight, and ten, and a science assessment in grades five, eight, and eleven. Because the writing and science assessments are not administered annually, their gain scores were not used in this study. The in reading and mathematics are tested at each grad e level, with increasing levels of cognitive complexity with each grade level Student scores on the SSS portion of the test are expressed as proficiency levels from one to five for general reporting and also as scale scores, which are translated to a Dev elopmental Scale Score (DSS). The DSS is used to track individual student progress from year to year. In addition to determining individual student performance, the gain scores on the DSS are also used to measure school and district progress under the stat e accountability program. The survey used to determine the level of implementation of model practices was developed by the National Association of Secondary School Principals (NASSP) for two

PAGE 113

113 national studies regarding successful middle level leadership an d effective practices for middle level learners (Valentine, Clark, Hackmann, & Petzko, 2002; Valentine, Clark, Hackmann, & Petzko, 2004). In preparation for the 2002 study, the research team developed an extensive survey instrument containing 170 items. Th e survey is divided into sections, including school and principal demographics, school leadership, school programs, school reform and professional development, instructional practices, curricular and co curricular programs, school organization, and leaders hip issues. Over 14,000 middle level principals participated in the 2002 study. Since this research focuses mainly on the implementation of effective middle level practices, only the portions of the survey corresponding to those practices were used. Data A nalysis The primary statistical analyse s program s used in this researc h are SPSS Statistics, version 19 .0 ( SPSS, Inc. 2010 ) and SAS 9.2 (SAS Institute Inc., 2008) The following sta tistical tools will be used to answer the research questions: Ho 1 : Mult iple Regression There is no relationship between the implementation of interdisciplinary teams and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in an urban school district when controlling for sch ool demographic factors of school size, grade level size, minority rate, SES rate, LEP rate, and ESE rate These hypotheses compare FCAT DSS achievement gain scores of students in the sample who were in the fifth grade in 2009 2010, and then in sixth grade the following year. Multiple regression using stepwise

PAGE 114

114 regression was used to determine the relationships between the six exemplary practices predictor variables and student gain scores. Backward elimination was used to eliminate the variables that had th e least effect on R 2 until only the most significant predictive variables remained. Ho 2 : Test of Differences There are no significant differences in the implementation of exemplary middle level practices in schools configured as K 8 compared to traditiona l middle schools with a 6 8 grade configuration. This hypothesis compares the degree of implementation of the exemplary middle level practices, including interdisciplinary teaming, common planning, advisory, heterogeneous grouping, exploratory programs, an d looping, of K 8 schools with 6 8 schools, as reported by five learning leaders within each of the two types of schools. A t Test for independent samples was used to compare the means for each of the exemplary practices in each type of school at the .05 l evel of significance. Ho 3 : Test of Differences There are no significant differences in student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students who remain in the same K 8 school from the previous year, when c ompared with students who transition from fifth grade to a 6 8 configured school.

PAGE 115

115 These hypotheses compare the same group of students who were in the fifth grade in 2009 2010 and then in sixth grade the following year. The first group of students remained in the same K 8 school; the second group of students transitioned from an elementary school in fifth grade to a 6 8 configured middle school. T tests for independent samples were calculated to compare FCAT SSS Developmental Scale Score achievement gain mea ns from the 2010 to the 2011 administration for sixth grade students by school type at the .05 level of significance. Ho 4 : Hierarchical Linear Modeling (HLM) There is no relationship between exemplary middle school practices and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students who attend K 8 schools implementing exemplary middle level practices compared to the same students attend traditional middle schools implementing the practices at the same level This hypothesis determines the degree to which the implementation of the exemplary middle level practices in each of the two sixth grade settings is predictive of academic achievement. Hierarchical Linear Modeling (HLM) was used to determine the correlat ion between the six exemplary practices predictor variables and the two grade configurations: K 8 and 6 8. HLM was chosen as a statistical tool because of the multi layer dimension of the data. HLM has the ability to detect variations in

PAGE 116

116 regression relatio nships by groups, and to test for main effects and interactions within and between levels of data (Willms,1999). Table 3 1. List of instructionally responsive practices for middle level learners Indicator Description Interdisciplinary Teams Teams of 2 t o 5 teachers that share a common group of students, and are located in close proximity to each other Common Teacher Planning Teams of teachers with a common group of students are assigned to a block of time to plan together on a daily basis Exploratory Courses Courses that provide required curricular opportunities for students to explore abilities, talents, and interests through diverse elective opportunities Advisor Program Student advisor activities that are regularly scheduled for fifteen minutes o r more on a daily basis Heterogeneous Grouping Scheduling students of mixed abilities into courses together in order to create a relatively even distribution of students in each classroom Exploratory Program Opportunities outside of the regular graded coursework that provide leadership roles, exploration in special interests, and socialization for students Looping Also called student teacher progression or multi year teaching, looping is the practices of keeping a small group of teachers together wit h the same group of students for two or more years. (Valentine et al., 2002)

PAGE 117

117 CHAPTER 4 ANALYSIS OF DATA Presented and discussed in this chapter are the statistical analyses of data collected in this study. The purpose of this study was determine if an y model middle level practices, including interdisciplinary teaming, common teacher planning, heterogeneous grouping, advisory groups, exploratory courses, and looping, contributed to the academic achievement of sixth grade students in urban school setting s. The study also compared the level of implementation of the model practices to determine which grade configuration setting was most likely to fully implement the practices. In addition, the study compared student achievement of sixth grade students in th e K 8 setting with that of students in a 6 8 grade configuration. Finally, the study examined the combined effects of model practices and grade configuration on academic achievement levels in reading and mathematics for sixth grade students enrolled in K 8 urban schools to those of students who attended a traditional 6 8 middle school in large urban school districts. To answer these questions, student performance data from the state assessment for the two prior years, along with student demographic informat ion, were requested from each of the participating school districts. Schools that house students in kindergarten through grade 8 and middle schools with grades six through eight were examined using sixth grade Florida Comprehensive Assessment Test data in both reading and mathematics. This assessment is criterion referenced and designed to measure student mastery of the Florida Sunshine State Standards. Further analyses were conducted to determine the degree of implementation of model level practices within each of the schools in the sample. These data were

PAGE 118

118 collected by use of an educator survey administered to a sample of administrators, guidance counselors, and teachers in each of the schools within the sample. Responses for each sample school site, by gra de configuration, were aggregated to determine a consensus regarding the degree of implementation by both school and grade configuration In addition, the study analyzed the effects of grade configuration and the implementation of model level practices on student achievement among sixth grade students when controlling for demographic factors including disability rate, LEP rate, minority rate, and free/reduced price lunch rate. Thus, the researcher sought answers to the following research questions: 1. Is ther e a relationship between any of the exemplary practices identified by NMSA/AMLE, including (a) interdisciplinary teaming, (b) common teacher planning time, (c) heterogeneous grouping, (d) student advisory, (e) exploratory courses, and (f) looping, and stud ent achievement as measured by student gains on the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students in K 8 and 6 8 middle schools in large urban school districts, when controllin g for school and student demographic factors? 2. Are there differences in levels of implementation of NMSA/AMLE identified exemplary practices in urban K 8 configured schools when compared to 6 8 configured middle schools? 3. Are there differences in academic achievement in reading and mathematics for sixth grade public school students in K 8 configured schools when compared to sixth grade students in 6 8 configured middle schools, as measured by the Florida Comprehensive Tests of Achievement Sunshine State Sta ndards (FCAT SSS) assessment system? 4. Is there a relationship between the level of implementation of exemplary middle level practices and academic achievement as measured by the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students enrolled in urban K 8 configured schools compared to similar students in 6 8 middle schools?

PAGE 119

119 Presentation of the Data Descriptive Data The population in this study was comprised of students, teachers, guidance c ounse lors, and administrators from 32 middle schools with a 6 8 grade level configuration and 32 K 8 schools. The student population included a total of 14,753 sixth grade students from four large urban school districts. Of those students, 4,230 students w ere enrolled in K 8 schools, and 10, 523 students were in the matched middle schools. From this student sample, 400 K 8 students and 998 6 8 students, for a total of 1,398 students, were excluded because they did not have FCAT DSS scores in reading or mathe matics for one or both of the years in the study; therefore, gain scores could not be computed. Another 63 K 8 students and 285 6 8 students were excluded because they attended more than one school during the 2010 2011 school year. Because of this mid year mobility, it would be difficult to attribute any gains or losses in reading or math to the configuration of the school or to the practices within th ose schools. Finally, another 10 K 8 students and 48 6 8 students were excluded because they were retained in sixth grade the prior year. Since the transition from fifth to sixth grade took place in a prior year, any differences related to grade configuration would have been present prior to the 2010 2011 school year. In all, a total of 1,804 students, consisti ng of 473 K 8 students and 1, 331 6 8 students, were excluded from the sample. This constitutes 12 .2 % of the total sample, with an 11.2 % reduct ion in the K 8 sample and a 12.6 % reduction in the 6 8 sample. The demographic data of the students excluded from the study can be found in Table 4 1.

PAGE 120

120 Table 4 1. Number of students excluded from study by demographic c haracteristics 6 8 Configuration K 8 Configuratio n Total Category (n = 1,331) (n = 473 ) (n = 1,804 ) Gender Female 599 209 808 Male 732 264 996 Race/Ethnicity American Indian/Alaskan 0 1 1 Asian 17 10 27 Black 404 97 501 Hispanic 633 289 922 Multi Racial 16 2 18 White/Non Hispanic 261 74 335 Limited English Proficient Less than 2 years (Current LEP) 284 131 415 More than 2 years (Former LEP) 221 90 311 None 826 252 1078 Free/Reduced Price Lunch Free 875 248 1123 Reduced Price 69 25 94 None 387 200 587 ESE (Excluding Gifted) Yes 236 94 330 No 1095 379 1474 The 32 K 8 schools had a total enrollment of 3 7,782 in the 2010 2011 school year, for an average of 1,145 students per school. Enrollment in the 32 6 8 configured schools was 33,846, for an average of 1,026 students per school. Although the average enrollment at K 8 schools was slightly higher than th e traditional 6 8 middle schools, the sixth grade cohort was much smaller due t o the increased number of grade levels in K 8 schools. The K 8 schools had an average of 132 students in the sixth grade, while the 6 8 schools had an average of 322 sixth grade rs per school. All of the schools in the study were located in metropolitan school districts with over 100,000 students enrolled in the public school systems.

PAGE 121

121 The degree of implementation of model level practices was determined by surveying educators withi n each of the sample schools. Using both an online s urvey tool and paper surveys, 269 educators were surveyed in the identified school sample. 126 of the educators surveyed were in 6 8 schools, while the remaining 143 were in K 8 schools. Participants were either identified by an administrator within in the school or by job responsibility listed o n the school website. Of the 269 surveyed, there was a total response rate of 72.6%, with196 educators responding. Of those responding, 101 of the educators were i n 6 8 schools, for a 6 8 response rate of 80.2%, while the re maining 95 were in K 8 schools, for a K 8 response rate of 66.4%. Statistical Compilation Student data were aggregated by school configuration and by total student population. Each subgroup, inc luding gender, race/ethnicity, li mited English proficient, free/reduced price lunch status, and primary exceptionality, were tested for group equivalency using the FREQ procedure in SPSS software. Descriptive statistics for both of the student groups for developmental scale scores was obtained using the MEAN procedure using SPSS software. The following table s represent the demographic data for the student population sample used in this study: Table 4 2 Aggregate d emographic data for student population by grade configuration and descriptive demographic data K 8 Configuration ( n = 3, 535 ) 6 8 Configuration ( n= 9, 192 ) Total (n= 12, 727 ) n % n % n % Free/Reduced Lunch Rate 2, 006 56.8 % 5,857 63.7 % 7, 863 61.8 % Minority Rate 2,926 83.8 % 7,68 3 83.6 % 10, 609 83.4 % SWD Rate 3 72 10.5 % 935 10.2 % 1,3 07 10.3 % LEP Rate 196 5.5 % 457 5.0 % 65 3 5.1 %

PAGE 122

122 Table 4 3 Number of student participants by grade configuration and gender, race/ ethnicity, limited English proficient, free or reduced price lunch rates and exceptionalit y K 8 Configuration (n = 3,535 ) 6 8 Configuration (n = 9, 192 ) Total (n = 12,727 ) Category n % n % n % Gender Male 1, 784 50.5 % 4, 671 50.8% 6, 455 50.7% Female 1, 751 49.5 % 4,5 21 49.2% 6, 272 49.3% Race/Ethnicity American Indian/Alaskan Na tive 2 0.1% 12 0.1% 14 0.1% Asian 7 4 2.1% 187 2.0% 26 1 2.1 % Black 6 78 19 .2% 2,077 22.6% 2,7 55 21.6 % Hispanic 2, 131 60.3 % 5,261 57.2% 7, 392 58.1 % Multi Racial 55 1.6 % 144 1.6% 199 1.5% White/Non Hispanic 595 16.8 % 1,509 16.4% 2,1 04 16.5 % Limited English Proficient Less than 2 years (Current LEP) 196 5.5 % 457 5.0% 65 3 5.1% More than 2 years (Former LEP) 1, 385 39.2 % 3,430 37.3% 4, 815 37.8 % None 1,954 55.3 % 5,305 57.7% 7, 259 57.0 % Free/Reduced Price Lunch Free 1, 756 49.7 % 5,026 5 4.7% 6, 782 53.3 % Reduced Price 2 50 7.1 % 831 9.0% 1, 081 8.5% None 1, 529 43.3 % 3,335 36.3% 4, 864 38.2 % Primary Exceptionality Gifted 51 4 14.5% 1,735 18.9% 2,2 49 17.7 % Speech Impairment (SI) 28 0.7% 63 0.7% 91 0.7% Language Impairment (LI) 10 0.3% 41 0.4% 51 0.4% Specific Learning Disability (SLD) 2 46 7.0 % 654 7.1% 9 0 0 7.1 % Intellectual Disability (InD) 8 0.2% 13 0.1% 21 0.2% Emotional/Behavioral Disorder (E/BD) 21 0.6% 82 0.9% 103 0.8% Deaf/Hard of Hearing (DHH) 2 5 0.7% 5 0.1% 3 0 0.2% Visually Impaired (VI) 3 0.1% 8 0.1% 11 0.1% Autism Spectrum Disorder (ASD) 1 2 0.3% 37 0.4% 49 0.4% Orthopedic Impairment (OI) 1 0.0% 7 0.1% 8 0.1% Other Health Impaired (OHI) 5 5 1.6 % 125 1.4% 18 0 1.4% Traumatic Brain Injury (TBI) 0 0.0% 2 >0.1% 2 >0.1% Hospital/Homebound (H/H) 1 >0.1% 2 >0.1% 3 >0.1% Individual s tudent data were retrieved from each of the respective school districts and were entered in the Statistical Analysis System (SA S ) Version software. Descriptive statistics were used to report the means and standard deviations for the

