Effects of metacognitive reading strategy instruction on EFL high school students' reading comprehension, reading strategies awareness, and reading motivation

Material Information

Effects of metacognitive reading strategy instruction on EFL high school students' reading comprehension, reading strategies awareness, and reading motivation
Wang, Min-Tzu
Place of Publication:
[Gainesville, Fla.]
University of Florida
Publication Date:
Physical Description:
1 online resource (253 p.)

Thesis/Dissertation Information

Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Curriculum and Instruction (ISC)
Teaching and Learning
Committee Chair:
Fu, Danling
Committee Members:
Lamme, Linda L.
Harper, Candace
Leite, Walter
Graduation Date:


Subjects / Keywords:
Foreign language learning ( jstor )
High school students ( jstor )
Learning ( jstor )
Nonnative languages ( jstor )
Reading ( jstor )
Reading comprehension ( jstor )
Reading instruction ( jstor )
Reading motivation ( jstor )
Reading strategies ( jstor )
Students ( jstor )
Teaching and Learning -- Dissertations, Academic -- UF
comprehension, efl, metacognitive, motivation, strategy
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Curriculum and Instruction (ISC) thesis, Ph.D.


Metacognitive reading strategy instruction (MRSI) has been valued in reading instruction in both L1 and L2 contexts and has been considered helpful for developing learners? reading awareness. This study explored the effect of metacognitive reading strategy instruction (MRSI) on Taiwanese EFL high school students? English reading comprehension, reading strategy awareness, and reading motivation. By using a sequenced mixed-method research methodology, five research questions were addressed in this study: (1) Does metacognitive reading strategy instruction affect high school EFL students? reading comprehension? (2) Does metacognitive reading strategy instruction affect high school EFL students? reading strategy awareness and reading motivation? (3) Does the effectiveness of metacognitive reading strategy instruction (MRSI) depend on general English reading proficiency levels? (4) What are the factors involved EFL high school students? English reading experience? (5) What features of metacognitive reading strategy instruction (MRSI) change EFL high school students? reading perceptions towards English reading? The study involved a 10-week intervention with 110 public high school EFL adolescents in southern Taiwan. During the course of the study, participants were randomly selected and randomly assigned to the experimental and control group. Data from pre/post measurements, classroom observations, group interviews, reading strategy journals were used to investigate the impact the intervention had on EFL adolescent students. Results from three post measurements indicated that metacognitive reading strategy instruction (MRSI) led to a measurable increase in EFL students? reading achievement after 10 weeks of metacognitive reading strategy instruction. EFL high school students in the experimental group showed statistically significant gains on three post measurements as compared with the control group which didn?t receive intervention on self-monitoring or self-reflection reading strategy. In addition, the result indicated that metacognitive reading strategy instruction (MRSI) had an impact on EFL high school students of all proficiency levels. Following the intervention study, group interviews with 24 subjects from the experimental group illustrated how EFL students? English reading experience has shaped their perception toward English reading. Moreover, interview data also indicated that the implementation of metacognitive reading strategy instruction (MRSI) changed both EFL adolescent students and the English teacher in some ways. For example, EFL students outgrew their previous passive English reading habit and became actively involved in English reading. Likewise, the English teacher, who participated in this study, realized that metacognitive reading strategy instruction is a way that should be integrated into EFL regular English class in order to help EFL students become independent readers and ultimately lifelong readers. Results of this study strongly suggested that metacognitive reading strategy instruction (MRSI) should be integrated into regular EFL reading classes. Limitations of this study and implication for future research as well as pedagogy are also discussed. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
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.
Thesis (Ph.D.)--University of Florida, 2009.
Adviser: Fu, Danling.
Statement of Responsibility:
by Min-Tzu Wang.

Record Information

Source Institution:
Rights Management:
Copyright Wang, Min-Tzu. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
LD1780 2009 ( lcc )


This item has the following downloads:

Full Text




2009 Min-Tzu W ang 2


To m y parents who taught me determination and to my family who provides me with inner strength 3


ACKNOWLEDGMENTS I would like to convey my most sincere gratitude to my advisor and committee chair, Dr. Danling Fu, who provides me with intellectual an d academic support at the times when I needed it the most. Her constant encouragement and guidance provides me with the impetus and confidence to complete this degree and th e years at the University of Florida. I am also deeply grateful to the other committee members, Dr. Candace Harper, Dr. Linda Lamme, and Dr. Walter Leite, for their patie nce, guidance, and criti cal suggestions. I am thankful to Dr. Lamme for her in spiration in introducing me to the charming world of Childrens Literature which motivates me to apply it to th e field of foreign language teaching. I am so grateful to Dr. Harper for her special expertis e in second language reading and her suggestions regarding research. I also a ppreciate Dr. Leite for his genu ine kindness and patient mentoring with the educational measurement, met hodology design, and statistical analysis. Appreciation is also extended to all other professors and friends who helped make my years at the University of Florida fruitable and en joyable. Moreover, I would like to thanks to all the high school students and the teacher who participated in this study. Their contributions were the source of the research. To my family in my country, I offer my appreciation for their continuing support and unconditional love. In particular I would like to express my grat itude to my parents and parentsin-laws for supporting me throughout my education. Finally, I would like to express my sincer e thank to my beloved husband, Hong-Long Chen, and my two boys, Nelson and Vincent, for their support and understanding all these years. Their love and consideration gave me courage to face challenges and never give up. Without the support of my family, I am not sure I would have been able to finish. 4


TABLE OF CONTENTS page LIST OF TABLES................................................................................................................. ..........9LIST OF FIGURES.......................................................................................................................12ABSTRACT...................................................................................................................................13CHAPTER 1 INTRODUCTION..................................................................................................................15Background of the Study........................................................................................................15A Research Focus: Metacognition in Foreign Language Reading.........................................19ESL Versus EFL Reading Strategies......................................................................................21EFL Reading Problems in the Context of English Education in Taiwan...............................22The Need for Effective Reading Strategies Instruction..........................................................26Rationale of the Study............................................................................................................30Purpose of the Study........................................................................................................... ....32Data Collection and Research Questions................................................................................34Hypotheses..............................................................................................................................35Significance of the Study........................................................................................................36Definitions of Key Terms....................................................................................................... 382 LITERATURE REVIEW.......................................................................................................39Introduction................................................................................................................... ..........39Reading Comprehension Models............................................................................................391. The Bottom-up Model.................................................................................................402. The Top-down Model..................................................................................................423. The Interactive Model.................................................................................................44Metacognitive Awareness in Reading Comprehension..........................................................46Review of Studies on Reading Strategies...............................................................................47Reading Strategies in the L1 Context..............................................................................48Reading Strategies and Motivation in the L1 Context....................................................52Summary of Reading Strategy Study in the L1 Context.................................................54Reading Strategies in the L2 Context..............................................................................56Descriptive Studies in L2 Context...................................................................................57Descriptive Studies in Metacognitive R eading Strategies in L2 Context.......................65Summary of Descriptive St udies in L2 Context..............................................................69Experimental Studies in L2 Context................................................................................69Summary of Experimental Studies in L2 Context...........................................................81Reading Strategy and Motivation in L2 Context....................................................................81Reading Strategy Studies on Chinese Speakers......................................................................82Remaining Gaps in Knowledge..............................................................................................86 5


3 METHODOLOGY .................................................................................................................88Purpose of the Study........................................................................................................... ....88Rationale for a Mixed-Method Design...................................................................................90Research Questions............................................................................................................. ....91Subjects...................................................................................................................................92Subjects Selection............................................................................................................92Rationale for Determining Sample Size..........................................................................92Classification of Englis h Reading Proficiency................................................................95Materials.................................................................................................................................95Textbooks........................................................................................................................95Reading Strategy Booklet................................................................................................96Reading Strategy Journal.................................................................................................96Measurements................................................................................................................... ......97Reading Comprehension Test in the General English Proficiency Test (GEPT)...........97The Survey of Reading Strategies (SORS)......................................................................99The English Reading Motivation Questionnaire (ERMQ)............................................100Establishing Translation Auth enticity for SORS and ERMQ..............................................102Research Design...................................................................................................................102Instructors......................................................................................................................104Treatment Delivery Accountability...............................................................................105Threats to the Study.......................................................................................................105Procedure..............................................................................................................................105Variables for Quantitative Data............................................................................................106Independent Variables...................................................................................................107Dependent Variables:....................................................................................................107Quantitative Data Analysis...................................................................................................10 7Qualitative Data Collection..................................................................................................10 8Participants in Interview................................................................................................108Interview and Data Analysis.........................................................................................109Summary of Methodology....................................................................................................1104 QUANTITATIVE RESULTS..............................................................................................112Introduction................................................................................................................... ........112Statistical Techniques......................................................................................................... ..113Descriptive Statistics of Participants....................................................................................113Outliers Deletion...........................................................................................................114Gender Distribution.......................................................................................................115English Reading Proficiency.........................................................................................115The preSurvey of Reading Strategies (SORS).............................................................116PreEnglish Reading Motiv ation Questionnaire (ERMQ)............................................118Descriptive Summary of Pretest Data...........................................................................120Descriptive Summary of the Posttest Data....................................................................122Reliability of Scores of the Measurements...........................................................................122Reliability of Scores of the Reading Comprehension Test ............................................123Reliability of Scores of the Survey of Reading Strategy (SORS)..................................123 6


Reliability of Scores of the English R eading Motivation Questionnaire (ERMQ).......124Assumptions of ANCOVA...................................................................................................125Homogeneity of Variances............................................................................................126Homogeneity of Regression Slopes...............................................................................126Linearity: the Relationship Between the Outcome and the Covariate is Linear...........127Normal Distribution of the Sample on Three Dependent Variables.............................127Hypothesis Test................................................................................................................ ....132Effect of Treatment on postEnglish Reading Comprehension Test .............................133Effect of Treatment on the postSurvey of Reading Strategies (SORS)........................135Effect of Treatment on postEnglish reading Motivation Questionnaire .....................139Effect of Treatment on Different Language Ability in Three Dependent Variables.....143Conclusion............................................................................................................................1475 QUALITATIVE FINDINGS................................................................................................149Introduction................................................................................................................... ........149(1) Problems in Reality........................................................................................................ .150Finding # 1.....................................................................................................................150Finding # 2.....................................................................................................................157Finding # 3.....................................................................................................................161Finding # 4.....................................................................................................................162(2) Why Metacognitive Reading Strategy Instruction?........................................................164Finding # 5.....................................................................................................................164Finding # 6.....................................................................................................................166Finding # 7.....................................................................................................................167Finding # 8.....................................................................................................................168Finding # 9 ....................................................................................................................170Conclusion............................................................................................................................1726 DISCUSSION AND CONCLUSION..................................................................................176Overview of the Study..........................................................................................................176Quantitative Findings............................................................................................................177Qualitative Findings........................................................................................................... ...180Discussion.............................................................................................................................183Findings in the Context of the Literature..............................................................................184Reexamine Key Features of MRSI.......................................................................................1871.Dialogues Between Teachers and Students...............................................................1872.Monitoring and Reflection.........................................................................................188Lessons Learned in this Study..............................................................................................1881.EFL Students Prefer More Inte ractions than Seat Work...........................................1882.The Need for Language Teachers Professional Development.................................1893.Let EFL Learners Choose Their Own Reading Materials.........................................190 4. Individual Learning Styles, Prefer ence, and Learner Resistance 188 Limitations and Recommendations for Future Research......................................................191Pedagogical Implications......................................................................................................1941. Values for Language Teachers..................................................................................194 7


2. Values for EFL Students ...........................................................................................197Conclusion............................................................................................................................198APPENDIX A DEMOGRAPHIC DATA SURVEY: STUDENTS PROFILE...........................................202B PARENT CONSENT ..........................................................................................................203C INFORMED CONSENT......................................................................................................205D READING STRATEGY BOOKLET & JOURNAL...........................................................210E READING STRATEGIC JOURNAL GUIDELINE............................................................222F SURVEY OF READING STRATEGY...............................................................................223G ENGLISH READING MOTI VATION QUESTIONNAIRE..............................................225H READING COMPREHENSION TEST...............................................................................227I INTERVIEW QUESTIONS.................................................................................................232J IRB FORM....................................................................................................................... ....233LIST OF REFERENCES.............................................................................................................234BIOGRAPHICAL SKETCH.......................................................................................................253 8


LIST OF TABLES Table page 3-1 Power and effect size calculated from previous experimental studies..............................94 3-2 Sample size estimation for a power level of 0.8 with Songs (1998) study with estimated e ffect size...........................................................................................................94 3-3 Power level and sample si ze with an effect of 0.32...........................................................95 3-4 Participants reading proficiency levels.............................................................................95 3-5 Steps for the implementation of the study.......................................................................103 3-6 Overview of research hypothesis and methodology........................................................108 4-1 Sample size before and after the treatment......................................................................114 4-2 Deleted outliers......................................................................................................... .......115 4-3 Gender distribution in each group....................................................................................115 4-4 Subjects GEPT level distribution...................................................................................116 4-5 Subjects pre-SORS distribution......................................................................................117 4-6 The most and the least frequently used reading strategy from the pretest data...............118 4-7 Participants pre-ERMQ distribution...............................................................................119 4-8 Participants most and least agreement on pre-ERMQ questionnaire.............................119 4-9 Descriptive summary of the pretes t score of the sample as a whole...............................121 4-10 Descriptive summary of th e pretest score on each group................................................121 4-11 Result of homogeneity of variance test between groups in terms of GEPT, preSORS,and pre-ERMQ......................................................................................................121 4-12 Independent t -tests on difference between groups on GEPT, pre-SORS, and preERMQ..............................................................................................................................121 4-13 Mean and standard deviation of three posttest data.........................................................122 4-14 Reliability of scores of reading comp rehension test in pr e-test and post-test..................123 4-15 Reliability of scores of the SO RS in the pre-test and post-test........................................124 9


4-16 Reliability of scores of the ERMQ for the pre-test and post-test.....................................125 4-17 Levenes test of equa lity of error variance......................................................................126 4-18 Correlations between covariat es and each dependent variable........................................127 4-19 Mean and standard deviation of postEnglish Reading Comprehension Test ................134 4-20 ANCOVA table on PostEnglish Reading Comprehension Test .....................................135 4-21 Mean and sandard deviation for postSurvey of Reading Strategy ..................................135 4-22 ANCOVA table on post Survey of Reading Strategy ......................................................136 4-23 Mean and standard deviation of GLOB subtest in postSurvey of Reading Strategy .......137 4-24 ANCOVA table on GLOB reading strategies in post-SORS..........................................137 4-25 Mean and standard deviation of PR OB reading strategies subtest in postSurvey of Reading Strategy ..............................................................................................................138 4-26 ANCOVA table on PROB readi ng strategies in post-SORS...........................................138 4-27 Mean and standard deviati on of SUP subtest in post-SORS...........................................138 4-28 ANCOVA table on supportive read ing strategies in post-SORS.....................................138 4-29 Mean and standard deviation English Reading Motivation Questionnaire.....................139 4-30 ANCOVA table on English Reading Motivation Questionnaire (ERMQ).....................140 4-31 Mean and standard deviation of intr insic reading motiva tion subtest in postERMQ .....140 4-32 ANCOVA table on intrinsic reading motivation in postERMQ .....................................141 4-33 Mean and standard deviation of extr insic reading motivation subtest in post-ERMQ ....141 4-34 ANCOVA table on extrinsic re ading motivation in post-ERMQ....................................141 4-35 Mean and standard deviation of the importance English read ing subtest in postERMQ ..............................................................................................................................141 4-36 ANCOVA table on the importance English reading in postERMQ ...............................142 4-37 Mean and standard deviation of reading efficacy subtest in postERMQ ........................142 4-38 ANCOVA table on readi ng efficacy in post-ERMQ.......................................................142 10


4-40 ANCOVA table on interacti on effect of treatment on la nguage ability in post-SORS test....................................................................................................................................144 4-41 ANCOVA table on interaction effect of treatment on langua ge ability in post ERMQ test....................................................................................................................................144 11


LIST OF FI GURES Figure page 3-1 Mixed-method research process........................................................................................91 4-1 PostReading Comprehension Test ..................................................................................128 4-2 PostSurvey of Reading Strategies (SORS) ......................................................................129 4-3 PostEnglish Reading Motivati on Questionnaires (ERMQ) ............................................130 4-4 Scatterplot of covariate agains t post-reading comprehension test...................................131 4-5 Scatterplot of c ovariate against postSurvey of Reading Strategies (SORS) ....................131 4-6 Scatterplot of c ovariate against postEnglish Reading Motivation Questionnaire (ERMQ) ............................................................................................................................132 4-7 Interaction between language level a nd group in post reading comprehension test........146 4-8 Interaction between language level and group in post SORS..........................................146 4-9 Interaction between language level and group in post ERMQ........................................147 5-1 Overall relationship among MRSI, readi ng motivation, and reading comprehension....172 5-2 The impact of metacognitive reading strategy instruction on the change of EFL adolescents reading comprehension, r eading motivation, and reading strategy awareness.........................................................................................................................175 12


Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECTS OF METACOGNITIVE READING STRATEGY INSTRUCTION ON EFL HIGH SCHOOL STUDENTS READING COMP REHENSION, READING STRATEGIES AWARENESS, AND READING MOTIVATION By Min-Tzu Wang December 2009 Chair: Danling Fu Major: Curriculum and Instruction Metacognitive reading strategy instruction (MRSI) has been va lued in reading instruction in both L1 and L2 contexts and has been considered helpful for deve loping learners reading awareness. This study explored the effect of metacognitive r eading strategy in struction (MRSI) on Taiwanese EFL high school students Englis h reading comprehension, reading strategy awareness, and reading motivation. By using a sequenced mixed-method research methodology, five research questions were a ddressed in this study: (1) Does metacognitive reading strategy instruction affect high school EFL students reading comprehe nsion? (2) Does metacognitive reading strategy instruction affect high school EFL students reading strategy awareness and reading motivation? (3) Does the effectivene ss of metacognitive read ing strategy instruction (MRSI) depend on general English reading proficie ncy levels? (4) What are the factors involved EFL high school students English reading expe rience? (5) What feat ures of metacognitive reading strategy instruction (MRSI) change EF L high school students reading perceptions towards English reading? The study involved a 10-week intervention wi th 110 public high school EFL adolescents in southern Taiwan. During the course of the st udy, participants were randomly selected and 13


random ly assigned to the experimental and cont rol group. Data from pre/post measurements, classroom observations, group interv iews, reading strategy journals were used to investigate the impact the intervention had on EFL adolescent st udents. Results from three post measurements indicated that metacognitive reading strategy instru ction (MRSI) led to a measurable increase in EFL students reading achievement after 10 weeks of metacognitive reading strategy instruction. EFL high school students in the experimental gr oup showed statistically significant gains on three post measurements as compared with the control group which didnt receive intervention on self-monitoring or self-reflection reading stra tegy. In addition, the re sult indicated that metacognitive reading strategy in struction (MRSI) had an impact on EFL high school students of all proficiency levels. Following the intervention study, group interviews with 24 subjects from the experimental group illustrated how EFL students English reading experience has shaped their perception toward English reading. Moreover, interview data also indicated that the implementation of metacognitive reading strategy instruction (MRSI) changed both EFL adolescent students and the English teacher in some ways. For example, EFL students outgrew their previous passive English reading habit an d became actively involved in English reading. Likewise, the English teacher, who participated in this study, realized that metacognitive reading strategy instruction is a wa y that should be integrated into EF L regular English class in order to help EFL students become independent readers a nd ultimately lifelong read ers. Results of this study strongly suggested that metacognitive reading strategy instruction (MRSI) should be integrated into regular EFL reading classes. Li mitations of this study and implication for future research as well as peda gogy are also discussed. 14


CHAP TER ONE INTRODUCTION Background of the Study Strategy training in language learning has been topical since 1970s and a large amount of work has been done on identifying the strategies used by both successful and less successful learners. Lately, there is an increasing inte rest in the role of metacognition on reading instruction. While previous reading research has focused on reading strategy use among good and poor readers, researchers are examining reader s awareness of strate gies during the reading process their metacognitive awareness. Meta cognitive knowledge or awareness is knowledge about learners themselves, the ta sks they face, and the strategies they use (Baker and Brown, 1984). Applying metacognition in the language learning field, it refers to the action that one uses for planning, organizing, evaluating, and monitoring of his or her langu age learning (OMalley and Chamot, 1990). Followed by metacognitive st udies in the L1 context, second language researchers have also drawn increasing attenti on on metacognitive strategies in second language learning, especially in the r eading domain. Many second language reading researchers have pointed out the positive correlation of proficient second language readers with more awareness of using appropriate reading st rategies in English reading ta sks (Barnett, 1988; Devine, 1984; Kern, 1989; Pardon, Knight and Waxman, 1986). Mo reover, several resear chers (Carrell, 1998; Cordero-Ponce, 2000; Sheorey and Mokhtari, 2001) assert that in order to make reading strategies effective in the reading process, metacognitive awareness or metacognition must be employed. This metacognitive awareness refers to knowledge of strategies as well as controlling this knowledge of acti on in the reading process (Carre ll et al., 1989). In response to this positive relationship between metacognitive reading strategies and reading comprehension, several second language instructor s began training second language learners with metacognitive 15


reading s trategies and the results suggested that metacognitive reading stra tegy instruction brings positive outcomes in language learners metacognitive awareness and reading comprehension. For example, in a study of four-day training th rough semantic mapping strategy, regardless of its mixed results, Carrell, Pharis, and Liberto (1989 ) found that metacognitive strategy training was effective based on statistically significant evidence fr om increasing scores in one of the posttests. Also, Auerbach and Paxton (1997) investigated whether giving lear ners opportunities in choosing their own research topics based on th eir interests could enhance their metacognitive reading awareness and expand their repertoire of readi ng strategies. The findings concluded that metacognitive training not only increased learners metacognitive awareness, but also made English reading more enjoyable since reading materials were based on their interests and thus they gained control ove r their second language reading proce ss. In addition to these direct effects of metacognitive reading strategy instruction on language learners awareness and reading comprehension, research has demonstrat ed that cognitive reading strategy training only results in small, short-term improvement in second language learners reading performance, while metacognitive reading strategy traini ng results in more stable and long-term comprehension gains (Carrell et al., 1998, Tang and Moore, 1992; Zhicheng, 1992, cited in Koda, 2005). Therefore, the most favorable goal for metacognitive reading strategy instruction is hoping that it can make ESL or EFL students become regulators of readi ng strategies and use reading strategies selectively and flexibly according to different reading tasks they face. In the meantime, being able to control over their reading process, ESL or EFL students are expected to use these strategies as resources to pursue the goal of English reading as personal interpretation or meaning making rather than confining their experience or learning to an understanding or acquisition of English li nguistic knowledge only (Carrell et al., 1998). 16


However, in spite of these persuasive and positive findings of metacognitive reading strategy instruction in the US context, many Eng lish teachers in Taiwan or other Asian countries such as Korea and Thai, still follow a tradit ional language teaching me thod, that is, grammartranslation method, through which vocabulary me morization and grammatical structures are emphasized (Huang, 2006; Merkelbach, 2006; Phak iti, 2003; Song, 1998) with the primary goal of helping EFL students to pass a nnual high school or college entr ance examinations. As a result, the typical English teaching and learning activ ities for Taiwanese EFL students in middle schools and high schools are nothi ng but memorizing English voca bulary and drill practicing grammatical rules. And, although this traditional English teac hing pedagogy seems to work well in yearly national entrance ex ams of high schools and colleges, evidence has shown that grammar-translation instruction keeps Taiwanese EFL students from developing critical thinking skills as well as positive at titudes toward English read ing (Chern, 2003; Cheng, 1998). According to the cooperative study measured by th e International Association for the Evaluation of Education Achievement (IEA) in the US and the National centr al University in Taiwan in 2005, Taiwanese students reading comprehension is low in comparison with world average. Moreover, evidence indicated that the average number of days of the school year they read was 24%, which was lower than the worldwide av erage of 40% (NCU News Network). Furthermore, another drawback of grammar-t ranslation teaching method lies in that it makes EFL students feel frustrated in English re ading since they spend time checking meanings from dictionaries and analyzing sentence structures, and yet dont get the main points from the reading materials, especially after all the translat ions are done. Gradually, they lose interests in reading English since it is hard to comprehend after so much effort and time has been spent. Consequently, most EFL students in Taiwan become unskilled r eaders, passive readers, and 17


dependent readers. W ith such negative E nglish reading experience, it also builds up a psychological hindrance or avoidance for EFL st udents toward English r eading. For instance, there is evidence that many Taiwanese college st udents have reported th at they either avoid taking classes in which English textbooks are assi gned or they hardly to uch the English reading materials once they have completed their high school education, unless they are enrolled in English related majors in college or ot her language institutes (Merkelbach, 2006). Nevertheless, as everyone knows, the majority of the latest resources and information on the World Wide Internet and professional magazines are written in English. In that sense, if one wants to be able to compete with others in this global market and expand their knowledge and world view, being a strategic E nglish reader is inevitable. In addition, reading is the very foundation for mastering of a second language, es pecially for EFL students whose exposure to English environment is far less than ESL students. Therefore, being aware of this global trend, the Minister of Education (MOE) in Taiwan has advocated that one of the objectives of English education should provide EFL student s with strategies in all language skills in order to make English learning successful and effective. As second language researchers has urged teaching productive reading strategies to ESL/EFL student s to motivate reading an d facilitate reading comprehension (Anderson, 2003; Chern, 1993; Eske y, 2002; Farrell, 2001; Grabe, 2004), this study plans to investigate whether providi ng EFL high school students with metacognitive reading strategy instructions (MRSI) in their re gular English reading classrooms will help them become more metacognitively aware of their Eng lish reading process and be motivated to read more English materials in order to not only su ccessfully acquire a second language, but also fulfill their personal interests and become lifelong readers. 18


A Research Focus: Metacognition in Foreign Language Reading The study of metacognition originated in the field of cognitive development, especially developmental psychology (Flavell, 1978) a nd information-processing cognitive psychology (Wagner and Sternberg, 1984). Flavell (1978) appears to be the first one, in early 70s, to define metacognition as knowledge of ones learning that consists of two elements: (1) metacognitive knowledge that is knowledge of cognition, and (2) regulation of cognition that is strategy use. Similarly, Brown (1981) describes metacogn ition as the deliberate conscious control of ones own cognitive action. (p. 453). In accord with Flavells (1978) definition, Baker and Brown (1984) further note that there are two ty pes of metacognitive activities: (1) those that concern a persons knowledge about his or her own cognitive resources and the compatibility between the learner and the learni ng situation, and (2) those that re gulate and modify the process of a cognitive activity. Taken as a whole, K oda (2005) summarized metacognition as knowledge of recognition which refers to a learners unders tanding and control of his or her own thinking and learning. When metacognition is applied to the reading process, effective readers, according to Baker and Brown (1984), thus are aware of and able to control the cogni tive activities they are engaged in during the reading process. In ot her words, effective reading usually involves metacognition. To illustrate how metacognition is involved in reading, Baker and Brown (1984) suggest that the following strategies are ty pically used by good read ers during the reading process: adjusting reading rate, skimming, or being aware of and revision the materials. To be specific, metacognitive reading strategies refer to procedures that one uses for monitoring his or her own reading processes including evaluating the effectiveness of cognitive strategies being used in reading. That is to say, if cognitive reading strategies are about knowledge of what strategy to use and how to apply it, then metacognitive reading strategies involve understanding 19


the rationale for applying a particular strate gy in a particular cont ext, and evaluating its usefulness in terms of effectiveness for that read ing situation. Auerbach and Paxton (1997) also maintain that strategic reading can only be effi cient when metacongitive reading strategies are actively used. Consistent evidence from previous studies has found that the major difference between skilled readers and less skilled readers lies in how much they are engaged in a self-regulated process, that is, the use of metacognitive st rategy of comprehension monitoring (Baker and Brown, 1984). From these findings, it is believed that less skilled readers are less strategic, largely because they fail to monitor the situ ation of their comprehension during reading. Comprehension monitoring is viewed as essential to the whole read ing process, from planned use of reading strategies to changes in strategies use in the reading process. Skilled readers use a different approach once they become aware that th eir current approach is not contributing to their reading comprehension. To su m up, characteristics of the metacognitive theory of reading stresses that skilled or profic ient readers actively evaluate their understanding in the reading process and apply or select appr opriate reading strategies to co nstruct meaning from the reading text. That is, skilled readers use strategies that they find effective for the kinds of tasks they need to accomplish in the L2 (Chamot, 2004). For th at reason, metacognitive aw areness is essential for reading comprehension and is crucial for read ers to possess and evaluate when and where to use the strategies they know. Chamot (2005) also proposes that it is important for language teachers to help ESL students develop thei r metacognition in language classrooms because metacognition help them select th e most appropriate strategies acco rding to a given task and their individual language learning preferences. 20


Moreover, the reading task that foreign la nguage readers face is far more complex and demanding than native language readers since read ing comprehension involves both the reader and the reading text, which is linguistically and culturally specific. For example, Farrell (2009) points out the reasons why English language learne rs have difficulty reading in English and he highlights areas such as first language and second language dissimilarities, age, learning styles, and cultural schemata. For that re ason, metacognition in foreign la nguage reading process is vital for foreign language learners to possess. Many st udies, both in L1 and L2, have found a positive correlation between the effective use of metacognitive skills and fluent reading as well as the fact that the main difference between skilled and unskilled readers is believed to be in the ability that skilled readers are engaging in deliberate activities that require th inking, flexible strategies use, and constant self-monitoring (Sheorey and Mokhtar i, 2001). Thus, it is ba sed on this theoretical background that the idea of present study was developed to investig ate the effect of metacognitive reading strategy instruction on EFL high school students reading process. ESL Versus EFL Reading Strategies Differences between ESL and EFL largely rela te to where English is taught. An ESL instructional environment is defined as an e nvironment where English is the primary language used in the society in which th e language is studied. For instance, learners studying in the US, England, or Canada are in an ESL envir onment (Anderson, 2003). By contrast, an EFL instructional environment is one where English is not the primary language of society in which the language is being studied. For example, learne rs in Korea, Japan, and Taiwan are in an EFL environment (Anderson, 2003). Under this conditi on, EFL students have fewer chances to apply English to situations outside of classrooms th an ESL students and thus have certain amount of influence on EFL students strate gies use. Riley and Harsch (1999) are a few of the researchers examining how the learning environment may influe nce strategy use. They conducted a research 21


project to co mpare the strategy us e of Japanese learners of Eng lish in ESL (in Hawaii) with EFL (in Japan) environments. Their findings s upported the notion that learners in an ESL environment use more strategies than learners in an EFL environment. The reason supporting this finding is that ESL student s have more opportunities to use English in daily lives and therefore have a greater need to use strategies to survive. By contrast, EFL students have much less opportunities to access English in their daily lives and thus us e fewer strategies. However, this doesnt mean that strategies arent important for EFL students. Th e limited opportunities in accessing English make English reading strategies equally important to EFL students because English written materials become major resource s and language input for EFL students to learn and acquire English. Reading is a valuable source of language input for language acquisition and it is the most durable of all L2 language ski lls and most cost-effective means of providing information regarding second language and culture (Bernhardt, 1993). Thus reading strategies should be emphasized in the initial stages of la nguage learning. EFL students need to know or learn how to read with ease in order to be motivated to read mo re English materials in order to build up their English ability. EFL Reading Problems in the Context of English Education in Taiwan The context of the learning situ ation and the cultural value of the learners society can be expected to have a strong influence on choice and acceptability of language learning strategies (Chamot, 2004; Farrell, 2009). Pr itchard (1990) and Grabe (1991) have also pointed out that reading strategy and strategy pref erence are socially and culturally constructed. The following section briefly describes how E nglish reading is learned and ta ught in Taiwan from a sociocultural perspective. Three major factors that contribute to English learning and teaching in Taiwan can be identified from literature review into the following categories: (1) examination22


oriented education system, (2) gramm ar-transla tion pedagogy, and (3) Chinese literacy tradition influence (Cheng, 1998). Long before, English has become one of the ma jor subjects to be tested in all yearly national entrance examinations of high schools and colleges in Taiwan (Chern, 2003). As a result, these major examinations have played a predominant role in lead ing English education. The content and questions appeared in written English exams are usually focused on testing EFL students linguistic knowledge such as vocabulary and grammati cal rules (Shih et al., 2000). Therefore, in order to help EFL students well-p repared for these examinations, the major English instruction and learning have di rected to acquire linguistic aspects of English (Hung, 1996, 2000; Hsu, 2003). In other words, memorizing gramma tical rules and vocabulary from textbooks become the major goals and tasks of English l earning (Sharp, 2002; Shih et al., 1999). As a result, when the center of English instruction fo cuses on linguistic aspect reading skills, on the other hand, has long been overlooked. In addition, English teachers often assume that once EFL students are equipped with gramma tical and structural rules a nd adequate vocabulary knowledge, then they will know how to read automatically. However, this assumption has became problematic as research has found that ce rtain EFL students with good grammatical and vocabulary knowledge still fail to construct meaning from a read ing text (Smith, 2004). From interactive reading theorists viewpoint, reading is a complex process which interactively involved the reader and the read ing text, instead of a process of word decoding (Eskey, 2005; Grabe, 2004). That is to say, reading, instead of word-by-word decoding, is a meaning-making process that integrates a reade rs prior knowledge and expectations with a writers intentions. Therefore, only with linguistic knowledge doe snt contribute to sound reading comprehension. 23