PAGE 123

123 dependent variables of FCAT Reading and Mathematics achievement with seven categories of students: ( a) all sixth grade students, ( b) students by sex, ( c) students by race/ethnicity, ( d) economically disad vantaged students, ( e) limit ed English proficient students, ( f) students with disabilities, and ( g) giftedness. The following table s present the descriptive data: Table 4 4 Mean gain scores and standard deviations for dependent variables by demographic fa ctors Demographic Category FCAT Reading Gain Scores FCAT Math Gain Scores n M SD M SD All students 12,727 102.7 213. 4 2 8.5 153. 2 School Type K 8 (Type 4) 3,535 126.5 215.0 49.0 151.5 6 8 (Type 2) 9,192 93.5 207.4 20.7 153.2 Gender Female 6,272 89.3 204.8 44.2 147.8 Male 6,455 115.7 220.8 15.8 157.4 Ethnicity/Race Am Indian/Alaskan 14 203.7 290.3 16.9 236.0 Asian 261 139.3 237.8 60.7 140.2 Black 2,755 70.9 216.3 16.8 169.6 Hispanic 7,392 114.1 209.3 27.9 1 50.6 Multi Racial 199 111.9 237.4 55.8 150.8 White 2,104 98.1 213.4 39.7 138.9 Free/Reduced Lunch Status Free 6,782 96.2 215.9 18.6 161.9 Reduced Price 1,081 116.6 203.6 24.5 152.7 Full Priced 4,864 108.7 211.8 43.3 139.2 Limited English Proficient Current LEP (<2 years) 653 184.2 317.0 61.8 231.8 Former LEP ( 2 years) 4,815 109.3 197.4 24.1 144.9 No LEP status 7,259 91.0 210.3 28.5 149.4 ESE Status Disabled 1,307 81.4 258.1 6.4 212.0 Nondisabled 11,420 105.1 207.6 33.5 144.5 Gifted Status Gifted 2,249 101.2 224.4 47.8 130.8 Not ide ntified gifted 10,478 103.0 211.0 24.4 157.3

PAGE 124

124 The five categories were not independent of each other, since there was significant overlap among the categories. Thus, a direct comparison among these groups was not appropriate. The degree of implementation of each of the model practices was determined based on educator survey. Using a web based electronic survey (SurveyMonkey 2011), t he surveys were delivered electronically from email addresses obtained from each of the school websites or provided by a sc hool based administrator. In addition, educators could elect to receive a paper copy of the survey. Of the 269 surveys that were distributed, 196 were returned, for a 72.6 % response rate. The descriptive frequencies were calculated using SPSS 19.0 software The data that represent the demographic composition of those completing the educator survey is found in Table 4 5 Table 4 5 Educator survey participants by school configuration and demographic information Total (n = 196 ) 6 8 Configuration (n = 101 ) K 8 Configuration (n = 95 ) Category n % n % n % Sex Female 146 74.5% 72 71.3% 74 78.7% Male 49 25.0% 29 28.7% 20 21.3% No Response 1 0.5% 0 0.0% 1 1.1% Age 20 29 13 6.6% 7 6.9% 6 6.3% 30 39 56 28.6% 30 29.7% 26 27.7% 40 49 67 34.2 % 36 35.6% 31 33.0% 50 59 42 21.4% 22 21.8% 20 21.3% 60 or older 16 8.2% 5 5.0% 11 11.7% No Response 2 1.0% 1 1.0% 1 1.1% Race/Ethnicity Asian 1 0.5% 1 1.0% 0 0.0% Black 39 19.9% 25 24.8% 14 14.7% Hispanic 61 31.1% 27 26.7% 34 35.8% Mu lti Racial 6 3.1% 3 3.0% 3 3.2% White 84 42.9% 42 41.6% 42 44.2%

PAGE 125

125 Table 4 5. Continued Total (n = 196 ) 6 8 Configuration (n = 101 ) K 8 Configuration (n = 95 ) Category n % n % n % Sex (continued) No Response 5 2.6% 3 3.0% 2 2.1% Years Experie nce First Year 2 1.0% 2 2.0% 0 0.0% 2 5 Years 20 10.2% 8 7.9% 12 12.6% 6 10 Years 39 19.9% 20 19.8% 19 20.0% 11 20 Years 71 36.2% 40 39.6% 31 32.6% 21 30 Years 46 23.5% 24 23.8% 22 23.2% 31+ Years 13 6.6% 5 5.0% 8 8.4% No Response 5 2.6 % 2 2.0% 3 3.2% Years at Present School 1 2 Years 45 23.0% 23 22.8% 22 23.2% 3 5 Years 66 33.7% 28 27.7% 38 40.0% 6 10 Years 41 20.9% 20 19.8% 21 22.1% 11 19 Years 31 15.8% 23 22.8% 8 8.4% 20+ Years 10 5.1% 5 5.0% 5 5.3% No Response 3 1. 5% 2 2.0% 1 1.1% Primary Responsibility All Grades K 8 19 9.7% 0 0.0% 19 20.0% Elementary (K 5) 4 2.0% 0 0.0% 4 4.2% Middle (6 8) 171 87.2% 100 99.0% 71 74.7% No Response 2 1.0% 1 1.0% 1 1.1% Highest Degree Earned 58 29.6 % 30 29.7% 28 29.5% 96 49.0% 49 48.5% 47 49.5% Post 41 20.9% 22 21.8% 19 20.0% No Response 1 0.5% 0 0.0% 1 1.1% Degree in Education No 46 23.5% 26 25.7% 20 21.1% Yes 149 76.0% 75 74.3% 74 77.9% No Response 1 0.5% 0 0.0% 1 1.1% Position in School Principal 18 9.2% 7 6.9% 11 11.6% Assistant Principal 22 11.2% 16 15.8% 6 6.3% Classroom Teacher 132 67.3% 68 67.3% 64 67.4% Counselor 15 7.7% 6 5.9% 9 9.5% Academic Coach 7 3.6% 2 2.0% 5 5.3% No Re s ponse 2 1.0 % 2 2.0% 0 0.0%

PAGE 126

126 Analysis of Research Questions Data analysis in this section focuses on each of the four research questions. Included are the null hypotheses developed to test each research question and the results of the data analysis. Research Questio n 1 : Is there a relationship between the exemplary practices identified by NMSA/AMLE, including ( a) interdisciplinary teaming, ( b) common teacher planning time, ( c) heterogeneous grouping, ( d) student advisory, ( e) exploratory courses, and ( f) looping, and student achievement as measured by student gains on the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students in K 8 and 6 8 middle schools in large urban school districts? Six corre sponding null hypotheses were developed to test the first research question. These hypotheses compare FCAT DSS achievement gain scores of students in the sample who were in the fifth grade in 2009 2010, and then in sixth grade the following year. The degre e of implementation of each of the identified practices was determined by educator survey. Each of the respondents rated the school on a scale from 1 to 5, with a score of 1 representing no implementation of the practice, and a score of 5 representing full implementation. The definitions on the surveys for each of the ratings were as follows: 1. No Implementation We do not have this characteristic in our school 2. Little Implementation We have this characteristic to a small degree 3. Partial Implementation W e implement this characteristic but not every aspect as described in the statement 4. Mostly Implemented We implement most of the characteristics as described in the statement 5. Full Implementation We implement every aspect of the characteristic as descri bed in the statement

PAGE 127

127 tailed significance at the .05 level were calculated in SPSS to determine preliminary correlations in reading and math gain s with each of the model practices. Four of the K 8 schools had insufficient survey responses regarding the implementation of the model practices; therefore, those data were not used in this analysis. These correla tions are reported in Table 4 6 and indicate that none of the practices correlate significant ly with reading or math gains before adjusting for demographic variables. Table 4 6 Nonparametric correlations of reading and math gain scores with model practices FCAT DSS Reading Gains FCAT DSS Math Gains r xy 2 tailed significance a r xy 2 tailed signi ficance a Interdisciplinary Teaming 0.197 0.139 0.142 0.286 Common Planning 0.026 0.845 0.130 0.329 Exploratory Courses 0.113 0.399 0.074 0.583 Advisory 0.251 0.057 0.086 0.520 Heterogeneous Grouping 0.082 0.540 0.206 0.121 Looping 0.040 0.763 0 .150 0.260 a C orrelations significant at the .05 level (2 tailed) Linear regression using stepwise regression in SPSS was used to determine the relationships between the six exemplary practices predictor var iables and student gain scores. Covariates inclu ded total school enrollment, school sixth grade enrollment, school free or reduced price lunch rate, and school minority rate. Ho 1 1: There is no relationship between the implementation of interdisciplinary teams and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in an urban school district when controlling for school demographic factors of school size, grade level size, minority rate, SES rate, LEP rate, and ESE rate

PAGE 128

128 The survey described the impleme Survey participants rated their implementation of this practice on a scale of 1 to 5, with 1 meaning no implementation of this p ractice and 5 defined as full implementation. Implementation scores of participants within the same school were combined into a school average. A linear regression test using stepwise regression in SPSS was performed using the degree of implementation of interdisciplinary teaming as the dependent variable, reading and math gains school enrollment, sixth grade enrollment, school minority rate school LEP rate, school ESE rate and school free and reduced price lunch rate as independent variables T he analys is revealed that there were no significant difference s in student academic achievement in mathematics and reading gains and interdisciplinary teaming, even when controlling for school enrollment, free and reduced price lunch rate, and minority rate The on ly significant predictor of the degree of implementation of interdisciplinary teams at the .05 l evel was sixth grade enrollment. Schools that had smaller sixth grade level sizes had a higher self rating of the implementation of interdisciplinary teaming. S ince no differences were found in math or reading gains the null hypothesis was retained. T hese data are shown in Table 4 7 and represent the results for the relationship between the independent variables and the dependent variable of interdisciplinary t eaming.

PAGE 129

129 Table 4 7 Sig nificance of interdisciplinary teaming and student achievement gains with descriptive characteristic tested through linear regression Partial Collinearity Statistics t Significance Correlation Tolerance Reading Gains 0.006 0. 996 0.001 0.938 Math Gains 0.401 0.690 0.054 0.866 School Enrollment 0.102 0.919 0.014 0.823 Free/Reduced Lunch Rate 0.165 0.869 0.022 0.998 Minority Rate 0.507 0.614 0.068 1.000 LEP Rate 0.025 0.980 0.003 1.000 ESE Rate 0.746 0.459 0.1 00 0.967 Sixth Grade Enrollment 2.188 0.033 --Ho 1 2: There is no relationship between the implementation of common teacher planning and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in an ur ban school district when controlling for school demographic factors of school size, grade level size, minority rate, SES rate, LEP rate, and ESE rate planning for teachers with th participants rated their implementation of this practice on a scale of 1 to 5, with 1 meaning no implementation of this practice and 5 defined as full implementation. Implementation scores of participants within the same school were combined into a school average. A linear regression test using stepwise regression in SPSS was performed using the degree of implementation of common teacher planning as the dependent variable, reading and math gains, school enrollment, sixth grade enrollment, school minority rate, school LEP rate, school ESE rate and school free and reduced price lunch rate as

PAGE 130

130 independent variables. The only significant predictor of the degree of implementation of common teacher planning at the .05 level was ESE rate Since no differences in student achievement were found, the null hypothesis was retained. T hese data are shown in Table 4 8 and represent the results for the relationship between the dependent variable of implement ation of common teacher planning and the independent variables of reading and math gain scores, and the school level independent variables. Tabl e 4 8 Sig nificance of common teacher planning and student achievement gains with descriptive characteristic tes ted through linear regression Partial Collinearity Statistics t Significance Correlation Tolerance Reading Gains 0.009 0.9 93 0.0 01 0.9 54 Math Gains 1. 156 0. 2 52 0. 1 54 0. 99 8 Sixth Grade Enrollment 1. 588 0. 118 0. 209 0. 967 Free/Reduced Lunch Rate 0.26 0 0. 9 79 0.004 0. 881 Minority Rate 1.209 0. 232 0. 161 0.99 3 Total School Enrollment 1.432 0.158 0.190 0.878 LEP Rate 1.086 0.282 0.145 0.958 ESE Rate 2.113 0.039 --Ho 1 3: There is no relationship between the implementation of heteroge neous grouping and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in an urban school district when controlling for school demographic factors of school size, grade level size, minority rate, SES rat e, LEP rate, and ESE rate students assigned to teams heterogeneously, either at random or purposefully, rather than by ability, achievement, or interest ntation of this practice on a scale of 1 to 5, with 1 meaning no implementation of this practice and 5 defined as

PAGE 131

131 full implementation. Implementation scores of participants within the same school were combined into a school average. A linear regression tes t in SPSS was performed using the degree of implementation of heterogeneous grouping as the dependent variable, reading and math gains, school enrollment, sixth grade enrollment, school minority rate, school LEP rate, school ESE rate and school free and r educed price lunch rate as independent variables. Stepwise regression yielded no significant results. In addition, the practice of heterogeneous grouping did not significantly correlate to any of the factors. Since no differences were found, the null hypot hesis was retained. T hes e data are shown in Table 4 9 and represent the results for the relationship between the dependent variable of heterogeneous grouping with the independent variables of reading and math gain scores as well as the other classificati on variables in relationship to the dependent variable. Table 4 9 Sig nificance of heterogeneous student grouping and student achievement gains with descriptive characteristic tested through General Linear Model procedure t Significance Reading Gains 1. 028 0.845 Math Gains 1.353 0.687 Total School Enrollment 0.275 0.671 Sixth Grade Enrollment 0.599 0.463 Free/Reduced Lunch Rate 0.837 0.762 ESE Rate 0.545 0.588 LEP Rate 1.295 0.201 Minority Rate 0.467 0.286 Ho 1 4: There is no relationship between the implementation of advisory groups and student achievement gains as measured by the FCAT SSS Developmental Scale Score

PAGE 132

132 for sixth grade students in an urban school district when controlling for school demographic factors of school size, grade lev el size, minority rate, SES rate, LEP rate, and ESE rate advisee Survey participants rated their implementa tion of this practice on a scale of 1 to 5, with 1 meaning no implementation of this practice and 5 defined as full implementation. Implementation scores of participants within the same school were combined into a school average. A linear regression test u sing stepwise regression yielded in SPSS revealed no significant differences in student academic achievement in mathematics and reading gains with the implementation of student teacher advisory groups or with any of the other school level variables Sinc e no differences were found, the null hypothesis was retained. In addition, the implementation of advisory groups did not significantly correlate with any of the additional factors. T hese data are shown in Table 4 10 and represent the results for the rela tionship between the independent variables and the dependent variables, 2011 reading and math gain scores, as well as the other classification variables in relationship to the dependent variable. Table 4 10 Sig nificance of advisory periods and student ach ievement gains with descriptive characteristic tested through General Linear Model procedure t Significance Reading Gains 1. 490 0. 143 Math Gains 0.8 33 0. 409 Total School Enrollment 1.165 0. 250 Sixth Grade Enrollment 0. 071 0. 944 Free/Reduced Lunc h Rate 0.7 08 0.4 82

PAGE 133

133 Table 4 10. Continued t Significance ESE Rate 1.704 0.095 LEP Rate 1.423 0.161 Minority Rate 1. 355 0. 182 Ho 1 5: There is no relationship between the implementation of exploratory course offerings and student achievement gains a s measured by the FCAT SSS Developmental Scale Score for sixth grade students in an urban school district when controlling for school demographic factors of school size, grade level size, minority rate, SES rate, LEP rate, and ESE rate The survey describ participants rated their implementation of this practice on a scale of 1 to 5, with 1 meaning no implementatio n of this practice and 5 defined as full implementation. Implementation scores of participants within the same school were combined into a school average. A linear regression test in SPSS was performed using the degree of implementation of exploratory cou rses as the dependent variable, reading and math gains, school enrollment, sixth grade enrollment, school minority rate, school LEP rate, school ESE rate and school free and reduced price lunch rate as independent variables. The only variable that yielded significant results in relation to the implementation of exploratory courses was minority rate. In this study, greater implementation of exploratory courses was negatively related to minority rate, with schools having the highest rates of minority students reporting the lowest implementation of this practice.

PAGE 134

134 The practice of exploratory course offerings did not significantly correlate to any of the other factors. Since no differences were found in reading and mathematics gain scores, the null hypothesis was retained. These data are shown in Table 4 11 and represent the results for the relationship between the dependent variable of exploratory courses with the independent variables of reading and math gain scores, as well as the other classification variable s in relationship to the dependent variable. Table 4 11 Sig nificance of exploratory courses and student achievement gains with descriptive characteristic tested through General Linear Model procedure Partial Collinearity Statistics t Significance Cor relation Tolerance Reading Gains 1.036 0.305 0.138 0.991 Math Gains 0.616 0.540 0.083 0.937 School Enrollment 0.217 0.787 0.037 0.992 Sixth Grade Enrollment 0.226 0.822 0.030 1.000 Free/Reduced Lunch Rate 0.494 0.623 0.066 0.531 ESE Rate 1.261 0.213 0.168 0.993 LEP Rate 0.652 0.517 0.088 0.948 Minority Rate 2.262 0.028 --Ho 1 6: There is no relationship between the implementation of looping and student achievement gains as measured by the FCAT SSS Developmental Scale Score for si xth grade students in an urban school district when controlling for school demographic factors of school size, grade level size, minority rate, SES rate, LEP rate, and ESE rate f participants rated their implementation of this practice on a scale of 1 to 5, with 1 meaning no implementation of this practice and 5 defined as full implementation.

PAGE 135

135 Imple mentation scores of participants within the same school were combined into a school average. A linear regression test in SPSS was performed using the degree of implementation of looping as the dependent variable, reading and math gains, school enrollment, sixth grade enrollment, school minority rate, school LEP rate, school ESE rate and school free and reduced price lunch rate as independent variables. The only variable that yielded significant results in relation to the implementation of looping was minor ity rate, with schools having the highest minority rates more likely to implement the practice of looping. The level of implementation of looping did not significantly correlate to any of the other factors. Since no differences were found in reading and ma thematics gain scores, the null hypothesis was retained. Th ese data are shown in Table 4 12 and represent the results for the relationship between the dependent variable of looping with the independent variables of reading and math gain scores, as well as the other classification variables in relationship to the dependent variable. Table 4 12 Sig nificance of looping and student achievement gains with descriptive characteristic tested through General Linear Model procedure Partial Collinearity Statistic s t Significance Correlation Tolerance Reading Gains 0.331 0.742 0.045 0.991 Math Gains 0.024 0.981 0.003 0.937 School Enrollment 0.061 0.952 0.008 0.992 Sixth Grade Enrollment 0.621 0.537 0.083 1.000 Free/Reduced Lunch Rate 0.069 0.945 0 .009 0.531 LEP Rate 1.354 0.181 0.180 0.948 ESE Rate 0.329 0.744 0.044 0.993 Minority Rate 2.020 0.048 --

PAGE 136

136 In summary, there was no relationship between the model middle level practices and student achievement in reading or mathematics, includi ng interdisciplinary teaming, common teacher planning, heterogeneous student grouping, student advisory periods, exploratory courses, and looping. The only factors that were related to the implementation of the practices were the number of students in the sixth grade within the school, school minority rate, and school ESE rate. Schools with smaller sixth grade enrollments had a higher reported implementation of interdisciplinary teaming. In addition, schools with higher rates of students with disabilities were more likely to report a higher implementation of common teacher planning. Minority rates were positively related to the practice of looping, with teachers staying with the same group of students for multiple years. Conversely, minority rates were nega tively related to the implementation of exploratory courses. Research Question 2 : Are there differences in levels of implementation of NMSA/AMLE identified exemplary practices in urban K 8 configured schools when compared to 6 8 configured middle schools? This hypothesis compares the degree of implementation of the exemplary middle level practices, including interdisciplinary teaming, common planning, advisory, heterogeneous grouping, exploratory programs, and looping, of K 8 schools with 6 8 schools, as r eported by survey participants who were educators within each of the two types of schools. Six corresponding null hypotheses for each of the corresponding practices were developed to test the second question. Group statistics were computed in SPSS for ea ch of the exemplary practices for both school types, w hich are presented in Table 4 13 Of the schools participating in the study, four of the K 8 schools had insufficient survey data with respect to implementation of the exemplary practices, so those scho ols were not included in the

PAGE 137

137 analysis of these data. With the exception of heterogeneous grouping, all means for reported implementation of each of the practices were higher in the K 8 schools than in the 6 8 schools. Table 4 13 Group statistics of exemp lary middle practices by grade configuration Exemplary Practice School Type n Mean Std Deviation Std Error Mean Interdisciplinary Teams 6 8 32 3.049 0.988 0.175 K 8 28 3.727 0.984 0.186 Exploratory Courses 6 8 32 3.591 0.761 0.134 K 8 28 3.746 0.784 0.148 Advisory 6 8 32 2.273 1.124 0.199 K 8 28 2.429 1.260 0.238 Heterogeneous Groups 6 8 32 2.734 0.914 0.162 K 8 28 2.649 0.851 0.161 Common Planning 6 8 32 2.451 1.163 0.206 K 8 28 2.981 1.197 0.226 Looping 6 8 32 1.867 0.783 0.138 K 8 28 1. 985 1.128 0.213 A t Test for independent samples was used to compare the means for each of the exemplary practices in each type of school at the .05 level of significance, which is presented in Table 4 14 Table 4 14 Comparison of means of implementati on levels of model practices by grade configuration for Equality of Variance t test for Equality of Means t df Sig a Mean Diff Std Error Diff 95% Conf. Interval F Sig Lower Upper Interdisciplinary .004 .948 2.654 58.00 .010 .677 .255 1.188 .166 Exploratory .014 .906 .775 58.00 .442 .155 .200 .554 .245 Advisory .546 .463 .504 58.00 .616 .155 .308 .771 .461 Het Group .592 .445 .371 58.00 .712 .085 .229 .374 .543 Common Plan .117 .733 1.739 58.00 .087 .530 .305 1.141 .081 Looping 3.150 .081 .473 58.00 .638 .117 .248 .614 .380 a two tailed

PAGE 138

138 case, equal variance is assumed. Ho 2 1: There are no significant differences in the impl ementation of interdisciplinary teaming in schools configured as K 8 compared to traditional middle schools with a 6 8 grade con figuration After performing a t Test for independent samples in SPSS, the analysis revealed that there were significant differe nces in grade configuration and the implementation of interdisciplinary teaming with K 8 schools reporting a higher implementation level than the 6 8 schools Since significant differences were found, the null hypothesis was rejected T hese data are shown in Table 4 1 4 Ho 2 2: There are no significant differences in the implementation of common teacher planning time in schools configured as K 8 compared to traditional middle schools with a 6 8 grade configuration. After performing a t Test for independent s amples in SPSS, the analysis revealed that there were no significant differences in grade configuration and the implementation of common teacher planning time. Since no differences were found, the null hypothesis was retained. These data are shown in Table 4 1 4 Ho 2 3: There are no significant differences in the implementation of heterogeneous student grouping in schools configured as K 8 compared to traditional middle schools with a 6 8 grade configuration. After performing a t Test for independent samples in SPSS, the analysis revealed that there were no significant differences in grade configuration and the implementation

PAGE 139

139 of heterogeneous student grouping. Since no differences were found, the null hypothesis was retained. These data are shown in Table 4 1 4 Ho 2 4: There are no significant differences in the implementation of student teacher advisory groups in schools configured as K 8 compared to traditional middle schools with a 6 8 grade configuration. After performing a t Test for independent samples in S PSS, the analysis revealed that there were no significant differences in grade configuration and the implementation of student teacher advisory groups. Since no differences were found, the null hypothesis was retained. These data are shown in Table 4 1 4 H o 2 5: There are no significant differences in the implementation of exploratory courses in schools configured as K 8 compared to traditional middle schools with a 6 8 grade configuration when controlling for SES and minority rate. After performing a t Test for independent samples in SPSS, the analysis revealed that there were no significant differences in grade configuration and the implementation of exploratory courses. Since no differences were found, the null hypothesis was retained. These data are shown in Table 4 1 4 Ho 2 6: There are no significant differences in the implementation of looping in schools configured as K 8 compared to traditional middle schools with a 6 8 grade configuration when controlling for SES and minority rate. After performing a t T est for independent samples in SPSS, the analysis revealed that there were no significant differences in grade configuration and the implementation of looping. Since no differences were found, the null hypothesis was retained. These data are shown in Table 4 1 4

PAGE 140

140 In summary, the grade configuration of the school was not predictive of the level of implementation of most of the model middle level practices, including common teacher planning, heterogeneous grouping, advisory groups, exploratory courses, and lo oping. However, the reported level of implementation of interdisciplinary teaming was significantly higher in K 8 schools than in 6 8 schools. Research Quest ion 3 : Are there differences in academic achievement in reading and mathematics for sixth grade pub lic school students in K 8 configured schools when compared to sixth grade students in 6 8 configured middle schools, as measured by the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system? To test the differenc es in student achievement gains by grade level configurations within the sample schools, two null hypotheses were developed. These hypotheses compare the same group of students who were in the fifth grade in 2009 2010 and then in sixth grade the following year. The first group of students remained in the same K 8 school; the second group of students transitioned from an elementary school in fifth grade to a 6 8 configured middle school. T tests for independent samples were calculated in SPSS to compare FCA T SSS Developmental Scale Score achievement gain means in both reading and mathematics from the 2010 to the 2011 administration for sixth grade students by school type at the 0.05 level of significance. Ho 3 1: There are no significant differences in student reading achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in a K 8 school compared to the same students in a traditional 6 8 middle school. Group means and standard deviations ar e shown in Table 4 15 and rep resent a comparison between the variable of grade configuration, and the dependent variable 2011 sixth grade reading achievement gains on the FCAT. An initial comparison of the

PAGE 141

141 means indicate the gain scores in K 8 schools were higher than in those in 6 8 configured schools. Table 4 15 Group means and standard deviations of sixth grade FCAT r eading gain scores from 2010 to 201 1 by school grade configuration Reading Gain Scores School Type N Mean Std. Deviation Std. Error Mean 6 8 Middle 9,192 93.53 21 5.026 2.243 K 8 Combination 3, 525 12 6.53 20 7.401 3. 488 A t Test for independent samples to compare the reading gain means in K 8 schools and 6 8 schools was conducted using SPSS Test for Equality of Variance wa s less than .05, equal var iances were not assumed. The analysis revealed that there were significant differences in grade configuration and re ading achievement gains at the .05 level, with the students in the K 8 schools scoring significantly higher than their peers in the 6 8 scho ols. The results from th e t Test are found in Table 4 16 Table 4 16 Independent Samples test for sixth grade FCAT reading gain scores from 2010 to 2011 by school grade configuration for Equality of Variance t test for Equality of Means Equal variances F Sig. t df Sig. a Mean Diff Std Er Diff 95% Conf Int Lower Upper Reading Gains assumed 4.988 .02 6 7.832 12 725 .000 33.005 4. 214 4 1.266 24.745 not assumed 7. 9 5 9 6624.3 .000 33.005 4. 147 41.135 24.876 a Two tailed Furt her analysis was conducted to determine the effects of the school demographic factors on student reading gains. Linear regression in SPSS was used with reading gains entered as the dependent variable, school configuration as the selection factor and total school enrollment, sixth grade class size, school minority rate,

PAGE 142

142 free/reduced price lunch rate, ESE rate, and LEP rate as the predictors. Those results are presented in Table 4 17 The only significant predictors of reading achievem ent gains at the .05 significance level were found in the 6 8 configured middle schools. Free/reduce priced lunch rate was negatively related to reading gain scores in middle schools, while minority rate was positively related to increased gain scores. There were no significan t predictors of reading gain scores in the K 8 schools. Table 4 17 Linear regr ession of reading gains grouped by school configuration and using school factors as predictors Model Unstandardized Coefficients Standardized Coefficients B Std Error Beta t Sig 6 8 Configuration (Constant) 35.337 39.388 0 .897 0 .378 Enrollment 0.032 0 .031 0 .337 1.022 0 .316 6 th Enrollment 0.137 0 .107 0 .442 1.278 0 .213 Lunch Rate 1.348 0 .562 0 .870 2.398 0 .024 Minority Rate 1.530 0 .595 0 .765 2.572 0 .016 LEP Rate 0.369 1.247 0 .055 0 .296 0 .770 ESE Rate 0.365 1.389 0 .051 0 .262 0 .795 K 8 Configuration (Constant) 135.166 43.497 3.107 0 .005 Enrollment 0 .065 0 .041 0 .585 1.581 0 .127 6 th Enrollment 0 .354 0 .301 0 .432 1.179 0 .250 Lunch Rate 0 .383 0 .339 0 .278 1.132 0 .269 Minority Rate 0 .250 0 .529 0 .112 0 .473 0 .641 LEP Rate 1.192 1.433 0 .164 0 .832 0 .414 ESE Rate 0 .117 1.159 0 .021 0 .101 0 .921 Ho 3 2: There are no significant differences in stu dent mathematics achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade

PAGE 143

143 students in a K 8 school compared to the same students in a traditional 6 8 middle school. Group means and standard de viations are shown in Table 4 18 and represent a comparison of the mean mathematics gain scores for sixth grade students in a K 8 setting and a 6 8 setting. Initial analysis indicated that the students in the K 8 schools had greater gains in math than their 6 8 peers, although the mean g ain scores for both grade configurations were much lower than the reading gain scores. Table 4 18 Group means and standard deviations of sixth grade FCAT math gain scores from 2010 to 2011 by school grade configuration Math Gain Scores School Type N Me an Std. Deviation Std. Error Mean 6 8 Middle 9,192 20.66 153.164 1.598 K 8 Combination 3, 535 49.00 151. 529 2. 549 A t Test for independent samples to compare the mathematics FCAT gain means in K 8 schools and 6 8 schools was conducted using SPSS Since Equalit y of Variance was greater than .05, equal variances were assumed. The analysis revealed that there were significant differences in grade configuration and math achievement gains at the .05 level, with the students in the K 8 schoo ls scoring significantly higher than their peers in the 6 8 schools. The results from th e t Test are found in Table 4 19 Table 4 19 Independent Samples test for sixth grade FCAT math gain scores from 2010 to 2011 by school grade configuration Test for Equality of Variance t test for Equality of Means Equal variances F Sig. t df Sig. a Mean Diff Std Error Diff 95% Confidence Interval Lower Upper Math Gains assumed 604 437 9.376 12 725 .000 28.337 3.022 34.261 24.413 not assume d 9.421 6472.5 .000 28.337 3.008 34.234 22.441 a Two tailed