Grammar-translation pedagogy has been repo rted as the dominant teaching method in English classes at most middle and high schools (Hung, 1994; Shih et al., 2000) and English reading strategies have not been addressed in most English classes in Taiwan (Merkelbach, 2006). By looking at most Taiwanese highs sch ool students English te xtbooks, it is common to find that Chinese translations are written betw een every line. Taiwanese EFL students are taught to pay close attention to word-level cues (mor phology and syntax), which is exactly the same as what they are expected to do when reading clas sical Chinese literature. That is, carefully analysis and translation of each word is strongly emphasized while guessing the meaning of each word is not encouraged. However, getting to read or understand an Eng lish text by this method takes an excessively long time and laborious. Over-reliance on Chinese translation may not only hinder EFL students English readin g abilities but it may also decrea se their interests in reading English (Chia and Chia, 2001). Ther efore, English reading experien ce like this is hard for an EFL student to maintain a high level of motivation or interest, let alone engage in an enjoyable reading experience and accomplishment. However, li ke L1 readers, motivation to reading is one of main factors to language acquisition and reading achieveme nt. In conclusion, grammartranslation teaching instruction has made word-by-word translat ion method the only strategy for EFL students to approach an English reading te xt, and as Schultz (1983) states, this teaching method results in a lack of contextual focus and immediate frustration as soon as the reader encounters an unknown word or comprehension breakdown. This disengagement is counterproductive to the purpose of English reading which allows EFL students to acquire new vocabulary words and understand grammatical patterns. English reading activ ities in Taiwan are not designed to develop reading sk ills, strategies, or develop high levels of reading ability. To sum up, lexical units and syntactical structures as well as word-byword translation are the focus 24


of reading instruction in Taiw anese English classroom s (Yang, 2003) and it results in several problems of which English teachers in Taiwan should be aware. Research has found that the majority of th e EFL students in Taiwan is word-bounded and relies on word-to-word translation when re ading English (Abbott, 2006; Parry, 1996; Yang, 1996, 2000, 2002). One possible explanation for this is th e influence of Chinese L1 literacy tradition and practice which has shaped teachers and stud ents views of teaching and learning English (Abbott, 2006; Farrell, 2009; Parry, 1996). As Farrell (2009) poi nts out L2 learners reading strategies and habits will be in fluenced not only by the text struct ure of their L1 but also by their beliefs about the reading process in general. In Taiwan, reading Chinese classical literature is required starting at the upper grad es in elementary schools. Students are frequently taught to translate each word in classical Chinese literature into modern Chinese language to derive its meaning. Therefore, the nature of Chinese literacy pr actice and learning of this kind has been transferred to English reading contexts and has more or less shaped Taiwanese students perception of how reading should be processe d (Farrell, 2009; Parry, 1996). Also, Chengs (1998) a qualitative research further confirmed the influence of L1 literacy on L2 reading behaviors and habits. His subjects, Chinese spea kers from Taiwan of a US university, offered evidence that English instruction in Taiwan was drill practice and gramma r translation oriented and thus has made the subjects in Chengs (1998 ) study have negative attitudes towards reading in English and their English reading approaches tended to be bottom-up. Being aware that English has become the global language across the world in business and politics as well as educati on, Taiwanese government has made English education a priority to enhance Taiwans economic competition. Starting in 1997, the Taiwanese government declared that English instruction would be extended beginning in the 3th grade instead of the 7th 25


grade as of 2001. However, receiv ing English instru ction four years earlier doesnt guarantee for successful language learning, what is important is that EFL students should be taught how to learn English strategically. Many college students in Taiwan, even though they are considered competent in English, still felt reluctant and held less positive attitudes towards reading in English (Cheng, 1998; Hung, 2000; Lin, 2003). This inability to read English effectively has not only caused Taiwanese students to experience barriers to academ ic success (Lin, 2003), but has also disadvantaged them in thei r career performance or in gain ing access to the latest trend of globalized information (Lin, 1996). Moreover, in responding to the widespread use of Communicativ e Language Teaching pedagogy in the second language field as well as the realization of the drawbacks of grammartranslation pedagogy in Taiwans English edu cation, the Ministry of Education (MOE) in Taiwan has called for a reform in all national entrance examinations from testing linguistic knowledge to examining EFL students overa ll use of language knowledge and global comprehension abilities. The pur pose of this movement is hoping to have certain amount of influence on English education pedagogy from grammatical focus to communicative and interactive pedagogy. However, this change in the content of English examinations seems to have produced little effect on the traditional t eaching methodology, which has rooted deeply in overall English teaching climate in Eng lish classes in high schools (You, 2004). The Need for Effective Reading Strategies Instruction As mentioned above, in an EFL context, th e chance for EFL learners to read English material is a lot more limited than ESL learners in the US. However, this doesnt mean that English reading abilities and skills should be overlooked or not necessary to be developed. In contrast, reading skills or ability should be valu ed with the following reasons. First, with limited opportunities to use English in ev eryday life, EFL students rely on written texts as a major 26


source of language input for English learning. F or in stance, Griffiths (2004) finds that reading is actually a useful strategy for language lear ning because it expands vocabulary and grammar usage, thereby assisting the development of bot h receptive (reading and listening) skills and productive skills (speaking and writing). However, the reading skills lear ned in EFL students first language is not necessarily will automatica lly transfer to EFL stude nts second or foreign language (Hassen et al., 2005; Farre ll, 2009). Therefore, reading strategy instruction should be known to EFL high school students. Second, English is one of the major languages in the world and most information stored in computerized data-bases world-wide is in the English language as well (Kaplan, 1982). For that reason, if one wants to be competitive in hi s or her filed of exper tise and expand his or her world view, the chances to read English informati on can be anticipated. Besides, with regard to the role of English in information access, no coun try can ignore the importance of English and still expect to compete professionally and ec onomically (Grabe, 1988). Accordingly, to be wellprepared to face the new challenges of the 21st century, English educators in Taiwan are advised to help their EFL students go beyond grammar an d vocabulary learning and become effective readers of English for academic and professi onal success through lifelong reading and learning. Moreover, English reading demands a lot of knowledge that EFL students might have insufficient knowledge to support them for readi ng comprehension. Considering the challenges of second or foreign language r eading, especially the c ognitive, cultural and linguistic aspects, all prove that second language readi ng is much more difficult and complicated than that in L1. Seeing that reading bear some knowledge specific to Western society and culture, EFL students are supposed to employ or select appropriate strate gies in order to be able to understand what they read and to achieve a satisfactory academ ic performance. Yang (2002) recommends that 27


English teachers in Taiwan should develop EFL students reading strate gies repertoire and m etacognitive awareness regarding to increase thei r control over their own reading process. With adequate metacognitive awareness, EFL students will be able to mon itor their own reading process and identify and solve problems when they encountered them by using the resources and strategies they have learned. While researchers in the U.S. has advocated for teaching productive reading strategies to ESL learners to motivate and facilitate reading (Anderson, 2001 and 2003; Eskey, 2002; Farrell, 2001; Grabe, 2004) English educators in Taiwan should know there is a need to incorporate r eading strategies instruction into regular English classes to foster reading comprehension among high school students as well as foster motivation and positive attitude towards English reading (Chu, 2000; Yang, 2000; You, 2004). According to a survey study from 212 EFL college students perspectives in relating to the s ituational factors that would motivate them to read English materials, Huang (2006) concludes from the data that EFL college students would have read more effectiv ely and efficiently and read more widely if reading strategies had been taught in middle and high schools. In ot her words, there is evidence showing that reading strategies were not widely taught in high schools ore middle schools in Taiwan and EFL students were hoping that readin g strategy should have be en taught to them. Therefore, how to assist EFL students in taking control of th eir own reading process while fostering success and positive attitudes toward EFL reading has become one of the most urgent tasks facing educators who teach English throughout Taiwan. Furthermore, motivational fact ors of English reading is of interest in this study because research findings have suggested that students affective factors in English such as attitudes, motivation, and self-esteem potentia lly contribute different levels of impact to English reading achievement (Drnyei, 2001; Griffiths, 2004). These affective factors need to be noticed here. 28


In view of the fact that Eng lish is taught as one of the m ost important academic subjects in schools, EFL students are forced to study English with no exception. Besides, as a subject matter for tests, English is taught in emphasizi ng testing EFL students knowledge of vocabulary and grammatical structures. Oftentimes, EFL stude nts get frustrated by the pressure of constant tests, learning English in this way is neither meaningful nor pleasant, but a constant cycle of unsuccessful experiences due to in effective strategies use, and he nce they are not motivated to read English, let alone extensively read outsi de the classroom. Howeve r, as mentioned above, English reading is the major resources for la nguage input to EFL stude nts. It is on this phenomenon that motivation is of interest in th is study. The assumption lies in that if EFL students are given the opportunity to learn reading strategies and are made aware of the reading process, they will be able to read more effectively and easily. If they have got over with initial reluctance of English reading and are having confid ence in their English reading ability and then motivating them to read extensiv ely could be anticipated. At the same time, it should also be noted that just providing EFL st udents with a list of reading strategies is not enough for them to beco me successful foreign language readers. It is just like if one wants to teach a kid how to fish, offering him a fishing rod is not good enough. He or she also needs to let the kid know how to use that fishing rod in the right situation. The fishing rod is just like cognitive reading strate gies, and when and how to use it is just like metacognitive reading strategies. Moreover, because everyone has his or her learning preference, in order to become motivated a nd selective strategy users, EFL st udents need to self-regulate or self-monitor their readin g strategy use (Anderson, 2001; Chamot et al., 2004). In other words, EFL students have to consciously or metacognitively know what and when to apply appropriate reading strategy when comprehension breaks dow n. Metacognition is the way that helps EFL 29


students b ecome the owners of their learning process and awareness of the metacognitive process and of their strategy re sources will give them ways to keep nourishing themselves through school. (Block 1992, p. 338). Therefore, EFL students need to be explicitly taught how to properly use reading strategies to mon itor their reading comprehension and identify the cause of their reading comprehension problems, and then successful learners could be expected. In conclusion, the goal of English reading inst ruction is to make English reading effective for EFL students and to change th eir attitudes towards English read ing. In addition, in order to develop EFL students global perspective and to increase their competitive edge, English education in Taiwan should not limit itself to preparation for a test. R eading for academic and professional purposes in English not only requires adequate langua ge proficiency but also proper training in reading skills and stra tegies. Therefore, it is hypothesi zed that metacognitive reading strategy instruction will help E nglish language learners become more aware of the reading process and use reading strategi es to compensate for their in sufficient language knowledge or cultural understanding. Through the employment of metacognitive reading strategy instruction in English class, EFL students should be able to improve their read ing comprehension and experience a higher level of competency to furthe r motivate them to read on a regular basis. Rationale of the Study Given the scarcity of experimental studies examining the effect of metacognitive reading strategy instruction with high school students in EFL contexts, reading theory and models in literature review section provi de the background to investigat e the impact of metacognitive reading strategy instruction on reading strategy awareness, reading comprehension, and reading motivation. The theoretical framework that guides this study is: (1) metacognition theory (Flavell, 1978), (2) constructivism (Vygotsky, 19 78), and (3) reflective practice (Dewey, 1911). 30


First, it should be noted that a lot of research in sec ond language reading stem s from research conducted in English as a first language setting. For that reason, the benefit in L1 reading strategy instruction cant be ignored a nd numerous first language training studies have demonstrated that struggle readers reading co mprehension improved after they received explicit instruction in metacognition (Baker, 2002; Cohen, 2003; Duffy, 2005; Grabe, 2004; Palincsar and Brown, 1984). Such research has sought to better understand the contexts in which comprehension strategies improve comprehensio n, the training procedures which are most effective, and the variables such as learners lang uage proficiency levels and L1 and L2 relations, which influence strategy instruction. Second, previous reading strategy training studi es were mostly ground ed in interactive and constructive or responsive reading theories. From the perspective of interactive reading theory, it argues that effective readers make use of top-down and bottom-up strategies interactively while reading depending on the read ing situation, task, and the text (Bernhardt, 1991; Carrell, Devine, & Eskey, 1988). Similarly, Grabes (1991) interactive perspective posits that fluent readers simultaneously apply highe rand lowerlevel skills to comprehend and interpret text. Goodmans (1992) transactive theo ry of reading and Pressley and Afflerbachs (1995) theory of responsive reading are also re levant here as they emphasize an interaction between the reader and the text That is to say, in order to maximize comprehension, readers must control their use of stra tegies through constant monitoring, and the reader must employ both top-down and bottom-up strate gies to construct meaning. Th erefore, since better readers apply metacognition more often and more effectiv ely than do those who read less, one might expect that, if the strategy is learned and taught, a simu ltaneous increase in reading comprehension would be evident. Moreover, e ffective reading strategy instruction should make 31


students aware of m ultiple reading strategies and know when and how to employ strategies effectively, and furthermore, allow readers to be more aware of their own reading behaviors and processes. Research in the area of metacognitive awareness and reading comprehension has shown that if a reader is aware of what is needed to perform e ffectively, then it is possible for him or her to take steps to meet the demands of a reading situation more effectively. If, on the other hand, a reader is not aware of his or her limitations as a reader or of the complexity of the task at hand, then the reader can hardly be expected to take actions to anticipate or recover from difficulties (Baker and Brown, 1984; Ca rrell, 1989; Devine, 1993). Purpose of the Study The purpose of this study is to investigate the effectiveness of integrating metacognitive reading strategy instruction (MRSI) to EFL high school students on thei r reading strategies awareness, reading comprehension, and readi ng motivation. Even t hough the teaching of reading strategies to improve st udents achievement has been the focus of a number of studies and has been positively associated with an in crease in academic achievement in ESL, the outcome is not conclusive (Taylor, Stevens, a nd Asher, 2006). Besides, the effectiveness of cognitive reading strategy training and met acognitive reading strategy training on second language reading comprehension is not firmly es tablished (Taylor, Stevens, and Asher, 2006). For instance, Taylor et al. (2006) in meta-analysi s of twenty-one research studies found that 68% of training studies were effec tive in improving L2 students r eading comprehension and thus there is no significant difference be tween training programs with c ognitive reading strategies and with metacognitve awareness training. Moreover, even though reading strategy training studies carried out in ESL or EFL c ontexts have proved to improve the L2 readers reading comprehension, they have been few in number (Barnett, 1988; Carrell at el., 1989; Chamot, 2005; Kern, 1989). Chamot (2005) points out that relatively small numb er of instructional language 32


learn ing strategy studies may be partially due to the inherent difficulties in conducting classroom research. Ideally, an intervention study should ha ve randomly assigned part icipants to either a control or an experimental group. Instruction in each group should be identical except for the presence of absence of the instruction being studied. Furthermore, up to now there has been limited research of reading stra tegy training to EFL students of high school levels. Thus, this area of reading instruction needs to be further investigated. The second purpose of this study is to see wh ether EFL high school st udents change their English reading motivation once they are trai ned with metacognitive reading strategies. Research done in motivational domain has mostly been targeted on L1 primary grade students; virtually little research has investigated the eff ect of reading strategies training on EFL students English reading motivation. Insufficient experi mental research on EFL secondary students English reading comprehension and motivation leaves educators without critical knowledge about the role of metacognition in EFL read ing and its possible impact on EFL high school students reading motivation. English reading motivation is especially important to EFL learners and it is the motivation that helps EFL learners do extensive reading other than textbooks and thus read to learn or read to acquire a sec ond or foreign language. Hence, it seems promising to do a study of integrating metacognitive reading strategy instruction (MRSI) into EFL high school English reading classes. According to Day and Bamfords (1998) motivation model for L2 reading, there are four major variables that determine the amount of L2 reading motivation, that is: (1) reading materials, (2 ) reading ability in L2, (3) attit udes toward reading in the L2, and (4) sociocultural environment. Metacognitive reading strategy instruct ion derived from Browns reciprocal teaching and explicit reading strategy training embedded the possible positive impact 33


on EFL high school students reading ability as well as reading attitude and thus has the potential to influence EFL students reading motivation to certain degree. Likewise, while this study trie s to investigate the effect of metacognitive reading strategy instruction (MRSI) on EFL high sch ool students readi ng comprehension and reading motivation, EFL high school students previous reading experience as well as their reaction toward MRSI experiment is also part of investigation of this study, with the inte nded goal of making connection between the reality and the theory. In other words, the purpose of this study is to explore the effects of metacognitive reading strategy instruction (MRSI) on EFL students metacognitive awareness in r eading, reading motivation, and reading comprehension of EFL high school students in Taiwan. In conclusion, this study plan to make use of quantitative and qualitative data with the attempt to see the e ffect of metacognitive reading strategy instruction and how this intervention change EFL students reading perception and reading motivation. By identifying EFL high school students previous reading experiences and reading problems along with the quantitative results can have a broade r application for researchers and L2 literacy instruction. Data Collection and Research Questions This research study followed a sequential mixed-method research design. As Farrell (2009) points out that there are th ree variables that cont ribute to differences in first and second language reading: (1) linguistic differences, (2) individual diffe rences and (3) sociocultural differences. Therefore, by usi ng mixed-methods research design, intervention study would get a more comprehensive result in terms of the im pact of MRSI on EFL highs school students English reading comprehension, English readi ng strategy awareness, and English reading motivation. An experimental study started out first and later fo llowed up with group interviews to examine the effect of metacognitive readi ng strategy instruction quantitatively and look for 34


causes and changes qualitatively with an attem pt to see the possible potent ials of benefits of metacognitive reading strategy instruction (MRSI) on EFL stude nts English reading. Five research questions of this study are as following. The first three research questions are used to guide the quantitative part of the study while the last two rese arch questions are used for the qualitative part of the study. 1. Does metacognitive reading strategy training affect high school EFL students metacognitive knowledge with re spect to their perceptions about reading and motivation to read English materials? 2. Does metacognitive reading strategy traini ng enhance high school EFL students reading comprehension? 3. Does the effectiveness of metacognitive st rategy training depend on previous level of general English language competence? 4. What are the factors contributing to EFL high school students read ing experiences and reading attitudes? 5. In what way can the features of metacongiti ve reading strategy inst ruction help current EFL English reading instruction? Hypotheses For this study, research hypotheses are as following: 1 aH: Perception about English reading will be greater for EFL high school students with metacognitive reading strate gy instruction than those students without metacognitive reading strategy instruction. 2 aH: Motivation to read will be hi gher for EFL high school students with metacognitive reading strategy instruction than thos e students without metacognitive reading strategy instruction. 3 aH: Reading comprehension will be greater fo r EFL high school students with metacognitive reading strategy instruction than those students without metacognitive reading strategies instruction. 4 aH: The effect of the metacognitive reading strategy instruction will be larger for EFL high school students for all langua ge proficiency levels. 35


Significan ce of the Study The Minister of Education (MOE, 1999) in Taiwan has announced that all 3th graders began their English instructi on in 2001. Taiwanese government believes that enhancing Taiwanese peoples English ability will also en hance Taiwans power to compete in a global society (Liou and Chen, 1998). Besides, like EF L students elsewhere in the world, Taiwanese students use the Internet to search for info rmation and communicate with people in other countries. Since English is one of the main languages for global communication and of the computer (Liou and Chen, 1998), it becomes important for Taiwanese students to be able to use and read English effectively. This study plans to investigate the effect of metacognitive reading strategy instruction on EFL secondary students metacognitive awareness in reading, reading comprehension, and reading motivation. Thus, the si gnificance of this stu dy lies in its potential for contributing to our deeper understanding of metacognitive reading strategy instruction on EFL readers, thus suggesting directions for future research. This study is al so significant in that it uncovers a general picture of English educatio n in Taiwan and its impact on EFL high school students reading attitudes through qualitative in vestigation regarding EFL high school students previous reading experience. By analyzing EFL student s reading experiences and the impact of traditional English education in Taiwan, this stud y thus gives a direction for English teachers and English education in Taiwan. Also, this study attempts to test the generalizab ility of the effect of metacognitive reading strategy instruction (MRSI) on ESL students to the field of foreign language reading in an EFL context. Although ex isting research shows that L1 elementary students and L2 college learners improve th eir reading comprehension from metacognitive reading strategy training, the extent to which this training is helpful to EFL secondary students is much less known. 36


Second, m any studies of English learning on Taiwanese students involved only descriptive analysis of English language learning strategy profiles in general, not teaching them in using reading strategies or metacognitive r eading strategies in particular (Huang, 2006; Lan and Oxford, 2003;Yang, 2003). Third, the length of the study, over two months, is longer than most other studies related to implementing metac ognitive reading strategy in struction. Fourth, of the studies reviewed in literature review, few of them had selected sample randomly (Chamot, 2005). This study tries to randomly select sa mple in hoping to eliminate the confounding variables to minimum degree. Fifth, this study will use read ing strategy journal, the idea borrowed from Anderson (2001), as a media for high school students to self-reflect on their English reading experience such as difficulties th ey encounter and problem solving strategies they use or plan to use. Rubi n (2003) also suggests using diarie s for instructional purposes as a way to help students develop metacognitive awareness of their own learning processes and strategies. By keeping a reading strategy journal, this study is hoping that it will help EFL adolescent students make conscious, informed decisions about their ow n reading and become more aware of their reading strategies use duri ng the reading process. Sixth, while Pressley (1998) found that by providing reading strategy to poor readers can improve learners reading motivation in L1 context, little experimental study exists to test the effect of reading strategy instruction on EFL students read ing motivation. Finally and most importantly, the effectiveness of metacognitive reading strategy instruction on L2 reading comprehension has not been firmly developed (Taylor, Stevens, and Asher, 2006); this study is intended to provide more evidence of metacognitive reading strategy in struction on EFL learning for Eng lish educators in Taiwan and will be of great interest to all those people or educators interested in the complex issues of learning to read in a foreign la nguage and pedagogical implication. 37


Definitions of Key Terms READING STRATEGIES: Oxford and Crookall (1989) defined strategies as learning techniques, behaviors, and problem-solving or st udy skills which make learning more effective and efficient. Applied to second language reading context, reading strategies are processes used by the learners to enha nce reading comprehension and overcome comprehension failures. COGNITIVE READING STRATEGY: Cognitive reading strategies used by learners to transform or manipulate the language while reading. In mo re specific terms, this includes note taking, formal practice of sounds a nd sentence structure, summarizing, paraphrasing, predicting, analyz ing, and using context clues. METACOGNITIVE READING STRATEGY: Metacognitive reading strategies are behaviors carried out by the learners to plan, arrange, and evalua te their own reading. Such strategies, include direct attention and self-eva luation, organization, setting goals and objectives, seeking practice oppor tunities, and so forth. In the context of reading, self-monitoring and correction of errors are further example of metacognitive reading strategies. (Singhal, 1999). MOTIVATION TO READ: Motivation to read was operationally defined as the scores obtained on English reading Motivation Questionnaire (Mori, 2002) in preand post-tests. Motivation to read has been defined by ot her researchers as those feelings that cause a reader to approach or avoid a reading situation. (Readence et al., 1989, p.102). Motivation to read depends upon the expectation of successful performance when trying reasonably hard and the value ava ilable rewards for success (Good and Brophy, 1991). In this study, motivation to read was operationally defined as the EFL stud ents score on a 30-item Likert-type instrument where one stands for very different from me and four stands for a lot like me 38


39 CHAPTER TWO LITERATURE REVIEW Introduction The purpose of this study is to test th e effects of metacognitive reading strategy instruction on EFL high school students Englis h reading perception, r eading motivation, selfevaluation skills, and reading comprehension. In or der to understand the context of this study, a review of related literature is presented in the following sections : (1) contemporary theories of L2 reading comprehension models are briefly summarized and discussed; (2) the role of metacognitive awareness in reading comprehension are addressed; (3) relevant research on reading strategies including de scriptive and experimental stud ies along with their findings and brief criticisms of any shortcomings are presente d; (4) a summary of overall contributions of previous reading strategy studies to current knowledge in the fi eld, including a description of strengths and weaknesses in existi ng research and conclude in a list of the research questions of the present study which are designed to fill those gaps. Reading Comprehension Models Before reviewing of the research literature relating to reading strategies, I briefly summarize three major reading models originally proposed from the perspectives of first language reading research that ar e usually discussed and applied to L2 reading theory. Current reading research has supported th e idea that both L1 and L2 read ers seem to go through similar cognitive processes (Alderson, 1984, Grabe, 1991, 2004; Eskey, 2005). These reading models have been influential in both L1 and L2 read ing research and can be distinguished from one another by its focus regarding how meaning is atta ined from print. For instance, the bottom-up model indicates that the reading process is guided by each word in the text and a reader decodes each word to obtain meaning. In contrast to the bottom-up model, the top-down model specifies


that the reading process is guided m ostly by a readers past experience and prior knowledge. The interactive model points out that the reading process is guided by an interaction between the text information and the readers previous know ledge as well as interaction between various reading strategies (Brunning, Shraw, and Ronning, 1999). In this study, metacognition is a new mode of instruction that focuses on the interactive nature of read ing, rather than a passive way of receiving information from the text through wo rd identification and task analytic learning. 1. The Bottom-up Model According to Gough (1972), the emphasis of this model is on print its elf. Readers start reading by recognizing the letters, word identification, a nd they gradually progress toward larger linguistic chunks to sentences, and eventually en ding in meaning. The whole reading process is basically word-based and readers construct th e meaning of a reading passage by decoding each word. Since this model emphasizes individual words in isolation, rapid word recognition is vital to the bottom-up approach (vanDuzer, 1999). This reading model believes that students who master this process quickly become proficie nt readers. However, students who are not successful at decoding become struggle reader s whose proficiency is interrupted by their inability to decode. Pressley (2000) claimed, sk illed decoders are able to recognize frequent letter chunks, prefixes, suffixes, and foreign root words rapidly and such ability can free more memory capacity in the brain for reading comprehe nsion. In contrast, le ss skilled readers put more effort into decoding words which leaves le ss processing capacity in the brain for reading comprehension. This notion has also been conf irmed by Breznitz (1997; c ited in Pressley, 2000) who concluded fast decoding impr oves reading comprehension. However, the bottom-up has been criticized that, bottom-up models suggest that all reading follows a mechanical pattern in whic h the reader creates a piece-by-piece mental translation of the information in the text, w ith little interference from the readers own 40


background knowledge. (Grabe a nd Stoller, 2002, p.32). In a ddition, this word-by-w ord decoding process causes slow and laborious read ing because short-term memory is overloaded, and readers forget easily what they have read when reading comes to an end (Adams, 1990). As a result, readers may only remember isolated f acts without integrating them into a cohesive understanding. Without cohesive un derstanding, readers wont evoke critical thinking. Without critical thinking, readers will lack the motivation to read on a regular basis. Therefore, the criticism of this model has been that it does not seem to consider the c ontribution of a readers active role and background knowle dge to reading comprehension. In other words, the linear nature (letters words sentences) of this reading model limits the scope of the reading process or envision the reading process as a one-way ma kes it fail to notice th e global or top-down processes (will explain in the next section) that take place during reading. Recognition of the limitations within the bottom-up model in explaini ng the reading process le d to the emergence of the top-down reading model. Application of the bottom-up model in L2. A growing body of research in L2 has supported the critical ro le of lower-level processing in reading comprehension (Bernhardt, 1986; Carrell, 1988; Eskey, 1988, 1997, 2005, Koda, 1992) a nd the lack of vocabulary maybe the greatest single impediment to fluent read ing by ESL readers. (Grabe, 1988, p.63). In a comprehensive review of L2 word-recognition research, Koda (1996) ag ain stressed the very significant role of word recognition in L2 r eading comprehension. What's more, in a study conducted the relationships betwee n the role of higherlevel syntactic and semantic processes and word recognition of sixty adult ESL learners in Canada, Nassaji (2003) found that lexical knowledge was strongly correlated with L2 read ing comprehension. However, reading in a second language bears some knowledge specific to that culture and society. Therefore, although 41


a num ber of researchers and studies have emphasi zed the role of lexica l knowledge in reading comprehension, some researchers maintain th at vocabulary knowledge is a necessary, but insufficient condition for the outcome of succe ssful reading comprehension (Koda, 1996). In other words, in order for true comprehension of a text to occur, a reader needs to possess other source of knowledge (Bernhardt, 1991, Carrell, 1988, Devine, 1987) and develop appropriate reading strategies (Anderson, 1991, 2001; Carrell, Pharis, and Liberto, 1989). 2. The Top-down Model Unlike the bottom-up model, the top-down mode l is a concept driven model where the readers background knowledge and expectations guide them to construct meaning from a reading text. As Eskey (2005) proposes, the top-down model empha sizes that the whole reading process is basically from brain to text (2005, p.564). That is to say, a reader star ts with certain expectations about the reading te xt derived from his or her b ackground knowledge and then uses his or her vocabulary knowledge they possess in decoding words to confirm and modify previous expectations (Aebersold and Field, 1997). In other words, a reading text itself has no meaning in the top-down reading model. It is the reader w ho constructs the meaning of the text by fitting it into his or her prior knowledge. According to Goodman (1967), who developed the top-down model, reading is a psycholinguistic gue ssing game and readers use their background knowledge to guess meaning. Smith (2004), who is al so in favor of the top-down model, claims that a reader plays a very active role in the pr ocess of translating prin t into meaning by using knowledge of a relevant language, knowledge of the subject matter, and knowledge of how to read to confirm or reject his or her hypotheses. The process of the top-down model is also called sampling of the text (Cohen, 1990). Describing the sampling process, Cohen (1990) maintains that a reader does not read all of the words and sentences in the text, but rather chooses certain words and phrases to comprehend the meaning of the text. Therefore, the top-down model 42


focuses on reading skills like m aking predictions a nd inference as well as guessing from context. The top-down model influences both L1 and L2 reading instruction in promoting the importance of prediction, guessing from context, and getting the gist of a texts meaning. Nevertheless, the top-down reading model has been criticized for its problem of overreliance on a readers prior linguistic and con ceptual knowledge and neglect the importance of the text (Eskey, 1973; Pearson, 1979). Moreover, the top-down model overlooks the possible difficulties that a reader may have or encounter with guessing or predicting the topic of text if the material is unfamiliar to him or her (Samuels an d Kamil, 1988). This is particularly true for second or foreign language learners. Up to this point, both the bottom-up and the top-down theories have been considered inadequate in te rms of describing a sound reading process. For the bottom-up theory, it was criticized for its failure to consider the readers role in the reading process, while the top-down theory relies too much on the readers prior linguistic and conceptual knowledge and negl ects the importance of the te xt (Eskey, 1973, 1986; Pearson, 1979). Thus, the inadequacy of both the bottomup and top-down models in explaining the reading process has led to the emergence of the interactive reading model. Application of the top-down model in L2 Since reading material s tends to be culturespecific, the top-down model take s into consideration that L2 readers may fail to understand a text if they do not posses or cannot access the appropriate cultural knowledge embedded in it. Reliance on top-down strategies at the expense of word identification skills might not contribute to comprehension. That is to say, limitations on cultural knowledge may cause distortion of the text meaning if the reader re lies on guessing from context and pr ediction (Eskey, 1988). It has long been argued that during the reading proces s, the readers langua ge knowledge, personal experiences and knowledge of the textual st ructure connect interactively to achieve 43


com prehension. Thus, Alderson (2000) specifically stresses that the w hole reading process is not an either/or selection between the bo ttom-up and top-down models, but involves the interaction between both approaches. (p.38). 3. The Interactive Model The interactive model combines features of the both bottom-up and top-down models and stresses the interrelationship between a reader and the text. It is now commonly accepted as the most conclusive picture of the reading process for both L1 and L2 readers (Anderson, 1999). Introduced by Rumelhart (1977), the interactive model suggests that ther e is an interaction between the bottom-up and top-down processes and this model advocates that neither bottom-up nor top-down models can by themselves describe the whole reading process. Rumelhardt (1977) says that both sensory and nonsensory come together at one place and the reading process is the product of simultaneous joint application of a ll the knowledge sources. (p.735). Grabe (1991) further describes the interactive th eory of reading as one that t akes into account the critical contributions of both lower-level processing skill (word identifica tion) and higher-level comprehension and reasoning skills (text interpretation). Theref ore, reading comprehension is the result of meaning construc tion between the reader and th e text, rather than simple transmission of the graphic information to th e readers mind (Eskey, 2005). Proponents of this model suggest that a skilled reader simultaneously synthesizes the information available to him or her from several knowledge sources of either bottom-up or top-down during the reading process. In addition, Stanovich (1980) brought the view of comp ensation into the interactive model by proposing that bottom-up and top-down pr ocesses compensate for each other in the reading process. In other words, when a reader lacks the appropriate content schemata for a certain text, he or she will rely on the bottom-up processes to compensate for the necessary background knowledge. The opposite could be true when a reader lacks the bottom-up skills 44


necessary to com prehend a text, he or she will re sort to high level proc esses. This phenomena explain for the process that poor readers tend to resort to high level processes more often than skilled readers given that the use of top-down processes seems to compensate for the poor readers limited ability of botto m-up processes (Stanovich, 1980). Application of the interactive model in L2. Because second or foreign language learners often find it challenging to understand the cont ext due to limitations with knowledge of the language and the culture unfamiliarity, most L2 reading specialists support the interactive reading model (Grabe, 2002, 2004; Eskey, 2005). Acco rding to Bernhardt (1990), in L2 reading, both text-driven and knowledge-driven processes ope rate simultaneously with varying degrees of success. The text-driven factors consist of word-recognition, phonemic decoding, and syntactic feature recognition, while knowle dge-driven operations involve intertextual perception, metacognition, and prior knowledge. All of these factors contribu te to successful L2 reading. Cook (2001) and Nassaji (2003) point out that even though readers may know all of the vocabulary and grammar, there are times that second language learners still cannot understand the text meaning. The difficulty seems to stem from the lack of social-cultural knowledge as comprehension is based on linguistic data (Bernhardt, 1991). Thus, background knowledge, in addition to the lower-level processing, has been view ed as another critical f actor that needs to be developed as part of the reading process. In th e reading process, the reader integrates the new information with the existing schemata (Anderson and Pearson, 1984; Carrell, 1983). Schema can be categorized as content schema and formal schema. For context schema, both Razi (2004) and Pritchard (1990) conducted studies in inve stigating the influence of cultural schema on reading comprehension. The results all showed that relevant cultural schemata obviously facilitated the reading process. In addition to re levant content or cultur al knowledge, familiarity 45


with text structure (formal schem a), is another factor affecting comprehension. Formal schema refers to the readers expectati ons about how information parts in a text are organized (Carrell, 1987) and this knowledge has been recognized as an important f actor in comprehension. For instance, Carrells (1984) study f ound that students from differen t language backgrounds recalled more information when the structure of the readin g task was close to the structure of their own language. In summary, reading comprehension is a complex cognitive process, and it seems more complex in an L2 context. Clearly, reading in an L2 is an active process involving various sources of knowledge such as relevant language knowledge, appropriate background knowledge and knowledge of text structure. In addition to the relevant linguistic, content, and formal schemata, L2 learners also need to be equipped with effective strategi es when approaching a reading task to compensate for insufficient knowledge in either language or content knowledge. Metacognitive Awareness in Reading Comprehension The theoretical framework that undergirds this study is metacognitive theory, the belief that self-monitoring or regulation is essentia l for reading comprehension (Flavell, 1976). Metacognitive awareness in r eading processes deal s with the knowledge that we perceive ourselves as readers, the reading tasks that we co nfront, and the reading st rategies that we apply so as to solve the tasks (Baker and Brown, 1984, as cited in Singhal, 2001). Generally speaking, metacognition in reading refers to the learner-performed actions such as planning, monitoring, or evaluating the success of a particular l earning task (OMalley and Chamot, 1994). Metacognitive awareness, therefore, also involves the awareness of whether or not comprehension is occurring, and the conscious app lication of one or more strategies to correct comprehension (Baumann, Jones, and Seifert-Kesse l, 1993). Several studies have been carried out to study the relationship be tween metacognitive awareness and reading comprehension. For 46


exam ple, Devines (1983) (as cited in Shi nghal, 2001) conducted a study on L2 readers conceptualizations of their L2 reading processes through interv iews. The results showed that older and more proficient readers tended to focus on reading as a meaning-making process rather than a decoding process. Meanwhile, the younger a nd less proficient readers appeared to do the opposite. In addition, conducting a study of L2 reading with 2 78 French language students, Barnett (1988) pointed out that the proficient ES L readers displayed more awareness of their use of strategies in reading Englis h than less proficient ESL reader s. Furthermore, Carrells (1989) study (as cited in Chern, 1993) also found support for positive relationships between readers metacognitive awareness of stra tegy use and their reading cap acity in both L1 and L2. More recently, Sheorey and Mokhtari (2001) (as cited in Mokhtari and Sheorey, 2002) discovered the relation between students readi ng capacity and strategy use while reading. In both ESL and U.S. college student groups, high-ability readers showed a higher level of awareness and strategy use than low-ability ones. Moreover, recent resear ch comparing the effectiveness of cognitive and metacognitive strategy training shows that expl icit instruction on cognitiv e strategies yields small, short-term improvements in reading performance, whereas training on metacognitive strategies results in more st able, long-term comprehension ga ins (cite in Koda, 2005; Carrell, 1998; Cohen, 2003; Tang and Moore, 1992; Zhicheng, 1992). Review of Studies on Reading Strategies As OMalley and Chamot (1990) point out, most research on strategies in both L1 and L2 contexts has focused on identifying and categor izing the reading strate gies that good or proficient readers employ in comparison to poor or less-proficient reader s. In the case of reading models, research on second language and fo reign language reading strategies has largely been informed by research done in L1 reading contexts with the majo rity of the participants were primary graders or remedial stude nts (Grabe, 1992). Given the di fferences in language learners 47


and contexts, it is reasonable to question the validity of using L1 reading research as a starting point of inquiry into L2 readi ng strategy use. A m ajor concern is the logical question of the extent to which less proficient or remedial L1 readers are equivale nt to less-proficient L2 readers (Grabe, 1992). In addition to the influence of a readers first language an d L1 literacy, his or her second language proficiency shoul d be taken into consideration. However, one cannot deny that reading is a cognitively learned proc ess for L1 readers, as it is of course for L2 readers as well. For that reason, L2 based research has relied so much on previous L1 work. The following is the brief summary of several key findings of L1 read ing strategy studies that are relevant to L2 reading. Reading Strategies in the L1 Context Block (1986) summarized several descriptiv e studies carried out between the mid 1960s and early 1980s that focused on the comprehension strategies of prof icient readers with English as their first language. She concluded that many of the strategies used by proficient readers were top-down and meaning focused strategi es, as described by psycholinguistics such as Goodman (1967) and Smith (2004) in the top-down reading comprehension model. Such studies have suggested, for example, that proficient L1 readers are more capable of monitoring their comprehension, are more aware of the strategies they use, and are able to use strategies flexibly by adjusting them to the text and purpose of a reading task (Blo ck, 1986). Additionally, proficient L1 readers have the following reading characteristics such as the abilities of differentiating between main points and details in a reading text, using textual clues to predict content and link information, r ecognizing discrepancies in a text, and solving the problems by using strategies to make such discrepancies und erstandable (Block, 1986). Research studies that focused on reading strategies point to the notion that the effective use of these strategies explains the deeper comprehension of proficient readers. 48


Other than the descriptive studies in the area of reading strategi es, there is also a considerable body of literature on reading strategy tr aining in a native language context. Overall, the types of reading strategy training among these st udies have been aimed at training L1 readers in summarizing texts (Brown, Campione, and Da y, 1981; Hare and Borchardt, 1984), using selfquestioning during reading (Wong and Jones, 1982), monitoring of understanding, and the use of specific fix-up strategies (Garner, Hare, Al exander, Haynes, and Winograd, 1984), activating prior knowledge and making inference (Han sen, 1981; Hansen and Pearson, 1983), using reciprocal teaching and explicit explanations (Brown and Palincsar, 1985). After reviewing several successful L1 reading comprehension st rategies training studies the common finding of these training studies is that direct instructi on in reading strategies has consistently positive results (Carrell et al., 1989, p.650). In other words, the key to successful training in these studies seems to be the ability to make inst ruction explicit enough to facilitate students development of metacognitive control of strategy use by providing clear and extensive explanations of the value of strategy use, a nd information on when and how to use them. Two influential bodies of research have documented how metacognitive strategy instruction can be integrated in to daily reading instruction: reciprocal teaching (Brown and Palincsar, 1985) and explic it strategy instruction (Duffy et al., 1986; Duffy et al., 1987). Brown and Palincsar (1985) developed reciprocal teach ing which focuses on direct instruction in comprehension-fostering and comprehension-mon itoring strategies. In a training project, 7th graders of L1 were taught to use four concrete comprehension strategies: summarizing the main content (summarizing), formulatin g potential test que stions (questioning), clarifying difficult parts of the text (clarifying), and predicting future content (pre dicting). During the training, the teacher worked with students and modeled how the strategies were used. Then, students 49


practiced this techn ique within a small group, and the teacher scaffolded until the students gradually took the responsibility for using those strategies on th eir own. Using such a technique directly impacted the independent test scores of poor readers who improved dramatically, going from below 40% correct to over 75% correct. This study demonstrated evidence to support the concepts that students can be taught reading strategies to im prove their comprehension and teacher modeling of specific reading strategies use in how to improve reading comprehension should be included in reading instruction st udy. The reciprocal teaching method was later replicated by L2 reading researchers such as Cotterall (1990 and 1993) in the US and in the EFL contexts conducted by Song (1998) in Korea an d Salataci and Akyel (2002) in Turkey. Dole, Brown, and Teathen (1996) conducted an other line of reading strategy research involving sixty-seven young L1 readers ranging from 5th grade to 6th grade. The authors compared a teacher-directed strategy in which th e teacher read prepared scripts designed to activate prior knowledge with inter active instruction in which the teacher and students together activated and discussed students background knowledge before read ing. Results indicated that at risk L1 readers who received interactive st rategy instruction made superior gains in comprehension performance over their peers who received traditional basal instruction, which is teacher-directed instruction. Also, an earlier study carried out by Hansen (1981) investigated the effects of two experimental methods on infe rential reading comprehension of twenty-four 2nd grade L1 children. The children were grouped into three experimental conditions: 1) the strategy group, 2) the question group, and 3) the control group. The Strategy Group was trained in prereading strategies and focused on integrating te xt information with pr ior knowledge before reading. The Question Group received training in answ ering questions which required inference practice between the text and prior knowledge. The Control Group received 50


trad itional story instruction accompan ied with the typical activity of literal inference probe. The instruction was applied to ten basa l-reader stories. Results of post-comprehension tests revealed that the performance of the children in both experi mental groups outperformed that of the control group. Standardized test scores and scores on an experimenter-designed test also favored the experimental groups. In conclusi on, when reading strategies are taught, readers performance in comprehension improved. A similar study about the inferential compre hension of good and poor fourth graders was investigated by Hansen and Pearson (1983). Forty 4th graders were assigned to one of four instructional groups, two groups of good readers (a n experimental and a control group) and two groups of poor readers (an experimental and a control group). The experimental treatment consisted of three parts: 1) making students aw are of the importance of drawing inferences between new information and prior knowledge structur e, 2) getting students to discuss some of their own experiences that were similar to even ts in the text, and 3) providing students with many inferential questions to discuss after read ing the texts. The results indicated that poor readers benefited significantly from this instruct ion, indicating that instru ctional procedures in reading do have positive effects on reading comprehension, especially for poor readers. Also, there are studies examining reading strategy instruction on specific strategies such as comprehension monitoring. For example, Miller, Giovenco, and Rentiers (1987) examined the benefit of self-instruction training on pe rformance of below average and above average readers. Forty eight 4th and 5th graders were tested on thei r ability to detect sentence contradictions in short expository texts after receiving either thr ee sessions of self-instruction or equivalent didactic instruction. Twenty-four ch ildren in the self-instr uction group were taught a series of self-statements designed to define the task, to select an a pproach to complete the task, to 51


evaluate the approach taken, to re inforce the learners efforts, and to check on the completion of the task. The task closely resembled metacogniti ve and monitoring reading strategies. Subjects in the control group were not exposed to the self -verbalization strategies. Results of think aloud protocols and post-reading questions on nine pass ages indicated that se lf-regulatory remediation strategy in reading was successful in enhancing students regula tory process while reading. Loranger (1997) conducted another similar study to determine whether students who were taught specific research-based strategies usi ng a transactional strategy instruction approach would improve in comprehension achievement a nd be more engaged during reading groups. The transactional strategies instruc tion refers to an approach that involves teaching readers to use comprehension strategies as they jointly construct interpretati ons of text. The emphasis is on making meaning of the text. Vocabulary instructi on occurs as needed when students encounter difficulty while reading. The particip ants in this study were thirty-two 4th grade students between ages of 8 to 10. The treatment gr oup was taught four comprehension strategies: predicting, questioning, clarifying, and summarizing which were parts of a reciprocal teaching method proposed by Brown and Palincsar (1985 ). The control group was conducted as a traditional reading group and the study lasted over eight weeks for a total of three hours of instruction per week. The readi ng sessions were videotaped and preand postinterviewed to determine knowledge of strategies performed by the students. Students were also required to keep response journals about their reading sessions. The findings favored the treatment group and Loranger (1997) concluded that transactional reading strategies trai ning is an effective method in improving students reading achievement. Reading strategies and moti vation in the L1 context In L1 settings, the strongest evidence of the direct impact of positive motivation on reading comes from Guthrie and his colleagues. In their studies, th ey demonstrated the impact of 52


reading engagem ent on both reading amount and reading comprehension (W igfield and Guthrie, 1997; Guthrie et al., 1997). They maintain that high motivation is highly correlated with the amount of reading. In addition, Schidfele (1999) demonstrated that personal interest (long-term intrinsic interest) is a signif icant predictor of comprehensi on and learning. Instruction of individual reading strategies has been shown to have a positive effect on reading comprehension, awareness of using strategies as well as motivat ion to read (Autrey, 1999; Black, 1995; Carriedo, and Alosno-Tapia, 1995; Cooper, 1998; Druitt, 2002; Guthrie et al., 1995, National Reading Panel, 2000). Another impact of teaching students how to use reading strategies to solve comprehension problems is on their motivation to read (Choochom, 1995; Guthrie, 2001; Guthrie and Alao, 1997; Hurst, 2004; Knoll, 2000). In an overview of engagement and motivation for reading, Guthri e (2001) proposed that: Engaged reading is a merger of motivation a nd thoughtfulness. Engaged readers seek to understand; they enjoy learning and they believe in their reading abilities. They are mastery oriented, intrinsically motivate d, and have self-efficacy. (2001, p.1). Guthrie and Alao (1997) posited that less succe ssful students lose their motivation to read because of unsuccessful reading experiences and l ack of confidence on th eir reading abilities. Based on this notion, the assumption of this study is that once students begin to know how to use reading strategies in English r eading and take more ownership of their reading, then motivation would improve (Snow, 2002). Motivation and reading strategies are related in some ways. For instance, research has found that a students motivation leve l impacts a students willingness of using comprehension strategies (Choochom, 1995). In a survey study examining the effect of motivational orientation and strategy use with ni nety students in the seventh, eighth, and ninth graders, Choochom (1995) c oncluded that intrinsically motivated students employed more strategies, exhibited greater frequency of self -regulation, and comprehended more from texts. 53


Besides, reading m aterials of hi gh interest in a particular read ing topic leads to high motivation, which leads to high comprehens ion (Hurst, 2004; Snow, 2002). Summary of Reading Strategy Study in the L1 Context Reading for understanding is a complex cognitive skill in which a reader constructs meaning by relating information from the text to prior knowledge. When a reading text can be easily understood or a reading task is simple, a reader can comprehend in the automatic way. But if a reading task is above ones reading le vel or difficult to understand, a reader needs to regulate his or her process of meaning construc tion more consciously (Baker and Brown, 1984). For that reason, reading instructi on in how to employ reading strategies is necessary for reading comprehension (Snow, 2002). Strategy-based readi ng instruction has always been regarded as very important in educational psychology and reading pedagogy in L1 learning situation, as learners internal monitoring and controlling of th eir learning processes are important to effective learning (Flavell, 1992; Pressley, 20 02). Research along this line also indicates that the process at the metacognitive level plays a crucial role in the development of intelligence and helps learners to take active control of their learning (Pressley and Afflerbach, 1995). As extension of the studies on the differences between good and poor readers, follow-up studies on the effects of reciprocal strategy instruction have also been reported to have positive effects on learners reading improvement (Loranger, 1997; Palincsar and Brown, 1984), and thus lend support to the benefit of strategy instruction. In all these L1 studies, L1 learners have demonstrated improvement in reading comprehension by virtue of the reading strate gy instruction program they have gone through. From the above discussion, it is clear that reading strategy instru ction does improve reading achievement of young L1 children as well as L1 students with reading problems. In general, these L1 reading strategy training studi es support the notion that students can be taught 54


to use strategies in helping them to comprehend better and read at ease. In addition, reading strategies should be explicitly and directly taught so students can understand the importance and application of reading strategies to reading. It is through ex plicit teaching of that reading comprehension and monitoring could be improved. Therefore, the goals for reading strategy instruction should be to enable students to sele ct appropriate strategies, familiarize themselves with all kinds of reading stra tegies, employ them to solve r eading problems (Presley, Johnson, Symons, McGoldrick, and Kurita, 1989; Wilhelm, 2001), and have them independently initiated by the students (Ellis, 1994; NRP, 2000; Snow, 2002) Moreover, effective reading does not rely on a single strategy, but incorporat es the coordination of several strategies (Snow, 2002; Pressley and Wharton-McDonald, 1997) and involves the constant, ongoing adaptation of many cognitive processes. (Williams, 2002, p.244), that is metacognitive reading strategy awareness. After reviewing 3 decades of the current res earch on strategic read ing instruction, Grabe (2004) summarizes the most c ontemporary discussions among L1 researchers center on the use of and training in multiple strategies to achieve comprehension (commonly including summarizing, clarifying, pred icting, imaging, forming questions, using prior knowledge, monitoring, and evaluating). As th e multiple strategies research suggests, most researchers now see that real value in teaching strategies as combined-strategies instruction rather than as independent processes or as processes taught independently of basic comprehension with instructional texts (Baker, 2002; Pressley, 2000). One should keep in mind that reading strategy instruction is not magic wand that will definite ly improve students reading comprehension; however, it is the students themse lves who should constantly reflect and evaluate their individual needs during the reading process. The action of constantly reflection and evaluation is the metacognition that readers should possess in order to be successful readers. Agreeing with this 55


conclusion from L1 studies, I believe that inco rporating reading strate gies instruction into English as a foreign language classes would also help to improve EFL students reading comprehension. Reading Strategies in the L2 Context A great deal of research on reading strate gy in the first languag e has revealed its importance in the reading proce ss and has applied to L2 reading domain. Researchers in L2 contexts conducted studies that were aimed not only at uncover ing possible reading strategies which learners used (Anderson, 1991; Block, 1986; Hosenfled, 1977), but also the effects of strategic reading instruction on reading improvement (Carrell et al., 1989; Janzen and Stoller, 1998). For instance, Hosenfled (1977) started w ith an investigation in finding the different reading behaviors and strategies between good readers and poor r eaders of French as a second language. Similar to the findings of L1 read ing study, there were evidence showing that the different strategies usage among go od and poor L2 readers. Later in the last th ree decades, the attention in second language reading research has been fo cused on understanding what proficient, skilled L2 readers typically do while reading, including identifyi ng the strategies they use and how and under what conditions they use those strategies. (Sheorey and Mokhtari, 2001, p. 423). In other words, studies were conducted to investigate successful and unsuccessful L2 readers reading strategy use through a variety of techniques such as th ink-aloud, interview, and questionnaire surveys. In addition to the ch aracteristics of good a nd poor L2 readers, L2 researchers have tried to examine all the va riables such as language backgrounds, learning preference, language proficiency, gender, cultur al backgrounds, etc., which might affect L2 readers strategy usage (Chamot, 2005). For ex ample, studies which have examined the relationship between gender and strategy use have come to mixe d and inconsistent conclusions (Ehrman, Leaver, and Oxford, 2003, 1990; Green and Oxford, 1995; Wharton, 2000). Ehrman 56


and Oxford (1989) and Green and Oxford (1995) have cam e to the same conclusion by discovering distinct gender differe nces in strategy use, while Ehrman and Oxfords (1990) failed to discover any evidence of differing language learning strategy us e between sexes. It might be concluded that although men and women do not always demonstrate differences in language learning strategy use, where differences are f ound women tend to use more language learning strategies than men. On the other hand, Whartons (2000) study found that men used more strategies than women (Wharton, 2000). As fo r the relationship between language learning strategies and the students prof iciency level is clearer. More proficient readers use a greater variety and often a greater number of readi ng strategies (Anderson, 2005; Green and Oxfrod, 1995; OMalley & Chamot, 1990; Wharton, 2000). Later in the early 1980s, several L2 researcher s also began training poor L2 readers to use some of the same strategies that skilled readers do with the belief that once poor L2 readers were taught reading strategies, then their reading comprehension would improve. However, due to the complex traits of L2 readers in terms of their diverse linguistic and non-linguistic factors as well as different research methodology being used, findi ngs from L2 research were unable to come to conclusive findings. The following discussion will categorize previous L2 reading strategies studies into descriptive studies and experimental studies. Descriptive studies in L2 context Taking cues from L1 reading researchers, second language re searchers began to investigate differences between proficient and less-proficien t readers in the late 1970s. Hosenfled (1977, cited in Carrell, et al., 1998) is widely acknowledge d as the first to explore the processing strategies of second language texts by good and poor readers of French. Her subjects were 9th graders learning French as a second language. They were asked to read in French but reported in English about their thinking processe s while reading French text. Through think57


aloud pro tocols, Hosenfeld found that successful r eaders of French as an L2 have the following characteristics: (1) keep the meaning of a passa ge in mind while reading, (2) read in broad phrases, (3) skipped unimportant words, (4) ha d a positive view in r eading. (Carrell et al., 1998, p.121). On the contrary, the unsuccessful reader s of French as an L2 have the following characteristics: (1) lost the meaning of sentences as soon as they decoded them, (2) read in short phrases, (3) viewed all words as important to to tal phrase meaning and thus rarely skipped words as unimportant, and (4) had a negative self-vie w as a reader. (Carrell et al., 1998, p.121). This preliminary study clearly described the strategies of good and poor L2 readers used to process the reading text. In spite of its contribution, however, this st udy has been questioned for not linking the strategy use to comprehension of specific paragraphs or the text as a whole. The data only focused on sentence-level comprehension rath er than overall comprehension of the entire text. Besides, think-aloud prot ocols, which did provide rich in sights into unobservable mental reading strategies, tended to reveal on-line processing instead of metacognitive aspects of planning or evaluating (Chamot, 2004). Years later, in a brief report of their research findi ngs, Knight, Padron, and Waxman (1985) used an interview and think-aloud protocol to report the frequenc y and types of strategy use of fifteen native English speaking (defined as good readers) and twenty-three mainstreamed Spanish bilingual (defined as poor readers) elementary school st udents. The result showed significant differences in terms of both the freque ncy of strategy use while reading English texts as well as the types of strategies used. In general, they found that monolingual English readers used several top-down strategies and twice as ma ny other strategies than bilingual readers used such as concentrating, noting deta ils, and self-questioning. Span ish bilinguals, on the other hand, used fewer metacognitive strategies than native English speakers. From this data, Knight, 58


Padron, and Waxm an (1985) concluded that th e bilingual students inability of using comprehension strategies was because they fo cused so much on decoding skills in reading English texts. Another possible explanation wa s that that they had not developed their own transferable strategies in their L1 due to early mainstreaming. This study offered insight into the effect of language backgrounds on the use of readi ng strategies. However, similar to Hosenfelds preliminary study, this study has been challenge d for not reporting the relationship between reading strategy usage and readi ng comprehension, either. What 's more, the comparison made between native English speakers and ESL reader s seems problematic because the L2 reading process is a lot more complicated due to other lingui stic and nonlinguistic factors. It is therefore recommended that comparison should be drawn between L1 and L2 individuals with similar reading behaviors. In order to accommodate for the previous two studies of not relating reading strategy usage to reading comprehension, Block (1986) chos e to focus only on less proficient readers who failed in a college reading prof iciency test before the study. Blocks subjects were six ESL students (three Chinese and three Spanish) and th ree remedial native English speakers of college level. Prior to entering college, the ESL st udents in her study had attended in American secondary schools for approximately the same am ount of time. Despite the fact that these students had similar language proficiency, Block found differences in strategy use within this group by using one-on-one interview / think-aloud t echnique. Subjects were asked to engage in thinking-aloud while reading two expository pass ages selected from an introductory psychology textbook and reported their reading process afte r each sentence. The process was recorded, transcribed, and scored, and then 20 multiple-choice comprehension questions were administered. Block then developed a coding scheme to classify reported strategies into two types of strategies: 59


(1) local and (2) general strategi es. Of the nine students in the study, the findings showed that the reade rs with higher comprehension scores on the retellings and multiple choice questions used general strategies of integrating new information with old information, distinguishing main ideas from details, referring to their background, and focusing on the textual meaning as a whole. On the other hand, readers with lower comprehension scores rarely distinguished main ideas from details, rarely referred to their background, infrequently focused on textual meanings, and seldom integrated old information with new information. In addition, Block found that ESL students used metacognitive strategies and monito red comprehension similar to native English speakers. Thus, conflicting with the previous two studies, this study demonstrated that language backgrounds did not account for the use of particul ar strategies. Moreover, there was evidence of individual difference and all th e readers made connections to their own experiences. What made a difference was that reader s with low comprehension scores failed to reconnect back to the original text. Since this study was conducted only with non-proficien t readers, it was later criticized that there is no way to know the role of language proficiency in reading strategy use. Sarig (1987) investigated the contribution of L1 reading strategies and L2 reading proficiency on L2 reading as well as the relationship between L1 and L2 reading strategies. Her subjects were ten Israeli female high school EFL students of Hebrew as an L1 and English as a foreign language. Sarig (1987) set out to compar e the strategies her subjects used in reading academic texts in each language, hypothesizing that L1 strategies transfer to L2 reading. The sample all had received eight years of formal instruction in English and represented low, intermediate, and high proficiency levels as determined by teacher evaluations and the results of proficiency test scores. Th rough think-aloud interviews, her subjects performed main idea analysis and overall text message syntheses using one text in thei r L1 and one in their L2. Then, 60


Sarig analyzed the frequencies and types of read ing behaviors of her subj ects in term s of what she called reading moves, or strategies. In add ition to identifying and cl assifying the types and frequency of strategies employed by her subjects, Sarig further analyzed the degree of effectiveness of each strategies used by her subjects in contributing to successful task performance depending on whether they promot ed or discouraged comprehension. Sarig identified a number of strategies and organized them into four types: technical-aid strategies such as skimming for the general idea, clarific ation and simplification strategies such as semantic decoding, coherence-detecting strategies such as using content or textual schemata to make sense of the entire text, a nd monitoring strategies. Her resu lts showed that the strategies employed by her subjects in reading L1 and L2 te xts were virtually identi cal in terms of types, frequency, proportion, and relative e ffectiveness for each language. Therefore, the findings of this study are accord with Blocks (1986) study in two aspects. First, both successful and unsuccessful readers use global strategies. In ot her words, the strategies were not inherently good or bad. Success in reading was shown to be a result of the quality of a readers unique usage of a combination usage of strategies, rather than the use of a certa in strategy. Second, the L2 reading process has a high degree of individuality. This study was unique in that strategies transfer did appear from L1 to L2 reading proce ss and the ability to tran sfer is not dependent on foreign language proficiency. In other words, when lacking of L2 proficie ncy, L2 students will rely on L1 reading strategy to compensate their L2 disadvantage. Anderson (1991) conducted another line of read ing research by choosing to investigate the differences in second langua ge reading strategy under two reading tasks-standardize reading test and academic reading task. Twenty-eight subjects in Andersons study were Spanish L1 students of mixed proficiency levels from va rious countries studying ESL intensively at an 61


Am erican university. Using a think-aloud protocol on two L2 reading tasks: 1) a standardized reading test with multiple choice comprehension que stions, and 2) an academic reading task of greater length followed by multiple choice questions), Anderson recorded the types and frequencies of strategies employe d by his research participants. He then performed simple regression analyses using the s ubjects proficiency levels as independent variables and both standardized test and academic reading task scores as dependent variables. Results showed that L2 language proficiency accounted for more of the variance on the standardized test than any strategy or combination of stra tegies. As for the academic reading task, language proficiency accounted for only a small percentage of score variance. A key finding was that there is no single set of processing strategies that signifi cantly contribute to success on these two reading tasks. Regardless the subjects different proficie ncy levels, it seems that the readers with high and low scores use of the same strategies wh en reading and when answering comprehension questions on both tasks. This finding led Anderson to conclude that strate gies were per se not intrinsically either successful or unsuccessful, but rather it is th e effective use of a strategy that makes comprehension successful. (1991, p.466). In general, subj ects who reported using more strategies tended to score highe r on both tasks and there was a si gnificant relationship between frequency of strategy usage and subjects scores No significant rela tionship was found between the numbers of unique strategies used and test scores. Therefore, Anderson concluded from his data that it is not sufficient to know what strate gies to use, but a reader must know how to use them according to their individua l styles and needs successfully. Block (1992) investigated the use of reading st rategies with regard to proficient and nonproficient readers. There were eleven native sp eakers and fourteen Chinese speakers of college level. They were further categor ized as 16 proficient readers (8 ESL and 8 native speakers) and 9 62


less proficient readers (6 ESL and 3 native speakers). She us ed a think-aloud m ethod to compare the comprehension-monitoring processes of na tive speakers and second language learners of English as they dealt with reference and voca bulary problems in an expository passage. The findings showed that ESL speakers with more English proficiency took more actions to solve problems and check solutions. Block reported that differences that existed in monitoring seemed due more to reading proficiency than to language backgrounds of the readers. (p.334). An important conclusion from this study was that the less proficient readers seemed to favor a local word processing strategy while the more prof icient readers tended to prefer a more global meaning based one. (p.334). This study thus indi cated a shift in strategy use based on language proficiency. Up to this point, the above studies seem to point out two very important findings: 1) good readers use strategies that best solve the probl ems, while 2) poor readers tend to use less effective and inappropriate reading strategies dur ing reading. However, the relationship between reading strategies and comprehe nsion was found to be more complicated than was suggested by these early studies. For instance, both Sa rig (1987) and Anderson (1991) questioned the traditional dichotomy of good and poor readers. They think there is no good strategy or poor strategy; instead, it is the read er who decides which strategy to use when comprehension breaks down in the reading process. Ande rsons study indicated that the sa me kinds of strategies were used by both high and low comprehension readers. Therefore, there is no one-to-one relationship between particular strategies and success or lack of success in r eading comprehension. Moreover, or perhaps in contrast to apparent group differences in terms of proficiency and general approaches to L2 reading seen in the studies of Anderson (1991), Block (1986), Hosenfeld (1977), Knight, Padron, and Waxman ( 1985), and Sarigs (1987) findings suggested a 63


great degree of variability between individual learners. Andersons (1991) findings and conclusion echoed Sarigs (1987) in two i mportant perspectives. First, strategy use was found to be highly individualis tic and no strategy was found to be in herently good when checked against subjects test scores. Second, succes sful use of strategies, that is using strategies strategically, requires a metacognitive approach to and control of the reading process. In closing, Anderson (1991) suggests that effective r eading strategies inst ruction aims to fost er language learners development as strategic readers, which is best achieved through e xplicit reading strategy instruction of not only the what and how of i ndividual strategy but, equally important the when and why as well. To be specific, that is the metacognitive awareness in the reading process. Similar to Sarig (1987) and Andersons findings, Kern (1997) carried out a case study of two American college juniors who learned French as a second language. The two students have different proficiency levels in French reading. The measurement consisted of a reading task interview. After analyzing thei r reading strategies, Kern found th at the two readers of different language proficiencies used similar reading strategi es, but they revealed differences in how they used these strategies in certain instances. Kern, once again, noted that no strategy has an inherently bad or good quality. The effectiveness of some strategies is dependent on a variety of contextual factors, including a readers purpose, la nguage competence, learning style, and L1 literacy background, as well as features of the particular text bei ng read (Kern, 1997). To sum up, no straightforward relationship appears to exis t between strategy use and reading ability. As research evidence indicate d, use of certain reading strategies does not always lead to successful reading comprehension, while use of other strategies does not always result in unsuccessful reading comprehension. (Carrell, 1991, p.168). In addition, strategies may not be inherently good or bad for a given reader. Rather, they may or may not promote 64


successful com prehension of a text, depending on th e particular reader, the particular text, the context in which the reading is going on, and the choice of other strategies in conjunction with the chosen one.(Cohen, 1987, pp. 132-133). Therefore, to be strategic readers, students not only need to know what strategies to use but they also need to be aware of when, why, and how to use these strategies according to their individual pref erence appropriately and effectively. This kind of knowledge is called metacognitive awareness or metacognitive reading strategies. The following section will discuss previous resear ch studies on metacognitive reading strategy. Descriptive studies in Metacognitive reading strategies in L2 context Research in the area of read ing strategies has recently st arted to focus on the role of metacognition. Researchers in L1 area like Flav ell (1992) and Pressley (2002) in particular argue for giving greater attention to the role of metacognition in helping students self-regulation of their own learning. They maintain that stude nts metacognition, i.e. their awareness of, and cognitive control and regulation over, learning, ca n enhance learning efficiency and self-efficacy (Vennman and Beishuizen, 2004). Metacognition has been defined as thinking about thinking (Anderson, 1999, p.72 and Carrell, 199 8, p.7) and as cognition about cognition (Carrell, 1998, p. 7). Sheorey and Mokhtari (2001) include both awareness and monitoring in their conceptualization of metacognition, which is defined as the knowledge of the readers cognition relative to the reading process and the self-control mech anism they use to monitor and enhance comprehension. (p. 432). Anderson (2002) briefly refers to metacognition as the ability to reflect on what you know and do and what you do not know and you do not do. (p.10). A number of research studie s with regard to metacogniti on (Barnett, 1988; Carrell, 1989; Sheorey and Mokhtari, 2001; Zhang, 2008; Kong, 2006) have examined the relationship between metacognition and reading strategi es. For instance, Devine (1984) conducted a case study with two participants of beginning and intermediate language prof iciency levels as defined by 65


University o f Michigan Placement test. Devi ne (1984) used oral reading interviews to investigate second languag e readers conceptualizations of their reading in a second language. Based on what language units they confessed to focus on, Devine classified subjects as sound-, word-, or meaning-centered. The finding found that different readers considered different aspects of reading as important. Results indicat ed that meaning-centered readers had good to excellent comprehension, while so und-centered readers were judged to have poor or very poor comprehension. In other words, the reader who used a meaning-centered approach demonstrated good reading comprehension. It was also observed that the use of a mean ing-centered approach can mitigate the effects of low language proficienc y, and thus allowing a reader to successfully transfer good first languag e reading strategies to the second language reading. Barnett (1988) conducted a study of forei gn language reading to investigate the relationships among reading comprehension, strate gy use, and perceived strategy use (strategy awareness). Two groups of fourth semester French students in college part icipated in Barnetts study which followed several steps. First, stude nts read an unfamiliar passage in French and rewrote the events in English. Second, before reading another unfamiliar story, they answered some general knowledge questions about it and then read it. Then, they answered a sixteen-item test to choose the best phrase, sentence, or para graph to continue each item (strategy use). At last, students were asked to answer a questionnaire about readi ng strategies that they thought they had used while reading (perceived strategy use). The results indicated that all three strategies (reading comprehensi on, strategy use and perceived stra tegy use) were significantly correlated for cognitively mature university-level readers of French as a foreign language. Moreover, results showed that as scores of e ffective strategy use increased, perceived effective strategy use increased, too. Ba rnett concluded that students who kept context in mind 66


com prehended more of their reading than studen ts who didnt use this strategy. Additionally, students who thought they used strategies effect ively read and understood better than those who did not think so. Barnett stated that stude nts who were taught st rategy use did show a significantly greater ability to read through cont ext than did their traditionally taught peers. (p.157). Carrell (1989) appears to have been the first L2 reading researchers to specifically focus on metacognition in second language reading strategy use, and to do so in a more quantitative fashion. In order to gather data on her subjects perceptions of their re ading abilities, repair strategies, preferred read ing strategies, and the difficulties th at they face when reading in both their first language and second lang uage, Carrell administered a que stionnaire to forty-five native Spanish-speaking intensive ESL students of inte rmediate and advanced proficiency and seventyfive native English speaking students studying Span ish as a foreign language of three different proficiency levels at a large Am erican university. In addition, Carrells subjects took a reading test in which they read two passages in their L1 and two passages in their L2 and then answered multiple-choice comprehension questions. Using simple regression analysis with questionnaire responses as her independent variables and L1 an d L2 test scores as her dependent variables, Carrell found that for both her ESL and her Spanish as a second language groups, local or bottom-up approaches to reading ( that is, decoding and sentences level strategies) negatively correlated with L1 test performance. For Spanish L1 students, the resu lts showed significant correlations with more global or top-down strategies reported, wh ich implied that these readers approached their reading in English in a similar fashion as their reading in their L1, Spanish. However, for English L2 students, Carrell obta ined the opposite results. For native English speaking students reading Spanish text, self-repor ted local approaches positively correlated with 67


test performance. Carrell explained these resu lts as being due to the relative difference in proficiency levels between the two groups as we ll as differences in learning environm ents. Specifically, Carrells ESL subjects were lear ning English for academic purposes in an immersion environment, making functional use of their second language, whereas her subjects reading Spanish L2 text were studying the langu age as a foreign language in an obviously nonimmersion environment and had only been studying for periods ranging from one to three years. Thus, as Carrell argues here, as well as in her later work, effective reading strategy instruction should seek to develop learners metacogniti ve awareness and knowledge of second language reading process through clear and explicit explanation, be cause it is possible that L2 readers will be able to transfer L1 readi ng strategies to L2 reading task s. Perhaps reading strategies instruction can be viewed as an attempt to jump start the transfer proces s and provide L2 readers with compensatory skills to construct meaning from texts before their bottom-up skills become automatized (Carrell, 1989). Sheorey and Mokhtari (2001) compared the metacognitive awareness of the reading strategies of ESL college student s with native speakers represented by American students. They sought to answer three questions : 1) are there any differences between ESL and US students in their perception of using strate gies? 2) are there any gender di fferences? and 3) is there a relationship between reported strategy and se lf-rated reading ability? Students provided information about their backgrounds including rati ng their reading ability. Then they answered the Survey of Reading Strategies (SORS) which is divided into th ree categories: 1) metacognitive, 2) cognitive, and 3) support stra tegies. Results showed that ESL students reported using more support strategies. In addition, both ESL and the US high reading ability speakers reported more use of support strategies than the low-reading ability ones. Also, the gender analysis showed that 68


fe male students in general reporte d using certain strategies more than the males. However, the researchers stated that because of the unequal numbers of females and males in the study, gender differences were not statistically significant. An important finding was that reading ability was significantly related to st udents reported usage of strategies. Sheorey and Mokhtari state that students who gave themselves a high rating on r eading ability, regardless of their language background, reported a higher use of all the reading stra tegies in the surv ey than did those students who gave themselves a lo w reading ability rating. (p. 446). Summary of descriptive studies in L2 context There are several points can be drawn fr om the above discussion regarding to the descriptive studies in the L2. Fi rst, proficient L2 readers are more focused on extracting meaning from texts and report greater frequency in (large ly top-down) strategy use than less-proficient readers. Second, less proficient readers tend to focus more at tention on decoding or bottom-up processes when reading a text. Third, reading strategies themse lves are neither inherently good or bad. Forth, proficient or less proficient L2 readers do not significantly differ in terms of the number and types of unique strategies used in reading. Fifth, whether language backgrounds account for the reading strategy use is inconclu sive. Last, but not leas t, metacognition, as it relates to a readers overall appro ach to the reading task, appears to be important for successful comprehension by means of planning, monitoring, and evaluating reading, and coordinating use of strategies. Experimental studies in L2 context As early as the mid-1980, Hamp-Lyons (1985) appears to have been the first second language researcher to compare th e relative effects of traditi onal and comprehension strategybased reading instruction with intermediate-level ESL students. Her subjects were twenty-four Asian college students. They were divided into three matche d groups, each with one of two 69


different teachers using differe nt teaching m ethods, but both using the same text. These groups were each taught by a different teacher, received traditional instructi on, with the third group receiving experimental text-stra tegic training. The posttest was id entical to the pretest and the memory effect was discounted as there was a 16 week interval between pretest and posttest. Other intervening variables such as ages, majors learner styles, time in the US and motivation were uncontrolled and thus assumed to be ra ndomly distributed. The results proved to be statistically significant in favor of the experimental group. Howe ver, as Kern notes, only eight out of twenty-four subjects received reading comprehension strate gy instruction in the study; the sample size was considered too small to statis tically significant to have relevance. Carrell, Pharis and Liberto (1989) examined th e effectiveness of re ading strategy training of semantic mapping (SM) and experience-text -relationship (ETR) methods on twelve students reading comprehension and both me thods resulted in the gained score of subjects reading comprehension. This study involved a heteroge neous group of twenty-s ix ESL students in a level four intensive ESL program at the Southern Illinois Univer sity. Two experimental groups were formed of which one received the seman tic mapping (SM) training and the other received the experience-text-relationship (ETR hereafter) training. Two control groups simply received the preand posttests. During four-day training, the semantic mapping (SM hereafter) group was given a series of reading passage s with questions used to stimul ate discussion and semantic maps were created. Meanwhile, the ETR group received the same passages as SM group and training activities included note taking, di scussion, comprehension questions and vocabulary activities that related to the texts. All subjects receiv ed a pretest prior to th e onset of training and a posttest nine days after the training. The tests included questions in varied formats and two out of the three passages on the test required the subjects to complete semantic maps. Scoring was 70


done accord ing to predetermined criteria. The re sults indicated that the control group did not have significant gains in the scores between their preand posttests on any four dependent measures which were multiple-choice questions open-ended questions, cloze semantic mapping, and open-ended semantic mapping questions. Each training group, however, did show significant gain scores in the open-ended questions, but not on multiple-choice questions. Carrell, Pharis and Liberto (1989) do caution, however, that such resu lts need to be replicated or supported by further research in this area. The cont ribution of this study is that, in general, the strategy training on the two methods was succes sful in enhancing second language reading comprehension. More specifically, both groups of students showed similar significant gains on one of the measures (open-ended questions). However, each group showed different gain scores on other measures (cloze semantic maps, and open-ended semantic maps). Carrell, Pharis and Liberto (1989) suggest that, based on the results adult students in second reading courses should benefit from the in clusion of explicit, co mprehension fostering metacognitive strategy training.(p. 669). In spite of the contributi on of this study to L2 reading strategy training field, this study has been questioned for the following aspects. First, there is no information regarding the validity of multiple choice questions. Second, results demonstrated that both semantic mapping and Experience-Text-R elationship training improved students reading comprehension. However, it is not clear to what extend thes e strategies are differ. Third, the researchers didnt explain why semantic mapp ing was used as part of pre-and post tests. Besides, including semantic mapping in the posttest clearly favored the group who was trained with this technique. Fourth, no delay test was ad ministered to see if the treatment effect was maintained. Fifth, the sample size of this study was too small (N=26) and whether the participants were randomly select ed or not was not clear. Sixth, since different teachers taught 71