PAGE 144

144 Further analysis was conducted to determine the effects of the school demographic factors on student math gains. Linear regression in SPSS was used with math gains as the dependent variabl e, school configuration as the selection factor and total school enrollment, sixth grade class size, school minority rate, free/reduced price lunch rate, ESE rate, and LEP rate as the predictors. Those res ults are presented in Table 4 20 There were no sig nificant predictors of math achievement gains at the 0.05 significance level in the both 6 8 and K 8 configured middle schools. Table 4 20 Linear regression for math gains grouped by school configuration and using school factors as predictors Model Unst andardized Coefficients Standardized Coefficients B Std Error Beta t Sig 6 8 Configuration (Constant) 80.483 47.858 1.682 0 .105 Enrollment 0 .020 0 .038 0 .182 0 .523 0 .605 6 th Enrollment 0 .036 0 .130 0 .102 0 .277 0 .784 Lunch Rate 1.136 0 .683 0 .638 1.663 0 .109 Minority Rate 0 .234 0 .723 0 .102 0 .323 0 .749 LEP Rate 2.696 1.516 0 .353 1.779 0 .087 ESE Rate 1.030 1.688 0 .125 0 .610 0 .547 K 8 Configuration (Constant) 125.111 52.735 2.372 0 .026 Enrollment 0 .057 0 .050 0 .3 88 1.146 0 .263 6 th Enrollment 0 .682 0 .365 0 .626 1.872 0 .073 Lunch Rate 0 .703 0 .410 0 .385 1.713 0 .100 Minority Rate 0 .951 0 .642 0 .321 1.482 0 .151 LEP Rate 2.388 1.737 0 .248 1.375 0 .182 ESE Rate 0 .107 1.405 0 .015 0 .076 0 .940

PAGE 145

145 Researc h Question 4 : Is there a relationship between the level of implementation of exemplary middle level practices and academic achievement as measured by the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for s ixth grade students enrolled in urban K 8 configured schools compared to similar students in 6 8 middle schools? This research question was developed to determine the degree to which the implementation of the exemplary middle level practices in each of th e two sixth grade settings is predictive of academic achievement. To test this question, six null hypotheses were developed for each of the model practices, with sub hypotheses for both reading and mathematics achievement gains in each of the two school gr ade configurations, K 8 and 6 8. The GLIMMIX Procedure, a general linear mixed model within SAS 9.2 for hierarchical linear modeling was used to determine the relationship between the six exemplary practices predictor variables reading and math achievem ent gain scores, and the two grade configurations: K 8 and 6 8. GLIMMIX was chosen as a statistical tool because of the multi layer dimension of the data. The general linear mixed model can be represented by : where Y is the ou tcome of observed response data, X is a set of fixed effects in the design, Z is the matrix of random effects, is the population intercept, is the coefficient for unknown fixed effects parameters, and represents the rando m errors (Ferrell, 2008). This model has the ability to detect variations in regression relationships by groups, and to test for main effects within and between levels of data in a hierarchical system (Willms 1999 ). Because student achievement data is nes ted within the identified model practices, which are in turn nested within both 6 8 and K 8 configured schools with varying levels of implementation of each of the model practices, GLIMMIX was determined to be the most effect method

PAGE 146

146 for studying the relati onship at any level of the analysis, while taking into account the variability of each level of the hierarchy. The lowest level of data for this analysis, L1, is student level data, including student achievement gains in reading and mathematics as well as all of the demographic information for each of the individual students. The second level of data, L2, is school level data regarding the level of implementation of each of the six model practices for exemplary middle level schools. Finally, the third leve l, L3, consists of the two groups of grade configurations of the schools in the sample, either a K 8 grade configuration or a 6 8 grade configuration. Ho 4 : There is no relationship between the level of implementation of the exemplary middle level practices and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students who attend K 8 schools implementing exemplary middle level practices compared to the same students attend traditional middle schools implementing the practices at the same level To determine the relationship between exemplary middle level practices, student achievement gains, and grade configuration, school and individual student data were first analyzed for fixed effects in SAS The fixed effect s model directly estimates (Allison, 2008, p. 1). Correlations between the estimates of unobserved differences with the observed variables, if any, are allowed in the fixed e ffects model. However, with fixed effects, the p values are higher and the confidence intervals are wider, due to the In this case reading and math scores on the FCAT DSS from 2010 and 2011 were each treated sep arately as the dependent

PAGE 147

147 variable. It is important to note that the Developmental Scale Scores provide directly comparable scores from different test administrations and across a number of years. Preliminary reading achievement correlations with effects ar e found in Table 4 21 and math achievement correlations with effects are found in Table 4 22 The Type III Tests of Fixed Effects shows the significance of each of the fixed effects in the model. The Kenward Roger option was used to adjust the estimated v ariance of the fixed effects. Since there were no significant differences in any of the model middle level practices for reading or math, the null hypothesis was retained. None of the model practice effects significantly correlated to achievement gain scor es in either reading or math. The Type III Tests of Fixed Effects for reading ach ievement are found in Table 4 23 and fo r math achievement in Table 4 24 Although none of the model practices significantly correlated with reading or math gain scores, th e research revealed other noteworthy findings For example, K 8 schools had higher gain scores than 6 8 schools in both reading and math. With respect to reading achievement, females had significantly lower gain scores than their male peers. However, the r everse was true with in math, with female math gain scores significantly higher than male sixth grade students. In addition, Asian and Native American students had higher reading gain scores, while black students did significantly worse in reading There w ere no significant correlations between race /ethnicity and math gain scores. Students on free lunch had significantly lower gain scores in reading than students on reduced price lunch or not on any subsidized lunch program with no significant findings bas ed on lunch status and math gains Interestingly, students who were currently limited English proficient had much higher gain scores in both reading

PAGE 148

148 and math than non LEP students, and non gifted had greater gains in both reading and math than students ide ntified as gifted. Students with disabilities had significantly lower gain scores in both reading and math than their non disabled peers. Table 4 21 Reading achievement gain scores and solutions for fixed effects Effect Estimate Std Error DF t Value Pr>It I School Type 6 8 (Type 2) 36.7386 9.8510 47.25 3.73 0.0005 K 8 (Type 4) 0 . Sex Female 28.5461 3.8842 11807 7.35 <0.0001 Male 0 . Race/Ethnicity Asian 34.9466 14.1706 11818 2.47 0.0137 Black 16.4223 7.2051 97 41 2.28 0.0227 Hispanic 5.3309 6.3039 11743 0.85 0.3978 Am Indian/Alaskan 119.5200 56.1299 11789 2.13 0.0332 Multi Racial 20.7691 16.0838 11416 1.29 0.1966 White 0 . Lunch Status Free 17.6038 7.1071 11806 2.48 0.0133 Reduced 0 None 9.4812 7.4334 11815 1.28 0.2022 LEP Status Current 80.4559 9.4029 11735 8.56 <0.0001 Former 4.3163 4.9009 11350 0.88 0.3785 None 0 . ESE Status No 27.7647 6.4613 11817 4.30 <0.0001 Yes 0 . Gifted Status No 11.3103 5.4500 11751 2.08 0.0380 Yes 0 . School Level Data Free/Red Lunch Rate 0.2976 0.3020 56.67 0.99 0.3286 Minority Rate 0.2901 0.4091 51.41 0.71 0.4815

PAGE 149

149 Table 4.21. Continued Effect Estimate Std Error DF t Value Pr>ItI Sch ool Implementation Lvls Interdisciplinary Teams 0.7036 5.4339 45.52 0.13 0.8975 Common Planning 4.3839 4.8014 49.32 0.91 0.3657 Exploratory 2.2938 6.4531 48.99 0.36 0.7238 Advisory 3.0051 3.6104 47.21 0.83 0.4094 Heterogeneous Groups 3.447 5 4.8681 44.48 0.71 0.4825 Looping 1.3810 4.9480 52.29 0.28 0.7813 Table 4 22 Math achievement and solutions for fixed effects Effect Estimate Std Error DF t Value Pr>ItI School Type 6 8 (Type 2) 45.2159 11.8957 46.16 3.80 0.0004 K 8 (Typ e 4) 0 . Sex Female 25.0068 2.7525 11786 9.09 <.0001 Male 0 . Race/Ethnicity Asian 16.2300 10.0488 11800 1.62 0.1063 Black 5.3343 5.1589 11565 1.03 0.3012 Hispanic 3.0273 4.4774 11817 0.68 0.4990 Am Indian/Alaskan 5 8.4134 39.7608 11777 1.47 0.1418 Multi Racial 1.9013 11.4485 11804 0.17 0.8681 White 0 . Lunch Status Free 4.3487 5.0431 11810 0.86 0.3885 Reduced 0 . None 8.7785 5.2691 11793 1.67 0.0957

PAGE 150

150 Table 4 22. Continued Effect Estimate Std Error DF t Value Pr>ItI LEP Status Current 50.2181 6.6783 11817 7.52 <.0001 Former 3.3432 3.4897 11799 0.96 0.3381 None 0 . ESE Status No 32.1575 4.5826 11803 7.02 <.0001 Yes 0 . Gifted Status No 12.4 432 3.8700 11816 3.22 0.0013 Yes 0 . School Level Data Free/Red Lunch Rate 0.08205 0.3530 49.89 0.23 0.8171 Minority Rate 0.4527 0.4861 48 0.93 0.3564 School Implementation Lvls Interdisciplinary Teams 2.7959 6.5825 46.13 0.42 0.6730 Common Planning 3.6393 5.7419 47.33 0.63 0.5293 Exploratory 13.2348 7.7239 47.33 1.71 0.0932 Advisory 3.1367 4.3416 46.95 0.72 0.4736 Heterogeneous Groups 11.3787 5.9240 45.64 1.92 0.0610 Looping 1.9873 5.8244 49.7 0.34 0.7344 Tab le 4 23 Type III Tests of Fixed Effects for reading achievement gain scores. Effect Num DF Den DF F Value Pr>F School Type 1 47.25 13.91 0.0005 Sex 1 11807 54.01 <.0001 Race/Ethnicity 5 10335 5.09 0.0001 Free/Reduced Lunch Status 2 11763 3.66 0.0258 LEP Status 2 11588 37.87 <.0001 ESE Status 1 11817 18.46 <.0001 Gifted Status 1 11751 4.31 0.0380 School Free/Reduced Lunch Rate 1 56.67 0.97 0.3286

PAGE 151

151 Table 4 23. Continued Effect Num DF Den DF F Value Pr>F School Minority Rate 1 51.41 0.50 0.4815 Inte rdisciplinary Teaming 1 45.52 0.02 0.8975 Common Planning 1 49.32 0.83 0.3657 Exploratory Courses 1 48.99 0.13 0.7238 Advisory 1 47.21 0.69 0.4094 Heterogeneous Grouping 1 44.48 0.50 0.4825 Looping 1 52.29 0.08 0. 7813 Table 4 24 Type III Tests of Fixed Effects for math achievement gain scores. Effect Num DF Den DF F Value Pr>F School Type 1 46.16 14.45 0.0004 Sex 1 11786 82.54 <.0001 Race/Ethnicity 5 11644 1.38 0.2282 Free/Reduced Lunch Status 2 11809 7.53 0.0005 LEP Status 2 11807 29.03 <.000 1 ESE Status 1 11803 49.24 <.0001 Gifted Status 1 11816 10.34 0.0013 School Free/Reduced Lunch Rate 1 49.89 0.05 0.8171 School Minority Rate 1 48 0.87 0.3564 Interdisciplinary Teaming 1 46.13 0.18 0.6730 Common Planning 1 47.33 0.40 0.5293 Explorato ry Courses 1 47.33 2.94 0.0932 Advisory 1 46.95 0.52 0.4736 Heterogeneous Grouping 1 45.64 3.69 0.0610 Looping 1 49.7 0.12 0.7344 Summary This study examined the relationship between grade configuration, reading and math gains, and model practices fo r sixth grade students in school districts with

PAGE 152

152 enrollments over 100,000. Two types of grade configuration were studied for middle level students: K 8 schools and 6 8 middle schools. Reading and math gains calculated using the difference of FCAT DSS from t he 2010 and 2011 administration for each student in the study. The six identified model practices included interdisciplinary teaming, common teacher planning, heterogeneous student grouping, exploratory courses, student advisory periods, and looping. The i mplementation of each practice was determined by self ratings via survey of staff members within each of the schools. Four general hypotheses were developed to test those relationships. The findings of this study reveal that there was no relationship betw een the implementation of any of the model practices and student achievement gain scores in reading and math There was one difference in the implementation of the model practices, with K 8 schools reporting a higher rate of implementa tion of interdiscipli nary teaming than their colleagues in 6 8 configured middle schools. There were no other significant differences in the implementation of model practices and grade configuration. There were significant differences in both reading and math gain scores by co nfiguration of the school, with K 8 students having significantly greater gains. Finally, when the combined effects of grade configuration and model practice implementation level were compared to student achievement gain scores, there were no significant d ifferences found, other than grade configuration alone, as previously stated. The study also revealed that other effects were correlated with achievement gain scores including student sex, LEP status, free lunch status, gifted status, race/ethnicity, and E SE status.

PAGE 153

153 CHAPTER 5 CONCLUSIONS Research has offered inconclusive results concerning the best organizational structure for early adolescents. Those favoring the middle school design believe that early adolescents need a specially designed school that b est meets their unique developmental needs Other educators however, argue that the transition that occurs believe that middle school students are better served by remaining in the stable setting that a K 8 design can provide Although there is some research regarding the relationship between the K 8 grade configurations and student achievement, there is little research regardin g this configuration on students in la rge urban districts, because until recently most of the K 8 schools were either private/parochial or located in more rural areas of the country. Despite the lack of research, many urban school districts are eliminating or reducing middle schools in favor o f the K 8 configuration. As this K 8 trend gains momentum in urban areas of the country, it has become more feasible to study the relationship between school configurations implementation of those practices that mi ddle school educators advocate and stude nt achievement for students that may be considered more at risk. The purpose of this st udy wa s to compare the academic achievement levels in reading and mathematics of sixth grade students enrolled in K 8 urban schools t o those of students who attend a tra ditional 6 8 middle school In addition, the study measures the degree of implementation of exemplary practices determined by the National Middle School Association (NMSA, 1982; NMSA, 1995) for the two grade configurations in

PAGE 154

154 relation to sixth grade studen t achievement Finally, this study determines if any of the se practices are predictive of student achievement in relation to the two grade configurations for the population studied Summary of Results This research studied a population of 12,727 sixth grad e students and 196 middle level educators from 32 K 8 schools and 32 6 8 middle schools. All schools were located with in 4 large urban districts located in the southeast United States. To meet the study qualifications, the districts had to have over 100,00 0 students enrolled during the 2010 2011 school year, and they must also have had combination schools that were both non charter and non magnet that housed students in grades kindergarten through eight. Of the seven large districts with in the identified st ate, only four districts had K 8 schools that fit these criteria. Using survey data from educators within the schools and student demographic and performance data provided by the districts, the four null hypotheses were tested. It is important to note that increased data requirements in this new age of accountability have created a strain on the human resources within the respective research and evaluation divisions of large districts, making external requests for data extremely difficult to fulfill. With respect to the level of implementation of the model practices of interdisciplinary teaming, common teacher planning, heterogeneous grouping, advisory groups, exploratory courses, and looping, there were no significant differences in sixth student performan ce in reading or mathematics on the Florida Comprehensive Achievement Test. There was, however, a significant difference in the reported level of implementation of interdisciplinary teaming, with the K 8 educators indicating a higher

PAGE 155

155 level of implementatio n. There were no differences in levels of implementation of any of the remaining model practices. Next, sixth grade performance data on the state assessment in reading and math were compared to determine if there were differences by school grade configurat ion. This test revealed significant student performance differences, with sixth grade students in K 8 schools performing significantly better in both reading and math. When student performance data was compared by both level of implementation of model prac tices and grade configuration, only grade configuration was significant with respect to student performance. The following research questions were addressed in this study. 1. Is there a relationship between any of the exemplary practices identified by NMS A /AMLE, including ( a) interdisciplinary teaming, ( b) common teacher planning time, ( c) heterogeneous grouping, ( d) student advisory, ( e) exploratory courses, and ( f) looping, and student achievement as measured by student gains on the Florida Comprehensi ve Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students in K 8 and 6 8 middle schools in large urban school districts when controlling for SES and minority rate ? From a group of 12,727 sixth grade students gain scores were computed from the 2010 and 2011 FCAT Developmental Scale Scores in both reading and math. The degree of implementation of each of the six identified practices was determined by educator survey from administers, academic coaches, and inst ruction staff within each of the schools with 19 6 educators responding (72.6% response rate) Nonparametric h 2 tailed significance at the .05 level were used to determine no significant relationships between any of th e model practices with reading or math gain scores. Additionally, a linear regression test was performed with each of the model practices as the dependent variable and school level factors entered

PAGE 156

156 as the independent variables. The only significant predicto rs found were a negative relationship between interdisciplinary teaming and school sixth grade enrollment size a positive relationship between common planning and school ESE rate, a negative relationship between exploratory courses and minority rate, and a positive relationship between looping and minority rate. Since none of the model practices indicate a significant relationship with reading or math gain scores, the null hypothesis was retained. The results of hypothesis 1 are summarized in Table 5 1. Ta ble 5 1. Summary of findings for hypothesis 1* Model Practice Achievement Area p= Status of Null Hypothesis 1 Interdisciplinary Teaming Reading Mathematics 0.139 0.286 Retained Retained Common Planning Reading Mathematics 0.845 0.329 Retained Retained Heterogeneous Grouping Reading Mathematics 0.399 0.583 Retained Retained Student Advisory Reading Mathematics 0.057 0.520 Retained Retained Exploratory Courses Reading Mathematics 0.540 0.121 Retained Retained Looping Reading Mathematics 0.763 0.26 0 Retained Retained *Null hypothesis 1 states that t here is no relationship between the implementation of exemplary middle level practices and student achievement gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students in an u rban school district when controlling for school demographic factors of school size, grade level size, minority rate, SES rate, LEP rate, and ESE rate. Question 2: Are there differences in levels of implementation of NMSA/AMLE identified exemplary practic es in urban K 8 configured schools when compared to 6 8 configured middle schools?