four groups, there is a possible threat to the validit y of the experim ent. Last but the most important, since ESL readers were trained for us ing semantic mapping to improve their reading behavior and comprehension in a classroom-based task, there was no evid ence proving that they would be able to use this method on future and independent reading. Kern (1989) conducted a semester-long study in volving fifty-three st udents of French as a second language at a large American college to determine the effect of reading strategy training on reading comprehension and word inference ab ility. The subjects were native speakers of English and enrolled in the third semester (inte rmediate-level) of French classes. Subjects were required to read longer authentic French l iterary text in order to successfully complete course requirements. Participation was voluntary, but training was conducted as part of normal French classes. The treatment group had two sect ions and the control grou p has three sections. In total, five different instructors were i nvolved in the study. The treatment group received explicit instruction in reading strategy use in addition to the normal course content, while the control group received no instru ction in reading strategy use, but covered the same materials as the treatment group. After the treatment, subjects were presented with a passage in French and were asked to report what they were thinking as they read each sentence, what they understood, what they didnt understand, how they determined the meaning of unfamiliar words, whether they made predictions or inferences, and wh ether they translated into English. Both comprehension and word inference measurements were derived from the reading task. Data analysis revealed that reading strategy trai ning has a strong positive effect on L2 reader comprehension gain scores. Those who had the most difficulty in reading a ppeared to benefit the most from reading strategy instru ction. Kern (1989) suggested th at mid and high ability readers may have already transferred more of their effect ive L1 reading strategies to the second language 72


reading task. As for the effects of such stra tegy training on word infere nce ability, the results were less clear. Overall, Ke rn (1989) reported significant im provement with foreign language readers of French over a semester of training with emphasis placed on word, sentence, discourse, and reading purposes analysis st rategies. Despite its positiv e result in improving reading comprehension, there are some gaps in this study. First, Kern (1989) didn t explain the possible effects of the confounding variable caused by five instructors w ho were involved in the study. Second, this study didnt demonstrate how these strategies were taught. Strategy instruction was also found to be beneficial to low-level re aders as demonstrated in a year-long qualitative study conducted by Jimenez and Games (1996). During a two-week period in a middle school, three bi lingual students in English a nd Spanish were taught how to engage in a think-aloud method wh ile reading. In addition, explic it metacognitive and cognitive reading strategies were taught in order to improve their poor read ing skills. Through the use of culturally relevant texts, the in struction designed to promote comp rehension was found to have a strong potential for promoting a nd fostering the reading ability of such students who were performing at low levels of literacy in the middle school grades. Song (1998) also took a metacognitive approach to teach reading strategies in her study with adult Korean EFL readers in order to investigate instructi onal effects on reading comprehension. Song attempted to incorporate L1 reading research c onducted largely with poor young readers and extends it to the adult second language cl assroom. The subjects were fifty tertiary-level Korean students of mixed language proficiency le vels majoring in liberal-arts. The study was designed with all s ubjects receiving the experimental treatment. The data was analyzed in terms of low, mid, and high la nguage proficiency groups as established on a treatment pretest. Strategy training took place over a 14-week semester and was integrated into 73


the regu lar classroom curriculum. The instruct ional approach employed was virtually identical to Brown and Palincsars (1984) reciprocal teachi ng (one-on-one based) in terms of procedures with some modifications to fit the large size of classroom setting. Four reading strategies (summarizing, questioning, predicating, and clarif ying) were taught simultaneously as a single four-step metacognitive approach to L2 reading. Learners were taught to read the beginning of a text and to first summarize the content of what had just been read. Next, questions were composed based on the text content. The third step was to predict the content of the following sections of the text. The last step was to evaluate the text content for compatibility with prior knowledge and identifying points that st ill needed clarification (Song, 1998). In teaching the above approach to her subjec ts, Song states that she also included explicit explanations in L1 (Korean) of the nature of strategic learning and strategic reading. Song states that, in addition to the four main strategies taught, other strategies such as skimming, contextual word-guessing and using text structure to aid comprehension were also introduced to students on an occasional basis. Data collection for the study was done by means of a test-retest format. The instrument which served as both pretest and po sttest was a multiple-choice reading proficiency test constructed for the purposes of the study. The test contained six short passages, each ranging from 302 -333 words in length. Each pa ssage had three types of question items: main idea, factual, and inference. Results of the pretest administration were used to classify subjects as low-, mid-, and high-proficiency for later analys is. Pretest was also used as the posttest which as was administered at the end of the seme ster-long training period. Data was analyzed by means of Repeat-Measure ANOVA, with test time and proficiency level as the independent variables, and test scores as the dependent variables. Both significant main and interaction effects were reported, with a si gnificant difference between over all pretest and posttest means 74


and training shown to benefit low and m id-prof iciency learners more than high proficiency learners. A follow-up ANOVA conducted on questi on-types showed signifi cant gains for main idea and inference questions. Song concludes that the treatment did have a significant positive effect on her subjects reading comprehension in terms of understand ing main ideas and making inferences, which benefit low and mid-proficiency learners. Nonsignificant gains for high proficiency learners were interpreted as evidence of prior strategy use. While there are some encouraging aspects to Songs studya relatively long tr aining period, explanations of cr itical informati on in subjects L1, which support the validity of the study, and the lack of any control or comparison group creates some doubts as to the cause of her subject significant gains. Some studies have found that direct instructi on of reading strategies were of greater help to students with lower proficiency in the L2. Based on the success of teaching students summarization strategies in L1, Co rdero-Ponce (2000) conducted a st udy to test the effects of L2 metacognitive strategy training in summarizatio n on the ability to comprehend and summarize expository texts. Thirty university level students enrolled at an intermediate French course were divided into the experimental and control groups. Testing incl uded pretest, immediate posttest, and delayed posttest with all of them involving two tasks written recall and summarization. The training was conducted on two periods of sixty mi nutes. The researcher introduced the following rules to teach summarization: collapse list, use topic sentences, get rid of unnecessary detail and collapse paragraphs. Results i ndicated that students significan tly improved their comprehension and recalled more ideas in the immediate posttest. In addition, training had positive effects on students ability to summarize French texts inco rporating the rules introduced to them in the immediate and delayed posttest. The author concluded that these summarization strategies can 75


be taught to college students with low levels of L2 proficiency to provide them with cognitive resources to rely on during comprehension. Cord ero-Ponce (2000) comments this training study with intermediate-level French students that such training programs may provide students with compensatory cognitive resources to rely upon during comprehension, thereby offsetting, to a certain degree, their limited L2 linguistic knowle dge and lessening the cognitive load. (p.346). Salataci and Akyel (2002) examined the imp act of teaching reading strategies to preintermediate Turkish EFL students. They used the experience-text-relationship and reciprocal teaching methods. The instruction lasted four weeks (three hours a week). The strategies introduced and practiced by st udents included: using prior knowledge, summarizing, finding main ideas, prediction, clarificati on, and some other repair strategies. The findings indicated that students use of bottom-up strategies such as using dictionaries and questioning meanings of word decrease when reading in English because they were not focused on word level understanding after the treatment. On the othe r hand, the instruction ha d a positive effect on students use of top-down strategi es when reading in English and Turkish. The strategies of prediction, summarizing, and using prior knowledge were used significantly more frequently. In addition, the use of metacognitive st rategies was higher when reading in English after instruction. Whats more, the reading comprehensi on scores increased after instruction. Another line of L2 reading strategy traini ng study has focused on providing L2 readers with knowledge of text structure. Research has found that different cultur es have different ways of representing ideas in written text and this difference often cau ses certain amount of impact on L2 readers reading comprehension while approachi ng English reading task. In response to this effect, Carrel (1985) conducted a training study with 25 high-int ermediate proficiency college ESL students of various L1 backgr ounds studying intensively at a large university in the US. 76


Over five one-hour class sessi ons in one week, subjects in the experim ental group (N=14) received instruction which raised awareness of four types of Englis h top-level structures (macrostructures) found in expository texts (comparison, causation, problem/solution, and collections of descriptions). Training for the experimental group initia lly centered on explicit and extensive explanations by the instructor conc erning the nature of re ading expository texts, the benefit of using the top-leve l structure strategy in supporting comprehension, and how to use the strategy with different toplevel structure texts. In ad dition, students were given study packets with instructor explanations as well as practice texts and exercises for subjects to work on at their own pace. Checklists were also included in the packet so that subjects could monitor and regulate their own learning. (p.736). During this period, a control group (N=11) read the same texts as the experimental group but engage d in various other linguistic and comprehension activities. Data collection instruments included pretest, posttest, and delay posttests on which subjects read two passages (one passage of comparison and one of collection of descriptions toplevel structure) and produced written recalls without referring to the original texts. Recalls were done in the subjects L2, English. For each test passage, subjects were also asked to identify the organizational plan the aut hors of each passage employed in writing through an open-ended question. All written recalls were scored for the qua ntity and quality (in terms of top-level vs. lower-level ideas) of idea s recalled, with a reporte d interrater reliability of r = .96. In addition, the organization of each subjects recall was noted to check for subj ects use of the original toplevel structure in the text in writing their r ecalls. Results of Chi-square and one-way ANOVA tests with the treatment as independent variable and text structure r ecognition, text structure recall use and posttest and delay posttest score (quantity and quality) as dependent variables, 77


showed sign ificant differences in terms of r ecognition and use of top-level structures on the experimental group. In addition, a significant difference in posttest scores was found between the experimental and control groups in favor of the former, which appears to have held for the delayed posttest as well (although ty pe of test employed was unspecified and statistics were not presented). Overall, the contribu tion of this study supports the no tion that the explicit instruction in the top-level structures of English texts can enhance ESL students comprehension and recall. It appears that based on the evid ence presented in this conclusi on is supported for the teaching of this particular strategy (use of knowledge of text structure) for strategy-based instruction as an instructional approach. However, this study has been questioned for the following aspects. First, this study didnt specify how subjects were taught to ac tually use a text-structure based strategy (procedural knowledge), as opposed to merely bei ng taught about text structure. Second, only 5 days of one hour per day training is a rather s hort training period (although this does appear to have produced significant results). Third, having subjects perfor m written recalls in their L2 rather than in their first language produced a pot ential violation of test content validity and likely had potential for producing confounding effects, although no differences we re indicated between treatment groups on the pretest. Forth, the samp le size (N= 25) of this study is small. Modeled closely to Carrells (1985) study with ESL students discussed above, Raymond (1993) studied the effects of text structure strategy training wi th French as a second language learners recall. Forty-three native English sp eakers of French as a second language of highintermediate proficiency levels were in the st udy. They ranged in ages from 18 to 23 and had completed five semesters (a total of 260 hours) of college level language study at a large Canadian university. Pa rticipants in the study volunteered to participate, but were paid for the 78


study. Subjects were divided into one experim ent al and one control group and were determined to be of equal proficiency by means of a pre-treatment standa rdized test. The study took place outside of the regular language class and was conducted by an outside instructor. During five one-hour training sessions spread over a twoweek period, the experimental group received strategy training in the identification and use of five French top-level structures found in expository texts (description, sequence, causation, problem so lution, and comparison) and accompanying signal words in order to promote recal l. Instruction for the structure was designed to be metacognitive and included explicit instruc tion in: what was the strategy, why the strategy should be learned, how to use the specific stra tegy and when to use it. Short quizzes were provided to help the subjects to evaluate the use of the stru cture strategy. During the five sessions, the experimental group rece ived strategy instruction, while another instructor taught the control group using the same texts as the expe rimental group for the same amount of time, but with standard questions and answers tasks. Data was gathered by means of preand posttests on which subjects read one of two texts with the problem-solution top-level structure and determined to be roughly equivalent in term s of difficulty through readability measures, counterbalanced and randomly distributed so that half of the subjects read a given text on the pretest and the other half read th e same text on the posttest. Wh en subjects had finished reading the text, they answered 10 Likert-scale questions regarding their percepti ons of text difficulty, memorability, affect, interest, background knowledge, and clarity of argument, organization, recommendations, content, and discussion of content (Raymond, 1993). After this was completed, subjects placed the text in an envelo pe and then recalled in L1 as much of the information in the text as possible in writing. Each subjects text read ing time and recall writing time were also recorded. The posttest, given one month after the end of training, used the same 79


for mat as the pretest. Recalls were scored using an idea unit protocol, with scores calculated as ratios with the sum of the number of idea units pr esent in a recall divided by the total number of units present in the original for each text. Because it was found that some of the subjects in both the experimental and control gr oup spontaneously used the text top-level structure in their pretest recalls (that is, before the experimental group receiv ed explicit training on the structure strategy), it was determined that there was a differe nce in subjects prior k nowledge in the use of the text structure and signal wo rds. Consequently, an analysis of covariance was first conducted on the data, with treatment condi tion as the independent variab les, posttest recall score as dependent variable, and pretest recall score as a covariate. A mixed design repeated measure ANOVA was then performed with treatment, text, and text time (pretest stands for Time I while posttest stands for Time II) as independent variables, and text recall mean (adjusted for pretest) as the dependent variable. In contrast to Carrells (1985) study, results of the analysis showed no main effect for treatment between groups. However, a with in-group two-way inter action was found between text and time and a three-way w ithin-group interaction was found between treatment, text, and time. That is, on the pretest, the two texts pr oduced significantly different mean recall scores (analysis of difficulty and prior knowledge Liker items revealed significant differences between texts), which was also true on the posttests, but the text means reversed their relative position. This means that there was a higher recall mean for the more difficult text on the posttest and a corresponding drop for the easier text. For th e experimental group, Raymond interpreted this interaction as the result of subjects need to consciously apply a text structure strategy on the more difficult text in order to comprehension and recall it and it was not the case with the comparatively easier text. For the most part no major problems were noted for this study. 80


However, as in the case of other intervention studies, the training period was short and m ay have contributed to the absence of main instructional effects. Besides, Raymond didnt offer any information regarding its validity in terms of the measurements used in the study. Moreover, using idea unit as a measurement of students reading comprehension might not really be accurate since idea unit is affected by memory. Furthermore, the sample was selected through cluster randomization and thus more participants were needed. However, a significant gain was found on one text and as noted above, it was actually found to be the one that was more difficult. Findings from this study also make clear that reading comprehension is a complex interaction between reader, test, and task and that instruction in reading strategies may not offer quick solutions. Summary of experimental studies in L2 context From the above discussion regarding to th e experimental studie s in reading strategy training, several points can be summa rized as the following. First, wi th regard to the benefits of strategy instruction, some results of the traini ng studies reviewed here are mixed. Second, with various design and methodological problems, it give s us reasons to doubt the claimed benefits reported in several of the studies. Third, there ar e beneficial effects of training, especially for readers of lower language proficiency. Fourth, dir ect and explicit teaching of reading strategies had a strong positive effect on L2 readers co mprehension. Last, metacognitive strategy training in semantic mapping and experien ce-text relationship is effective in enhancing L2 reading comprehension. Reading Strategy and Motivation in L2 context In L2 settings, there is lit tle research specifically on th e relation between motivational variables and reading comprehens ion. Most L2 motivation resear ch focuses more generally on language abilities. Drnyei (2001) provides an excellent overview of mo tivational factors and 81


their influences on L2 learning. In H uangs (2006) survey study with 212 Taiwanese college students, the findings indicated that they would have read a lot more or extensively if they were taught reading strategies in high schools. Therefore, based on th ese studies with regard to the correlation between motivation and reading compre hension in L1 setting and L2 setting, it is believed that EFL students reading motivation would rise after the metacognitive reading strategy instructi on are implemented. Reading Strategy Studies on Chinese Speakers Most studies found for this literature review with regard to Chinese speakers learning English are descriptive in nature Several western L2 reading re searchers have studied Chinese EFL learners and found that Chinese EFL readers are unable to use their conceptual abilities to the fullest potential due to difficulty in transfer of reading skills from L1 to L2. As such, because the traditional method of learning is to memorize, they were unable to use the more abstract process strategies, such as guessing contextual meaning to attain f luent levels of reading skill. (Field, 1985, p.175). For example, Chinese EFL reader s reading strategies were greatly different from those of their American counterparts (Kohn, 1992). According to Kohns observation, American readers tend to read rapidly, while Ch inese readers tended to reading slowly. This comment has later been criticized of being from Kohns perspectives beca use he didnt ask his students how they themselves conceptualized their knowledge of or actual use of reading strategies. However, later a study conducted by Parry (1996) confirmed Kohns conclusion. Parry (1996) analyzed 25 Chinese trainee-teacher s written journal entries and indicated that Chinese tended to use bottom-up strategies more than top-down strategies and this tendency was closely linked to their L1 l iteracy tradition and their understan ding of the reading process. In contrast, Zhangs (2008) st udy investigated Chinese EFL read ers perceived use of reading strategies with an EFL readi ng strategies inventory (subject N=312 from China). His findings 82


suggested that, by and large, the Chinese readers actually use both local a nd global strategies for m eaning-construction. The results of this study differed because readers who demonstrated higher levels of comprehension reported using more global strategies such as guessing meaning through inferences, while the readers who demonstrated lower levels of comprehension reported using more local strategies such as detailed word meaning. Other than descriptive studies regarding to Chinese EFL learners English reading characteristics, there are studies investigate the effects of othe r variables, such as language proficiency levels, text stru ctures, and background knowledge on reading comprehension. For example, Chern (1993) carried out a study investigating Chin ese readers metacognitive awareness on text structure in both English a nd Chinese. This study tested the hypothesis that more proficient language learners are more aware of the strategies they use in reading. Subjects were twenty-eight native speakers of Manda rin Chinese who learned English as a Second Language in college level. Subjects recruited from fourth-year English majors were categorized as good readers while subjects drawn from first-year in other academic areas were categorized as poor readers. Subjects were first interviewed abou t their reading habits in Chinese and English. They then read two manipulated passages, one in English and one in Chinese. In each passage, eight function words were removed and replaced with nonsense words, ch aracters, or character combinations. After reading, the subjects were interviewed about how they resolved difficulties in reading the passages. It wa s found that, in general, the subjects were more bothered by difficult words in English than by those in thei r native language. When confronted by unfamiliar words, good ESL readers used context clues to fi gure out meaning while reading in English. In contrast, poor readers translated them in Chinese and focused on lexical features in English. Results were consistent with previous research and thus concluded that the awareness of using 83


strategies to get m eaning in readi ng in L2 appeared to distinguish good readers from poor readers. While in this study the poor reader s appeared to be more aware of strategies in tackling words than in dealing with global co mprehension, it reveals the need to emphasize the skills for holistic reading and it is advisable to enhance students awareness of their own reading processes and develop their ability in selec ting appropriate reading strate gies through instruction. Zhang (2008) conducted another study to investigate EFL college students metacognitive knowledge of readin g strategies in an acquisition-poor environment in China. He suggested that the available stud ies on Chinese EFL readers have not adequately addressed the issue of metacognitive reading st rategies and their English read ing proficiency. Therefore, he tried to see the relationships between types of metacognitive knowledge of reading strategy use and language proficiency levels. Ten EFL univers ity students were selected from a sample of 312 participants in China. Semi -structured interviews, mainly in Chinese were used under the framework of Flavells (1987) metacognitive st rategy model to elicit students metacognitive knowledge. The interviews were recorded and tr anscribed and three categories of metacognition were defined: 1) person, 2) ta sk, and 3) strategic knowledge. Th e findings showed that Chinese students metacognitive knowledge of reading st rategies closely resembles their English proficiency and L2 reading ability interacted w ith L2 proficiency level. That is, those who achieved high scores on metacognitive knowledge s howed a clearer awareness of strategy use. And those who scored lower didnt realize that they needed to a dopt different reading strategies to solve certain problems in r eading. The findings of this stud y suggest the need for teaching EFL Chinese learners English reading strategies. Kong (2006) investigated the connection be tween L1 and L2 reading through think aloud and interview methods. This study aimed to e xplore the differences between Chinese good and 84


poor readers in their strategy use in H ong Kong. Eight 7th grade students were selected, four good readers and four poor readers, and they pe rformed a think-aloud task and an interview throughout the study. The findings of this study indi cated that Chinese good readers used more strategies and had better ability and knowledge of strategy use than did poor readers. In addition to the cognitive deficiencies, poor readers were also found to have poorer intrinsic motivation than do good readers. The combined problems of poor reading ability and motivation made them reluctant to process the text at a deeper level and they gave up easily when they encountered reading difficulties. Another study carried out by Abbott (2006) i nvestigated whether top-down or bottom-up reading strategies fa vor in certain cultural or linguistic groups. The resu lts of verbal report data were collected from Arabicand Mandarin-spea king English as second language (ESL) learners to identify the reading strategies involved in performing thirty-two reading questions. Then a confirmatory approach to differe ntial item functioning was used to determine whether bottom-up and top-down items functioned differentially for e qual-ability Arabic and Mandarin ESL learners. Results revealed systematic group performance differences in four bottom-up and three topdown strategy categories. Items involving break ing a word into smaller parts, scanning, paraphrasing, and matching were favored by Mandarin speakers, whereas items involving skimming, connecting, and inferring were favored by Arabic speakers. Little experimental studies about reading strategies of Chinese speakers can be found for review. One study, conducted by Chen (1995), expl ored the effect of previewing and providing background knowledge on Taiwanese college studen ts reading comprehension in American short stories. Approximately 240 college freshmen were randomly assign ed to four treatment groups and read two short stories. Before read ing each story, one group listened to a 200-word 85


preview, a second group listened to backgr ound knowledge, and a third group listened to both the preview and the background know ledge presentation. The fourth group read each story without any assis tance. The results on shortanswer questions and multiple choice posttests showed strong positive effects of the previewing and combined strategies treatments and weaker positive effects of the background knowledge treatment. The findings of this study are in line with the previous study as they both suggest that previewing and pr oviding readers with background knowledge of a reading text can increase their reading comprehension. Remaining Gaps in Knowledge From the previous literature review, there ha s been a heavy reliance on L1 research to L2 reading research, with an insuffi cient number of EFL reading stra tegy training studies, especially with regard to learners metacognition. In addition, according to a meta-analysis conducted by Taylor, Steven, and Asher (2006), the effects of reading strategies training on L2 reading comprehension showed that there is no statisti cally significant differen ce between training that did and did not use metacognitive reading strategies in L2 reading comprehension results. In other words, the results of th e experimental studies on the re ading strategy training are not conclusive. As suggested by Ca rrell et al. (1998), strategy training which focused on learner metacognitive development is more effective than other types of training. Also, recent research comparing the effectiveness of cognitive and met acognitive strategy training shows that explicit instruction on cognitive strategies yields small, short-term improvements in reading performance, whereas training on metacognitive strategies results in more st able, long-term comprehension gains (cite in Koda, 2005, Carrell, 1998; Chamot, 2004, 2005; Cohen, 2003; Tang & Moore, 1992; Zhicheng, 1992). While previous research as reviewed here has helped to shed light on various issues related to EFL r eading strategy training and resear ch, more experimental studies regarding metacongitive reading strategy are needed in this area of research. Moreover, not much 86


research has focus on the possible impact of read ing strategy training on EFL students reading motivation. Based on the expectancy-value theo ry proposed by Day and Ba mford (1998, cited in Mori, 2002), metacognitive reading strategy inst ruction involving recipr ocal teaching method which has the potential to inspire EFL stude nts motivation and improve their reading comprehension. If EFL students feel the ease in English reading, then it would motivate them to get involved and read more on a regular basis. And if EFL students are r eading more, then their reading achievement and language prof iciency level can be expected. In conclusion, as discussed in the prev ious section, though much is known about how good and poor readers do in the field of second language reading or fo reign language reading, not much is known about the effect of readi ng strategy training on EFL students reading comprehension and reading motivation, especially with the adolescent populati on. It is for these two reasons that the current study was planned to undertake. 87


88 CHAPTER THREE METHODOLOGY Purpose of the Study The purpose of this study was to examine the effect of metacognitive reading strategies instruction (MRSI hereafter) on secondary EFL students reading comprehension, reading strategies awareness, and motiv ation to English reading. Meanwhile, secondary EFL students English past reading experience and their reaction to this kind of interventi on were also part of this studys interest. The questions of interest guiding th is study were: 1) what are the effects of metacognitive reading strategy in struction (MRSI, her eafter) on secondary EFL students reading comprehension, reading strategy awareness, and r eading motivation? 2) what factors contribute to EFL students reading behavior ? 3) in what ways did the inte gration of MRSI in EFL class change the participants? In this study, an experiment al explanatory study using a mixed methods approach was conducted for data collection and to quantitatively and qualitativel y examine the effect of MRSI on EFL students reading experien ce by integrating MRSI into English class in a urban public high school of southern Taiwa n. A sequential mixed method approach was used to better understand the whole impact of MR SI by first quantitatively conduc ts the experiment and later qualitatively explain the real life context in which is occurred (Yi ng, 2003). It is my intention to not only establish the pre and post SORS and ER MQ measures but also examine and present observations of EFL students and their instructor behavior s in the reading program. The sequential method enables me to conduct quantita tive data first and follow-up qualitative interview and later integrate the information when interpreting the overall results (Creswell, 2003). The preliminary assumption was that MRSI would help improve secondary EFL students reading comprehension, reading strategy awareness, and reading motivation.


MRSI is the central issue for this stud y because of its previous success reported in the literature from multiple studies in the field of the L1 and L2 reading instruction. Even though little research regarding EFL stude nts response to MRSI was avai lable at the time of planning and upon completion of this study, the purpose of th e study was not to compare the effectiveness of MRSI with other types of EF L reading instruction. Instead, the focus of this study is to explore EFL students reading problems and why it is necessary for MRSI to be incorporated into regular English reading instruction. Moreover, qualitative data were collected af ter the experiment to further explore the effectiveness of this metacognitive reading st rategies instruction. Based on a comprehensive literature review in this area, there is a common agreement th at explicit reading strategy instruction does improve reading comprehens ion of native speakers and second or foreign language readers (Block, 1986; Brown and Palin csar, 1985; Carrell et al., 1998). However, insufficient experimental research has been done to investigate the e ffects of metacognitive reading strategies on EFL/ESL st udents reading strategies awar eness, motivation to English reading, reading comprehension, and even fewer st udies have examined its effects on adolescent EFL language learners, a population beyond Piagets Formal Operations stage (11 or 12 years old to adult) that are capable of thinking logically and abstractly. For that reason, it is assumed that, metacognitive reading strategies instructi on, a training focuses on consciousness-raising in reading process, is integrated into a regular English class in an EFL context can produce more successful comprehension in English as a fore ign language and help EFL learners become independent and lifelong readers. This chapter descri bes the methods and procedures used in conducting the study which includes (1) rationale for the research design, (2) research questions, 89


(3) site and participan ts, (4) materials, (5) va riables, (6) data collection procedures, and (7) statistical and descript analyses used in analyzing the data. Rationale for a Mixed-Method Design When planning this study for the first time, I set out to employ only quantitative method. However, it was not until I read the participants several comments in their journals that I realized the unexpected phenomenon that most EFL secondary students have been through were worthy of exploration through quali tative methods. The qualitative findings were able to play more than just a complementary role in explaini ng and elucidating the qu antitative results. For instance, Linlof and Taylor (1998 ) said that qualitative data, inte rview data in particular, are often used to validate test hypothesis in the field. For that reason, I decided to adjust the research method by changing from purely quantitative de sign to a mixed-method design. The rationale lies in that, according to Seliger and Shohamy (1989), the complexity of L2 acquisition makes it almost impossible to investigate L2 learning from a single perspe ctive. For that reason, it has been suggested that researchers need to draw on multi-disciplinary knowledge to provide insights into the phenomena of L2 teaching and learning in order to consider appropriate methods and tools to explore different perspectives of L2 acquisition. Therefore, this study employs a mixedmethod design of integrating bot h approaches-quantitative and qualitative research methods, which complement each other to provide a much more detailed and comprehensive picture of which is being explored. The strongest advantag e of quantitative methods is its ability to obtain results that generalize to the population. However, quantitative can be limited as they tend to measure a limited number of outcomes and thus specific issues arent ab le to be probed more deeply (Drnyei, 2003). For these reasons, it is suggested that researchers apply qualitative methods to capture what statistical measures ma y not able to sufficiently explain, such as the reason of why and how some students succeed or fail in their learni ng (Wisker, 2001). 90


In the current study, I followed what Cres well (2003) calls a sequential explanatory model, a type of m ixed-method design in wh ich quantitative data collection would be undertaken prior to qualita tive data collection. In other words, the priority of this study was placed on the quantitative data (e.g., the reading achievement test and two questionnaires given to the whole sample), and the qualitative data (e.g., group interviews conducted with a subsample and classroom observations) were meant to expl ain and elucidate the quantitative data, thus deepening the understanding and interpretation of the results. An example of Creswells (2008) sequential explanatory model is pres ented in the following Figure 3-1. Figure 3-1. Mixed-method Research Process Research Questions For this study, the following research questions were addressed: 1. Does metacognitive reading strategy training affect high school EFL students metacognitive knowledge with re spect to their perceptions about reading and motivation to read English materials? 2. Does metacognitive reading strategy traini ng enhance high school EFL students reading comprehension? 3. Does the effectiveness of metacognitive st rategy training depend on previous level of general English language competence? 4. What are the factors contributing to EFL high school students reading experiences and reading attitudes? 91


5. In what way can the features of m etacogniti ve reading strategy inst ruction help current EFL English reading instruction? Subjects Subjects Selection The participants in this study were 118 11th grade students in a public high school located in Kaohsiung city, which is in southern Taiwan. The participants were selected randomly from a pool of approximately 400 students in 11th grade in that school. Am ong these students, sixty-six were male and fifty-two were female aged from 15-17. This high school was selected for this study based on the following criteria First, this school is neit her in the top 30%, nor in the bottom 30%, according to Joint High School Entrance Examination scores of 2007. For this reason, the student population is more reflectiv e of the general population. Second, this high school follows the general curriculum standard s mandated by the Ministry of Education (MOE) in Taiwan and the school is officially recognized by the MOE. Third, the students admitted into this high school are required to take the national standardized Basic Competence Test (BCTEST) and have to meet the admission standards. In that sense, the student s of this high school generally had at least low-to-intermediate English language proficiency level. Furthermore, entering this high school depends on students test scores de spite of their socio-economic backgrounds. The subjects nativ e language is Mandarin and all the subjects have received approximately 8 years of formal English instruct ion since grade 3. All the students study English as a mandatory subject of five class hours (50 minutes) per week. Participants background information was collected via the Personal Data Questionnaire (see Appendix A). Rationale for determining sample size Gall et al. (2003) suggested that it is important for a research to determine an appropriate sample size in order to detect the effect in an experiment. Besides, Ol ejnik (1984) states that 92


there are fou r factors commonly us ed to determine sample size: significant level, statistical power, analysis procedure, and effect size. Therefore, significance level alpha ( ) in this study is set at 0.05 with two tails while th e statistical power is .80, which is considered as a desirable standard power. Based on the effect size and power of past research to estimate the size of effect (See Table 3-1), I determined the samp le size of this study by using G*Power1 (Faul, Erdfelder, Lang, & Buchner, 2007), a general program used for conducting sample size, a priori and post hoc power analysis. In Table 3-1, both Raymonds (1993) and Co rdero-Ponces (2000) studies have very small effect sizes (0.067 and 0.102, co rrespondingly) and that would require a very large sample size to replicate. On the other hand, Carrell (1985) and Carrel, Phrais, and Liberto (1989) have large effect sizes (0.353 and 0.207 co rrespondingly) that wo uld require a small sample size. In this study, I used Songs (1998) study as the basis for sample calculation because it had a moderate effect size (0.32) which would require a moderate sample size. In other words, I use estimated effect size to calculate how many participants I need to detect the effect and to achieve the desired leve l of power. Table 3-2 shows the estimated effect size as well as sample size estimated for a power level of 0.8 in Songs (1998) study. From Table 3-2, Songs study had an effect si ze of 0.32 with a sample size of 50 which led the study to have a power level of approximately 0.49. For that reason and from G*Power program, I determined that at least a minimum of 45 subjects per group could allow me to detect a statistically significant main effect of metac ognitive reading strategies instruction (MRSI). In other words, at least 90 subjects shou ld be included in this study to lead to the power level of 80. 1 Source: Faul, F., Erdfelder, E., Lang, A.-G., & Buch ner, A. (2007). G*Power 3: A flexible statistical power analysis for the social, behavioral, and biomed ical sciences. Behavior Research Methods, 39, 175191. 93


Table 3-1. Power and Effect size calculat ed from previous experim ental studies Research Topic N, Mean, SD, df, and F test Effect Size Carrell, P.L. (1985) Facilitating ESL reading by teaching text structure, TESOL Quarterly 19, 727-757. N=25 (Exp=14; Con=11) df = 1 F (1,23)=14.63 5 hours training in 5 days Both Exp and Con had Preand post testing and Exp also has second post-test 3 weeks later. Omega square = 0.353 One-Way ANOVA Carrell, Pharis, & Liberto (1989) Metacognitive strategy training for ESL reading, TESOL quarterly, 23, 647-678. N=26 (Exp1=9 ; Exp2=9; Con1=3; Con2=5) F (2, 20)= 4.40 main effect on treatment on open-end posttest measure, F (1, 20)= 0.31main effect on learner style on open-end posttest measure F (2, 20) = 3.99 interaction b/w treatment and learning style 4 days training Both Exp and Con had Preand post testing with 9 days interval. Omega square = 0.207 Omega square = 0.011 Omega square = 0.234 2 way ANOVA Raymond, P.M.(1993), The effect of structure training on the recall of expository prose for university students reading as a second language, Modern Language Journal, 77, 445-458. N=43 (Exp=21; Con=22) df=1 F (1, 41)= 4.10 significant 5 hours training in two weeks Both Exp and Con had Preand post Omega square = 0.067 ANCOVA ( the correlation b/w pre and post was 0.58) Song, M. (1998), Teaching reading strategies in an ongoing EFL university reading classroom, Asian Journal of English Language teaching, 8, 41-54. N=50 (one group) F=24.60, df=1, p=.0001 significant on the effect of training F=4.53, df=2, p=0.015 significant on the interaction of training and language level 42 hours training in 14 weeks Pretest and post test 2-way ANOVA Omega square = 0.32 Omega square = 0.124 Cordero-Ponce, W.L.(2000) Summarization instruction on foreign language comprehension, Reading research and instruction, 39(4), 329-350 N=30 (Exp=15; Con=15) df =1 F (1, 28)=4.44 2 days training Both Exp and Con had Preand post and Exp had 3 week delay posttest One-way ANOVA Omega square = 0.102 Table 3-2. Sample size estimation for a power level of 0.8 with S ongs (1998) study with estimated effect size. Effect Effect Size (Omega square) Total sample size Main effect Interaction effect 0.32 0.124 80 130 94