PAGE 157

157 Educators from 32 6 8 configured schools and 28 K 8 schools responded to a survey regarding the level of implementation of each of the model middle level practices within t heir schools. A t Test for independent samples was used to compare the means of the self reported implementation levels for K 8 and 6 8 schools. Of the practices, only interdisciplinary teaming had significantly different levels of implementation, with the K 8 schools reportedly implementing the practice more fully. Therefore, all null hypotheses were retained except for Ho 2 1, regarding differences in the implementation of interdisciplina ry teaming, which was rejected. A summary of the results of hypothesis 2 are found in Table 5 2. Table 5 2. Summary of findings for hypothesis 2* Model Practice p= Status of Null Hypothesis 2 Interdisciplinary Teaming 0.010 R ejected Common Planning 0.087 Retained Heterogeneous Grouping 0.712 Retained Student Advisory 0. 616 Retained Exploratory Courses 0.442 Retained Looping 0.638 Retained *Null hypothesis 2 states that t here are no significant differences in the implementation of exemplary middle level practices in schools configured as K 8 compared to traditional mid dle schools with a 6 8 grade configuration. Question 3: Are there differences in academic achievement in reading and mathematics for sixth grade public school students in K 8 configured schools when compared to sixth grade students in 6 8 configured middl e schools, as measured by the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system? To test this question, differences in two years of reading and math developmental scale scores from 2010 and 2011 were calculat ed for 3,535 sixth grade students in K 8 schools and 9,152 students in schools with a 6 8 configuration. T Tests for independent samples were calculated to compare the achievement ga in means by school type at the

PAGE 158

158 .05 level of significance. Both reading and math achievement gain scores were found to be significantly higher in the schools with the K 8 grade configuration, therefore the null hypotheses were rejected for both reading and math. The results of hypothesis 3 are summarized in Table 5 3. Table 5 3. Summary of findings for hypothesis 3* Subject p= Status of Null Hypothesis 3 Reading <0.001 Rejected Mathematics <0.001 Rejected Null hypothesis 3 states that t here are no significant differences in student achievement gains as measured by the FCAT SS S Developmental Scale Score for sixth grade students in a K 8 school compared to the same students in a traditional 6 8 middle school. Question 4: Is there a relationship between the level of implementation of exemplary middle level practices and academic achievement as measured by the Florida Comprehensive Tests of Achievement Sunshine State Standards (FCAT SSS) assessment system for sixth grade students enrolled in urban K 8 configured schools compared to similar students in 6 8 middle schools? This res earch question tested whether the combined factors of grade configuration and implementation of exemplary middle level practices in each of the two sixth grade settings were predictive of student achievement. Achievement gain scores for 3,535 sixth grade s tudents in K 8 schools and 9,192 sixth grade students in 6 8 schools were compared using GLIMMIX, a general linear mixed model in SAS Scores were compared by school configuration and level of implementation of each of the model practices to determine if there was a significant relationship between factors. Type III Tests of Fixed Effects showed that although several of the school and student demographic factors, including school configuration tested in question 3, related significantly to student achievem ent gain scores, none of combinations between grade

PAGE 159

159 configuration and model practices were statistically significant. The findings for hypothesis 4 are summarized in Table 5 4. Table 5 4 Summary of findings for hypothes is 4* Effect Subject p= Status of Nu ll Hypothesis Interdisciplinary Teaming Reading Math 0.8975 0.6730 Retained Retained Common Planning Reading Math 0.3657 0.5293 Retained Retained Exploratory Courses Reading Math 0.7238 0.0932 Retained Retained Advisory Reading Math 0.4094 0.4736 R etained Retained Heterogeneous Grouping Reading Math 0.4825 0.0610 Retained Retained Looping Reading Math 0.7813 0.7344 Retained Retained *Hypothesis 4 states that t here is no relationship between exemplary middle school practices and student achieveme nt gains as measured by the FCAT SSS Developmental Scale Score for sixth grade students who attend K 8 schools implementing exemplary middle level practices compared to the same students attend traditional middle schools implementing the practices at the s ame level. Discussion The purpose of this research was to identify practices and grade configurations that may assist in increasing the academic achievement for middle level studen ts in an urban setting. The schools were specifically chosen from large urb an school districts to secure a representative sample of students in each of the two grade configurations that provided the typical demographics of students who may be considered more at risk. It was hoped that this research could provide insight for pract ices that might be effective for typically low performing middle schools. To that end, t his research provides several important implications for application in middle level education.

PAGE 160

160 Grade Level Configuration In the initial analysis it would appear that reconfiguring to a K 8 design may have some merit, at least for sixth grade student population studied in this research Other studies have found similar increased student achievement in schools with broader grade spans ( Alspaugh, 1995; Frankin & Glascock 1998; Offenberg, 2001; Tucker & Andrada, 1997; Wren, 2003 ). Because of other limiting factors in this research, however, there is little evidence to suggest that the increased student achievement gains were due, either wholly or in part, to the grade con figuration of the school. One of the limiting factors in this study was the SES profile of the K 8 school population. The districts chosen for this research are among the largest public school systems in the nation and represent a diverse body of students and teachers. By selecting these urban school districts, it was expected that the K 8 sample would be reflective of the demographics of the district as a whole. In the study, e ach K 8 school was matched to a 6 8 middle school as closely as possible by fre e/reduced price lunch and minority rates. This was a difficult process however, since many of the K 8 schools had much low er free/reduced price lunch rates than the middle schools within the same district Even though there were many more 6 8 configured s chools, the majority of the middle schools had much higher poverty rates than their K 8 counterparts To better illustrate this disparity, free and reduced price lunch rat es from one of the districts were examined by gra de configuration of the school. The re was a huge difference in the percentage of students on free or reduced price lunch in the K 8 schools as compared to the students in the schools with a 6 8 grade configuration. For example 92% of the middle schools in this district had more than half o f the students on

PAGE 161

161 free or reduced price lunch compared to only 54% of the K 8 schools. In comparing the number of Title 1 schools within the same district 88% of the middle schools qualified for Title 1 services, but only 46% of the K 8 schools had Title 1 eligibility. The intention of this study was to provide insight into promising practices for middle school students in urban areas who may be considered more at risk. Although the achievement gain scores in this study were controlled for SES and race/et hnicity, it appears that there is a differential distribution of K 8 schools that corresponds with socio economic patterns Because academic achievement is related most closely to SES status, and the K 8 schools in this study seem to be disproportionately located in more affluent sections of metropolitan areas, generalizing the results to other large urban districts or even other schools within the same district, would be a mistake. It is also not clear whether these gains will remain steady over the cour se of the three years of middle school or if the gains will begin to even out as the middle school years progress As Abella (2005) found in his research of a large public school district in the southeast the gains the K 8 schools enjoyed in sixth grade h ad nearly evened out at the completion of the students eighth grade school year The student achievement differences were also significantly related to grade level size, with students in s chools with smaller sixth grades faring better in reading gains th an their peers with larger sixth grade classes. The sixth grade level sizes for the K 8 schools in this study were considerably smaller than the middle schools which is consistent with other research on grade span configuration (Ecker, 2002; Renchler, 200 2). Earlier research also demonstrated a relationship between smaller grade level

PAGE 162

162 sizes and increased academic achievement ( Bickel & Howley, 2000; Mertens et al., 2001 ). The K 8 configuration draws students from 9 different grade levels (10 if the school includes prekindergarten), so grade level sizes need to be kept to approximately 100 to maintain school capacities of around 900 students. In order for a middle school with three grade levels to achieve the same enrollment, each grade level would need appr oximately 300 students. This means that the middle schools have grade levels approximately three times the size of K 8 schools. If student achievement is related to grade level size, the K 8 model would have an inherent advantage. There are many factors t hat may contribute to increased achievement in schools with smaller grade levels. Because of their smaller grade level size, the K 8 schools may configure their schedules differently. For example, K 8 schools may use two teacher teams more frequently than 6 8 schools. Often times, this two teacher team arrangement consists of a reading and a math teacher, who typically split of the remaining subjects between them based on interest certification or expertise. This may result in teaching time that is more fo cused on reading and math, which would account for the higher performance in those accountability areas (Blair, 2007). Because of the differences in grade level size and the differential placement of K 8 schools in more affluent areas of the school distri cts, the results in this study provide inconclusive evidence regarding the relationship between student achievement and grade configuration. Exemplary Middle Level Practices With respect to the implementation level of model practices, there is no real evid ence to support that the six identified practices contribute to student achievement

PAGE 163

163 gains. In addition, there is very little difference in the level of implementation of each of the practices between the grade level configurations. It is interesting to no te that one of the arguments for maintaining the sixth through eighth grade middle school model is the ability to implement developmentally appropriate practices. In practice, however, the K 8 model reports a higher overall level of implementation of those practices than their middle school colleagues, although most of those differences are not significant, and are not related to student achievement. In the 1980s when the middle school movement was in full swing, the emphasis for middle schools was on build ing relationships. The practices that were put in place during those years were designed to create a greater sense of belonging and cohesiveness for early adolescents during this time of physical and emotional transition. While quality instruction was stil l a priority, developing quality relationships with students was considered equally important. With the new age of accountability, however, there is a laser like focus on student achievement at every grade level and middle level students are not given a p ass because they are in a stage of transition While educators are certainly still concerned about the emotional well being of their students, the bottom line is proficiency in the accountability areas on the state tests. The implementation of double block s of courses intended to boost test scores has meant that many exp loratory programs have gone b y the wayside. Time spent in advisory programs has been exchanged for instruction in remedial math Heterogeneous grouping has been replaced by reading groups so rted by proficiency level. Principals today are faced with tough choices regarding how to best spend their most valuable resource instructional time.

PAGE 164

164 Because of the increased pressures of accountability, m ost school leaders are investing in those areas t hat bolster student achievement rather than their self esteem. One point worth noting is that level of implementation was determined by a survey of th e educators within the schools. Although the survey instrument provides definitions for each of the practi ces, there may be inconsistencies between educators these differences in understanding or perception, some educators may have under or over rated the implementation of the practices w ithin their respective schools. To develop a more accurate, consistent picture of the degree of model practice implementation, a team of experts would need to personally visit each school for a period of time and evaluate each practice on a consistent rubr ic. As other grade levels begin to adopt these practices, particularly in the high school small learning communities, it may be a worthwhile pursuit for further investigation. Limitations of the Research There were several limitations to this research. Th e study was limited to the Florida Comprehensive Achievement Test SSS data in reading and mathematics reported to the state of Florida for the 2009 2010 and 2010 2011 school years. Although the researcher recognizes that sixth graders can be educated in a variety of grade level configurations, this study was limited to two configurations: ( a) sixth grade students in schools with sixth through eighth grade, and ( b) pre kindergarten or kindergarten through eighth grade configurations. In addition, the study w as limited to a comparison of public K 8 and middle schools in urban districts larger than 100,000 students in the state of Florida participating in the state accountability system and may or may not be representative of

PAGE 165

165 other K 8 and middle schools in the United States. The study is delimited to the data provided by the research and evaluation departments in each of the identified districts of the participating schools. Only middle schools and K 8 schools identified by the Florida Department of Education ( FLDOE, 2011) were included in this study because an ample number of K 8 and comparable middle schools were available in that region, and all students took identical achievement tests. Because they may represent a special population, K 8 and middle schools identified as charter schools, magnet schools, or schools of choice were not included in this study. The level of implementation of each of the model practices was determined by a self assessment via survey of a sampling of educators within each of the sam ple schools. Information from educators within the school about implementation of practices was delimited to a single set of questions related to degree of implementation provided in an online or paper survey. Although results were averaged by several part icipants within the school to get a better degree of accuracy, the extent to which the survey responses by individual participants in each of the respective schools represent the actual practices within the schools may limit the quality of the findings. Th e study was further limited to only those students who had achievement scores for both the 2009 2010 and the 2010 2011 school years. In addition, the study only included students in the K 8 design who had achievement data for the previous school year withi n the same K 8 school. Therefore, this study fails to account for the achievement of students who transfer to or from different school systems. Schools with high mobility rates would tend to confound the number of school to school transitions. Achievement scores were limited to gains from fifth to sixth grade only. The study was

PAGE 166

166 based on the assumption that the student achievement data were appropriate measures for comparing academic achievement of middle schools and K 8 schools. A further assumption was th at comparable kind and quality of curriculum and instruction were present in both configurations of schools. Implications This study utilized data from the district, school, and individual student levels. Because of the layered nature of the data, t his res earch provides several interesting implications at all three levels of accountability Implications for Students The practices designed to ease the transition from elementary to middle school do not appear to contribute to increased achievement outcomes fo r students. For schools that abandon these practices, either due to limited resources or lack of effectiveness in terms of student achievement, it will be interesting to note other effects on students. Outcome measures such as student attendance and discip linary infractions may be areas affected as a result. The elimination of a school to school transition inherent in the K 8 design has uncertain effects on student achievement. Special groups such as students with disabilities, students in high poverty are as, or limited English proficient students may benefit from the continuity in educational services Focused research on special populations within the two configurations may provide additional insight on the advantages and disadvantages of each design. An examination of students who stay in public school and those who choose to leave may account for differences in aggregate student performance Achievement data that compares the students who remain in public school and those that leave after fifth

PAGE 167

167 grade may provide additiona l insight into the dip in student achievement that typically occurs in the year after the elementary to middle school transition. Implications for Schools Due to accountability legislation, schools have been forced to scrutinize the effe ctiveness of their practices like never before. Exploratory courses and advisory periods may be luxuries that schools can no longer afford, in terms of both time and money. In many schools, s tudents in need of intensive remediation in reading and math are now provided specialized instruction that formerly would have been labeled as tracking. Common teacher planning is being replaced with departmentalized professional learning communities designed to increase student learning in state accountability areas. Because there is no evidence that these model practices lead to increased student learning, school leaders will need to make hard choices regarding which practices to abandon and which to retain. It will be interesting to note the research in high schools over the next decade, as they move to smaller learning communities and begin to implement many of these same practices. Grade configuration also has important implications for instruction at the school level. The certifications of the teachers within each of the grade level configurations, the ability to recruit and retain quality teachers, and the organization of the teachers within the schools all play an important role in outcomes for students. One area that may warrant further investigation is teacher certification in the two grade configurations. Typically, teachers in elementary schools have certifications in grades K 6, while middle school teachers have secondary certifications in subject areas. Hig hly Qualified Teacher legislation has imposed additi onal mandates on teacher

PAGE 168

168 certification. Because the majority of the children in K 8 schools are elementary students, it would be reasonable to assume that the majority of teachers have elementary certifications. To increase flexibil ity of teacher assignmen t within the school, K 8 schools may hire a greater percentage of teachers that have both elementary education and secondary subject area certifications. This may be an advantage for the K 8 design. Teacher r ecruitment and retention may be another area of difference for the two grade configurations. If K 8 schools are disproportionately place d in more affluent areas, their ability to attract and retain high quality teachers may be enhanced. As an added advantage, e lementary schools have historically had an easier time of recruiting teachers t han middle schools. In addition, t eachers may choose to stay longer in K 8 schools. Further research regarding the number and quality of applicants to K 8 schools and their tenure within those schools, in comparison to m iddle schools may reveal patterns of interest. Finally, there may be differences in organizational structures of the two school configurations. With smaller grade levels, K 8 schools may opt for two or three teacher teams with a focus on core academic ar eas. Fewer classroom transitions during the day could lead to increased instructional time that ultimately results in increased student achievement. Areas of interest for further research include amount of time focused on instruction by grade configuration instructional time disaggregated by accountability areas, and differences in grade level teacher configurations. Implications for School Districts As stated previously, there seems to be a differential distribution for K 8 school placement that correspon ds to SES rates. It would appear that K 8 schools are being

PAGE 169

169 placed or reconfigured in more affluent areas within the metropolitan regions. This could be happening for a variety of reasons. New schools are often built to relieve overcrowding in high growth suburban areas In this era of extremely limited resources, it would make more financial sense to build a new school configured to the K 8 design, rather than retrofit an existing facility. The K 8 reconfiguration could simultaneously relieve overcrowding in both elementary and middle schools of the adjacent boundary. Because they draw a wider span of grade levels, K 8 schools are more typically located within neighborhoods and are closer to student homes, which may relieve some of the transportation costs associated with the larger schools that bus students from a wider boundary. Districts may also be using the K 8 configuration to keep students in public schools longer. While many parents express satisfaction with their neighborhood elementary school, qui te often they are less enthusiastic about the middle school to which their child is assigned. Elementary schools are often smaller and closer to the homes of their students. Because middle school grade level sizes are approximately three times larger than the K 8 model, they typically have much broader school boundaries that encompass two or three elementary schools. During the transition from elementary to middle school, some parents are choosing smaller private schools, parochial schools, or even choosin g to home school their children. Because there are associated costs with each of these choices, this option is not available to all parents. Although charter schools are a cost free option for all families, at this time there are very few quality charter s chools available. Of the high

PAGE 170

170 quality charters, admission is quite competitive, further limiting the options for low income families. In contrast, the narrower K 8 boundaries allow children to stay in the neighborhood schools for an additional three years. This proximity leads to greater personalization and invites increased parental participation. In addition, the natural transition between fifth and sixth grade is eliminated, whic h has been an historic decision point for parents who choose to leave public school. This makes sixth grade just another year at the neighborhood school. Parents who are already satisfied with the school and have had leave. Parental pressure m ay also be a contributing factor to the disproportionate distribution of K 8 schools in more affluent areas. Parents in mid to high socioeconomic neighborhoods are better prepared to navigate the bureaucracy of large school districts and are more likely to collectively advocate for their schools. Particularly in hard economic times, when middle class budgets are tight and private school tuition may be an uncomfortable stretch to family finances the K 8 configuration is an attractive option for parents. By creating K 8 schools, districts may be stemming the exodus from public to private schools for children from middle and high income families. Since academic achievement is closely tied to socioeconomic factors, this could provide a boost for districts stru ggling to meet accountability targets. If the intended purpose of moving to the K 8 design is increased student achievement, districts need to be mindful of where these schools are placed. Because of their narrower boundaries, K 8 schools can add to segreg ation and have the

PAGE 171

171 appearance of being selective schools of choice in more affluent communities. If the configuration of a school has the potential of raising academic achievement for students, then those schools should be placed in the areas of greatest n eed. Further study regarding district policies procedures and motivations for school configuration and placement would shed light on this complicated area. Summary There are a number of large urban school systems throughout th e county that are moving or plann ing to move from the middle school 6 8 design to the K 8 configuration schools (Look, 2002a; OPPAGA, 2005; Pardini, 2002; Patton, 2005). The K 8 grade configuration may be a popular trend in middle level education that may provide some lift in terms of student achievement however that increase may well be due to other factors such as grade level size or socioeconomic differences rather than the configuration of the school. Regarding the model practices of interdisciplinary teaming, student advisory, common teacher planning, exploratory courses, heterogeneous grouping, and looping, there was little difference in the level of implementation between the two grade configurations. T here is also no evidence that any of the model practices studied in this de sign lead to increased student academic achievement. While these practices may be important in providing support to students during their transitional years, of even greater importance to schools in this era of accountability is the ir impact on student ach ievement. Greatly reduced budgets and pressure to perform will continue to have an impact on how schools spend their time and align their resources.

PAGE 172

172 APPENDIX A UNIVERSITY OF FLORID A IRB CONSENT

PAGE 173

173 APPENDIX B DISTRICT CONSENTS TO CONDUCT RESEARCH

PAGE 174

174

PAGE 175

175

PAGE 176

176

PAGE 177

177

PAGE 178

178 APPENDIX C MIDDLE LEVEL EDUCATO R SURVEY PARTICIPANT CONSENT FORM

PAGE 179

179 APPENDIX D MIDDLE LEVEL EDUCATO R SURVEY

PAGE 180

180

PAGE 181

181

PAGE 182

182

PAGE 183

183 LIST OF REFERENCES Abella, R. (2005, Sept). The effects of small K 8 centers compared to large 6 8 schools on student performanc e Middle School Journal, 37 (1), 29 35. Alexander, W. M., & McEwin, C. K. (1989a). Schools in the middle: Status and progress Columbus, OH: National Middle School Association. Alexander, W. M., & McEwin, C. K. (1989b). Schools in the middle: Progress 1 968 1988. A report on trends and practices. Reston, VA: National Association of Secondary School Principals. Allison, P. D. (2008). Fixed effects regression methods in SAS Retrieved October 27, 2011, from http://www2.sas.com/proceedings/sugi31/184 31.pdf Alspaugh, J. W. (1995). A comparison of four enrollment groups of K 8 and K 12 Missouri rural school districts Paper presented at the a nnual c onference of the National Rural Education As sociation Salt Lake City, UT. ( ERIC Document Reproduction Service No. ED389501) Alspaugh, J. W. (1998). Achievement loss associated with transition to middle school and high school. Journal of Educational Research, 92 (1), 20 25. Alspaugh, J. W., & Harti ng, R. D. (1995). Transition effects of school grade level organization on student achievement. Journal of Research and Development in Education, 28 (3), 145 149. Alt, M., Choy, S., & Hammer, C. (2000). In the middle: Characteristics of public schools with focus on middle schools (NCES 2000 312). Washington, DC: US Department of Education National Center for Education Statistics. Anderman, E. (2002). School effects on psychological outcomes during adolescence. Journal of Educational Psychology, 94 795 809 Anfara, V, & Lipka, R. (2003). Relating the middle school concept to student achievement. Middle School Journal, 35 (1), 24 32. Arhar, J. (1994). Personalizing the social organization of middle level schools: Does interdisciplinary teaming make a differ ence? In K.B. Borman & N.P. Greenman (Eds.) Changing American education : Recapturing the past or inventing the future? 325 350. Albany, NY: State University of New York Press. Association for Middle Level Education (AMLE) (2011). A name change for NMSA: A ssociation for Middle Level Education Westerville: OH: Author. Retrieved August 5, 2011 from www.nmsa.org

PAGE 184

184 Aycock, C. E. (2005). The effect of interdisciplinary teaming on student achievement in middle level schools. (Do ctoral Dissertation, University of Southern Mississippi, 2005). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3180132) Balfanz, R. & MacIver, D. (2000). Transforming high poverty urban middle schools into strong learning institutions: Lessons from the first five years of the Talent Development Middle School. Journal of Education for Students Placed at Risk, 5 (1/2), 137 158. Balfanz, R., Spiridakis, K., & Neild, R. C. (2002). Will converting high poverty middle schools to K 8 schools f acilitate achievement gains: A research brief for the school district of Philadelphia Philadelphia: Philadelphia Education Fund. Retrieved January 5, 2009 from http://ww w.philaedfund.org/pdfs/Converting_High Poverty_Schools_2002.pdf Baltimore City Public School System Division of Research, Evaluation, and Accountability. (2001, November). An examination of K 5, 6 8 versus K 8 grade configurations: A research study cond ucted for the new board of school commissioners. Baltimore, MD: Author. Banks, R. (2004). Middle school Urbana: University of Illinois at Urbana Champaign, Clearinghouse on Early Education and Parenting. Barth, J. (2001). The investigation of the relat ionship between middle school organizational health, school size, and school achievement in the areas of reading, mathematics, and language. Dissertation Abstracts International, 62 (5), 1642. Beane, J. & Lipka, R. (2006). Guess again: Will changing the gr ades save middle level education? Educational Leadership, 63 (7), 26 30. Beaton, A., Mullis, I., Martin, M., Gonzalez, E., Kelly, D., & Smith, T. (1996). Mathematics achievement in the middle school years. Boston: TIMSS Study Center. Becker, H. (1987). A ddressing the needs of different groups of early adolescents: Effects of varying school and classroom organizational practices on students from different social backgrounds and abilities. (Report No. 16). Baltimore, MD: John Hopkins Center for Research on Elementary and Middle Schools. Bedard, K. & Do, C. (2005). Are middle schools more effective? The impact of school structure on student outcomes. Journal of Human Resources, 40 (3), 660 682. academic achievement motivation: An experimental study. Journal of Educational Psychology,82 (4), 664 670.

PAGE 185

185 Betts, J. R. & Shkolnik, J. L. (2000). The effects of ability grouping on student achievement and resource allocation in secondary schools. Economics of Education Review, 19 (1), 1 15. Bickel, R. & Howley, C. (2000). The influence of scale on school performance: A multi level extension of the Matthew Principle. Education Policy Analysis Archives, 8 (22), 1 32. Bickel, R., Howley, C., Williams, T., & Glascock, C. (2001). High school size, achievement equity, and cost: Robust interaction effects and tentative results. Educational Policy Analysis Archives, 9 (40). Black, D. (2009). A quantitative study of school configuration and the ach ievement of sixth grade students in Alabama. (Doctoral Dissertation, Alabama State University, 2009). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3362648) Blair, N. C. (2007). A comparative study of the effects of grade configurati on on middle school and K 8 school value added scores (Doctoral Dissertation, Tennessee State University, 2007). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3260215) Blyth, D. A., Simmons, R. G., & Bush, D. (1978). The transition in to early adolescence: A longitudinal comparison of youth in two educational contexts. Sociology of Education 51 (3), 149 162. Boren, D. L. (1997). Professional middle level teacher preparation programs: A needs assessment survey. (Doctoral Dissertation, O hio University, 1997). Ann Arbor, MI: ProQuest Information and Learning Company. (UMI No. 9840278) Bracey, G. (1997). Setting the record straight: Responses to misconceptions about public education in the United States Alexandria, VA: ASCD. Brown, D. F. (2001). The value of advisory sessions for urban young adolescents. Middle School Journal, 32 (4), 14 22. Brown, E. (2004). The effect of number of school to school transitions on district performance. (Doctoral Dissertation, Ohio University, 2004). Disse rtation Abstracts International, 65 (9), 3224. (UMI No. 3149014) Burton, J., Horowitz, R., & Abeles, H. (2000). Learning in and through the arts: The question of transfer. Studies in Art Education, 41 (3), 228 257.

PAGE 186

186 Bush, B. L. (1999). Factors that support and impede full institutionalization of the middle school concept. (Doctoral Dissertation, University of Florida, 1999). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 9956553) Butzlaff, R. (2000). Can music be used to teach reading? Journal of Aesthetic Education, 34 (3/4), 167 178. Byrnes, V. & Ruby, A. (2007). Comparing achievement between K 8 and middle schools: A large empirical study. American Journal of Education, 114 (1), 101 135. Caccamo, J. M. (2000). Sharing the vision: Hea lthy, achieving students: What can schools do? Journal of School Health, 70 (5), 2168 2218. Carnegie Council on Adolescent Development. (1989). Turning points: Preparing American youth for the 21 st century. New York: Author. Carter, S. & Carter, C. (2000) How principals can attract teachers to the middle grades. Schools in the M iddle, 9 (8), 22 25. Caswell, C. (2003). Personalizing the school environment: Teacher based advisory programs that support student adjustment and academic outcomes. (Doctoral Diss ertation, Rhode Island University, 2005). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3115623) Coladarci, T. & Hancock, J. (2002a). Grade span configuration. Journal of Research in Rural Education, 17 189 192. Coladarci, T. & Han cock, J. (2002b). The (limited) evidence regarding effects of grade span configuration on academic achievement: What rural educators should know Retrieved from ERIC database. (ED467714). Connors, N. A. (1992). Teacher advisory: The fourth R. In J. L. Irv in (Ed.), Transforming middle level education (162 178). Needham Heights, MA: Allyn & Bacon. investigation into characteristic differences between middle grade school c onfigurations at Milwaukee Public Schools. Retrieved August 7, 2011 from Milwaukee Public Schools Web site: http://mpsportal.milwauke e.k12.wi.us/portal/server.pt/comm/research_and_evalu ation/340/research_reports/41172 Cook, P., MacCoun, R., Muschkin, D., & Vigdor, J. (2007). Should sixth grade be in elementary or middle school? An analysis of grade configuration and student behavior Retrieved July 30, 2011 from http://www.pubpol.duke.edu/research/papers/SAN0701.pdf