Table 3-3. P ower level and sample size with an effect of 0.32 Effect Size Sample size Power 0.32 51 0.49 0.32 54 0.50 0.32 0.32 81 99 0.70 0.80 Classification of English reading proficiency In order to compare differential effects of metacognitive reading strategies on different English reading proficienc y levels, the subjects in this st udy were divided into three reading levels (high scorers, intermediate scorers, and low scorers) according to their pretest scores on the GEPT test. Participants who scored 60 or be low were designated or placed in the low-scorer group (25.4 % of the partic ipants), participants whose scores fell in the range of 60-70 were designated as intermediate-scorer group (42.4 % of the participants), and those who scored above 70 were designated as high-scorer gr oup (32.2 % of the participants). Table 3-4. Participants read ing proficiency levels Group N Mean SD High scorers 30 85.83 6.92 Intermediate scorers Low scorers 52 36 65.19 43 5.045 9.879 As shown in Table 3-4, the mean score of the high-scorers was higher by 20.64 points than those intermediate-scorers and the mean score of interm ediate-scorers was higher by 22.19 points than that of low-scorers. Materials Textbooks All students participating in the research project used the same textbook required for all 11th graders. The textbook is a co llection of short stories or ar ticles compiled from different 95


resources by Taiwanese authors. Each lesson is supplem ented w ith short reading comprehension questions (largely literal) a nd rather extensive vocabulary list containing L1 glosses and explanations, grammar highlights, and sentence patte rns and drill activities. The textbook is best described as a basal reader and is used exclus ively for intensive in-class reading exercise and vocabulary as well as grammar learning. The la st three lessons in the textbooks, which are slightly above the students current grade level and are not taught in a normal school semester due to the time constraint, were selected to us e in this study. In addi tion, similar grade level reading materials published by other textbook publishe rs were distributed to all the subject as outside class practice or take home assignments. Reading materials were the same regardless of the treatment or the comparison group. Reading strategy booklet A reading strategy booklet (Appendix D for the ex ample) was given to each subject in the treatment group and all the strategies listed inside were explained and modeled by the experimenter and practiced by the subjects in the treatment group thr oughout the experiment. They kept this booklet for future reference and we re expected to familiarize themselves with all of the strategies listed in the booklet. Reading strategy journal All of the subjects in the experimental group were required to keep a Reading Strategy Journal (See Appendix D for the example) while inside and outside of the class to record their reading strategy use during English reading. This j ournal lists specific questi ons to facilitate and guide them on how to record their reading expe rience and what strategies they use during the reading. The subjects were encouraged to use wh atever language they were comfortable with. This idea is borrowed from Andersons recommendation (2002), a way of developing students metacognition in their reading process and is designed to facilitate EFL students reading 96


strategy reflection. Developing m e tacognition during the reading pr ocess is important since L2 researchers have suggested that being exposed to reading strategies may not ensure success in language learning (Skehan, 1989), especially if the learners do not metacognitively connect their strategy knowledge and language use (Vann a nd Abraham, 1990). Thus, one potentially effective way of encouraging EFL high school st udents to aware and make a connection between reading strategy knowledge and read ing strategy usage is to have them to reflect on their own experiences while English reading, and hopefully th is reflection process will lead them to take more control over their own English reading and become independent readers. Measurements The measurements used to collect quantitative data for the dependent variables include the reading section of the General English Proficiency Test (GEPT, hereafter), the English Reading Motivation Questionnaire (ERMQ, hereafter), and the Survey of Reading Strategy Questionnaire (SORS, hereafter). All three measuremen ts were collected before and after the treatment. Reading comprehension test in the General English Proficiency Test (GEPT) To obtain a measure of the students base line knowledge in their English reading competency, the reading section in General English Proficiency Test (GEPT) was administered to each participant prior to the beginning of th e study. In 1990, the Ministry of Education (MOE) in Taiwan announced the need fo r the country to develop a relia ble and standardized test to evaluate the English proficiency of students and other populations. A quasi-official institution named the Language Training and Testing Center (LTTC), which was certified for administering TOEFL and other language tests in Taiwan, was appointed to design and administer the General English Proficiency Test (GEPT for short and hereafter). According to LTTC (2003), the GEPT focuses on its emphasis on four language skills with 30% of listening skill, 35% of reading skill, 97


18% of speaking skill, and 17% of writing skill. Read ing skill has the highe st percentage am ong these skills. The test itself is a criterion-referenc ed test and has the follo wing given levels: a) the beginning level, b) the intermediate level, c) the high intermediate level, d) the advanced level, and e) the superior level. According to the given levels, students who pass the GEPT beginning level have the English ability of understanding basic conversations about time and place, reading simple instruction on a daily basi s, and writing simple sentences Students who pass the GEPT intermediate level have the English ability of understanding general conversations, such as public announcements and weather reports, and ca n talk to English native speakers in simple English. They are also expected to reading short stories, private letters, and fliers, and can deal with their career needs in the use of English. For students in the high intermediate or low advance level, with the equivale nt scores of 550 on the TOEFL, it is not a problem for them to understand English public speeches and broadcast ne ws report. In addition, they are capable of reading English documents, meeting records, an d business reports, expr essing their viewpoints freely in English and conducting bu siness presentations in English are also expected. Major universities in Taiwan have re quired students to pass the inte rmediate level as one of the standards for admission, and the high intermedia te level as a requirement for graduation. The GEPT as a single standardized English profic iency test endorsed by the MOE in Taiwan has gained credibility and attention, and further guided the English instruction at schools at all levels. The reliability of GEPT reading secti on is 0.72 with N=100 (LLTC in Taiwan, 2003). In this study, five reading passages taken from intermediate level of GEPT preparation books were chosen (Appendix H) and examined by the experimenter to establish the content validity of the reading test. Then the test was administered to all the subjects prior to the experiment and at the end of the experiment. 98


The Survey of Reading Strategies (SO RS) This is a questionnaire used to test students metacognitive reading strategies awareness. The Survey of Reading Strategies2 (SORS) is adapted by Mokhtari and Sheorey (2002) and based on Mokhtari and Reichards (2002) Metacognitive-Awareness of Reading Strategies Inventory1 (MARSI), which was first developed for native speakers in the US. According to Mokhtari and Sheorey (2002), the SORS is intend ed to measure the type and frequency of reading strategies that adolescent and adult ESL students perceive th ey use while reading academic materials in English. (p.4). The SO RS was developed and pilot-tested by Mokhtari and Sheorey (2002) with 147 ESL students in the US and found consistent results relative to instruments overall reliability (with Cronbachs alpha = 0.89). Th is indicated a reasonable lev of consistency in measuring awareness and pe rceived use of reading strategies among ESL students. In total, there are 30 items in the SO RS with a Likert scale to measure three broad categories of reading strategies: a) global read ing strategies (13 items), b) problem solving strategies (8 items), and c) support strategies (9 items). According to Mokhtari and Sheo (2002), global reading strategies (GLOB for short) are defined as those intentional, carefully planned techniques by which learners monitor or manage their reading, such as having a purpos in mind, previewing the test as to its length and organization, or using typographical aids and tables and figures. (p.4). Problem solving strate gies (PROB for short) are classified as the actions and procedures that readers used when problems develop in understanding textual information; examples includes adjusting ones speed of reading when the materials becomes difficult or easy, guessing the meaning of unknown words, and rereading the text to improv the el rey e e 2 Mokhtari, K. & Sheorey, R. (2002). Measuring ESL Students' Awareness of Reading Strategies. Journal of Developmental Education 25, no. 3, 2-10. 1 Mokhtari, K., & Reichard, C. A. (2002). Assessing st udents metacognitive awareness of reading strategies. Journal of Educational Psychology 94, 249-259. 99

PAGE 100

com prehension. (p.4). Support strategies (SUP for short ) are classified as basic support mechanisms intended to aid the reader in comp rehending the text such as using a dictionary, taking notes, underlining, or hi ghlighting textual information. (p.4). The SORS used in this study was translated into Chinese and each statem ent in the SORS has both English and Ch versions (see Appendix F). Subjects in this study filled out this questionnaire before and a training in order to measure the possible differences or changes in their perceived strategy use. Reliability of the SORS was analyzed and test ed its internal consistency as measured by Cronbachs alpha and reported in Chap inese fter the ter four of this study. The English Reading Motivation Questionnaire (ERMQ) The English Reading Motivation Questionnaire4 (ERMQ) was developed by Mori (2002), with the intention to measure EF L learners English reading motiva tion. In order to investigate EFL Japanese students reading motivation, Moris adapted ERMQ from Wigfield and Guthries (1995, 1997) Motivation to Reading Questionnaires (MRQ) to better fit the environment of English as a foreign language setting. According to Mori s study, foreign language reading motivation closely resembles more general forms of motivation as laid out in expectancy-value theory as proposed by Day and Bamford (1998, cited in Mori, 2002). Expectancy-value theory of reading motivation model cons ists of four reading component s which may influence language learners decision to read in a second or fo reign language. These four reading components include reading materials, reading ability, reading attitudes, and sociocultural environment. This is a 30-items 4-point Likert scale questionnaire. Many of the questionnaire items were written referring to the theory of reading motivati on proposed by Wigfield and Guthrie (1995, 1997). Although Wigfield and Guthrie identified and included 11 components in their MRQ, three 4 Mori, S. (2002). Redefining motivation to read in a foreign language. Reading in a Foreign Language, 14(2). Retrieved October 30, 2002, from 100

PAGE 101

com ponents, competition in reading, reading reco gnition, and social reasons for reading were removed by Mori as these didnt seem to relevant to the participants in a foreign language context. This irrelevance is due to the fact that Wigfield and Guthries motivational scales were specifically developed for primary school students learning to read in their L1 and thus some items appearing in the MRQ were not considered directly applicable to EFL students based on the results of Moris studies (2000, 2001, and 2002). The internal cons istency estimate of reliability for this questionnaire was .93 (N=447) which demonstrated that the questionnaire has high reliability value. Same as the SORS in this study, the ERMQ was translated into Chinese and each statement has both English and Chinese versions (see Appendix G). There are four subscales in th e ERMQ: a) intrinsic value of reading, b) extrinsic utility value of reading, c) importance of reading, and d) reading efficacy. The response format was 1 stands for very disagree, 2 stands for disagree, 3 stands for agree, and 4 stands for very agree. Intrinsic value of read ing refers to reading curiosit y, reading involvement, reading avoidance, and reading challenge. This subscale can reveal de grees of students interest in reading in English or their per ception of enjoyment involved in reading in English. Extrinsic utility value of reading refers to motivation to engage in a task in or der to obtain external rewards such as good grades. Importance of readi ng refers to a students perceived usefulness of reading. And reading efficacy refers to an indivi duals sense of efficacy and beliefs about their ability in terms of reading in English. The ER MQ was administered two times in the study. The pre-ERMQ test was conducted at the begi nning of the study, while the post-ERMQ was conducted after the implementation of metacognitiv e reading strategy training. The purpose was to explore probable changes in the participants motivation toward English reading. In terms of 101

PAGE 102

reliability of ERMQ, data collected from pretreatment and after treatment were analyzed and tested its internal consistency as measured by Cronbachs alpha and reported in Chapter four. Establishing Translation Authenticity for SORS and ERMQ I translated the SORS and the ERMQ into Chinese, the subjects native language (See Appendix F and G) in order to ensure the subjects understanding of the items on the questionnaires. The translations of these two questionnaires were confirmed by several bilinguals at UF. To confirm that the Chinese version of the questionnaires elicited the same information as the English versi on, I sent both the initial English and Chinese versions separately to native speakers of Mandarin who are pursuing doctoral degrees in the Co llege of Education at UF (n=2), and Taiwanese professo rs who graduated from universities in the US (n=2). They were asked to complete each questionnaire as if th ey were the participants in the study. The two administrations were two week apart and the period was short so that the verification group didnt have a long period of time between the tw o administrations which might result in changes in their perceptions about the que stionnaire. However, the period was not so short that it would allow them to remember their re sponses to the previous version of the questionnaire. Responses to the two versions of the quest ionnaires were analyzed to check their compatibility with each other. The results came out c onsistent and thus satisfied. Research Design The design of the study has been structured to adapt strengths of previous research and attempt to overcome weakness believed to be important in extant reading research. First, the composition of metacognitive reading strategies instruction extend the work of Brown and Palincsars (1984) reciprocal teaching along with the use of Reading Strategy Journals consciousness raising, and self-reflection on the r eading task. Second, most of the previous reading strategy training studies either use small sample size, quasi-experiment design, or 102

PAGE 103

without com parison group, this study used a tr ue experimental equi valence control group pre/post test design as shown in the following tabl e. In addition, in order to eliminate all the threats to internal validity, this study used a random assignment to equalize the comparison groups. The steps to implement this st udy were as the following Table 3-5. Table 3-5. Steps for the implementation of the study Screening (pretreatment Aug,31st, 2007) Group Pre-test TreatmentPosttest (Nov, 17th, 2007) Post treatment GEPT reading section A Experimental O1 X1 MRSI O2 Interviews with 24 students GEPT reading section B Control O1 X2 Normal Class O2 The treatments included metacognitive readi ng strategy instruction (X 1) and the teachers normal routine of instruction (X2). The metacognitive reading strategy instruction (MRSI hereafter) served as the experi mental treatment (X1), while th e teachers normal routine of instruction served as the comparison treatment (X2). The pretest (O1) consists of the SORS questionnaire, the ERMQ questi onnaire, and reading comprehens ion test. These assessments were conducted during the week prior to initiation of the metacognitive reading strategy instruction during the fall semester of 2007. The experimental group (Group A) received metacognitive reading strate gy instruction while the cont rol group (Group B) followed a traditional English reading instruction routine without metacognitive reading awareness training. The content in each class is equivalent and was de livered by the two instructors. In addition, the measurements were the same for each group. The posttest (O2) occurred at the end of the study, 10 weeks later. It consists of the same measurements as O1. Prior to the administration of the experime nt, I obtained permission from the school (See Appendix J), the parents (parents consent form See Appendix B), and the subjects themselves 103

PAGE 104

(See Appendix A). The data collection took plac e during the fall sem ester in the year of 2007 starting in the last of August and ending in November 13th. In this study, there were 59 subjects in the control group and 59 subjects in the experi mental group. All of the subjects were informed that they came to English Reading class, which was in addition to their regular English classes. They were called out for the treatment during the recession time from 12:30 pm to 1:20 pm on Monday and Wednesday for the experimental gro up, and Tuesday and Thursday for the control group, respectively. The duration of the experi ment was two lessons per week for each group over ten weeks. As previously stated, th e subjects were randomly selected from 11th grade with approximately 400 students and randomly assigned to each group. For the experimental group, subjects were explicitly taught Englis h reading strategies as listed on the Reading Strategy Booklet (see Appendix D) and they were required to keep a Reading Strategy Journal (see Appendix D) to self-reflect on th eir reading process, reading experience, and monitor their English reading strategies usage. For the contro l group, on the contrast, they received all of the same reading materials as the experimental gr oup and they were also taught English reading. The only difference between these two groups lies in the level of consciousness-raising in terms of reading strategies usage, which was stressed in experimental group. Instructors In order to eliminate the pitf alls of an experimental study, two instructors were included in this study. As Barber (1973) pointed out, research studies would be less biased if the investigator who plans the study and who has an investment in the outcome is not the same person who has responsibility for the data anal ysis. (p.400). For that reason, one English teacher from that high school was recruited to pa rticipate in the study due to her interest and willingness to try a different approach to English teaching and thus she was the instructor for the experimental group. This teacher, Ms. Lin, holds a bachelors degree in English education and a 104

PAGE 105

m asters degree in American and English literature from Kaohsiung Normal University in Taiwan. She has been teaching English in the high school for over 12 years including 2 years in middle school. Based on her experience and bac kgrounds, it was determined that she possesses sufficient knowledge and ability to teach reading stra tegies to the experimental group. The control group was taught by the researcher. Treatment delivery Accountability In order to ensure that treatment was de livered appropriately and accurately to both groups, both instructors sat in each others class and discussed before a nd after the classes. I prepared lesson plans and handouts outlining how to implement metacognitive reading strategy instruction for her. Threats to the study According to Seliger and Shohamy (1989), the influence of extraneous variables was less powerful in EFL than in ESL settings, due to the amount of language input. For that reason, the design of this study had eliminated the possibil ity of any variables to the lowest degree. Moreover, as Gall et al. ( 2003) stated that the main threat to in ternal validity is the possibility of group differences on posttests due to preexisting differences in the groups, rather than the effects of the treatment. Therefore, in order to control for this threat, they recommend analyzing covariance. In order to control for preexisting student conditions, pr etest of reading test in GEPT was treated as covariates th roughout ANCOVA procedure. Procedure The steps to implement this study are as following: 1. 118 students were randomly selected and assigned to two groups one experimental group (59) and one control group (59). 2. Before the treatment, all subject were ad ministered the SORS questionnaire, ERMQ questionnaire, and reading comprehension test. 105

PAGE 106

3. The experim ental group received metacognitive reading strategy awareness training along with reading strategy journal keeping. R eading strategy instruction included teacher modeling (what to apply, how to apply, and why to apply) and explicit explanation of the strategies, and scaffolding of student learni ng the strategies. The subjects practice the strategy taught, guided practice using the strategy with grad ual release of responsibility, and independent use of the strategy. 4. The core of the training consisted of practicing basic reading strategies, such as 1) finding the main idea of a paragraph, recognizing to pic sentences, distinguishing the main idea from supporting details; 2) cl arifying, such as concentrating on key words and guessing their meaning from the context; 3) summarization, such as key people, key place, key information or key ideas; use semantic map to visualizing; 4) pred icting, such as based on what is already known and how it related to what might happen next; and 5) questioning, such as what was the main idea, what was happening, what would you do if?. In addition, the training also contai ned interactive group activities that invited the subjects to observe their own reading pr ocess and that of their p eers, including observing and discussing the strategies they apply to understa nd. In order to enc ourage the students to apply the strategies presented during the training in their independent reading, after each lesson, the subjects were given reading assignments to work on at home and record their reading in the read ing strategy journal. 5. The control group followed the normal English teaching routine on which grammar drills, vocabulary, and translation were stressed. As the subjects all came out for English reading class, the control group were also taught reading skills used for comprehension but group sharing, teacher modeling, and reading strategy reflection were not emphasized. 6. The experiment lasted for ten weeks (r oughly two and half-months) with Monday and Wednesday for the experimental group and Tu esday and Thursday for the control group. All subjects still have their regular English classes (five hour s per week) with their regular English teachers. 7. After the treatment, all subjects took the postt est with the same quest ionnaires (the SORS and the ERMQ) and the read ing comprehension test. 8. In order to attain qua litative data from the EFL student s overall reaction toward this metacognitive reading strategy instruction, 24 students from the experimental group were called out for group interview in the nap time between 12:30 to 13:10 pm. Interview questions mainly focuses on their previous re ading experience and how do they feel about this English reading class, that is the core instruction of this study. Variables for Quantitative Data The following section lists and describes the me aning of several variables used for data analysis. 106

PAGE 107

Independent variables In this study, the independent variables (nominal variables) are the teaching methods the metacognitive reading strategy instruction (MRSI) and the participants English reading proficiency levels. The particip ants pre-GEPT scores were used to classify their English reading competence (into high, intermediate, and low proficiency groups) and as a covariate in statistical analysis. Dependent variables: The dependent variables (continuous variables) used in the study were (a) post-test of metacognitive reading strategy ques tionnaire; (2) post-test of read ing comprehension test; and (3) posttest of English reading motivation. All of the independent variable s and dependent variables were used to test the hypotheses from H to H 1 a4 aQuantitative Data Analysis Overall, the independent variables in this study are metacognitive reading strategy instruction and subjects reading proficiency levels as m easured by pretest. The dependent variables in this study are the scores of the postSORS, the postERMQ, and the post reading comprehension test scores. ANCOVA was used to test hypotheses from H01 to H with the pretes t score as the covariate and the post score of the SORS, the ERMQ, and reading comprehension score as the dependent variables. ANCOVA was considered to be best for this analysis based on the following reasons. Firs t, ANCOVA would adjust the post-test means on the basis of the pre-test means, and then compar e these adjusted means to see if they differed significantly. Second, as Johnson (2001) stated, in experiment al studies that use ANCOVA, causality can be inferred when significance occurs. For instance, the independent variable can be 04 107

PAGE 108

said to significantly influence or affect the de pendent variable in som e, unless there are some uncontrolled extraneous variables. Third, the use of the pre-test reading comprehension score as a covariate also provides a more cons ervative statistical analysis. And at last, in order to see if it is appropriate to allow the inclusion of the c ovariate data, the correlation between the pre-test reading comprehension score and the post-test reading comprehe nsion score was calculated in this study. All analysis was performed using SPSS statistic package for Window 12.0. Table 36 summarizes the measurements and data analysis for four research hypotheses. Table 3-6. Overview of Res earch Hypothesis and Methodology Research Hypothesis Measurement Data Analysis 1 aH Reading comprehension will be greater for high school students with metacognitive reading strategies training than those without metacognitive reading strategies training. Pretest reading Comprehension and posttest of reading comprehension ANCOVA 2 aH Motivation to reading English will be greater for high school students with metacognitive reading strategies training than those without metacognitive reading strategies training. Pretest of ERMQ and posttest of ERMQ ANCOVA 3 aH Perception about reading strategies will be greater for high school students with metacognitive reading strategies training than those without metacognitive reading strategies training. Pretest of SORS and posttest of SORS ANCOVA 4aH The effect of metacognitive strategy training will be larger for students with lower language competence. Pre and Posttest of reading comprehension, SORS and ERMQ ANCOVA Qualitative Data Collection Participants in interview Nunan (1992) points out that qualitative inform ation is often crucial for the interpretation of quantitative data. For that reason, qualitative data was coll ected through face-to-face group interviews conducted by the res earch. 24 students participated in semi-structured open-ended interview (Patton, 1990, p. 284) conducted by the rese arch on group of four basis in a casual and relaxed atmosphere. The interview questions (see Appendix I) were wr itten out in advance 108

PAGE 109

exactly the way they were to be asked during th e interview an d each group was asked essentially the same questions in the same order. Accord ing to Patton (1990), there are several advantages in favor of using open-ended interviews. For ex ample, this approach can reduce the possibility of bias that results from havi ng different interviews for differe nt participants, including the problem of collecting more information from some people than from others. Meanwhile, asking the same questions of each participant can also ensure that data is complete for each person on the topics addressed in the interview. More over, the interview is highly focused so that interviewee time is carefully use d. Therefore, this method is es pecially appropriate when the research wants to obtain the same information fr om each participant within a limited period of time. Finally, the data analysis also becomes easier because the researcher can locate each respondents answer to a partic ular question rather quickly. However, the standardized openended interview is not without w eaknesses. First of all, it do es not allow the interviewer to pursue topics or issues that were not anticipat ed when the interview was written. Second, the researcher is not able to use different lines of questioning with different people based on their unique experiences. In other words, this ap proach does not take into account individual differences or situational changes (Ethrman, L eaver, Oxford, 2003). Therefore, to compensate for these weaknesses, I preserved some flexibil ity to ask probing questions whenever it was necessary to explore certain subj ects in greater depth or to purs ue new topics that were not included in the original interview instrument. Themes were determined and reported in a descriptive format. Interview and data analysis In order to understand in depth the students experience s of the implementation of metacognitive reading strategy inst ruction in English reading class, a face-to-face standardized open-ended interview with the students was conduc ted as a further data collection method after 109

PAGE 110

the experiment was completed. In conducting the interviews, I m yself served as the interviewer. The language used in the interview was students first language, Mandarin. The interviews were scheduled in their recession time from 12:30 pm to 1:10 pm and were conducted in the teachers restroom beside the teachers office. The cons ent of each participating student was obtained. The students came in groups of four and there were 24 students participating in the interview. In advance of each interview, I explained the purpose of the study and an interview question list was provided. Each interview was audio taped and transcribed into English for further analysis (Creswell, 1998). All names have been changed to protect the identity of the teacher and her students. Data were analyzed using a version of the ground theory approach, in which multiple themes emerged through repeated reading of the data with the intention to explain the advantages for the quantitative data. Summary of Methodology In summary, this chapter provided a deta iled description of the methodology employed and the data collection in the present study adopted both quant itative and qualitative methods in order to explore EFL secondary participants ex perience with the metacognitive reading strategy instruction. The rationale of choosing a mixe d-method design was given. The quantitative methods used were one reading comprehension test and two questionnaires. These investigated the effect of metacognitive reading strategy instruction on EFL secondary students English reading comprehension, reading st rategies awareness, and Englis h reading motivation. With respect to the qualitati ve data collection, the participants reading strategy journals were collected and analyzed and the participating teacher as well as 24 participants were group interviewed regarding their ex periences toward English read ing. Participants were group interviewed regarding their prev ious reading experience and their experience toward this 110

PAGE 111

experim ent. The results and findings will be reporte d in Chapter four for quantitative finings and Chapter five for qualitative findings respectively. 111

PAGE 112

CHAP TER FOUR QUANTITATIVE RESULTS Introduction The purpose of this chapter is to provide a presentation of statistics used to examine the independent variables and depende nt variables related to the null hypotheses. The independent variables in this study were metacognitive reading strategy instruction versus the regular English reading instruction. The difference between these two instructions lies in that the former one emphasizes the importance of students autonomy in terms of their Englis h reading process when reading comprehension breaks down and monitoring th e effectiveness of thei r reading strategies uses. The dependent variables in this study we re the post-English reading comprehension test, the postEnglish Reading Motiva tion Questionnaire (ERMQ) and the postSurvey of Reading Strategy (SORS) In order to present the quantitative results of these analyses in a clear and coherent manner, the discussion of the statistical re sults are organized as the following order: (1) Statistical techniques us ed in this study; (2) The descriptive statistics of the sample in this study; (3) The reliability of the SORS, the ERMQ, and th e reading comprehension test in this study; (4) The assumptions of ANCOVA is also addressed and examined; (5) Results for null hypotheses H01 of this study; (6) Results for null hypotheses H02of this study; (7) Results for null hypotheses H03of this study; (8) Results for null hypotheses H04of this study; 112

PAGE 113

Statistical Techniques The purpose of this study was to examine th e effects of metacognitive reading strategy instruction on EFL secondary students English reading comprehension, English reading strategy awareness, and motivation to E nglish reading by using an experimental design. Nonetheless, random assignment to the experimental and c ontrol groups may not guara ntee perfect linguistic balancing groups; therefore, analysis of covariate (ANCOVA) was used to control for differences in reading comprehens ion measured prior to the treatment. Data pertaining to the independent and dependent variab les were analyzed using the Statistical Package for the Social Science (SPSS) version. From hypotheses H through H a series of ANC OVA examined the main effect of teaching approach (matacognitiv e reading strategy instruction), as well as interaction effect between Eng lish reading proficient levels and treatment. For hypothesis 1, ANCOVA was performed to determine if ther e was any significant difference on the post reading comprehension test when the covariate, GEPT score, wa s held constant. For hypothesis 2, ANCOVA was performed to determine if there was any significant difference on the postSurvey of Reading Strategies (SORS) when GEPT score was held constant. For hypothesis 3, ANCOVA was performed to determine if thes e was any significant difference on the postEnglish reading motivation questionnaire (ERMQ) when GEPT score was held constant. For hypothesis 4, ANCOVA was pe rformed to determine if there wa s an interaction between reading proficiency levels and treatm ent on dependent variables. 01 04Descriptiv e Statistics of Participants Originally there were 59 students in each gr oup (see Table 4-1). However, five subjects (case number 20, 25, 61, 65, and 118) failed to provide complete information on either post reading comprehension test or th e two post questionnaires. In other words, the data set was 5 113

PAGE 114

observations short. Therefore, these five cas es w ere deleted based on listwise deletion (Allison, 2001) which omits cases that do not have da ta on all variables in the variable list of the analysis. In this data set, 5 subjects (4%) failed to provide completed information, leaving a total sample size of N= 113 subjects from the data were collected. The contro l group lost 2 subjects (N=56, 3%) while the experimental group lost 3 subjects (N=57, 5%). Based on the small percentage of missing data, it is decided that deletion of these 5 cases was appropriate. Summary of the sample size before and after the experiment were shown in Table 4-1. Table 4-1. Sample size before and after the treatment N before experiment N left for analysis Experimental Group 59 (two incomplete data and one outlier) 56 Control Group 59 (two incomp lete data and two outliers) 54 Total 118 110 Note: case number 20, 25, 61, 65, & 118 were deleted due to incompletion Outliers deletion Furthermore, three outliers were detected in the data, case number 25 and 57 in the control group and case number 73 in the treatment group as presented in Table 4-2 below. The outlier was determined to be deleted based on their previous performance on high school entrance exam score in English, English midterm ex am score, and GEPT score in reading section. If the subjects post-reading comprehension test was far below their previous English achievements such as entrance exam score, Englis h mid-term exam score, and GEPT score, then his or her score on post reading co mprehension test was considered as an outlier. This deletion resulted in unequal group sizes of n = 54 for th e control group and n = 56 for the experimental group. 114

PAGE 115

Table 4-2. Deleted outliers Case # Group Gender EESE ETE GEPT Post-RC 25 0 (Control) Male 67 50 59 10 57 73 0 (Control) 1 (treatment) Male Male 49 60 80 50 65 55 10 0 Note: C= control group; T=treatment group; EESE= hi gh school entrance exam score in English; ETE=English mid-term exam; GEPT=general E nglish Proficiency Test; PostRC=post reading comprehension test Gender distribution Of the 110 participants in the study, 61 ( 55.5 %) were male, and 49 (44.5%) were female (see Table 4-3). The control group had a lowe r percentage of male students (51.9%) than experimental group (58.9%), and a higher per centage of female st udents (48.1%) than experimental group (41.1%). Table 4-3. Gender distribution in each group Control Group Experimental Group Total Gender n % n % n % Male 28 51.9 33 58.9 61 55.5 Female 26 48.1 23 41.1 49 44.5 Total 54 100 56 100 110 100 English reading proficiency The mean score of the GEPT for the total sample is 70.88 with standard deviation of 11.55. Based on this mean score, the criterion for determining the reading proficiency level is: mean score (70.88) +/one standard deviation (11.55). Therefore, t hose who scored over 82 were classified as ranking 3 or high, the score fell between 60 and 82 were classified as ranking 2 or intermediate, while those sc ores fell below 59 were classifi ed as 1 or low. Twenty-two students (20%) scored below 59, 67 students ( 60.9%) students scored between 60-82, and 21 students (19.1%) scored above 83. In the control group, 9 students (16.7%) score below 59, 33 students (61.1%) students scored within 60 to 82, and 12 students (22.2%) scored above 83. In the treatment group, 13 students ( 23.2%) scored below 59, 34 stude nts (60.7%) students scored 115

PAGE 116

between 60 and 82, and 9 students (16.1%) scored above 83. See Table 4-4 for English reading proficiency levels distribution. Table 4-4. S ubjects GEPT level distribution Control Group Experimental Group Total GEPT score ranking n % n % n % 1 or low (below 59) 9 16.7 13 23.2 22 20 2 or intermediate (60-82) 33 61.1 34 60.7 67 60.9 3 or high (above 83) 12 22.2 9 16.1 21 19.1 Total 54 100 56 100 111 100 Note : GEPT = general English proficiency test The pre-Survey of Reading Strategies (SORS) Students reading strategies awareness was assessed thr ough their responses on the Survey of Reading Strategies (Mokhtari and Sheorey, 2002) before and after the treatment. As noted in Table 4-5, the mean score of pretest fo r the SORS for the entire sample was 3.32 with a standard deviation of 0.51. Overall, the particip ants reading strategy aw areness on the pre-test ranged from 1.41 to 4.13 with a mean score of 3.32. For the control group, overall students reading strategy awareness had a mean of 3.316 on the pre-test. For the experimental group, students reading strategy awar eness as measured by the SORS had a mean of 3.32 on the pretest. Independent t-test t (108) = 0.5, p > .05 yielded non-sign ificant mean difference between the experimental group and control group. This indicates that the two groups reading strategies awareness were statistically similar before the tr eatment. In other words, both groups had similar reading strategy usage before the experiment. In responding to the frequency use of reading strategies, Mokhtari and Sh eoreys (2002) outlined a scale range of 1-5: High strategy use = 3.5 and above Medium strategy use = 2.5 to 3.4 Low strategy use = 2.4 or below 116

PAGE 117

Therefore, according to the re sults, the participants in th is study had a m ean (3.3) of overall reading strategy use on the 5-point Likert s cale. In general, both groups were medium reading strategy users. Table 4-5. Subjects pre-SORS distribution Score Control Group Experimental Group Total SORS n % n % n % 2.5 or lower 6 11.1 3 5.4 9 8.2 2.5-3.4 24 44.4 30 53.6 54 49.1 3.5 or higher 24 44.4 23 41.1 47 42.7 Total 54 100 56 100 110 100 Note : SORS=the Survey of Reading Strategies As there are three subtests in 30 reading strate gies items in the SORS, it is also of interest to know which were the most frequently used an d the least frequently reading strategy used by EFL secondary students before the treatment. The results are shown in the following Table 4-6. As shown in the Table 4-6, the most intere sting finding here is that EFL secondary students use note-taking, paraphrase, thinking and translate from their native language most of the time. All of these strategies fell at the high-use range. This finding reflected that the traditional English teaching method of Grammar Translation Approach is still dominant in Taiwans most English classrooms and teacher-cen tered English language education. Likewise, I also identified four least freque ntly used reading strategies. A ll of these least frequently used strategies fell toward the bottom of low-use ra nge. The least frequently used strategies are: critically thinking, questioning, a nd read aloud strategies. This finding was not much a surprise since EFL language learners st ill follows traditio nal rote memorizati on pattern, the entire education system does not really promote initia tive-taking, critical thin king, self-direction, or self-regulation on the part of stude nts in learning as a whole. Speaking ability is not valued in normal English education as well. For that re ason, questioning, critical thinking and read aloud 117

PAGE 118

reading strategies are appeared as the leas t used reading strategies am ong EFL high school students. Table 4-6. The most and the least frequently used reading strategy from the pretest data The most frequently used readi ng strategy by EFL secondary students Strategy no. Strategy MeanStrategy Category 2 I take note while reading to help me understand what I read 4 Support reading strategy 18 I paraphrase (restat e ideas in my own words) to better unde rstand what I read 4 Support reading strategy 20 I use typographocial features like bold face and italics to identify key information 3.9 Global reading strategy 29 When reading, I translate from English into my native language (Chinese) 3.96 Supporting reading strategy 30 When reading I think about information in both English and my mother tongue (Chinese) 3.96 Supporting reading strategy The least frequently used readi ng strategy by EFL secondary students 21 I critically analyze and evaluate the information presented in the text. 2.21 Global reading strategy 5 When text becomes difficult. I read aloud to help me understand what I read 2.11 Support reading strategy 6 26 I think about whether the content of the text fits my reading purpose I ask myself questions I like to have answered in the text 2.4 1.94 Global reading strategy Support reading strategy Pre-English Reading Motivation Questionnaire (ERMQ) Participants English reading motivation was assessed through Moris English Reading Motivation Questionnaire (2002). The mean score for the pre-ERMQ was 78.84 with standard deviation of 10.84. The score in the pre-ERMQ questionnaire range from 48 to 101. For the control group, EFL students motivation to read in English had an overall mean score of 79.02 on the pre-test. For the experimental group, student s motivation to reading English had an overall mean score of 78.66. Even though control group has higher score than the experimental group, the independent t-test t (108) = 0.175, p >.05 yielded non-significant. Therefore, these two groups were statistically similar on ER MQ score before the treatment. 118

PAGE 119

Table 4-7. P articipants pre-ERMQ distribution Score Control Group Experimental Group Total ERMQ n % n % n % 72 or lower 13 24.1 12 21.4 25 22.7 73-90 33 61.1 35 62.5 68 61.8 90 or higher 8 14.8 9 16.1 17 15.5 Total 54 100 56 100 110 100 Note: ERMQ=English Reading Motivation Questionnaire There are four subtests in ERMQ questionn aire. Among thirty reading motivation items, it is also worthy of looking at th e most agreed and the least agr eed reading motivation answered by the participants in this study. Results are summarized in the following Table 4-8. As shown in Table 4-8, it is interesting to find that most participants agree that English is really important in their future job or further e ducation. However, most participants didnt have confidence in their English reading ability and th ey dislike long and challenged reading text and thus lack of intrinsic motivation to read English. In other words, the participants in this study all realize its importance in English reading and they seemed to be aware of their English reading problems. Qualitative data in chapter five will explain further in detail regarding the participants English reading motivation. Table 4-8. Participants most and le ast agreement on pre-ERMQ questionnaire The most agreed reading motiv ation by EFL secondary students no. Statement Mean Statement Category 10 I would like to get a job that uses what I studied in English reading class. 3.3 Extrinsic reading motivation 3 Learning to read in Engl ish is important in that we need to cope with internationalization 3.6 Importance of English reading 18 Learning to read in English is important because it will be conductive to my general education 3.3 Importance of English reading The most disagreed reading motivation answered by EFL secondary students 22 I enjoy the challenge of difficult English passage 1.83 Intrinsic reading motivation 17 English reading is my best subject 1.9 Reading efficacy 8 11 Long and difficult English reading passages interest me I am good at reading in English 2.1 2.02 Intrinsic reading motivation Intrinsic reading motivation 119

PAGE 120

Descriptiv e summary of pretest data Descriptive summary of the whole sample da ta is listed in the following Table 4-9 and descriptive summary of each group is listed in the Table 4-10. All scores of GEPT test, the SORS pretest, and the ERMQ pretest were ra w scores. As sample were random selected, independent-samples t -test were conducted to confirm a ny preexisting difference between these two groups in terms of their English reading pr oficiency level, readi ng strategies uses as measured by the Survey of Reading Strategies (SORS) and their English reading motivation as measured by the English Reading Motivati on Questionnaire (ERMQ) The use of independent sample t-test requires the test of the homogene ity of variance; therefore, Levenes test was performed to test whether the assumption of hom ogeneity of variance was met. The results of Levenes tests were not significant at p > .05 indicating that the diffe rence between variance is zero in each pretest (see Table 4-11 for Levenes test). In other words, the pretest had significantly equal variance in th e experimental group and the cont rol group. Consequently, the assumption of homogeneity of variance was met and the use of independent t -test is appropriate. Table 4-10 contains group means fo r categorical variables (e.g. nu mber of subjects) as well as numeric variables (e.g. GEPT scor e; pre-SORS score, and preERMQ score) for the treatment group and control group. Results of independent t-test on three pretests were show n in the above Table 4-12. For GEPT test, t (108) = .368, p >.05, which was not significant; for pre-SORS test, t (108) = -.05, p > .05, which was not significant, and for pre-ERMQ test, t (108) = 0.175, p > .05, was not significant as well. Overall, although the mean score of GEPT and pre-SORS of the control group were slightly higher than the treatment group, the differences between these two groups were not significant. In summary, from the pr etest data showed above, there is no significant difference among groups by their English reading proficiency, pre-SORS test, and preERMQ 120

PAGE 121

test. In other words, these two groups were sim ilar in te rms of their E nglish reading proficiency level, their reading strategy uses, and their English reading motivation before the experiment was conducted. Table 4-9. Descriptive summary of the pr etest score of the sample as a whole N Mean Sd. GEPT 110 70.88 11.55 Pre-SORS Pre-ERMQ 110 110 3.3 78.8 0.51 10.68 Note: GEPT=general English proficiency test; pre-SORS=pretest score on Survey of Reading Strategies; pre-ERMQ=pretest score on Eng lish Reading Motivati on Questionnaire. Table 4-10. Descriptive summary of the pretest score on each group E or C N Mean Sd. control 54 71.30 11.74 GEPT experimental 56 70.48 11.44 control 54 3.316 0.517 Pre-SORS experimental 56 3.320 0.520 Pre-ERMQ control experimental 54 56 79.02 78.66 11.08 10.37 Note: GEPT= general English profic iency test; pre-SORS=pretest score on Survey of Reading Strategies; pre-ERMQ=pretest score on English Reading Motivation Questionnaire; Sd.= standard deviation. Table 4-11. Result of homogeneity of variance test between groups in terms of GEPT, pre-SORS, and pre-ERMQ Levenes Test for Equality of variance F Sig. GEPT .000 .933 Pre-SORS Pre-ERMQ .665 1.152 .417 .268 Note : GEPT= general English profic iency test; pre-SORS= pretest score on Survey of Reading Strategies; pre-ERMQ=pretest score on Eng lish Reading Motivati on Questionnaire. Table 4-12. Independent t -tests on difference between groups on GEPT, pre-SORS, and preERMQ t df Sig. (2-tailed) Mean difference GEPT .368 108 .713 .814 Pre-SORS Pre-ERMQ -.05 0.175 108 108 .960 .861 -.005 .358 Note: GEPT=general English proficiency test; pre-SORS=pretest score on Survey of Reading Strategies; pre-ERMQ=pretest score on E nglish Reading Motiva tion Questionnaire. p<.05 121

PAGE 122

Descriptiv e summary of the posttest data Overall, participants reading comprehensi on posttest scores ranged from 40 to 100 with a mean score of 73.3. For the co ntrol group, the mean score for pos t reading comprehension test is 69.81, while it is 76.79 for the experimental group. The participants read ing strategies scores on the post-test ranged from 1.8 to 4.73 with a m ean score of 3.45. For the experimental group, participants reading strategies awareness had an overall mean score of 3.54 on the post-test. For the control group, participants re ading strategies awareness had an overall mean score of 3.36 on the post-test. The participants English reading motivation sc ore range from 48 to 117. The mean score for the control group on post-ERMQ is 82.28, while the mean score for experimental group on post-ERMQ is 87.39. Table 4-13 below listed the summary of the posttest scores on each dependent variable of each group. In general, the experimental group outperformed the control group on three dependent variables. Statistical significance will run for anal ysis later in this chapter. Table 4-13. Mean and standard de viation of three posttest data T or C N Mean Sd. Mean difference control 54 69.81 14.21 Post RC treatment 56 76.79 12.95 6.98 control 54 3.36 0.53 Post-SORS treatment 56 3.54 0.65 0.18 Post-ERMQ control treatment 54 56 82.28 87.39 11.3 13.3 5.11 Note: C= control group; T=treatment group; EESE= hi gh school entrance exam score in English; ETE=English mid-term exam; GEPT=general E nglish Proficiency Test; PostRC=post reading comprehension test. Reliability of Scores of the Measurements As discussed in Chapter three, the validity and reliability of three measurements used in this study were already establis hed by previous research. (LTT C in Taiwan, Mokhtari, K. and Sheorey, R., 2002; Mori, 2002) However, the test ing environment and population of this study 122

PAGE 123

were different from previous researches. For that reason, the reliability of each measurement was tested with the participants of this study to check whether it was also reliable in this experimental condition. Cronbach alpha one of the most commonly reported standard reliability estimates in the language testing litera ture, was used to check reliability or internal consistency of each measurement and was presented in order in the following section. Reliability of scores of the reading comprehension test Reading comprehension test was selected from GEPT with 20 ques tions in the reading section. The reading section co vers diverse topics and genres (See Appendix H) Subjects responses on the test were recorded as correct and incorrect (0/1) to avoid the effect of the treatment. The result in Table 4-14 indicates a high reliability of reading comprehension test. In the pretest, 119 answer sheets were colle cted and the result of Cronbachs alpha was .741. For the posttest, 113 answer sheets were obtai ned and the result of Cronbachs alpha was .837. According to Kline (1999, cited in Field, 2005), th e general accepted value of .8 is appropriate for cognitive tests such as intelligence tests, while for the ability test, a cut-off point of .7 is more suitable. Therefore, the read ing comprehension test of this study, whose Cronbachs alpha = .73 and .83 were considered reliable for this study. Table 4-14. Reliability of scores of reading co mprehension test in pr e-test and post-test Reading comprehension test alpha N Pretest Posttest .741 .837 119 113 Reliability of scores of the Survey of Reading Strategy (SORS) The reliability of scores of the SORS and its subscales were also calculated in pre-test and post-test administration for the entire sample with N=114 by using Cronbachs alpha The result indicates that this instrument as a whole and the subscales has moderate to high reliability 123

PAGE 124

f or the sample tested (See the following Table 4-15). The overall is .881 for the pretest and .911 for the posttest. The Cronbachs alpha of each subscale in the pretest data were .803, .742, and .675 respectively. And the Cronbachs alpha of each subscale in the posttest data were .778, .814, and .736 respectiv ely. As the overall Cronbachs alpha in pretest and posttest were all above .7 and thus indicated good reliability and thus a reliable instrument for surveying reported readi ng strategy use in this study. Table 4-15. Reliability of scores of the SORS in the pre-test and post-test SORS Alpha N Pretest Total .881 114 Subscale 1 (GLOB) .803 114 Subscale 2 (PROB) .742 114 Subscale 3 (SUP) .675 114 Posttest Total .911 114 Subscale 1 (GLOB) .778 114 Subscale 2 (PROB) Subscale 3 (SUP) .814 .736 114 114 Note: SORS=Survey of Reading Strategies; Subs cale 1 (GLOB) = Global Reading Strategies; Subscale 2 (PROB) = Problem-Solving Strate gies; Subscale 3 (SUP) = Supporting Reading Strategies Reliability of Scores of the English Reading Motivation Questionnaire (ERMQ) The reliability of scores of the ERMQ and its subscales was also calculated for the pretest and post-test administration with the entire sample of 113 by using Cronbachs alpha The instrument as a whole and the subscales had moderate to high reliability for the sample tested (See the following Table 4-16). The result indicate s that this instrument as a whole and the subscales has moderate to high reliability for the sample tested (See Table 4-16). The overall is .918 for the pretest and .915 for the posttest. For all the subscales in the pretest, the Cronbachs alpha was .855, .734, .739, and .786 respectively. Si milarity, all the subscales in the posttest the Cronbachs alpha yielded .879, .747, .707, and .850 respectively. As the 124

PAGE 125

overall were all above .8 in the pret est and posttest, and thus indi cated a good reliability of the English Reading Motivation Questionnaire (ERMQ ) and thus a reliable instrum ent for the purposes of surveying reported Englis h reading motivation in this study Table 4-16. Reliability of scores of th e ERMQ for the pre-test and post-test ERMQ Alpha N Pretest Total .918 113 Subscale 1 .855 113 Subscale 2 .734 113 Subscale 3 .739 113 Subscale 4 .786 113 Posttest Total .915 113 Subscale 1 .879 113 Subscale 2 .747 113 Subscale 3 Subscale 4 .707 .850 113 113 Note: ERMQ= English reading motivation questionnair e; Subscale 1 is intrinsic motivation; Subscale 2 is extrinsic motivati on; Subscale 3 is the importance of English; while subscale 4 measure reading efficacy. Assumptions of ANCOVA This study intended to use ANCOVA to analyze the measurement outcomes by taking account of confounding variables to get a purer measure of the effect of the experimental manipulation. The advantages of using ANCOVA are that it increases statistical power and reduces bias by equating groups on one or more c ovariates. In this study, the covariate is the GEPT score before the treatment and the dependent variables are: 1) post-reading comprehension test, 2) the post-SORS score, and 3) the postERMQ score after the treatment. The use of ANCOVA requires several assumptions. Hen ce, performing ANCOVA, it is worthy of evaluating the data to determine whether the assumptions of ANCOVA are met. The following section describes whether this study has met a ll the ANCOVA assumptions: 1) homogeneity of 125

PAGE 126

variance; 2) hom ogeneity of re gression slopes; 3) normal distribution of the sample on three dependent variables; and 4) the relationship betw een the outcome and the c ovariate is linear. Homogeneity of variances The assumption of homogeneity of variance means that the group variances should be the same throughout the data (Field, 2005). Levenes test tests the hypothesis that variances in the groups are equal. That is, the difference between the variances is zero. Table 4-17 summarizes Levenes tests on three dependent variables. The results yielded that for post reading comprehension test, F (1, 108) = 1.13, p >.05. For the post SORS, F (1, 108) = 0.776, p >.05, and for the post ERMQ, F (1, 108) = 1.533, p >.05, all showed at a level of non-significance (p > .05) and thus indicated that the group variance s were equal in three measurements and hence the assumption of homogeneity of variance has met. Table 4-17. Levenes test of equality of error variance F df1 df2 Sig Post RC 1.134 1 108 .289 p >.05 Post SORS Post ERMQ 0.776 1.533 1 1 108 108 .380 p >.05 .218 p >.05 Note: Post RC=post reading comprehension test; pos t SORS=post Survey of Reading Strategies; post ERMQ=post English readi ng motivation questionnaire Homogeneity of regression slopes The assumption of homogeneity of regression slope means that the relationship between covariate and the outcome variables should be the same in all groups. Therefore, this assumption was examined by running the ANCOVAs with the interactions and checking whether they were significant. The result of the interaction effect of treatment by covariate in the post-reading comprehension test is not significant at F (1, 106) = 0.019, p > .05. In the post-Survey of Reading Strategies the result of interaction effect of treatment by covariate is not significant at F (1, 106) = 2.597, p > .05, either. In the postEnglish Reading Motivation Questionnaire the 126

PAGE 127

result of interaction effect of treatm ent by covariate is F (1, 106) = 0.615, p > .05, which is not significant, either. Besides, because there were no interactions between the covariate and the treatment on each dependent variable, the assu mption of homogeneity of regression slopes was met. Linearity: the relationship between th e outcome and the covariate is linear This assumption means that the relationship between the covariate and the outcome variables should be the same in all of the groups The correlation between covariate (GEPT) and three dependent variables are shown in the foll owing Table 4-18. The conventions proposed by Davis (1971) were used to indicate the magnitude of the correlations. Correlations between 0.50 and 0.69 are substantial, and correlations be tween 0.70 and 0.99 are very high. Therefore, a substantial correlation was observed between GEPT and the post-reading comprehension test ( r = .643). Also, high correlations were discovered between GEPT the covariate, and the postSORS ( r = .809) as well as the post-ERMQ ( r = .781). Table 4-18. Correlations between covari ates and each dependent variable Post-reading comprehension test Post-SORS Post-ERMQ GEPT (covariate) .643(**) .809(**) .781(**) Note : GEPT=general English proficiency test; po st-SORS=posttest score on Survey of Reading Strategies; post-ERMQ=posttest score on E nglish Reading Motivation Questionnaire. (**) correlation is significant at the 0.01 level N=110 Normal distribution of the samp le on three dependent variables The following figures (4-1, 4-2, and 4-3) pr esented the distributions of the three dependent variables-the post-English reading comprehension test, the post Survey of Reading Strategies and the post English Reading Motivation Questionnaire In addition, the tables for Skewness and Kurtosis are under each figure. The results of the figures and tables demonstrated 127

PAGE 128

norm al distributions from the data in which were sampled. Therefore, the assumption of normal distribution on dependent variables was met. Statistics Post-Reading Comprehension Test 110 0 -.170 .230 -.326 .457 Valid Missing N Skewness Std. Error of Skewness Kurtosis Std. Error of Kurtosis Table of Skewness and Kurtosis in postReading Comprehension Test Figure 4-1. Post-Readin g Comprehension Test 128

PAGE 129

Statistics Post Survey of Reading Strategies 110 0 -.246 .230 .228 .457 Valid Missing N Skewness Std. Error of Skewness Kurtosis Std. Error of Kurtosis Table of Skewness and Kurtosis in post Survey of Reading Strategy (SORS) Figure 4-2. PostSurvey of Reading Strategies (SORS) 129

PAGE 130

Statistics Post English reading motivation questionnaire 110 0 -.050 .230 .093 .457 Valid Missing N Skewness Std. Error of Skewness Kurtosis Std. Error of Kurtosis Table of Skewness and Kurtosis in postEnglish Reading Motivati on Questionnaire (ERMQ) Figure 4-3. PostEnglish Reading Motivati on Questionnaires (ERMQ) 130

PAGE 131

Figure 4-4. Scatterplot of Covariate against post-r eading comprehension test Figure 4-5. Scatterplot of Covariate against postSurvey of Reading Strategies (SORS) 131

PAGE 132

Figure 4-6. Scatterplot of Covariate against postEnglish Reading Motivation Questionnaire (ERMQ) Therefore, assumption of linear ity between the covariate and the three outcomes was met. Figure 4-4, Figure 4-5, and Figure 4-6 showed the scatterplot of covariate against three dependent variables by groups where 0 represents the experimental group while 1 represents the control group. The three scatterp lots show that there is a positive relationship between the covariate and the three outcome variables. Hypothesis Test The research questions in this study restated as: 1) what is effect of metacognitive reading strategy instruction on EFL secondary students English reading comprehension; 2) what is effect of metacognitive readi ng strategies instruction on EF L secondary students English reading strategies awareness; 3) what is effect of metacogniti ve reading strate gy instruction on EFL secondary students English reading motivation; and 4) Is there any interaction effect of metacognitive reading strategy in struction and participants Eng lish reading proficiency? In other words, this research was designed to check whether metac ognitive reading strategy instruction (MRSI) could influence EFL seconda ry students English reading comprehension, 132

PAGE 133

reading strategies awareness, English readi ng m otivation, and whether the impacts of MRSI varies across difference language pr oficiency levels. In addressing these research questions, this study aims to test the foll owing four null hypotheses: H Reading comprehension will be the same for high school s tudents with metacognitive reading strategy instruction (MRSI) with t hose without metacognitive reading strategy instruction. 01H Perception about reading strategies will be the sam e for high school students with metacognitive reading strate gy instruction with those w ithout metacognitive reading strategies instruction. 02H Motivation to reading English will be the sa m e for high school students with metacognitive reading strategy instruction with those wit hout metacognitive reading strategy instruction. 03H The effect of metacognitive strategy instructio n will be lar ger for EFL students with lower language competence. 04 Results of these hypotheses were analyzed through ANCOVA and were presented in the following sections. Effect of treatment on post-English reading comprehension test The first dependent variable is the pos ttest of English reading comprehension as measured by GEPT reading section. Analysis of covariate (ANOCA) was conducted to examine between-subjects effect of treatment on post-English reading comprehension test If the F-test is significant, then the probability of the effect bein g detected will be high. The mean and standard deviation as well as the result of ANCOVA ar e presented in Table 4-19 and Table 4-20 respectively. Results of the ANCOVA, as shown in the following Table 420, revealed that the covariate, general English prof iciency test, was significantly related to the participants postreading comprehension test, F (1, 107) = 88.83, p <.05. The effect size (Eta Squared p 2 ) = .43 of covariate also indicate s that after controlling th e effect of the treatment, the covariate explains 45% of variable of the post reading comprehensi on scores and thus it has large effect on the post 133

PAGE 134

reading comprehension test. That is, after controlling for the effect of the treatm ent, the covariate explains 45% of the variables of post reading comprehension scores. Table 4-19. Mean and Standard Deviation of postEnglish Reading Comprehension test Experimental Group Control Group N 56 54 Mean SD Mean SD Pre test GEPT (covariate) Post-reading comprehension test (DV) 70.48 76.79 11.43 12.95 71.30 69.81 11.74 14.21 Note : DV=dependent variable In addition, there was a signi ficant effect of the treatment on post-English reading comprehension score after controlling for the e ffect of participants English reading ability, F (1, 107) = 15.65, p < .05. The effect size (Eta Squared p 2) = .08 of the treatment (MRSI) indicates a small to medium effect on post English readi ng comprehension. The values of effect size are useful because they provide an objective and standa rdized measure of the im portance of an effect, the treatment effect in this study. To put it brie fly, there was a significant difference between the experimental group and the control group on the post reading comprehension test after the treatment (MRSI) when covariate GEPT was held constant. The impact of this significant difference for the change in the post reading comp rehension test was also analyzed with Cohend statistical effect size. That is in order to obtain an objective and standardized measure of the difference between the experiment al group and control group, Cohen d effect size was calculated according to the following formula. pooledMMd /21 and 2/2 2 2 1 pooled The result of Cohens d calculation was 0.51, which indicated that the experimental group scored 0.51 standard deviation higher than th e control group. This indicated that there was a large effect of the treatment (MRSI) and it is educationally significantly. Therefore, there was 134

PAGE 135

enough evidence to reject H and concluded that, after 10-week MRSI training, the experim ental group outperformed the control group in English reading comprehension. 01Table 4-20. ANCOVA table on Post-Eng lish Reading Com prehension Test Source SS df MS F p Eta Squared GEPT (covariate) 9035.4 1 9035.4 88.83* .000 .43 Treatment 1591.47 1 1591.47 15.65* .000 .08 Error 10884.18 107 101.72 Total 21255.46 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares Effect of treatment on the post-Survey of Reading Strategies (post-SORS) The second dependent variable was the postSORS test as measured by the Survey of Reading Strategies The result of this measurement aims to examine the effect of metacognitive reading strategy instruction (MRSI) on EFL sec ondary students English reading strategies awareness. Therefore, an Analysis of Cova riate (ANCOVA) was used to answer this null hypothesis with the treatment as th e independent variable. Postte st score of the SORS is the dependent variable while the pret est score of the reading comprehe nsion test is the covariate. Results of ANCOVA revealed that there was a significant effect of treatment (MRSI) on postSurvey of Reading Strategy test, F (1, 107) = 10.28, p <.05. The effect size (Eta Squared p 2 ) = .10 also indicated a small to medium effect on SORS. Desc riptive statistics for these two groups in terms of the postSORS are in the following Table 4-21. Table 4-21. Mean and Standard deviation for postSurvey of Reading Strategy Group Mean Std. N Control 3.36 .53 54 Experimental 3.53 .65 56 135

PAGE 136

Table 4-22. ANCOVA table on post Survey of Reading Stra tegy Source SS df MS F p Eta Squared GEPT (covariate) 25.7 1 25.7 225.179* .000 .66 treatment 1.173 1 1.173 10.28* .002 .03 Error 12.21 107 0.114 Total 38.73 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares Results of the ANCOVA, as shown in Table 422, revealed that th e covariate, general English proficiency test, was significantly related to the participants post-SORS, F (1, 107) = 225.179, p <.05. Therefore, the assumption of AN COVA was met and the use of ANCOVA to test this hypothesis is ap propriate. The effect size of covariate (Eta Squared p 2 ) = .66 indicating a large effect of the covariate on the postSORS Moreover, there was a significant effect of treatment (MRSI) on post-SORS after controlling for the effect of particip ants English reading ability, F (1, 107) = 10.28, p < .05, with th e effect size (Eta Squared p 2) of .03. The result of ANOCA table indicated that the treatment (MRS I) made a significant difference between the experimental group and the control group on postSurvey of Reading Strategy test when covariate GEPT was held constant. The Table 4-22 also reveals a small effect ( p 2 ) of the treatment. The result of Cohens d was 0.31 standard deviation, wh ich demonstrated that the experimental group scored 0.31 higher than the control group, thus there was a medium to large effect of the treatment and it is educationa lly significantly. Ther efore, there was enough evidence to reject H02and concluded that the experimental group outperformed the control group after 10-week MRSI on their English reading strategy awareness and thus indicated that MRSI has positive effect on EFL secondary stude nts reading strategy awareness. 136

PAGE 137

As there wer e three subtests in the Survey of Reading Strategy each subtest in postSORS was also examined to check the effect of treatment (MRSI) on each subtest, global reading strategies (GLOB), problem-solving strategies (PROB), and supportive reading strategies (SURP). Results are presented in the following tables. Table 4-23. Mean and Standard deviation of GLOB subtest in postSurvey of Reading Strategy Group Mean Std. N Control 3.33 .53 54 Experimental 3.57 .61 56 From the Table 4-23 above, the mean scor e of the experimental group was 3.57 and the mean score of the control group was 3.33. According Mokhtari and Sheoreys (2002) classification, the experimental gr oup is high reading strategies user while the control group is moderate reading strategies user in terms of Global Reading Strategies (GLOB) after 10-week instruction. As shown in the below Table 4-24, the analysis of ANOCA yielded an F ration of 14.02, p< .05 which was statistically significant. Also, the obtained e ffect size Cohens d of 0.12 was also educationally meaningful. In other words, the treatment, metacognitive reading strategy instruction (MRSI), made a significant differenc e between groups in terms of the awareness of Global Reading Strategies. Table 4-24. ANCOVA table on GLOB reading strategies in post-SORS Source SS df MS F p Eta Squared GEPT (covariate) 20.55 1 20.55 148.44* .000 .56 treatment 1.94 1 1.94 14.02* .000 .05 Error 14.81 107 0.138 Total 36.89 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares 137

PAGE 138

Table 4-25. Mean and Standard deviation of PROB reading strategi es subtest in postSurvey of Reading Strategy Group Mean Std. N Control 3.50 .67 54 Experimental 3.61 .81 56 Table 4-26. ANCOVA table on PROB r eading strategies in post-SORS Source SS df MS F p Eta Squared GEPT (covariate) 36.72 1 36.72 173.64* .000 .61 Treatment 0.677 1 .677 3.202 .076 .01 Error 22.624 107 0.211 Total 59.71 109 Note : Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares As shown in the above Table 4-26 in terms of problem-solving reading strategies subtest, the ANCOVA analysis yielded an F ration of 3.202, p>.05 which was not statistically significant. Meanwhile, the obtained effect size of 0.01 was not educationally meaningful, either. In other words, the treatment, metacognitive reading strategy instruction (M RSI), didnt make a significant difference between groups in terms of the awareness of Problem-solving Reading Strategies. Table 4-27. Mean and Standard devia tion of SUP subtest in post-SORS Group Mean Std. N Control 3.28 .58 54 Experimental 3.41 .70 56 Table 4-28. ANCOVA table on Supportive reading strategies in post-SORS Source SS df MS F p Eta Squared GEPT (covariate) 24.66 1 24.66 128.47* .000 .54 Treatment 0.753 1 .753 3.925* .05 .04 Error 20.541 107 0.192 Total 45.682 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares 138

PAGE 139

As shown in the above Table 4-28, the analysis yielded an F ration of 3.925, p< .05, which was statistically significant. Likewise, th e obtained effect size of 0.04 was a really small effect. In other words, the treatment, metac ognitive reading strategy instruction (MRSI), did made a significant difference between gr oups in terms of the awareness of Supportive Reading Strategies. In summary, the treatment, the metacognitive reading strategy instruction (MRSI) has effects on global reading strategies and s upportive reading strategies. And this result indicating that, by integr ating the metacognitive reading strate gy instruction (MRSI) into English reading classes, EFL high school students we re involved in the r eading activiti es through cooperative and group learning activities and they think more and deeper while reading. Effect of treatment on postEnglish Reading Motivation Questionnaire (postERMQ ) The third dependent variable is post-ERMQ as measured by English Reading Motivation Questionnaire. To answer the third hypothesis, an anal ysis of covariate ( ANCOVA) was used to answer this null hypothesis with treatment (MRSI) as the independent variable and posttest score of the English Reading Motivation Questionnaire as the dependent variable, and pretest score of GEPT as the covariate. Descriptive statistics for these two groups in te rms of the postERMQ are in the following Table 4-29. Table 4-29. Mean and standard deviation English Reading Motivation Questionnaire Treatment or control Mean Sd. N Control 82.28 11.30 54 Treatment 87.39 13.29 56 Total 84.88 12.56 110 Results of the ANCOVA, as shown in the following Table 4-30, revealed that the covariate, general English prof iciency test, was significantly related to the participants postERMQ, F (1, 107) = 198.19, p <.05. The effect size of covariate ( Eta Squared p 2) = .62 indicating a large effect of the covariate on post ERMQ. Moreover, there was a significant effect 139

PAGE 140

of treatm ent on post-ERMQ after controlling for the effect of participants English reading ability, F (1, 107) = 17.187, p < .05, with a medium to large effect size ( Eta Squared p 2 ) of .19. The result of ANOCA table indicates that th e treatment (MRSI) made a significant difference between the experimental group and the control group on postEnglish reading motivation questionnaire (ERMQ) when covariate GEPT was he ld constant. The result of Cohens d was 0.4, which was a medium to large effect for the treatment, indicati ng that the result is educationally significantly. Therefore, there was enough evidence to reject H03and concluded that the experimental group out performed the control group after 10-week MRSI in their English reading motivation. Table 4-30. ANCOVA table on English Reading Motivation Questionnaire (ERMQ) Source SS df MS F p Eta Squared GEPT (covariate) 10703.53 1 10703.53 198.19* .000 .62 Treatment 928.211 1 928.211 17.187* .002 .19 Error 5778.66 107 54.006 Total 17201.464 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares As there were four subtests in English reading motivation questionnaire (ERMQ)intrinsic reading motivation, extrinsic reading mo tivation, the importance of English reading, and reading efficacy. Each subtest in post-ERMQ was examined and presented in the following section. Table 4-31. Mean and Standard deviation of intrinsic reading motiv ation subtest in postERMQ Group Mean Std. N Control 2.68 0.41 54 Experimental 2.85 0.48 56 As shown in the below Table 4-32, the ANCOVA analysis yielded an F ration of 12.684, p< .05 which was statistically significant. Me anwhile, the obtained effect size of 0.11 was a small to medium effect. In other words, the treatment, metacognitive reading strategy instruction 140

PAGE 141

(MRSI), did m ade a significant difference betw een groups in terms of the intrinsic English reading motivation. The effect of treatment in ex trinsic reading motivation subtes t was examined as following. As shown in the Table 4-34 below, the analysis yielded an F ration of 8.833, which was statistically significant. Meanwh ile, the obtained effect size of 0.08 was a really small effect. In other words, the treatment, metacognitive reading strategy instruction (MRSI), did made a significant difference between groups in terms of the extrinsic Eng lish reading motivation. However, it is probably not educationally mean ingful with such a small effect size of 0.08. Table 4-32. ANCOVA table on intrin sic reading motivation in postERMQ Source SS df MS F p Eta Squared GEPT (covariate) 12.83 1 12.83 160.127* .000 .58 Treatment 1.016 1 1.016 12.684* .001 .11 Error 8.57 107 .08 Total 22.173 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares Table 4-33. Mean and Standard deviation of Ex trinsic reading motiva tion subtest in postERMQ Group Mean Std. N Control 2.92 .517 54 Experimental 3.08 .516 56 Table 4-34. ANCOVA table on extrinsic reading motivation in post-ERMQ Source SS df MS F p Eta Squared GEPT (covariate) 16.745 1 16.745 147.78* .000 .57 Treatment 1.016 1 1.016 8.833* .004 .08 Error 12.124 107 .113 Total 29.603 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares Table 4-35. Mean and Standard deviation of th e importance English r eading subtest in postERMQ Group Mean Std. N Control 3.15 .48 54 141

PAGE 142

Experim ental 3.35 .46 56 As shown in the above Table 4-36, the analysis yielded an F ration of 6.996, p < .05 which was statistically significant. Likewise, th e obtained effect size of 0.05 was a small effect. In other words, the treatment, metacognitive re ading strategy instruction (MRSI), did made a significant difference between groups in terms of the importance of English reading. In other words, the treatment had an effect on the par ticipants perception regarding the importance of English reading. However, the e ffect size indicated that this wa s not educationally meaningful. Table 4-36. ANCOVA table on the impor tance English reading in postERMQ Source SS df MS F p Eta Squared GEPT (covariate) 5.072 1 5.072 28.351.78* .000 .20 Treatment 1.252 1 1.252 6.996* .009 .05 Error 19.142 107 .179 Total 25.295 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares. Table 4-37. Mean and Standard deviation of reading efficacy subtest in post-ERMQ Group Mean Std. N Control 2.43 .73 54 Experimental 2.62 .53 56 Table 4-38. ANCOVA table on r eading efficacy in post-ERMQ Source SS df MS F p Eta Squared GEPT (covariate) 16.997 1 16.997 66.45* .000 .38 Treatment 1.249 1 1.249 4.88* .029 .03 Error 27.368 107 .256 Total 45.309 109 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares As shown in Table 4-38, th e analysis yielded an F ration of 4.88, which was statistically significant. Similarly, the obtaine d effect size of 0.03 was a small effect. In other words, the treatment, metacognitive reading strategy instruction (MRSI), did made a significant difference between groups in terms of the r eading efficacy of English reading. 142

PAGE 143

In summ ary, the treatment, metacognitive read ing strategy instruction (MRSI) has effect on the participants in the experimental group on their perception regard ing intrinsic reading motivation, importance of reading, and reading ef ficacy. In other words, through metacognitive reading strategy instruction, the high school EF L students English reading motivation was higher than before, especially on their intrinsic reading motivati on and reading efficacy. More qualitative data on this issue will discuss further in chapter five. Effect of treatment on different langu age ability in three dependent variables Previous research literature findings all showed that it a ppeared that those who with lower language proficiency level be nefit the most from the reading strategy instru ction (CorderoPonce, 2000; Kern, 1983; Song; 1998). This stud y, on the other hand, has different result. The results of interaction of treatment by language ab ility were not at a significant level in each dependent variable. The following Table 4-39 pr esents ANCOVA table on the interaction effect of the treatment (MRSI) on langua ge abilities on postreading comprehension test and the obtained F ration was 0.567, p > .05, which was not si gnificant. Table 440 presents ANCOVA table on the interaction effect of the tr eatment (MRSI) on language abilities on postSORS test, and the obtained F ratio was 0.96, p > .05, which was not si gnificant. At last, Table 4-41 presents ANCOVA on the interaction effect of the treatment (MRSI) on language abilities on post -ERMQ test, and the obtained F ra tion was 0.204, p >.05, which was not significant, either. In other words, the effect of treatment (MRS I) didnt make a significant difference on diverse reading abilities across three measurements. Th at is to say, the treatment (MRSI) had similar effect on high, medium, and low reading levels Therefore, there was no enough evidence to reject 04HTable 4-39. ANCOVA table on interac tion effect of treatm ent on la nguage ability in post-reading comprehension test 143

PAGE 144

Source SS df MS F p Group 1143.815 1 1143.815 10.321* .002 Lanability Group*Lanability 7899.02 125.613 2 2 3949.51 62.807 35.638* .567 .029 .569 Error 11525.498 104 .256 Total 613300 110 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares Table 4-40. ANCOVA table on interaction effect of treatment on language ability in post-SORS test Source SS df MS F p Group 1.713 1 1.713 10.969* .001 Lanability Group*Lanability 21.301 .300 2 2 10.651 .150 68.197* .960 .000 .386 Error 16.242 104 .156 Total 1349.219 110 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares Table 4-41. ANCOVA table on interac tion effect of treatment on la nguage ability in post ERMQ test Source SS df MS F p Group 941.163 1 941.163 12.735* .001 Lanability Group*Lanability 8600.367 30.216 2 2 4300.183 15.108 58.185* .204 .000 .815 Error 7686.147 104 73.905 Total 809743 110 Note: Type III sum of squares. *p<.05; SS = Sum of squares, df = degree of freedom; MS = mean squares However, a closely examination on the following figures (Figure 4-7, 4-8, and 4-9) tells us that, even though the treatment (MRSI) didnt make a statistically significant effect on three language levels on three post measurements and the effect did exist between the treatment group and the control group. For instance, the following figures (Figure 4-7, 4-8, and 4-9) highlight three important points. First, the treatmen t groups post-reading comprehension test, postSORS and postERMQ were always higher than the control groups post-reading comprehension test, post-SORS and postERMQ This means that participants who received metacognitive reading strategy instruction (MRSI) have higher scores than those who did not in terms of reading 144

PAGE 145

com prehension, reading strategy awareness, and reading motivation. Second, regardless of GEPT score, participants who had the treatm ent (MRSI) outperformed than those who didnt have the treatment (MRSI). This means that the treatment (MRSI) di d have influences on participants reading comprehension, reading strate gy awareness, and Englis h reading motivation. Finally, postreading compre hension test shows greater increase with those of intermediate English level. In other words, for those who with intermediate English level students, metacognitive reading st rategy instruction (M RSI) help them the most in comparison with high and low proficiency level students. For high proficiency level students, their English abilities have been good before the treatment a nd thus the improvement is limited in comparison to the intermediate level student s. As for lower proficiency le vel students, metacognitive reading strategy instruction also helps them in improving their reading comprehension to some degree but the improvement wasnt as obvious as those intermediate level students. Likewise, PostSORS score shows greater increase with those of high English level students. For high proficiency level students, since their reading ability has been good before the treatment, their reading strategy awareness became unconscious befo re the treatment. And they were reminded and were informed to a list of reading strategi es and thus have higher scores on post SORS. As to the post ERMQ, the figure shows slightly increase with those of intermediate and high English level students in comparison to low proficiency level, though the difference wasnt at all significant. 145

PAGE 146

Figure 4-7. Interaction between language leve l and group in post reading com prehension test Figure 4-8. Interaction between lang uage level and group in post SORS 146