PAGE 187

187 Cooksley, M. (2010). The student achievement decline at sixth grade and the practices impacting achi evement. (Doctoral Dissertation, University of Northern Iowa, 2010). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3424826) Cooney, S. & Bottoms, G. (2003). Middle grades to high school: Mending a weak link Atlanta, GA: Southern Reg ional Education Board. Craig, J. A. (2006). Grade configurations. School Planning & Management 45 (3), 42 43. Daniels, E. (2005, April). On the minds of middle schoolers: Students in their first year of middle school talk about teachers, peers, homework and stress. Educational Leadership, 62 (7), 52 54. Darling Hammond, L. (1997). Doing what matters most: Investing in quality teaching. In Carnegie Corporation of New York, Turning Points 2000: Educating adolescents in the 21 st century New York: Teachers College Press. Darling Hammond, L. (2000). Teacher quality and student achievement. Educational Policy Analysis Archives, 8 (1). DeJong, W. S. & Craig, J. (2002). Age appropriate schools: How should schools be organized? School Planning & Management, 41 ( 6), 26 32. Dorman, G., Lipsitz, J, & Verner, P. (1985). Improving schools for young adolescents. Educational Leadership, 42 (6), 4 49. Dove, M. J., Pearson, L. C., & Hooper, H. (2010). Relationship between grade span configuration and academic achievement Journal of Advanced Academics, 21 (2), 272 298. Dunne, D. W. (2001). When size matters: Making big schools feel small. An Education World e interview with Paul S. George. Education World Retrieved July 16, 2005, from http://content.educationworld.com/a_issues/issues153.shtml Eccles, J. S., Lord, S., & Midgley, C. (1991, August). What are we doing to early adolescents? The impact of educational contexts on early adolescents. Amer ican Journal of Education, 99 521 542. Ecker, M. (2002). Middle schools still matter. School Administrator, 59 (3), 30 31. Education budget cuts: Schools face fiscal cliff as stimulus money runs out. (2011, March 6). The Huffington Post Retrieved August 8, 2011 fro m http://www.huffingtonpost.com

PAGE 188

188 Epstein, J. L. (1990). What matters in the middle grades grade span or practices? Phi Delta Kappan, 71 438 444. Epstein, J. L., & MacIver, D. (1990). The middle grades: Is grade span the most important issue? Educational Horizons, 68 88 94. Epstein, J. L., McPartland, J. M., & MacIver, D. J. (1991). The Hopkins enhancement survey of NELS:88 middle grades practices (Codebook and Data Collection Instruments). Wa shington, D.C.: U. S. Department of Education, Office of Educational Research and Improvement. Erb, T. O. & Stevenson, C. (1999). From faith to facts: Turning Points in action What difference does teaming make? Middle School Journal, 30 (3). Felner, R. D., Jackson, A. W., Kasak, D., Mulhall, P., Brand, S., & Flowers, N. (1997). The impact of school reform for the middle years: Longitudinal study of a network engaged in Turning Points based comprehensive school transformation. Phi Delta Kappan, 78 (7), 52 8 532, 541 550. Felner, R. D., Seitsinger, A. M., Brand, S., Burns, A., & Bolton, N. (2007). Creating small learning communities: Lessons from the project on high performing learning e nhancing, learning contexts. Educational Psychology, 42 (4), 209 221. Ferrell, J. (2008). A comparison of the mixed procedure and the Glimmix procedure Paper presented at the SAS Global Forum virtual conference. Retrieved October 27, 2011, from http://www2.sas.com/proceedings/sugi31/189 31.pdf Fibkins, W. L. (1999). Stronger advisory periods can stem the spread of school violence. Middle Ground, 3 (2), 41 43. Florida Department of Education (FLDOE). ( 2011a ). 2010 2011 School Accountability Reports Retrieved July 18, 2011 from http://schoolgrades.fldoe.org/ Florida Department of Education (FLDOE). (2011b). Florida Comprehensive Assessment Test Retrieved July 9, 2011 from http://fcat.fldoe.org/ Flowers, N., Mertens, S., & Mulhall, P. (1999). The impact of teaming: Five research based outcomes. Middle School Journal, 31 (2), 57 60. Franklin, B. J., & Glasco ck, C. H. (1998). The relationship between grade configuration and student performance in rural schools. Journal of Research in Rural Education, 14 (3), 149 153.

PAGE 189

189 Franklin, J. (2001). Trying too hard? How accountability and testing are affecting constructiv ist teaching. Education Update, 43 (3), 1, 4 5, 8. Freeman, C. (2005). A comparison of K 8 schools, junior high schools, and middle schools with respect to student academic achievement and attendance in North Carolina. (Doctoral Dissertation, Western Carol ina University, 2005). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3194231) Fuller, B. D. (2006). The result of looping on the Mississippi Curriculum Test in a middle school. (Doctoral Dissertation, Mississippi State University, 20 06). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3211236) Galassi, J. P., Gulledge, S. A., & Cox, N. D. (1997). Middle school advisories: Retrospect and prospect. Review of Educational Research, 67 (3), 301 338. Gamoran, A., & Hanni gan, E. C. (2000). Algebra for everyone? Benefits of college preparatory mathematics for students with diverse abilities in early secondary school. Educational Evaluation and Policy Analysis, 22 (3), 241 254. George, P. S., & Alexander, W. M. (1993). Exemp lary middle school. Orlando: Harcourt College Publishers. George, P. S., & Lounsbury, J. H. (2000). Making big schools feel small: Multiage grouping, looping and schools within a school Columbus, OH : National Middle School Association. George, P. S., & Oldaker, L. L. (1985/1986). A national survey of middle school effectiveness. Educational Leadership, 43 (4), 79 85. George, P. S., & Shewey, D. (1994). New evidence for the middle school. Columbus, OH : National Middle School Association. Gewertz, C. (200 4, May 19). City districts embracing K 8 schools. Education Week, 23 (37), 1 20. Goldschmidt, P., & Wang, J. (1999). When can a school affect dropout behavior? A longitudinal multilevel analysis. American Educational Research Journal, 36 (4), 715 738. Gome z, M. O. (2010). School culture and performance at different middle level structures. (Doctoral Dissertation, University of California at Riverside, 2010). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3433911) Goodwin, J. (2003). Ur Dissertation Abstracts International, 64 (05), 1589. (UMI No. 3091252)

PAGE 190

190 Gray, C. (2004). The effects of interdisciplinary teaming on the preparation of adolescents for high school. (Doctoral Dissertation Union University, 2004). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3147073) Gregory, B. S. (2009). The impact of looping on academic and social experiences of middle school students. (Doctoral Dissertation, Walden University, 2 000). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3369795) Grolnick, W. Kurowski, C., Dunlap, K., & Hevey, C. (2000). Parental resources and the transition to junior high. Journal of Research on Adolescence, 10 465 488. Hackman, D & Valentine, J. (1998). Designing an effective middle school schedule. Middle School Journal, 29 (5), 3 13. Haycock, K. & Ames, N. (2000). Where are we now? Taking stock of middle grades education Keynote Address at the National Educational Research Pol icy and Middle Grades: Linking Research and Practice, July 24 25, 2000, Washington, DC. Heller, R., Calderon, S. & Medrich, E. (2003). Academic achievement in the middle grade s: What does the research tell us? A review of the literature Atlanta, GA: Southern Regional Education Board. Herman, B. (2004). The revival of K 8 schools. Phi Delta Kappa Fastbacks 519, 7 37. Hertzog, C. J., & Morgan, P. L. (1999). Making the transit ion from middle level to high school. High School Magazine, 6 (4), 26 30. Hetland L. (2000). Listening to music enhances spatial temporal reasoning: Evidence Journal of Aesthetic Education, 34 (3/4), 105 148. Hochman, D., Tocci, C ., & Allen, D. (2005, April 11 15). Advisory programs in high school restructuring Paper presented at the annual meeting of the American Education Research Association, Montreal, Quebec. Hoffer, T. B., (1992). Middle school ability grouping and student a chievement in science and mathematics. Educational Evaluation and Policy Analysis, 14 (3), 205 227. Hough, D. L. (1989). Vertical articulation for the middle grades Riverside, CA: University of California, California Educational Research Cooperative. (ERI C Document No. 315896). Hough, D. L. (1995). The elemiddle school: A model for middle grades reform. Principal, 74 (3), 6 9.

PAGE 191

191 8 school truly applies best middle level practices and deserv es the designation. School Administrator 62 (3), 10 11, 13 14. Hough, D. L., & Sills Briegel, T. (1997). Student achievement and middle level programs, policies, and practices in rural America: The case of community based versus consolidated organizations Journal of Research in Rural Education, 13 (1), 64 70. Howley, C. B. (2002). Grade span configurations: Where 6 th and 7 th grades are assigned may influence student achievement, research suggests. School Administrator, 59 (3), 24 29. Hoy, W. K., & Sabo, D (1998). Quality middle schools: Open and healthy Thousand Oaks, CA: Corwin Press. Ingels, S., Abraham, S., Rasinski, K., Karr, R., Spencer, B., & Frankel, M. (1990). National education longitudinal study of 1988 base year student component data file us Washington, DC: National Center for Educational Statistics. Irvin, J. L., Valentine, J. W., & Clark, D. C. (1994). Essential elements of a true middle school: What should be versus what is. Middle School Journal, 26 (1), 54 58. Jackson, A. & Davis, G. (2000). Turning points 2000: Educating adolescents in the 21 st century. New York: Teachers College Press. Jenkins, D., & McEwin, C. (1992). Which school for the fifth grader? Programs and practices in three grade organizations. Middle School Jo urnal, 23 (4), 8 13. Johnson, A. M. (2002). Implementation of characteristics and provisions of developmentally responsive middle schools. (Doctoral dissertation: University of Southern Mississippi, 2002.) Dissertation Abstracts International, 63 3433. J ohnson, D. J. (2002). Academic achievement of rural students as they transition from elementary school to middle school (Doctoral dissertation, University of South Dakota 2002). Ann Arbor, MI: ProQuest Information and Learning. (UMI No. 3048762) Johnso n, T. (2005). An examination of the relationship between the type of teacher certification of middle school teachers and student math, reading, and writing achievement. (Doctoral Dissertation, University of Michigan, 2005). Ann Arbor, MI: ProQuest Informat ion and Learning Company (UMI No. 3164080) Juvonen, J., Le, V N., Kaganoff, T., Augustine, C, & Constant, L. (2004). Focus on the wonder years: Challenges facing the American middle school Santa Monica, CA: Rand.

PAGE 192

192 Kaiser, J. (2000). Are high stakes tes ts taking control? Schools in the Middle, 9 (7), 18 21. 8 versus middle school grade configurations: Assessing the benefits of K 8 schools. NWREL 11 (3), retrieved electronically on January 30, 2009, from http://www.nwrel.org/nwedu/11 03/research/ Koos, L. (1920). The junior high school. New York: Harcourt, Brace, & Howe. Krugman, P. (2011, February 28). Leaving children behind. The New York Times p. A23. Lashway, L. (2000). Educators reorganize schools to foster a sense of community Retrieved July 13, 2011, from http://eric.uoregon.edu/trends_issues/organi zation/index.html Lee, V. & Smith, J. (1995). Effects of high school restructuring and size on early gains in achievement and engagement. Sociology of Education, 68 241 270. Lee, V. E., Smith, J.B., Perry, T. E. & Smylie, M. A. (1999). Social support academic press, and student achievement: A view from the middle grades in Chicago. Chicago: Consortium on Chicago School Reform. Lipsitz, J. (1984). Successful schools for young adolescents New Brunswick, NJ: Transaction Books. Lipsitz, J., Mizell, M H., Jackson, A. W., & Austin, L. M. (1997). Speaking with one voice: A manifesto for middle grades reform. Phi Delta Kappan, 78 (7), 533 540. Look, K. (2002a). The great K 8 debate Philadelphia: The Philadelphia Education Fund. Retrieved January 20, 200 9, from http://www.philaedfund.org/notebook/TheGreatK8Debate.htm Look, K. (2002b ). K 8 schooling: A retreat? A mandate? An intervention? (Doctoral Dissertation, University of Pennsyl vania, 2002). Dissertation Abtracts International, 63 (5), 1649. Lounsbury, J. H. (1996). Personalizing the high school: Lessons learned in the middle. National Association of Secondary School Principals Bulletin 80 (584), 17 24. Lounsbury, J. H., & John ston, J. (1988). Life in the three sixth grades. Reston, VA: National Association of Secondary School Principals.

PAGE 193

193 Lounsbury, J. H., & Vars, G. F. (2003). The future of middle level education: Optimistic and pessimistic views. Middle School Journal, 35 (2), 6 14. Lucas, S. E., & Valentine, J. W. (2005). Grade configuration (Research Summary #1) National Middle School Association. Retrieved on April 22, 2005 from http://nmsa.org/research/ressum1.htm Lutz, P. A. (2007). The relationship between selected ch aracteristics of middle school teachers and eighth grade mathematics achievement on a state assessment. (Doctoral Dissertation, Marywood University, 2007). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3275694) MacIver, D. J., & Epst ein, J. L. (1993). Middle grades research: Not yet mature, but no longer a child. The Elementary School Journal, 93 519 533. MacLaury, S. & Gratz, Z. (2002). Advisories led by trained facilitators: Their impact on middle school students. Research in Mid dle Level Education Online, 26 (1), 1 10. Manning, M. L. (2000). A brief history of the middle school. Clearinghouse 73 (4), 192. Manning, M. L. & Saddlemire, R. (1996). Implementing middle school concepts in high schools. Clearing House, 69 (6), 339 342. Mason, D. A., Schroeter, D. D., Combs, R. K., & Washington, K. (1992). Assigning average achieving eighth graders to advanced mathematics classes in urban junior high. Elementary School Journal, 92 (5), 587 599. McCown, C. & Sherman, S. (2002). Looping f or better performance in the middle grades. Middle School Journal, 33 (4), 17 21. McEwin, C. K., & Alexander, W. M. (1990). Middle level programs and practices in elementary schools: Report of a national study Columbus, OH: National Middle School Associati on. McEwin, C., Dickinson, T., & Jacobson, M. (2004). How effective are K 8 schools for young adolescents? Middle School Journal, 37 (1), 24 28. McNeil, M. (2009). Budget pain dampening K 12 efforts: Governors, state legislatures seek ways to limit damage Education Week 28 (16), 1. McPartland, J. M., Coldiron, J. R., & Braddock, J. H. (1987). School structures and classroom practices in elementary, middle, and secondary schools. Report No. 14. Baltimore, MD: The John Hopkins University Center for Researc h on Elementary and Middle Schools.

PAGE 194

194 Mertens, S. & Flowers, N. (2003a). Middle school practices improve student achievement in high poverty schools. Middle School Journal, 35 (1), 33 43. Mertens, S. & Flowers, N. (2003b). The impact of middle grades certif ication on student performance: The proof is in the practices Paper presented at the Symposium on Middle Level Teacher Preparation sponsored by National Middle School Association, Charlotte, N.C. Mertens, S., Flowers, N., & Mulhall, P. (1998). The Middle Start Initiative phase 1: A longitudinal analysis of Michigan middle level schools. Urbana, IL: University of Illinois. Mertens, S., Flowers, N., & Mulhall, P. (2001). School size matters in interesting ways. Middle School Journal, 32 (5), 51 55. Miami D ade County Public Schools (M DCPS). (2011). M DCPS New School Prototypes. Retrieved August 7, 2011, from http://facilities.dadeschools.net/prototypes.asp Miller, T. L. (2008). Straddling the fence: The relationship of elementary versus achievement. (Doctoral Dissertation, University of Nevada at Reno). Ann Arbor, MI: ProQuest Information and Learning Company ( UMI No. 3339131) Mills, R. (1998). Grouping students for instruction in middle schools Champaign, IL: ERIC Clearinghouse on Elementary and Early Childhood Education (ERIC ED 419631). Mizell, H. (2005). Grade configurations for educating young adolescen ts are still crazy after all these years. Middle School Journal, 37 (1), 14 23. Moore, D. W. (1984). Impact of school grade organization patterns on seventh and eighth grade students in K 8 and junior high schools (ERIC Document Reproduction Service No. ED 245346). Moss, T. C., Jackson, D. J., & Jackson, R. G. (1979). Do fifth graders belong in the middle school? Middle School Journal, 10 (2), 3, 30 31. K. M., Chr ostowski, S. J., & Smith, T. A. (2000). TIMSS 1999 international Mathematics and Science Study at the eighth grade. Chestnut Hill, MA: International Study Center, Lynch School of Education, Boston College.

PAGE 195

195 National Association of Elementary School Principals (NAESP). (1984). Standards for quality elementary schools: Kindergarten through eighth grade. Reston, VA: Author. National Middle School Association (NMSA). (1982). This we believe Columbus, OH: Author. National Middle School Association (NMSA). (1995). This we believe : Developmentally responsive middle schools Columbus, OH: Author. National Middle School Association (NMSA). (2003). This we believe: Successful schools for young adolescents. Westerville, OH: Author. National Middle School Association (NMSA). (2011). A name change for NMSA: Association for Middle Level Education Westerville: OH: Author. Retrieved August 5, 2011 from www.nmsa.org. National Middle School As sociation & National Association of Elementary School Principals (NMSA & NAESP). (2000). Supporting students in their transition to school: A position paper jointly adopted Westerville, OH: Author. Nave, B., Miech, E., & Mosteller, F. (2000). A lapse in standards: Linking standards based reform with student achievement. Phi Delta Kappan, 82 (2), 128 132. Neighbors, M. D. (1998). An investigation of the relationship of middle school practices toral Dissertation, Auburn University, 1998). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 9835351) Neill, M. W. (1999). An analysis of the implementation of the essential elements in accredited Montana middle schools. (Doctoral Dis sertation, University of Montana, 1999). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 9939782) Newman, B., Lohman, B., Newman, P., Myers, M., & Smith, V. (2000). Experiences of urban youth navigating the transition to ninth grade. Y outh and Society, 31, 387 416. to 8 schools compared to middle grades schools: The power of smallness in urban middle grades schooling. Middle School Journal, 32 (4), 23 29.

PAGE 196

196 Office of Program Poli cy Analysis & Government Accountability (OPPAGA). (2005, January). K 8 schools may help school districts improve student performance OPPAGA Report No. 05 02. Tallahassee, FL: Author, an office of the Florida Legislature. Retrieved electronically on Januar y 20, 2009, from http://www.oppaga.state.fl.us/reports/pdf/0502rpt.pdf Paglin, C., & Fager, J. (1997). Grade configuration: Who goes where? Portland, OR: Northwest Regional Educational Laboratory. Pardini, P. (2002). Revival of the K 8 school. School Administrator, 59 (3), 6 12. Pate, P. E., & Muth, K. D. (2003). Perspectives on the middle school movement: Snapshots from the past. Middle School Journal, 35 (2), 15 22. Patton, C. (2005 ). The K 8 bunch: K 8 schools are growing in popularity across the country. Do they really lead to fewer discipline problems and better academic performance? District Administration 41 (2), 44 47. Peak, L. (1996). Pursuing excellence: A study of U. S. eig hth grade mathematics and science teaching, learning, curriculum, and achievement in international context: Initial findings from the Third International Mathematics and Science Stud y Retrieved August 7, 2011, from National Center for Education Statistics US Department of Education Institute of Education Sciences Web site: http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=97198 Phillips, J. (2003). Powerful learning: Creating learning co mmunities in urban school reform. Journal of Curriculum and Supervision, 18 (3), 240 258. Poliner, R. A., & Lieber, C. M. (2004). The advisory guide: Designing and implementing effective advisory programs in secondary schools Cambridge, MA: Educators for Social Responsibility. Ranta, R. (2001). Implementation levels of, supports and barriers to, and influence of the essential characteristics of effective middle schools in New Jersey Blue Ribbon middle schools. (Doctoral Dissertation, Seton Hall University 2001). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3026733) Raymond, M. (2005). The effect of exploratory classes and other factors on student achievement in Colorado middle schools. (Doctoral Dissertation, University of Colorado at Denver, 2005). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3199617) Reed, C, Dickinson, G., & Cuozzo, C. (2000). Middle school catchers in the rye: The Research in Middle Level Education Annu al, 23 35 48.

PAGE 197

197 Reeves, K. (2005). Figuring and reconfiguring grade spans: While most remixes result from enrollment challenges, some changes start with academics in mind. School Administrator 62 (3), 16 21. Reising, B. (2002). Middle school models. Clear inghouse, 76 (2), 60 61. Renchler, R. (2000, Spring). Grade span National Association of Elementary School Principals 16(3), 1 4. Renchler, R. (2002). School organization: Grade span, trends, and issues Eugene, OR: ERIC Clearinghouse on Educational Ma nagement. Retrieved from ERIC database (ED472994). Richardson, A. R. (2008). An examination of teacher qualifications and student achievement in mathematics. (Doctoral Dissertation, Auburn University, 2008). Ann Arbor, MI: ProQuest Information and Learnin g Company (UMI No. 3333147) Ritzenthaler, B. K. (1993). An investigation of key programs and practices of the middle school concept in institutionalized and non institutionalized middle schools in Florida. (Doctoral Dissertation, Florida State Universit y, 1993). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 9334268) Rockoff, J. E. & Lockwood, B. B. (2010). Stuck in the middle. Education Next, 10 (4), 68 75. Roeser, R., Midgley, C., & Urdan, T. (1996). Perceptions of the psychologic al school: The mediating role of goals and belonging. Journal of Educational Psychology, 88 408 422. Rudolph, K. D., Lambert, S. F., Clark, A. G., & Kurlakowsky, K. D. (2001). Negotiating the transition to middle school: The role of self regulatory processes. Child Development, 72( 3), 929 946. Russell, J. (1997). Relationships between the implementation of middle level program concepts and student achievement. Journal of Curri culum and Supervision, 12 (2), 169 185. Sailor, P. (1986). The effects of various middle grade configurations Austin, TX : Austin Independent School District. ( ERIC Document Reproductio n Service No. ED297417) SAS Institute Inc. (2008). SAS 9.2 Enhanced Logging Facilities Cary, NC: Author. Sociology of Education, 72 (4), 216 233.

PAGE 198

198 Schmidt, W., McKnight, C., Jakwerth, P., Cogan, R., & Houang, R. (1999). Facing the c onsequences: Using TIMSS for a closer look at the United States mathematics and science education. Dordrecht: Kluwer. Schmitt, V. (2004). The relationship between middle level grade span configuration, professional development, and student achievement. Re search in Middle Level Education Online, 27 (2), 1 17. Schrieber, J. B. & Chambers, E. A. (2002). After school pursuits, ethnicity and achievement for 8 th and 10 th grade students. Journal of Educational Research, 96 (2), 90 100. Shoffner, M. F. & Williams on, R. D. (2000). Facilitating student transitions into middle school. Middle School Journal, 31 (4), 47 51. Simmons, R. G., & Blyth, D. A. (1987). Moving into adolescence: The impact of pubertal change and school context. New York: Aldine DeGruyter Slavi n, R. E. (1993). Ability grouping in middle grades: Achievement effects and alternatives. Elementary School Journal, 93 (5), 535 552. Sparks, K. (2004). The effect of teacher certification on student achievement. (Doctoral Dissertation, Texas A&M Universit y, 2004). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3172168) SPSS, Inc. (2010 ). SPSS Statistics GradPack for Windows, Version 19 .0 Chicago: Author. Spurgeon, S. (2003). The relationship between a middle school implementation an d student performance. (Doctoral Dissertation, Saint Louis University, 2003). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3102935) Stevenson, C., & Erb, T. (1998). How implementing Turning Points improves student outcomes. Middle S chool Journal, 30 (1), 49 52. Strauss, V. (2005, June 7). A year of physical, emotional passage: After decades of experimenting, schools debate whether to put students with younger or older kids. Washington Post A10. Stroud, L. B. (2002). To group or no t to group: A qualitative study of middle school making processes concerning ability level grouping. (Doctoral Dissertation, East Tennessee State University, 2002). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 931 5947)

PAGE 199

199 SurveyMonkey (2011). http://surveymonkey.com Palo Alto, CA: SurveyMonkey.com, LLC. Sutton, A. & Soderstrom, I. (1999). Predicting elementary and secondary school achievement with school related and demographi c factors. Journal of Educational Research, 92 (6), 330 338. Swan, B. A. (2006). Middle school mathematics teacher certification, degree level, and experience, and the effects on teacher attrition and student mathematics achievement in a large urban distri ct. (Doctoral Dissertation, University of Central Florida, 2006). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3233679) Thompson, K. L. (2007). Do urban K 8 teachers believe they were properly prepared to teach in urban schools? A c ase study of those who have left, those who have just begun, and those who have stayed. (Doctoral Dissertation, Cardinal Stritch University, 2007). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 3259722) Toepfer, C. E., Jr. (1990a). M iddle level school grades and program development. Schools in the middle: A report on trends and practices Reston, VA: National Association of Secondary School Principals. (ERIC Document Reproduction Service No. ED 326999). Toepfer, C. E., Jr. (1990b). R evisioning middle level education: A prelude to restructuring schools. Educational Horizons, 68 95 99. Toukoushian, R. K. & Taylor, C. (2005). Effects of socioeconomic factors on public high school outcomes and rankings. Journal of Educational Research, 98 (5), 259 272. Tucker, C. G., & Andrada, G. N. (1997). Accountability works: Analysis of performance by grade span of school Paper presented at the annual meeting of the American Education Research Association, Chicago, IL. (ERIC Document Reproduction Service No. ED411278) Turque, B. (2008, July 22). Doubts linger on Pre K 8 strategy: School chancellor pushes program as part of D.C. reform. Washington Post B 1, Retrieved electronically on January 30, 2009 from http://www.washingtonpost.com/wpdyn/content/article/2008/07/21/AR200807210 2678_pf.html United States Department of Education. (2008). Fiscal year 2009 performance budget performance plan. Retri eved January 30 2011, from http://www2.ed.gov/about/reports/annual/2009plan/fy09perfplan.pdf

PAGE 200

200 United States Department of Education (2003). Fact sheet on the major provision s of the conference report to H.R. 1, the No Child Left Behind Act Retrieved from http://www2.ed.gov/nclb/overview/intro/factsheet.html United States Department of Education. (2009). A pproved state accountability plans Retrieved January 30, 2009 from http://www.ed.gov/admins/lead/account/stateplans03/index.html United States Department of Education, National Center for Education Statistics. (2010). The Condition of Education 2010 (NCES 2010 028), Indicator 17. Vaccaro, D. A. (2000). A comparative study of middle school and K 8 school value added gains in the areas of reading, language, math, science, and soc ial studies. (Doctoral Dissertation, University of Tennessee, 2000). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 9973519) Valentine, J. W. (2000). United States middle level grade organizational trends [Online]. Retrieved April 2 4, 2005. from http://www.mllc.org/docs/USMLTrends Valentine, J. W., Clark, D. C., Hackmann, D. G., & Petzko, V. N. (2002). A national study of leadership in middle level schools, volume 1: A national study of m i ddle level leaders and school p rograms. Reston, VA: National Association of Secondary School Principals. Valentine, J. W., Clark, D. C., Hackmann, D. G., & Petzko, V. N. (2004). Leadership for highly successful middle level principals, v olume II : A national s tudy of l eadership in m iddle level s chools Reston, VA: National Association of Secondary School Principals. Vaughn, K. (2000). Music and mathematics: Modest support for the oft claimed relationship. Journal of Aesthetic Education, 34 (3/4), 149 166. Viadero, D. (1995, January 25). Changes in attitude. Education Week pp. 27 29. Viadero, D. (2008, January 16). Evidence for moving to K 8 model not airtight. Education Week 27 (19), 1. Wallis, C., Miranda, C, & Rubiner, B. (2005, August). Is middl e school bad for kids? Time, 166 48 51. Washington, S. G. (2000). The effects of interdisciplinary teaming on middle school climate and student achievement. (Doctoral Dissertation, University of Missouri Saint Louis, 2000). Ann Arbor, MI: ProQuest Info rmation and Learning Company (UMI No. 9988422)

PAGE 201

201 The Washington Post. Waugh, R. (2004). Stuck in the middle. Scholastic Scope, 53 (2), 21. Weibling, C. J. (1997). A compa rison of national middle school association criteria for excellence in three blue ribbon award winning middle schools. (Doctoral Dissertation, Southwestern Oklahoma State University 1997). Ann Arbor, MI: ProQuest Information and Learning Company (UMI No. 9809818) Weiss, C. C. & Kipnes, L. (2006). Reexamining middle school effects: A comparison of middle grades students in middle schools and K 8 schools. American Journal of Education, 112 (2), 239 272. White, G. (1993). Revolution in the middle: Recastin g the middle level learning system. Middle School Journal, 25 (1), 8 12. Wihry, D., Coladarci, T., & Meadow, C. (1992). Grade span and eighth grade academic achievement: Evidence from a predominantly rural state. Journal of Research in Rural Education. 8 (2), 58 70. Willms, J. D. (1999). Hierarchical Linear Modeling as an analytical tool. In G. J. Cizek (Ed.), Handbook of Educational Policy (pp.473 493). London: Academic Press. Wise, S. R. (2000 ). The effect of middle school organization upon student ach ievement (Doctoral Dissertation, Indiana State University, 2000). Ann Arbor, MI: Bell & Howell Information and Learning Company. (UMI No. 9965723) Wren, S. D. (2003). The effect of grade span configuration and school to school transition on student achie vement. (ERIC Document Reproduction Service No. ED479332) Yakimowski, M. & Connolly, F. (2001). An examination of K 5, 6 8, and K 8 grade configurations. Report prepared for the Board of School Commissioners. Baltimore: Division of Research, Evaluation, a nd Accountability, Baltimore City Public School System. Ziegler, S. & Mulhall, L. (1994). Establishing and evaluating successful advisory program in a middle school. Middle School Journal, 25 (4), 42 46.

PAGE 202

202 BIOGRAPHICAL SKETCH Dana Kriznar graduated from Iowa State University in 1986 with a degree in mathematics. After graduation, she accepted a teaching position at John Gorrie Junior High School in Jacksonville, Florida, where she taught math for four years, before transferring to Mandarin Mid dle School, in the same county. Dana earned her degree in Educational Leadership at the University of North Florida, and was then appointed as an assistant principal at Mandarin Middle School. She also served as vice principal of Twin Lakes Middle School, principal of Thomas Jefferson Elementary School, and principal of Kirby Smith Middle School, all in Jacksonville, Florida. She earned her Ph.D. from the University of Florida in the fall of 2011, and currently serves as the Executive Director of M ultiple Pathways and Support Services for Duval County Public Schools, a division dedicated to finding additional supports and alternative pathways to graduation for nontraditional students. Dana is the daughter of Maureen A. Rodas and Dr. George F. Schuchmann, deceased. She is married to Brad Kriznar, a retired U. S. Navy Senior Chief who, after serving in the military for 21 years, now teaches middle school math in Duval County Public School s. They are the proud parents of two teenage sons, Bradley Austin and Tyler Joseph. In her free time, Dana enjoys camping, reading, traveling, Gator football and spending time with her dog, Buster