PAGE 147

Figure 4-9. Interaction between lang uage level and group in post ERMQ Conclusion The goal of the quantitative part of this st udy was to investigate th e possible effects of integrating metacognitive read ing strategy instruction (MRSI) into regular English reading classes on EFL secondary students English read ing comprehension, English reading motivation, and perception about reading strategies. Meanwhile the effect relates to EFL students different reading proficiency levels are also part of an in terest of this study. This chapter has presented the statistical results of th is study and the findings are as follo wing. First, the experimental group experienced a significant improvement in thei r post reading comprehension when comparing with the control group, which demonstrates the evidence that, integrating metacognitive reading strategy instruction (MRSI) into EFL reading cl asses did significantly increase secondary EFL students reading comprehension. Second, there was also a significant difference between the two groups in their postSurvey of Reading Strategy (SORS). That is, the experimental group 147

PAGE 148

used m ore reading strategies after the treatment, and the salient increasing use of reading strategy was Global Reading Strategies as defined by SORS. That is to say, the experimental group were not confined themselves in usi ng the strategy of word-by-word tr anslation; instead, they agreed that they used more Global Reading Strategies after the experiment. Third, there was a significant difference between two groups in their postEnglish Reading Motivation Questionnaire (ERMQ). Participants in the experime ntal group showed greater boost in the subtests of Intrinsic Reading motivation and Reading Efficacy and slightly increase in their Extrinsic Reading motivation Last, the effect of metacogn itive reading stra tegy instruction (MRSI) didnt reach a statistically significant diffe rence on different English proficiency levels. That is, the result was consistent with previous studies and further proved that metacognitive reading strategy instruction can help all EFL or ESL students regardless th eir English proficiency levels. In conclusion, significant differences were found between the experimental and control group in English reading comprehension, reading strategy awareness, and English reading motivation as measured by the three posttests sugg esting that achievement in the experimental group improved as a result of exposure to MRSI. Th is means that integra ting MRSI proved to be an effective way for improving EFL high school students reading co mprehension, reading strategy awareness, and reading motivation. Besides, MRSI improved good and average EFL students performance and the performance of the low-level students. This finding is consistent with findings of prior studies using (Corde ro-Ponce, 2000; Kern, 1989; Song, 1998) other types of MRSI in reading comprehension. 148

PAGE 149

149 CHAPTER FIVE QUALITATIVE FINDINGS Introduction My predominant interest in secondary EFL students originated from my personal experience teaching EFL adolescents in the past. Fr om my experience, I real ized that there is a need to teach high school EFL students how to read in English r eading materials. However, it was not until I started my doctoral program at UF that I began to learn about metacognitive learning and how they are related to successful learning. I believed th at, with the increasing values placed in English these days, English teachers in Taiwan are doing their best in instruction and preparing their EFL students with the desira ble English capacities for their future academic performance and future life or career. Howeve r, many high school EFL students, who have been learning English for at least 8 years in average (refer to the average year s of the participating students in this study), still try very hard and put so much effort in learning English and yet are struggling or having fears when reading someth ing in English. What is the problem? In this study, that is my intention to furt her look into this situa tion, the situation that quantitative research cant answ er through the measurements, by k eeping a skeptical attitude to question research assumptions and research-based effective practice, such as metacognitive reading strategy instruction (MRSI for short) in this study. Although the sta tistical results in the previous chapter has showed th e evidence of MRSIs effectiveness on reading comprehension, reading strategy awareness, and r eading motivation, I still consider that these statistic results should be double checked through another approach before accepting the fact Data collected for analyzing in this chapter are: students reading strategy journals, interviews, classroom observations, and casual communica tion between the participating teacher, the one who taught in the experimental group and me. The research ques tions in this phase th at guided me to probe

PAGE 150

further are: (1) W hat are th e factors bring about high school EFL students reading experience and reading behaviors? and (2) What interven tion features helped secondary EFL students successfully learn reading strategies that enable d them to comprehend the reading text better and are willing to engage more in English reading? This chapter first describes qualitative findings organized by these two research questions that reflect the previ ous English readi ng experience of secondary EFL students and then the changes that has been made throughout the study period. (1) Problems in Reality Finding # 1 Traditional English instruction makes vocabulary and grammar become reading goals for secondary EFL students. To EFL students, English reading is simply to check the unknown vocabulary and grammatical or sentence structure analysis. 1-1. Vocabulary instruction The role of vocabulary in reading comprehension is undeniable in both L1 and L2 reading resear ch (Droop and Verhoeven, 2003; Grabe, 2004; Nation, 2002; Nassaji, 2003; Stanovich, 2000). However, reading instruction only emphasizes the importance of vocabulary wit hout sound instruction of its application does little help to reading comprehension. As discussed in Literatu re Review in relation to the bottom-up reading comprehension model, vocabulary knowledge is a necessary but insufficient condition for the outcome of successful reading comprehension (K oda, 2005). In view of its important role in reading comprehension, research on effective practice in improving EFL students vocabulary knowledge has suggested targeting both dimens ions of breadth and depth of vocabulary knowledge (Ordonez et al., 2002). In other words, as vocabulary knowledge is important for the reading comprehension for ESL or EFL students, it is better to learn it with engaging context rather than out of context (Fu, 1995; Baumann a nd Kameenui, 1991). In this study, nearly all the interviewees stated clearly that what they concern the most in E nglish reading is their shortage knowledge in vocabulary, phrases, and grammatical knowledge. The reasons behind this issue are as following. First, the way high school EFL students learned English vocabulary was 150

PAGE 151

isolated and fragm ent. That is, a new word is learned as a word with one meaning only. Typically, in English reading instruction, new wo rds are picked out prior to reading a text. English teachers directly tell their students the meaning of a new word, and then EFL students would grab a pen and a piece of paper, repeatedly write the word down, and then self-test to see if it is memorized. The reading activity wouldn t start until all the new words have been memorized. This kind of vocabulary learning method has made EFL students easy to forget what they have learned because they don know how a word is used in the context. For instance, one student commented, I forgot the words soon after the quizzes. A nd I cannot recognize them whenever I came across them in reading English books. Second, neither is vocabulary l earned in context, nor is th e pronunciation of a new word the focus of the vocabulary instruction. One interviewee said: I seldom know how to sound the word out; English teachers seldom teach us or let us practice pronunciation. I just copy it couple times on the paper, then I would gradually know how it looks like and I will know how to spell it. When she (the English teacher) tests us vocabulary, she said the Chinese m eaning and we spell th e corresponding English word out. Echoed with the quantitative findings in Chapter Four, read aloud is the least frequently used strategy among EFL high scho ol students. What this interviewee meant is that, without knowing how to sound out the word or its pronu nciation, he still can memorize the vocabulary and pass vocabulary quizzes. This kind of learni ng style regarding voca bulary learning probably comes from their experience of learning their first language, Mandarin Chinese. English is an alphabetic system and its basic unit consists of its pronunciation and word structure (spelling) and these two features also map phonological and morphological identities. Chinese, on the other hand, is a logographic system, of which stru cture provides few or no clues about their 151

PAGE 152

pronunciation. Therefore, to som e EFL student s of Chinese as a native language, English vocabulary is viewed or recognized as a pictu re or a logo, and not knowing how to sound out English word still can make them pass English pape r tests. To them, they are very likely to rely on their L1 reading strategies when reading English. Thus, readers with logographic script in their L1 will try to apply a visual strategy to learn English words. Besides, English teachers seldom expect th eir EFL students to know how to sound out a word. As long as high school students can get high scores in the exams, it seems that correct spelling is more valued than pronunciations in mo st Taiwanese English teachers points of views, especially when speaking is not the focus in ma ny major exams, such as college entrance exams. Ironically, despite EFL students limited ability in pronunciation of English words, EFL students are still taught to read aloud when there is a comprehension breakdown during reading. When responding to the question regarding English reading strategies they learne d during the interview, Lee said: reading strategy? UmmI think the strate gies I learned didnt help much. They (English teachers) told me that, whenever you encounter difficulties during the reading, unknown words or confusing sentences, just r ead that sentence out loud couple times, then you will get it. But, I tried this method, and I dont thin k this strategy or solution works at all, honestly, I dont know how it works to teachers but not to me. From what Lee mentioned, if EFL students w ithout the skills or ability of sounding out words were taught to read aloud when comprehe nsion breakdown during th e reading process, it would be problematic and would make EFL studen ts even more frustratedstruggling for word recognition as well as sounding out the words. In conclusion, vocabulary building in L1 and L2 has won a lot of attentions in reading research literature and it is a common consensus that vocabulary knowledge and reading comprehension are highly co rrelated. For that reason, effective instructions on vocabulary building are important in L2 reading. Similarly, effective 152

PAGE 153

instructions on vocabulary building are m uch mo re crucial for EFL readers whose exposure to spoken English is limited. In other words, E nglish teachers should provide their language learners with engaging text during vocabulary instruction and strategies related to ascertaining the meaning of unknown words, as well as general vocabulary building. All of these are part of and essential to a strong program in reading comprehension instruction. 1-2. Reading instruction Speaking of English reading, th e interviewees naturally refer English reading to checking vocabulary, grammar c onstruction, or sentence structure analysis. EFL high school students spend most of their reading time and effort on looking up new words in the dictionary to find their meanings. They vi ew English reading as acquiring vocabulary and learning about grammar and sentence structure. Few of them refer it to access to information or meaning making. For instance, when responding to the question of how do you read when reading in English? Most typical res ponses would be like Hsians remarks: I will start reading by skimming unknown word s in the reading text, underline or highlight them, and then look them up in th e dictionary, write down their meanings, and if there are several meanings, I just choose the first one. After all the meanings are checked out, I begin to read it word by word. Another student, Leo, has similar reading process as the above student described. Actually, this kind of reading process has been re ported several times from different interviewees. Leo also said: Leo: Other than checking the meaning out first, I also analyze the sentence structure. My teacher taught me, after all the unknown word has been looked up in the dictionary; I should begin to analyze the sentence structure to get the meaning. The researcher: what do you mean by analy ze the sentence struct ure? Can you explain more? Leo: Ihmm, for instance, I break the sentence into several parts acco rding to its parts of speech. For instance, all the verbs will be circled (using hand gesture), nouns will be ( ) (using hand gesture), relative pr onouns will be underlined, and [ ] (using hand gesture) are used for ad jectivessomething like that. 153

PAGE 154

Once again, from the evidence above, we can see that English reading is taught as vocabulary decoding and grammatical analysis to most EFL student s. Moreover, what strikes me the most was the responses from two stude nts who share the simila r perceptions about a good English reader is. My teacher told me that if you want to be very good at English reading, you have to start with memorizing the whole dic tionary, from A to Z. Another student agrees with him by adding th at he did try to me morize the dictionary years ago with an attempt to become a good reader in English: My teacher told me that too. She said that all good English readers go through the process of spending time in memorizing the whole dictionary. I was actually convinced of that point, and started to memorize word s in the dictionary beginning with letter A, .but I gave up eventually, and I made it to half of A part. The above comments so far revealed one thing. That is, to many EFL highs school students, English reading is all about vocabulary and grammar. No one mentioned or was aware of how thoughts or ideas were perceived, organized and presente d in English, or how and where to locate topic sentence or to identify the suppor ting data, etc. In other words, all of these comments above indicated that EFL students are actually not ta ught how to read. From my previous experience and observations of my coll eagues teaching, the typical teaching approach is: vocabulary sentence structure or analysis translations along with grammatical sentence analysis some literal comprehension questions. Clear ly, the way they were taught is not for reading comprehension or meaning making. Inst ead, their perception about English reading focused more on vocabulary and grammatical structure than meaning making or reading comprehension. This findings also echoes with Parry (1996) and Kohns (1992) conclusions in that, Chinese EFL readers tend to use bottom-up strategies more than top-down strategies 154

PAGE 155

and this tendency was closely linked to their L1 literacy tradition and th eir understanding of the reading process. However, reading English in this way w ould only m ake EFLs spend more time and energy and of course, more frustra tion and less desire to approach E nglish reading materials. It is no wonder that a lot of intervie wees would view English read ing as time-consuming and tedious and meaningless. One interviewee, Jay, said: Jay: I dont like to read in English, I don t read it other than teachers request or assignment, To me, its painful and very time consuming, especially when you spend time checking vocabulary out in the dict ionary, and you found them with many definitions and dont know which one to choose from, still dont get it even though you know every meaning of each word. The researcher: so did you ask othe rs for help regarding this? Jay: Asking for help is no use; I knew what they would answer me. Sometimes I tried to express my problems in English reading, they would say, thats the way what English reading is about, just keep going, remember, the key to be good at English is read more, listen more, speak more, and write more of course, I know this motto, but the problem is that, sometimes when you spend time reading, and it turns out you are not necessary understand the whole text, and it is very frustrated, and thus, sometimes I choose to give up. Jays comments Asking for help is no us e points out one thing, that is, low-level support from the teachers cause a sense of helple ss and loss of confidence (Gan et al., 2004). On the other hand, EFL students are blamed for not reading enough and not memorizing enough vocabulary and thus make their En glish reading difficult. They are constantly told that their reading problems lie in their insufficient vocabulary knowledge and the key to achieve successful comprehension is to read more, as mu ch as possible. Thus, diligent EFL students kept memorizing vocabulary out of the context and kept forgetting what they have memorized. They kept focusing on grammatical features and ignor ing meaning making in the reading process and thus kept finding that this reading job meani ngless and laborious. Then, this kind of English learning seems going nowhere to them. It is like Mathew effect to them in view of the fact that 155

PAGE 156

English reading is an arduous work to them a nd they dont have control over their learning, and as a result, they build up a ps ychological hindrance toward Eng lish reading, or even choose to give up as if give-up is the way they can control over their learning. As May said: I keep memorizing vocabulary and I forget them quickly, or sometimes I memorize it, when it appears in a sentence, then there is possib ility that I dont know what it is. Another case, Chu explained: A lot of time I feel so frust rated in reading, sometimes I spend so much time looking up vocabulary and sometimes you dont have much time for that, so many other subjects I have to deal with in school, I would rather give up and leave it to English teacher and wait for their explanatio n or translation. According to researchers in Taiwan (Che ng et al., 2004; Lu, 2004), the English threshold of known vocabulary is 3,500 for EFL students to comprehend. In other words, EFL students who possess more than 3,500 sight words would f ace little difficulties when reading appropriate reading text. EFL high school stud ents in this study, if under regul ar schooling from elementary school to high school, should have learned at least 3500 vocabulary. Therefore, with this amount of vocabulary knowledge, English reading inst ruction should go beyond just vocabulary and grammar. Constantly focusing on meaningless vocabulary learning and grammar instruction does little help to EFL high school students read ing comprehension because they dont see how meaning is constructed in the text. Moreover, memorization makes students become lazy in thinking critically, in some contexts; students do not guess the meaning of an unfamiliar word even that is within their ability. According to the interactive reading model as discussed in Literature Review in Chapter Two, reading is an interactive process be tween a reader and the text (Grabe, 2004; Eskey, 2005). Th erefore, reading strategy should be taught to make them feel the ease of reading and construct meaning from the text and willing to read more on a regular basis. One student commented that he seldom read something in English. He said: 156

PAGE 157

I seldom read something in English. I know I have to read a lot to improve my English reading comprehension. But I think my vo cabulary knowledge is too less. I think I might start to read when my vocabulary is built up. The above comment indicated th at those EFL students excuse s of not reading a lot are because they have limited vocabulary knowledge. However, research said that in order to retain vocabulary, EFL students need to meet the words in a variety of contexts anywhere from 5-16 times. For that reason, instead of meaningle ss vocabulary learning out of context, EFL high school students should reading extensively to get access to vocabulary to build up their vocabulary. And one effective way to motivate them to read extens ively is to teach them reading strategy to let them feel the ease when approach ing to English reading materials. In addition, EFL students need to realize that English reading is not just de coding or grammatical structure analysis. As Tierney (2005) stat es, ...learning to read is not only learning to recognizing words, it is also learning to make sense of text.(p.51) To EFL high school students, this is especially important because reading is a valuable source for language in put and it is the most costeffective means of acquiring another language an d culture (Bernhardt, 1993). Thus, reading strategy, rather than vocabulary and grammar or translation, should be emphasized in the initial stage of language learning to avoid fossilization. Finding # 2 Teacher-centered approach has strengthe ned the habit of heavy dependence on teachers and made independent reading impossible. Meanwhile, social interaction is hardly seen in English classes. One fact emerging from the data is that EF L students heavily relied on language teachers help, explanation and translati on, with English reading. Based on social constructivism, the development of reading is more than a matter of linguistic and psychological processing. Instead, the social context of learning to read played an important role in the later development of reading strategies. Therefore, language learning and behaviors are soci ally shaped (Vygotsky, 1978) and so are English reading habit and behavior. Or asanu (1987) pointed out that the acquisition of 157

PAGE 158

literacy skill norm ally reflects social and instru ctional environment in which it take place. Therefore, English teachers reading instru ction could have a tremendous impact on EFL learners reading behaviors. For instance, most EFL high school students in this study stated that they prefer to read under English teachers guid ance of word by word translation or sentence by sentence analysis. However, detailed translati on and explanation or anal ysis has reinforced the belief that one has to understand the precise mean ing of every word and every sentence in order to comprehend the reading text. In other word s, translating strategy and sentences analysis guided by English teachers only promote dependenc y, not autonomy to EFL high school students. With such heavy dependence on English teacher s for problem-solving in reading revealed several issues here. First, what if such res ources like translation and analysis from English teachers arent around EFL high school students a nymore? Does it mean that EFL high school students will not read once they graduate from high school? Sec ond, EFL students arent equipped with the reading strate gies for being independent read ers. The consequence is, they become passively learners who lack of initiativetaking in reading for self-engagement, which is considered as an important step for language lear ners to acquire a language, especially in an EFL setting where exposure to English input is lim ited. To most EFL high school students, English reading is supposed to be assigned for them by English teachers. For instance, one student said: I never read outside of the cl ass. I only read in the classr oom. The time I spend reading English at home are mostly in the area of grammar review and memo rizing vocabulary. As for reading, unless the teacher assigned us to read, I seldom read. Or, I will read only if I have a lot of time, you know, I might encounter a lot of vocabulary that need time to check them out. But I prefer teachers led us to read, through analyze and translation, then it would be more effective. In that ca se, you wont waste time looking up in the dictionary and the result is guaranteed. 158

PAGE 159

From the statement above, it clearly demonstr ated that EFL high school students were so used to passively being guided by their English te achers in the course of English reading process and were in favor of reading with English teach ers instead of reading independently. It seems that they wont feel secure or efficiency when they are asked to read independently. If they do read independently, they turn to the help of the translation. One student said: If I have to read, I will read the translation or the Chinese version in the back first. (Note: some English reading materials in Taiwan ha s translation in the back of the book). In that way, I wont lose the real meaning. But most of time, I read grammar-related magazine at home. It seems that, to them, without Chinese translation, they dont know how to read independently, and the only thing that they can do individually for improving their reading ability is roterepetition in vocabulary memorization and grammar drill practice. To English teachers in Taiwan, due to the grea t pressure of the examinations as well as parents and schools expectations of the college acceptance rate, they were forced to take a direct way to explain language di screte points, and constantly encourage their students to learn by memorization or rote-repetition. Moreover, English teachers believed that classroom activities of teacher-centered is the best way th at would allow them to efficiently control the class time on what should be taught and covere d as guided by the mandated curriculum and prescheduled term exams. The participant teacher explained after the MRSI experiment: I know I have to allow time and opportunities fo r students to negotiate and to interact one another to make English learning meaningful and interesting, but if I do so, then there would be no much time left for me, you know we English teachers, have to cover what they need to know and learned for the exams. I have to keep reminding myself that what schools and parents want is all about the exam scores. Clearly, English teachers in Taiwan dont have much choice in terms of making an effective instruction or interactive learning enviro nment. What they have to do is prepare their students for mandated curriculum and examinations Consequently and gradually, how English 159

PAGE 160

teachers tau ght their EFL students how to read has left them the impression regarding English reading as reading by word-by-wo rd translation and sentence-bysentence structure analysis to get the meaning from the reading task. More than that, with English teachers constantly thorough and detailed explanations on vocabular y and grammatical structure during the reading process has virtually given EFL students the messag e and indication that they are still incapable of reading alone or independently because there ar e still a lot of linguistic knowledge that they need to learn before to the stage of compre hension. As a result, EFL students dont have confidence in reading independently because they believe that they are still lack so much knowledge in English grammar and vocabulary to r ead independently and successfully. They are not aware that knowing all the vocabulary isnt equal to comp rehending the text as a whole picture. What they learned is translation of each sentence, yet not the entire message of the text. They are used to passively absorbing inform ation provided by teachers and become passive learners, which hinders successf ul language learning as well. Influenced by Chinese culture for years, Confucian learning culture favors teacher-led activity in most Taiwanese cl assrooms. EFL students in this study were not used to the interaction with their peers and teachers. Most English teachers according to the participating teacher, did not see social inte raction as a learning tool or as imperative in building EFL students reading comprehension. Like most teachers in Taiwan, Ms. Lin thought that teacherdirected instruction is a good way to maintain a quiet orderly cla ss environment, especially with a large size of forty or more stude nts in a class. In contrast to the researchers, such as Pappas (1991) and Sipe (2000), who has found positive evidence in their research regarding childrens social interaction and building comprehension, the participating teacher in this study believed that minimal social interaction or little social interaction can keep stude nts focused. Moreover, it 160

PAGE 161

seem s that having order in the classroom enabled teachers to teach and without this order there might have chaos. One of the interviewees, Chi stated how he usually behaved in English classes: Talking is not allowed in the class. We have to sit quietly with intensive note-taking. The teacher always grades our textbooks in the end of semester because she wants to see if we write down what she said in the class. Th e more you take notes on the margin of the textbooks, the higher grade you will get. The above comments reveal that teaching activities in English class te nd to be in the form of teacher lecturing and students note-taking. Part of EFL student s English grades is depending on the amount of notes they take in the textbooks. That is, EFL students they have to record and copy whatever English teachers teach in the class. Social interaction or discussion with teachers and peers, on the other hand, is hardly happened or never been evaluated as part of students grade and performance. Finding # 3 Meaning-making is not valued in reading instruction and reading strategies taught in English classes are all about test-taking strategies. It is common that when reading strategies are taught in English class, they tend to be testoriented. EFL students are traine d to read in an analytical ma nner in noticing language features, not the content itself. As a result, EFL students dr aw their attention to la nguage-based strategies. English reading to them is a pr ocess of decoding, translation, and getting the answers right rather than meaning-making. When interviewees were asked if they were taught English reading strategies before, their responses naturally refer to test -taking strategies. For example, Lin stated that she was taught to read the questions first: My teacher told me to read the questions first. Then go back to the reading text and find the answer out. In that case, you dont have the read the whole passage and you can score high. And, Huai said, 161

PAGE 162

I was taught that the answer for question one is mostly in paragraph one and the answer for question two is m ostly in paragraph two. And for the main idea it is often the first sentence. I think it is a very good strategy be cause I could get the answer right even I dont know what the passage is about. So if my parents or my classmate asked me how was the exam or what was the exam about, I always told them I dont know. But I always get the score which is not bad. Under the pressure of test outcomes and being appreciated by schools and parents, English teachers hardly have time or energy to teach other aspect s of the language but anticipated test items or content. As a resu lt, English teachers would rather teach their EFL students how to score high in the reading comp rehension test. They use teacher-directed instruction with the belief that it is the best for their students. Most of the time, they focuses on testtaking strategies such as di rects EFL students to read the ques tions first and then go back to the text, and check if all the an swers to the comprehension questi ons has been correctly obtained. It seems that the content of the text is not important but the right answer to each question is more important. The strictness of the teaching and learning materials and techniques can also be attributed to the entrance examinations of hi gh schools and colleges, wh ich are held twice per years over the past forty years. This situation encourages EFL students to read the language for testing, not for meaning construction or language acquisition. Finding # 4 Reading materials in the textbook are not interesting to most EFL students. With mandated national curriculum, the read ing texts in the textbook are selected and compiled to best represent vocabulary and the gramma tical features. This kind of reading texts is mostly organized in the order of teaching language structures and not appealing or interesting to EFL adolescents. As a result, it prevents EFL students from developing good reading habits and interests. Several interviewees revealed their desi re to read something they like or are interested in, such as sports and video games for boys and fa shion or celebrity for girls. They all found that reading materials in reading te xtbook are boring. According to Schiefeles (1999) study in 162

PAGE 163

conducting motivation f or reading in L1 setting, he stressed that personal interest is a significant predictor of comprehension and learning from te xts. One student, Xiao, expressed that his interest in reading have dec lined since middle school as evid enced by the following transcript excerpt: I used to like reading when I was in the elementary school. The teacher in Busiban (the name generally referring to the private language institutes in Taiwan) gui de us read a lot of funny short stories with colorful pictures on each page. At that time the vocabulary was what I can handle. But now, I have to study what I am not interest ed in the textbook and read it thorough for the quizzes and exams. Be sides, it is so boring in the class, taking notes and drill practice. Xiaos comment echoed Yoshimuras (2000) res earch findings in that students usually have an intense interests in lear ning at the beginning stage. But gradually, they lose interest due to the monotonous teaching methods of grammar and tr anslation. In Xiaos case, he used to like English and enjoy reading short stories before mi ddle school, but the situation changed as he got into secondary level, not only because of the reading materials but also the learning activities. English teachers should be aware of adolescents growing need for autonomy that they wish they could read something like popular culture in their daily lives. When responding to the questions regarding what they like to read about, high sch ool students all expressed similar sentiment that they want to read something related to their lives This finding is consistent with the previous study (Hurst, 2004; Snow, 2002) as discussed in Chapter Two. S now (2002) and Guthrie et al. (1995) all suggest that reading ma terials of high interest in a pa rticular reading topic leads to high motivation, which also leads to high comp rehension. Therefore, to motivate EFL high school students to read more, it is recommended that metacognitive reading strategy instruction (MRSI) combine with interesting and age-appropriate reading texts. 163

PAGE 164

(2) Why Metacognitive Read ing Strategy Instruction? The previou s section reveals the message th at current English reading instruction in school settings in Taiwan are not teaching EFL students how to read or cultivate them appreciation toward English reading. Most EFL students are passive lear ners or readers with little confidence in their voca bulary and grammatical knowledge. They are dependent readers with little confidence when assi gned to read independently. In addition, both language teachers and EFL students have the misconceptions regardi ng English reading. That is, English reading is a tool for learners to gain linguistic knowledge and is for exams that they have to encounter rather than meaning making or th e resources that they can gain information from. After 10 weeks of implementing metacognitive reading strategy instruction on high school students in Taiwan, the result of this study has ascer tained that metacognitive reading strategy instruction (MRSI) had benefited EFL high school students and th e participating teacher in some ways. Finding # 5MRSI has promoted EFL students strategy awareness, which in turn, increased their willingness to actively partic ipate English reading activities. According to Flavels (1979) theory, metacogni tive reading strategy includes two parts. One is cognitive reading strategi es, that is, general reading st rategies refer to prediction, questioning, clarification, and summarizations. Th e other one is reading strategy regulation. By regulation, it means that a reader, once familiarize d himself with a series of reading strategies, knows how to use strategies strategi cally. In other words, a reader will constantly monitor his or her reading process, and when there is a comprehe nsion breaks down, he or she will take actions to fix the problem. Therefore, simply knowi ng those reading strategi es is not good enough for reading comprehension, metacognitive reading stra tegy instruction (MRSI) fi rst familiarizes EFL students with a series of general reading strategi es and then English teachers have to guide them and demonstrate how these reading strategies work when comprehension breaks down during the 164

PAGE 165

reading p rocess. Later, after the teachers modeling of reading strategy use, he or she will give EFL students ample opportunities to practice this kind of reading st rategy on their own. Last, the teacher divides EFL students into groups and gra dually lets EFL students take control over their reading strategy uses through modeli ng and sharing with peers. In the present study, after gene ral reading strategy instruction and modeling of combined strategy use in several readings in the textbook, EFL students were able to demonstrate their improved thinking skills in English reading process. They were able to predict, to generate questions, summarize, and inferenc e which is the most difficult part for them. As EFL students gain a sense of accomplishment or a sense of su ccess, their self-confidence as English readers developed, so did their interests in reading not only to participate in the small group strategy practice but also to respond for the teacher and peers. The par ticipant teacher noted: was good to see how they st art to volunteer in class to distinguish main ideas from unnecessary details. I still remember clearly th at in the beginning of this study, they were not used to present themselves in front of th e class, and whenever they were asked to do reading strategies in groups, they seemed uns ecured. It took almost three weeks to make them alive in the class One student, Yo, also commented: I think, after this reading program, I began to think more during reading, I dont waste my time in search of unknown words, instead, I th ink more about the content, and I found my thinking ability has gone be yond vocabulary and grammar, and I found English reading actually is not that hard Another student also said: I like the questioning strategy, I never know I ca n question myself or my classmates like that, before I am always the one being ques tioned, I always reminded myself to wellprepared to be tested and to be questioned, it is neat, and I feel like I am in the role of teacher questioning myself. Besides, because I have to question others, so I have to really understand the reading content, or I dont know where to question about. These responses reveal that after MRSI expe riment, EFL students ac tually gain control over their English reading. Furthermore, with deeper mental processi ng, they are able to 165

PAGE 166

understand the content to generate the questions, and thus in tur n, they gain confidence on their own learning. Finding # 6 MRSI encourages invo lvement and social interactions which benefit EFL students of all language proficiency levels. Skilled read ers use reading strategies flexibly and interchangeably to construct meaning while reading (Anderson, 2001; Chamot, 2005). Th ey understand how, where, and when to use each strategy. In contrast, poor readers may use th ese strategies ineffectively or in isolation. They seldom monitor their thinking while re ading, and have a superficial understanding of strategy use. Therefore, by soci al interaction, low proficiency students would benefit a lot when they see or guided by their high proficiency peer s. Many theorists believed that a metacognitive skill starts from expert-novice interaction (Baker, 2004), a view that originates in the social interaction theory of Vygotsky (1978). Th erefore, modeling, dialogue, feedback, and many opportunities for practice need to be introduced into EFL classroom to help EFL students learn through an inductive approach. One of the features of MRSI, re ciprocal teaching, is that it allows students to mutually share their readi ng experience and see how others tackle reading problems through talking and discussing. This approach enhances students reading comprehension competence by directly engaging them with texts, especially through student-to student or teacher-to-student dialogues in the classroom (Harris, 2003). In addition, by sharing and peer modeling, EFL students of lower proficie ncy can benefit from seeing how others tackle the problems in the reading and how they mon itor their reading comprehension successfully. Likewise, for students of higher pr oficiency, their self-efficacy is re inforced when they explained their thinking or try to verbali ze their thinking to thei r peers with lower English proficiency. In addition, this study also demonstrates the fact that EFL students are actually in favor of learning with others. For example, one student said: 166

PAGE 167

What I like most in this program is group shar ing activity. Before, most of our class time was spent on seat work and note taking, and a lot of time I doze off during the class. But group work, on the other hand, it is like our show time, and we talk and discuss when we read along and share whatever come out in our mind, and its fun to see how others brain work. It is apparent that EFL stude nts like this kind of learning environment. They no longer feel reading class is boring and are willing to make full use of learning opportunities with their peers. Social interaction did play an important role in the development of using metacognitive reading strategy. Finding # 7MRSI makes EFL students outgrow from rote memorization to more active learning behaviors and increase comprehension. When EFL high school students were doing group reading, th ere was evidence showing that cooperative learning facilitated active and strategic r eading. Unlike typical and traditional whole-class mode, the use of small group and pair work followed the whole class discussion enhanced EFL high school students involvement and participation. In this study, after twoweek training on familiarizing general cognitive reading strategies and teacher modeling, the participant teacher, Ms. Lin, divided the class into 9 groups with assigned roles and asked them to read a short story Mammoth in their workbook. EFL students we re using their native language, Chinese, in the discussion. As OMalley et al. (1985) point out, native language can be a source of metacognitive reading strategy, es pecially it helps clarify and e xplain some lexical items when interact with self or peers duri ng the reading. It appeared to me that they liked this kind of activity and some comprehension difficulties we re solved while everyone in the group were attempted to comprehend and to self-regulate their l earned reading strategies. The following example was recorded from classroom observation and was translated into English. There was a sentence in the reading task They had tusks that were very long. Th e tusks curled up at the ends. Mammoths used their tusks to scrape the snow off grass and plants so th ey could eat them. 167

PAGE 168

Student A: They had tu sk tusk that were very long. What is tusk ? What is curled ? What is scrape ? Oh, my god, so many vo cabularies in just two sentences, how am I going to guess from the context? Student B: I think its something like ivory. Student C: I think so, too. Becauseyou see, Mammoth s used their tusks to scrape the snow off grass and plants so they could eat them. I think he must use some kind of tool and move the snow away. We know what off means. Students B: and Mammoth is a kind of anim al and the tool they can use to move the snow away should be part of their body. Student A: so it is possible that Mammoth uses its leg to move the snow away, not ivory. Student C: but there are some clues that cant be refer to legs, like, very long and up Student A: hmm, it makes sense. Therefore, tusk means ivory The above example demonstrated that, dur ing group reading, many students applied a variety of reading strategies via small group st rategy instruction and practice and tried to construct meaning from text. In this case, it appeared that EF L high school students were taking turns trying to solve the vocabulary problems, and they finished each others thoughts and constructed meaning together. Furthermore, they challenged and supported each others thinking. Different from their typical classroom activitie s, small group activity has made EFL high school students read with ease. Besides, MRSI made EFL students outgrow from rote memorization. In this case, if the language teacher or EFL students just check the meaning of tusk from the dictionary, then the result is that th ey might just forgot it weeks later. Finding # 8Reading strategy journals are mirro rs to individual EFL students and language teachers clues for indivi duality in language learning. Anderson (1991) carried out a st udy on individual differences in reading strategy use and found that there was no single set of processing strategies that signifi cantly contributed to reading comprehension. That is, there are no so call guaranteed stra tegies for guaranteed comprehension. Therefore, to be a strategic r eader, EFL students need to constantly monitor their own reading process, awar e their problems and modify a nd choose the right strategy to 168

PAGE 169

solve it. Research also stresses that com bined effects of cognitive and metacognitivie strategy instruction were effective in enhancing readi ng comprehension. Teaching just metacognitive strategies and not their connec tion to cognitive strategies doe s not seem to improve reading comprehension (Garner, 1994; ONeil, 1992). Thus, the bridge that connects these two strategies, cognitive and metacognitive strate gies, is through reading stra tegy journal which constantly requires EFL students to self-monitoring or self-re flect on their English reading process. One student commented on reading strategy journal: I think keeping a journal is a good way to monitor and remind us which strategies to use after the assigned read ing. Sometimes you know a lot of strategies, but you just dont remember to use it while reading, therefore, wh en I keep a reading strategy journal, its kind of forcing me to refresh my own reading process, and by rethink my reading process, I also remind myself of some strategies I forgot to use. Research in the area of metacognitive read ing strategy on L2 read ing comprehension has repeatedly stressed the importance of equipping L2 readers with a li st of reading strategies and letting L2 reader use it strategi cally during the reading process. But how? Being exposed to and using reading strategies may not ensure success in language le arning (Skehan, 1989), especially if the learners do not metacogniti vely connect their strategies knowledge with strategy use (Vann and Abraham, 1990). Therefore, one potentially effective way of encourag ing learners to notice or make connection is to have them self refl ect on their own experiences while reading or learning. In addition, as everyone has his or her own learning preference, reading strategy journal can be a way for English teachers to s ee and know how their EFLs read. The belief that learners dont learn the same way or in a consistent manner has been stressed in the previous research, even with the same cultural background (Fu, 1995). By reading EFL students reading strategy journal, English teachers can get some insights and know the problems of their individual EFL students. 169

PAGE 170

Also, L2 reading is a lot more com plex than L1 reading, not only it is linguistically and culturally demanding, but also when there is individuality being involved. Researchs like Block Anderson, and Sarig (1987) all point out that L2 reading process has a high degree of individuality. Therefore, read ing strategy journal is a good way fo r language teachers to detect students individuality. Finding # 9 Once secondary EFL students find English reading is rewarding and enjoyable, their motivation for readi ng and reading engagement boost. Most interviewees (85%) stated that they prefer to choose their own reading materials. However, to measure precisely the extent to which that choice of reading materials improves reading engagement, due to the limited time, is out of the scope of this study. The participating teacher identified an important turning point, she said: I think they really got engaged because I he ard some of them discussing about forming a book club to read Harry Potter for this coming winter break, and some of them gather and vote for the books that they want to read, and it did surprised me because Harry Potter is really long, well, I dont know, I hope that they wont back out due the large vocabulary demand. Reading engagement has been defined by Guth rie and Wigfield as, the interplay of motivation, conceptual knowledge, strategies, and social interaction duri ng literacy activities. (Wigfield et al., 2004, p. ix). Pr omoting reading engagement and reading comprehension are the central tenets on which this experiment is based. In order to attain and maintain these central tenets, there must be support for cognitive stra tegies and knowledge construction during reading and support for motivational development of learners. (Wigfiel d et al., 2004, p.3). The result of this experiment revealed that reading engage ment and comprehension could be enhanced by integrating MRSI into English cl asses along with self-selected r eading materials. Engagement and comprehension in reading are positively linked. It is difficult for a reader to comprehend material from text without certain level of enga gement and in order to be engaged in reading 170

PAGE 171

there is an intersection between reading strategies and m otiva tion. Engaged readers monitor their comprehension by asking questions. M onitoring comprehension is essential when developing in-depth knowledge. Students growth in reading co mprehension is significantly influenced by the amount of engaged reading in which they participate. Substantial evidence suggests that when teachers create an environmen t that supports and allows reading engagement to be extensive and fulfilling, students read ing comprehension and achievement increases (Guthrie & Cox, 2001 as cited in T. Guthrie, Wigfield et al., 2004). Moreover, through metacognitive reading strategy instruction, an interactive t eaching method that can be a motivational learning factor beneficial to EFL st udents literacy development. Because reading competence is embedded within the motivation construct (Alvermann, 2002; Guthrie, 2000), and because explicit strategy instruction and guide practice develop reading skill, reading strategy instruction was taught throughout the intervention, beginning with strategy instruction, teacher modeling, guide practice, and independent readi ng. In the post-treatment group interview, students offer various thoughts about their expe rience regarding metacognitive reading strategy instruction. For instance one student stated: I thought it was fun, it was actua lly more fun than sitting in class and listening to teachers lecture and doing boring things. Another student commented: I thought it was really interesti ng to me because Id never done that before by visualizing your thinking. It encourages me to do more th inking like this. Besides, I feel like to read something not in the textbook. I think I am a better reader now because I found by constantly thinking about my reading pro cess, I understand more, and I remember the content of that article longer. It stays in my mind longer. As the statements above, participants in the experimental groups have been inspired and found the joyfulness of English reading. To them, English class is no longer sitting there and listening to teachers lecture or constantly jotting down in formation on the blackboard or 171

PAGE 172

constan tly being tested in each English class. They were delighted with the group sharing activity and they were happy that they can freely talk about the readi ng materials. And by reading together with the goal of meaning making over vocabulary check ing, they think more, understand more, and feel that they are involved. And as they understand more, the content of the materials stays in their mind longer than be fore. EFL high school students were also be motivated and planned to read th e books they are interested in. Figure 5-1. Overall relationshi p among MRSI, reading motivation, and reading comprehension Conclusion This chapter presented the qualitative findi ngs resulting from a research study based on examining EFL students previous reading e xperience and how metacognitive reading strategy 172

PAGE 173

instruction (MRSI) has changed their perspectives to English reading. T he research questions guiding this study phase were: 1) What are th e factors bring about EFL students reading experience? 2) What MRSI features help EF L students in changing their English reading perspectives? Unlike children who depend mostly on adults and thus try to act to please their teachers, adolescent students are in the cr itical human development stage that they increasingly seek to exercise some control over their circumstances. In that case, if teenagers are unable to gain control over language learning, then there is a chance that they might choose to give up the reading task and giving-up is the way that th ey could control. This phenomenon is evident by several interviewees responses in the study. Moreover, as show n in the interview data, EFL high school students in this study we re all aware of their English reading problems or difficulties, they seemed to lack of a strategic plan to overc ome these obstacles in English reading. Therefore, awareness of which reading strategies to use a nd the ability to select appropriate strategies depending on the problems encountered duri ng reading are the goals of this study. After 10-week of metacognitive reading strate gy instruction, EFL high school students expressed their positive reaction toward the intervention. They are a lot more confident than before when approaching English reading and will ing to engage more reading during the winter break. One of them even mentioned that she would like make use of metacognition to other subject areas. In addition, the pa rticipating teacher was also infl uenced by metacognitive reading strategy instruction and further made her recons ider her teaching approaches. This empirical study suggests that when EFL students receive e xplicit instruction in metacognitive reading strategy, their get more positive attitudes on Englis h reading and their reading comprehension get better. Based on the above information, I construct the following model to illustrate the impact of 173

PAGE 174

m etacognitive reading strategy in struction on EFL high school students English reading. (Figure 5-2) 174

PAGE 175

Perception of English reading Acquire language knowledge Test-oriented Negative Reading for meaning More positively Reading Comprehension Read for answer comprehension questions Dont know what the content is about Read for test: focus on preposition, transition words Know what is read and content retain longer in mind Focus on the meaning instead of read for test Reading Motivation Wont read except teachers requests or guidance Lack of confidence in reading Gain confidence in reading Plan to read for pleasure on self-selected reading materials Reading Strategy Awareness Passive decoding and grammatical analysis Bottom-up Precisely understand each word in the sentence Active meaning construction More critical thinking involved Learn questioning and overall understanding Figure 5-2. The im pact of metacognitive reading strategy instruction on the change of EFL adolescents reading comprehension, r eading motivation, and reading strategy awareness 175

PAGE 176

176 CHAPTER SIX DISCUSSION AND CONCLUSION Overview of the study This study was designed to investigate the e ffect and influence of metacognitive reading strategy instruction on high school EFL students reading comprehension, reading motivation, and reading strategy awareness. The theoretical framework for the study has been established by the review of relevant literature in Chapter Two. In Chapter Two, literature review indicated that it is hard to draw generalizati ons across studies because of th e large variations in the way metacognitive reading strategies have been em ployed. Therefore, this study is viewed as explanatory in the field of met acognition in EFL reading process. In this study, an explanatory sequential mixed method research design was conducte d. That is, this study first implemented an experimental design and followed up with small gr oups interviews. In the experimental phase, there was a control group a nd experimental group randomly selected from the whole 11th grade in a public high school in southern Taiwan and were randomly assigned to the experimental and control group. In total, there were 110 high sc hool students participated and completed in the study with 56 students in the experimental group and 54 in the c ontrol group. Three measurements were administered at the beginning of the school year and the sample was statistically ascertained th at these two groups were similar before the experiment. In other words, no preexisting difference between these two groups were found in te rms of their English reading proficiency, reading motivation, a nd reading strategy awareness. The experiment lasted a 10-week period cl ass consisting of 50 mi nutes of metacognitive reading strategy instruction (MRSI for short) pe r section. Instructional time for the reading strategy instruction was allocated in the nap time from 12:30 pm to 1:20 pm and two times a week. Participants were all taught with genera l reading strategies (w ord attack, summarizing,

PAGE 177

question ing, and predicting). Moreover, th e experimental group was given metacognitive reading strategy instruction with a focus on c onsciousness raising and se lf-monitoring during the participants reading processes in the study pe riod. On the other hand, for the control group, they didnt receive the training on metacognition in their English reading processes. I worked with the participant teacher and co-planned the activities for the instruction. In the 10th week, three posttests were administered to all the subjects in the cont rol and the experimental group to investigate the effectiveness of metacognitive reading strategy instru ction on EFL high school students reading comprehension, reading motivati on, and reading strategies awareness. In the end of study, I also interviewed 24 students from the experimental group w ith an intention to understand their English reading experience and how their English reading attitudes changed through MRSI intervention. This chapter will first start with the findings of the quantitative data and later qualitative data. Then discussion regarding this study will be presented later. Quantitative Findings In quantitative analysis, ANCOVA was used to test four research hypotheses H01 to H with the p retest score GEPT representing as a co variate while three posttest scores representing as dependent variables. The findings of each null hypothesis are briefly presented in the following order. 04H Reading comprehension will be the same for high school s tudents with metacognitive reading strategy instruction with those wit hout metacognitive reading strategy training. 01As shown in the quantitative results in Chapte r F our, this hypothesis was rejected because there was a significant effect of treatment on post reading comp rehension test. (F = 88.83, p<.05, df =107). This result was also consistent with empirical evidence that supports the implementation of metacognitive reading strategy instruction on improving reading 177

PAGE 178

com prehension in both L1 and L2 reading research (Anderson, 1991; Dole, Brown, & Teathen, 1996; Salataci & Akyle, 2002; Song, 1998; Lorang er, 1997; Brown and Palincsar; 1985). In addition, a multiple regression analysis indicate d that the treatment, metacognitive reading strategy instruction, has an e ffect on all language proficienc y levels. In other words, metacognitive reading strategy instruction could increase EFL high school students reading comprehension, regardless their English proficiency levels. In summary, after 10-week MRSI training, the experimental group outperformed the control group in English reading comprehension. H Motivation to reading English will be the sa m e for high school students with metacognitive reading strategies trai ning with those without metacogniti ve reading strategies training. 02The quantitative data related to this pe rspective was collected via Moris (2002) English Reading Motivation Questionnaire (ERMQ) This questionnaire was ad ministered two times as preand postERMQ. The quantitative data analysis demonstrated that significant difference was found for the postEnglish Reading Motivation Questionnaire (F=17.187, p<.05, df =107) regarding EFL students motivation toward English reading after metacognitive reading strategies instruction. In other words, the treat ment of metacognitive read ing strategy instruction did make a significant difference between two groups in terms of th eir English reading motivation, especially in the s ubtests of the intrinsic motivation and reading efficacy. Even though research has advised the implementation of metacognitive reading st rategy instruction in improving students reading ability, few empirical studies have been conducted in an L2 context to examine the effects of metacognitive reading strategy instruction on reading motivation in the EFL context. To be specific, as the data indi cate, the participants in the experimental group gained higher score in two subtests: 1) intrinsi c reading motivation, and 2) reading efficacy. In other words, the EFL high school students were mo re motivated than before and they were more 178

PAGE 179

confident regarding English read ing. The findings of this stud y provided empirical support for the positive im pact of metacognitive reading st rategy instruction on EFL high school students motivation toward reading in English. H Perception about reading strategies will be the sam e for high school students with metacognitive reading strategies training with those without me tacognitive reading strategies training. 03The quantitative data related to this perspec tive was collected via M okhtari & Sheorey (2002) Survey of Reading Stra tegy questionnaire (SORS) and the questionnaire was administered two times as preand post SORS. The quantitative data analysis demons trated that significant difference was found in postSurvey of Reading Strategy questionnaire (F=10.28, p<.05, df =107) regarding EFL students reading strategies awareness in English reading after metacognitive reading strategy instruction. Similarly, the significant differences were found in the subtest in Global reading strategy and Suppor tive reading strategy. As metacongnitive reading strategy instruction expected EFL students to have a whole understanding of a reading task by questioning and summarizing, EFL high school student s think more than before and thus their scores in Global reading strategy (GLOB) sectio n were higher than before. Besides, group sharing and interaction were encouraged in metacognitive reading st rategy instruction, EFL students have a positive learning environment and thus their score in S upportive reading strategy section (SUP) was higher than pretreatment. In summary, the treatment, the metacognitive reading strategy instruction (MRSI) has effects on global reading stra tegies and supportive reading strategies. And this result indicating that, by integrating the metacognitive reading strategy instruction (MRSI) into English reading classes, EFL high school students were involved in the reading activiti es through cooperative and group learning activities and they think more and deeper while reading. 179

PAGE 180

H The effect of metacognitive strategy traini ng will be large r for students with lower language competence. 04Previous research has found that m etacognitive reading strategy benefit more for students with lower English proficiency levels (Corde ro-Ponce, 2000; Kern, 1989; Jimenez and Games, 1996; Song, 1998). The results of inte raction of treatment by various language abilities were not at a significant level on each dependent vari able. Therefore, there was no enough evidence to accept this hypothesis. In other words, the e ffect of treatment didn t make a significant difference on diverse reading abilities. That is to say, metacognitive reading strategy instruction (MRSI) has similar effect on EFL high school students of different language proficient levels in this study. In other words, regardless of their diverse language proficienc y levels, metacognitive reading strategy instruction (MRSI) can actually help all of the EFL secondary students of varied language proficiency levels. Different from their traditional English learning and teaching climate, EFL students of all proficiency levels pr ofited from this instruction because they were able to explore different reading strategies, evaluate readi ng strategy effectiveness, and eventually choose their own set of effective reading strategies according to their own learning styles and thus EFLs have almost equally am ount of improvement in terms of their English reading comprehension, reading strategy awareness, and Engl ish reading motivation. Qualitative Findings The qualitative data analysis revealed abunda nt information regarding EFL high school students previous reading experience and its eff ect on their reading processes. The findings told us that EFL high school students reading experiences were clos ely related to the classroom environment (Day and Banford, 1998) as well as th e teaching methods and focus, especially in an EFL context where language input and exposure to reading are limited (Yoshimura, 2000). 180

PAGE 181

Such past reading experiences had not only infl uenced EFL students English reading behavior, but also their attitud es towards English readi ng. Qualitative data in dicated that English education in Taiwan mostly cons ists of the following characteristics: test-oriented, grammar and vocabulary are stressed, and teacher-centered. For instance, the qualitative data indicated that EFL high school students English teachers spend time focusing on linguistic knowledgetranslation and grammatical analysis. Therefor e, this kind of wordby-word translation and sentence-by-sentence analysis have influen ced on EFL high school students English reading habits. A typical routine for an E nglish reading class in Taiwan is like this: 1) EFL students pick out the unknown words first, check the meaning out in a dictionary, then go back to the reading text.; 2) then they begin to analyze sentence structure, circling all the subjects and verbs, underlining adjectives and adverbs, highlighting preposition or pre positional phrases ; and then 3) they begin to read with clues from translati on and grammatical structures. After reading, the language teacher asks several literal comprehe nsion questions to ensure that EFL students understand the content. No further actions, su ch as probing deeper meaning construction or critical thinking skills such as how do you know and why, were involved. As a result, English reading habits like this have directed EFL high school students to read on a word-byword basis at the expense of the broader understa nding of the content of the material they are reading. English reading, to most EFL high school students, is time -consuming and laborious process, and thus the hold negative attitude toward English reading. As EFL students progress through school, with the increasing amount of reading content and vocabulary as well as increasingly complicated sentence structur es, EFL students begin to struggle with the demanding English reading text and gradually lost interest in reading. 181

PAGE 182

However, as data revealed, som e EFL high school students were actually aware of their English reading problems and difficu lties, but they seemed to lack a strategic plan to overcome those reading obstacles; they also blamed th emselves for not having enough vocabulary and grammatical knowledge. Furthermore, data also revealed that a lot of factors seemed to discourage EFL high school students from reading w ith joy and independent reading. In addition to the ineffective reading strate gies, there was evidence of othe r factors that bring about EFL students English reading. For instance, Englis h teachers instruction methods, test-oriented instruction, teacher-centered approach, and borin g reading materials. Metacognitive reading strategy instruction (MRSI) is a teaching approach which stresses explicit reading strategy instruction, teacher modeling, small group interaction, and journal reflection. This research proved that MRSI can offer EFL high school stud ents with positive experiences toward English reading in English classes. It is also delightful, after the ex periment, to find that most EFL students gained confidence toward their reading pr ocess. They felt that they had more control over their reading, and their reading motivation in creased. As they were motivated, they were willing to try further processes outside of a classroom setting. For instance, at the end of the study, they were planning to read something toge ther outside of the class. MRSI helped EFL students develop confidence and efficacy in readi ng, thus impacting their reading motivation so they engaged more reading. Moreover, qualitative results from this study indicated that real world interaction and interesting texts promote active student engagement in the English reading process. For example, at the outset of the experiment, the participants in this study we re initially not comfortable in asking each other questions and answering each ot hers questions and especially uncomfortable with questioning their peers or speaking out loud about their thinking for sharing. As the 182

PAGE 183

experim ent progressed over 10 week period, EFL high school students gradually demonstrated a noticeable difference in their comfort. They we re more involved and gained confidence in sharing reading strategies they used in all the assigned reading. Therefore, the findings suggest that teachers in the field of language teaching should reconsider traditional teacher-centered instruction which has been isolated from real learning and positive experience, and increase students interaction in the classroom. In conclusion, learning builds on experience a nd learning is internally motivated, taskand problem-centered (Vygostky, 1978). Results fr om the interview data after the experiment reveal that MRSI increased EFL students productivity, in creased students motivation, decreased students frustration, a nd increase higher retent ion rates, when self-reflection reading activity is allowed. This study adds to th e growing body of literature suggesting that metacognitive reading strate gy instruction (MRSI) can increase not only secondary EFL students reading comprehension, but also tran sform them to be more motivated and more engaged in English reading. Discussion This section examines my findings within the context of the current literature on metacognitive reading strategy in struction. I begin by discussing the areas of my findings that are slightly different from the research in EFL reading comprehension instruction. I then reexamine the major features of MRSI presente d in Chapter Two and discuss where my findings are situated within the context of the MRSI literature. I also reflect the lesson I learned in this study and present my recommendations for improvemen t if the intervention should be replicated. This chapter closes with limitations of the st udy and ongoing questions for future research. 183

PAGE 184

Findings in the Context of the Literature 1. Metacognitive reading s trategy instruction increased the reading achievement of secondary EFL students over a short period of time. While existing research says that metacogni tive reading strategy in struction takes long time to influence students reading comprehension (Block, 2002), my findings show that explicit instruction in metacognitive read ing strategy instructi on over a 10-week period can also increase secondary EFL learners reading comprehension. In addition to measureable gains on their pre/post reading comprehension assessment, EFL st udents in this study were able to show the evidence of their increasing met acognitive strategy awareness as the study went on. It has been conventional belief that such changes in a shor t period of time are unlikely. This study had a different result. In two and half months of MRSI intervention, this study demonstrated a statistical difference in reading comprehension of most secondary EFL students, regardless of their L2 language proficiency. Moreover, it was skeptical that MRSI could improve EFL students reading engagement or motivation in the first place, but this study revealed that once EFL students are able to take ove r their learning process and when English reading to them is no longer decoding and sentence structure analysis, they are actually willing to engage themselves into the reading materials of their individual interests. 2. Metacognitive reading strategy instruction has positive effect on EFL students of difference proficiency levels. Metacongitive reading strategy instruction stresses familiarizing EFL students with a list of different reading strategies through teach ers explicit explanation of their importance, teacher modeling and planning, comprehension monitoring, and effectiveness evaluation based on different reading task and indi vidual needs because differences in strategy use vary according to the tasks and language proficiency (Cha mat, 2004; 2005; Cohen, 2003). In other words, metacogntion is the awareness that EFL students choose their own set of reading strategies to 184

PAGE 185

cope with their diverse reading problem s. As a result, regardless of their difference proficiency levels, EFL students all demonstrated improving scores on their reading comprehension, reading awareness, and reading motivation after they learned how to self-reflect and to self-evaluate their reading process. 3. Metacognitive reading strategy instructio n also has positive e ffect on language teachers. Very little research ex ists that explain how language teachers learn about how to teach metacognitive reading strategy instruction in their teacher preparation programs. Studies in L1 language shows that teachers struggle with this abstract concept and have difficulty providing instruction in this area (Block and Pressley, 2002). However, Ox ford et al. (1990) claimed that the metacognitive reading strategy instruction (MRSI) training has some positive effects on teachers: Teachers who use strategy training often b ecome enthusiastic about their roles as facilitators of classroom l earning. Strategy training makes them more learner oriented and more aware of their students needs. Teachers also begin to scrutinize how their teaching techniques relate or fail to relate to their students learning strategies and sometimes teachers choose to alter their instructional patt erns as a result of such scrutiny (Oxford et al., 1990., p. 210) Data from this study revealed that, the part icipant teacher, despite the fact that there was only one evidence in the data, has begun to aware of and reth ink her own teaching approaches as a result of this study. As Ms. Lin said: I realized that I had a very superficial unde rstanding of metacognition, what it really means and how to implement it into my teach ing, I remembered I read it in textbooks, heard about it in psychological class, maybe, back in colleg e, but I never know students need this knowledge in order to be actively involved in their own learning. After all, they are the ones who have the responsibil ity of their learning and whosuffer under educational pressure (laughing), I think I know, as least, I know I should challenge students thinking more instead of asking them passively to memorize whatever in the textbook and whatever I taught. Now, I thi nk a real language teacher should not only teach linguistic aspect knowledge but teach language learners how to learn and explore more on their own. 185

PAGE 186

In this study, Ms. Lin showed her personal growth in term s of m etacognition in reading strategies at the end of the st udy, and this point is important to L2 reading research. Now that language learning and teaching is socially sh aped (Vygostky, 1978), when language teachers self knowledge and perceptions change, improvement in their ESL/EFL students in terms of reading skills, reading comprehension, a nd reading motivation can be expected. 4. Metacognitive reading strategy instructio n should be integrated into language classroom. There are issues of whether strategy instru ction should be integrated into and taught concurrently with the language course, or whether to provide a separate how to learn course independent of the language course (Chamot, 2005). Although the studies reviewed in chapter two all included strategy instructi on as part of language class, it has been argued that strategies taught in a language class are less li kely to transfer to other tasks and that it may not be practical to prepare all language teachers to teach strate gies (Gu, 1996). This study reveals evidence that if reading strategy instruction is integrated into a regular EFL classroom, there is possibility for EFL students to transfer metacognition to othe r subject areas. One female student commented in the interview saying that she felt metacogn ition helps the reading content she read stay longer in her mind (that is higher retention) and she said she pla nned to use metacognition with other subjects. Despite the fact that only one such comment appeared in all interviewees and might not be representative and hardly be gene ralized to the whole popul ation; however, as the students being interviewed were relatively small in this study, such evidence also demonstrated that there still stands a chance that metacogn ition can help EFL students in other learning process as well. 186

PAGE 187

Reexamine Key Features of MRSI As previous discussed, four findings of this study provi de evidence for existing L2 research on EFL learners reading and the findi ngs about the core features of MRSI that confirmed existing L2 reading res earch. In Chapter Two, I discuss the major features of MRSI is reciprocal teaching, peer shar ing, self-reflection and self ev aluation. Through group learning, EFL students gradually take responsibilities on th eir own learning. Moreover, MRSI offers EFL students opportunities to monito r and self-reflect on their r eading practice by examining the effectiveness of their strategies. In this following section, I reexamine each of these key features of MRSI in the cont ext of my findings. 1. Dialogues between teachers and students This study was also conducted within the fram ework of social constructivism, a theory of learning theory, which emphasizes th at students must interact with new concepts or new learned skills in both the physical and social world in order to make meaning and understand others as well as being understood (Fosnot, 1996). Cons tructivism is not a teaching method but an educational theory that encourages questioning, problem generating, and problem solving. That is, constructivist teachers design tasks and learning environment that allow students to question, internalize and reshape, or tr ansform new information (Brook s & Brooks, 1993). According to Vygotskys theory, certain metacognitive actions such as monitoring ones own learning along with all higher-order cognitive functions, such as analyzing and synthe sizing, are internalizing through social interactions with more competence adults or peers who provide the learner with scaffolding, that is, assistance th at is gradually removed when no longer needed by the increasing independent learner. EFL high school students commented that they like MRSI, especially group sharing and group work, bette r than just seat work. Under an encouraging and supportive 187

PAGE 188

learning atm osphere, EFL students gained positive experience in helping others, self-growth, and as independent learners. 2. Monitoring and Reflection Both language teachers and EFL high school students should spend some time to pause and reflect on their own teaching or learning process. Reflection can help EFL students generate questions from their reading experience, monitor e ffectiveness of the strategies used, and result in modification of certain strategies. Meto cognitive reading stra tegy instruction (MRSI) encourages EFL high school stude nts to constantly reflect on and examine their own reading process through reading strategy jo urnal keeping and peer sharing. Engaging constantly in active reflection process provides th e very foundation of thinking an d learning, either for English teachers to reexamine their teaching instruction or for EFL students to experience their learning situations. Lessons Learned in this Study 1. EFL students prefer more inte ractions than seat work Constructivists philosophy has asserted that learning is an active, ongoing and meaningmaking process through which student s are able to construct thei r own knowledge based on their current or prior experiences (Brunner, 1986). Prin ciples of cooperative learning as a model of teaching are suggested to enhance intellectua l discovery, particularly as related to metacognition in reading strategi es. In the interview, EFL high school students all agreed that they enjoyed a lot in metacognitive reading stra tegy instruction not only because it was novel to them but it was learners-centered. They mentioned that they appreciated the opportunity to work together with their classmates, they were happy to see how their peers solve problems, find the main ideas, guess meanings of unknown words, question and summarize. As Huang explained: 188

PAGE 189

It was a great tim e to get into groups with our friends and talk about things in the reading and think. We have never had the chance to do it before, a lot of class time is devoted to teacher lecturing as well as endless note-tak ing and quizzes. I know that our teacher works very hard, before she always, like quick, let s hurry up, we have to cover this and that because term exam is coming, and time is running out, Harrison (2002) argues that coll aborative pair and group work is a valuable turn point step in shifting from the teacher as expert during the modeling phase to the point at which learners are able to use the strategies indepe ndently (Harris and Gren fell, 2004). The importance of a social community in which meanings are constructed together a nd reading strategy are shared, is also underlined in L2 studies by D onato and McCormick (1994) and Lehtonen (2000). Moreover, collaboration is a key element in creating and sustaining group motivation (see Drnyei & Csizer, 1998). Takeuchi Griffith, and Colye (2007) ma intain that language teachers should set the language classrooms as strategic learning communities where learners will be encouraged to make a maximum use of different reading strategies. 2. The need for language teachers professional development In order to control for the teacher effect of the experiment, I, as a researcher, avoided being the teacher of the experimental group and in vited Ms. Lin as my co-investigator. For that reason, I had to share and explai n to her how the experimental procedure should go. I designed all the materials, set the sche dule and explained the reading st rategies thoroughly, and how they should be taught and modeled to the participant EFL st udents. While I was trying to plan the whole reading instruction program with the participant teacher at the beginning of the experiment, she asked me: Metacognition is a very abstract concept to me, let alone teach my students. How am I going to model this to the students? What do you mean by think about thinking? I have heard this idea a lot in college. How are you going to e xplain to students that? How are you going to make it concrete to EFL student s? I personally, being honest, really dont know what it is, how am I going to teacher them to use it? 189

PAGE 190

From this response, I realize that some language teachers are not really having sound foundation or knowledge base about metacognitive reading strategy inst ruction. Thus, language teachers should be trained in strategy instru ction and how to implement strategy instruction inside their classrooms. Also, there is a need for and policy makers or administers to offer learning opportunities to language teachers, such as workshops. Only when they have command of their knowledge with regard to metacognitiv e reading strategy would they know how to teach it to their EFL students. 3. Let EFL learners choose thei r own reading materials The EFL students interviewed all preferred to read what they want to read. They all agreed that the reading materials in the textbook were not interesting and longed to read something related to fashion, pop culture, life, experiences, or whatever was in their area of interest. This study, due to the time constraint, failed to offe r EFL high school students opportunities to select their own reading materials. Howe ver, according to the participan t teacher of the experimental group, she overheard several groups of students considering form ing an English reading club over the winter break. We are not sure whether MRSI inspired them to form book clubs or if it was due to the joyful feelings gained from readi ng discussion in this study that motivated them to do further reading activities. Though the outcome was left unexplored, th e students plans still offer insight for language teachers that reading mate rials used in the class are really important, as they might reduce or increase EFL students reading motivation. This point is also consistent with the previous study conducted by Snow (2002) and Hurst (2004) that reading materials of high interest in a particular topic leads to hi gh motivation and thus lead to high comprehension. 4. Individual learning styles, pref erence, or learner resistance One of the interviewees in the experimental group commented that he didnt think the reading strategies taught in the class was helpful to him. When the researcher asked further about 190

PAGE 191

how he read, he just answered, I have m y ow n way of reading, and he refused to reveal anything more. Obviously, there was a sign of lear ner resistance in this MRSI experiment. The reason behind this might be that there was a mi smatch between the goals and expectations on the parts of the teacher and the student. Canagara jah (1993) and Tsang (1999, cited in Huang, 2006) investigated the issue of learne r resistance in different socio-cultural and political contexts and both identified a link between resistance and produc t-oriented learning. Ts engs (1999, cited in Huang, 2006) study also suggests that reflective learning might not be well received by learners in an examination-oriented educational system because learners did not feel the type of autonomy offered to them would bring about any favorable changes in th eir learning or they didnt ask for autonomy but for short-term visible re sults. In addition to learner resistance, there might be the possibility of indivi dual learning styles or preferences. It could be that the resistant students might not have found the reading strategies taught in the study helpful in meeting their goals or needs. Therefore, language teacher s should be aware of EFL students individual learning styles or preferences before and during metacognitive reading strate gy instruction implementation (Anderson, 2002; Ethman, Leaver, & Oxford, 2003) Limitations and Recommendations for Future Research As I was trying to interpret the results of quantitative and qualitative data, several questions and doubts came into my mind, which could be the directio ns for future research. Also, some limitations of this study need to be addressed. First, since this study was conducted in southern Taiwan with adolescent EFL students, it is hard to determine whether the results obtained from the study can be generalized to ES L context in other coun tries, whether in the USA or elsewhere around the world. Even though it was a significant challenge for me to seek permission from school and administrators, teache rs, parents, and students consent for this study, research of this kind is necessary to the fi eld to study the effects of metacognition reading 191

PAGE 192

strategy instruction on real students and future studies can expand across sections with ESL or EFL learners or languag e teachers of various native languages. Second, secondary EFL students L1 literacy sk ills were not examined in this study. However, research literature (Alderson, 1984; Bernhard t, 2005; Clarke, 1979; Koda, 2007) in the L2 reading field has pointed out that there is a transfer and in teraction effect of ESL or EFL learners L1 literacy skill on L2 reading skills, and this unexplored ar ea in this studyEFL students L1 literacy skillthus leaves a question mark about wh ether L1 reading strategies or reading skills transfer to or interfere with L2 reading, as what current research on bilingual populations shows (Lindsey, Manis, and Baile y, 2003; Alderson, 1984). Therefore, another possible direction for future research could us e a more comprehensive model including both L1 literacy levels and L1 reading st rategies and L2 proficiency leve l and L2 reading strategies in order to explore the differential roles of L1 and L2 in predicting L2 reading comprehension. Third, as the existing literature has shown, most studies on MRSI used statistical method in investigating its effectiv eness on reading comprehension. Only few studies regarding metacognitive reading strate gy instruction on reading comp rehension of ESL/EFL were qualitative-oriented (Jim nez, 1997). Therefore, more qualita tive research on this aspect is needed in this field. Despite little evidence in this study, there existe d facts of individual differences (such as individual l earning styles and reading strategy transfer), which can only be examined through qualitative methods. Therefor e, further research of this kind should incorporate case study and ethnogra phic methods to get better insight of overall findings for groups of learners and the effect on individuals w ithin the groups. Fourth, more attention should be paid to language teachers professional development on the training of metacognitive reading strategy inst ruction. Meanwhile, more research should also 192

PAGE 193

be done so that adm inisters and policy makers acknow ledge that it is necessary to form effective learning communities in a professional climate a nd further provide supports to language teachers in the implementation of metac ognitive reading strategy model. Fifth, time is a limiting factor in conducting research in public school classrooms. With mandated curriculum and the pressure to pass th e college entrance exams, language teachers have limited time to freely teach students what they think might help students in language learning. Also, whether there would be a pos itive outcome when EFL students are giving opportunities to choose or read what they want to read based on their interest was unable to be covered. Sixth, whether this kind of metacognitive read ing strategy instruction (MRSI) has a longlasting effect on secondary EFL students Eng lish reading process is left unknown, either. Research reviewed in Chapter Tw o has indicated that cognitive re ading strategy has a short-term effect while metacognitive read ing strategy instruction has a long-term effect on students reading achievement. Therefore, fu ture longitudinal research, such as delayed post-test or followup questionnaires, on metacognitive reading strategy instruction and its effect on language learners reading comprehension a nd reading motivation are desired. Finally, a true experiment faces a lot of ch allenges, especially when an experimental research is conducted in public schools. As I did my best to control for the external validity of this experiment, I failed to control for its in ternal validity, that is, the instructors of the experimental and control group we re different. Therefore, futu re research conducting this kind of experiment should try to take th e internal and external validity in terms of the instructors into consideration. 193

PAGE 194

Pedagogical Implications This research study has helped highlight some important pedagogical values for EFL teachers and EFL students. Carrell (1998) explains that skilled readers learn how to do this complex thing we call reading by doing it repeated ly, over long periods of time, with a lot of different text, and lot of different opportunities to practice applying strategies, and monitoring their process and evaluating the effectiveness of di fferent strategies for themselves in different reading situation. (p.17). Traini ng on this aspect should be a long term project for EFL students to become conscious of the reading strategi es and use them independently based on their personal learning preference and needs. 1. Values for language teachers It is a fact that each learner within the same classroom may have di fferent learning styles and varied awareness of the use of strategies. The more the language teachers learn about EFL students individual difference, the more the language teache rs gain a sense of how many different ways they can understand the comple x system of language learning and teaching (Ehrman et al., 2003). Just as Cohen (2007) and Macaro (2001) put it, only when the language teachers know what strategies students are usi ng and how they are using them in different contexts, can they better understand the sour ces of EFL students problems with reading strategies and be able to d ecide on students learning needs and adjust teaching procedures accordingly. Therefore, language teacher are sugges ted to reflect on their in struction and to learn which strategies their students use. Sometimes language teac hers are not aware of or are mistaken about their students use or preferred us e of reading strategies. Uncovering information about their students strategies use may provide so me useful insight for language teachers to change or modify the way they plan, teach and in teract in a timely manner with students. This point is particularly important for language teachers with large-size classrooms. With a large-size 194

PAGE 195

class, Eng lish teachers might not have time to di scuss each students reading strategies uses. Instead, language teachers can ask their students to keep journals to reflect either on the reading content or reading strategy that used or even how they solve their read ing problems encountered in the reading process. Reading their journals offer language t eachers insight as to how their students read. In addition, teachers written comm ents on reading strategy journals not only give EFL students encouragement on stra tegy use but also offer them mo tivation to continue explore more reading strategies. Moreover, this kind of one-to-one written communication, in addition to respecting individual diversity, al so help EFL students receive appropriate support for diverse learning styles. Reading strategy journals provide access to the often-hidden processes that EFL learners use to accomplish their reading goals. What is more, metacognition can be integrated into even the earliest stages of reading instruction (Clay, 1991). Providing children with instruction in metacognition and comprehension-monitoring before they are flue nt readers can help them develop ways of responding to text that will serve them well in future reading endeavors (Baker, 2005). Teachers should also be aware that rote or drill practice and memorization skills will keep students from developing cognitive skills such as problem solv ing, critical thinking, and effective communicate. If students are denied the opportunities to develop such advanced skill, it will place them in an increasing disadvantage group in the changing economy of the 21st century. Exam-oriented instruction will be trading long-term benef its to EFL students for short-term gains on standardized tests. If EFL students are motivated to read more, then their language ability will improve and thus they will do well on all kinds of exams. The implications for teaching are that EFL high school students need to explore diffe rent reading strategies, experimenting and evaluating their effectiveness, and eventually choosing their own set of effective strategies, 195

PAGE 196

accord ing to their individual needs. In addition, all EFL readers can profit from learning how to use metacognitive strategies to plan, monitor, and evaluate themselves throughout their reading process. Another key aspect of metacognitive reading strate gies is the ability to facilitate transfer and promote greater learner autonomy. Research in L1 settings has shown that strategy transfer is often difficult, but that exp licit instruction and the developm ent of metacognitive awareness promote strategy transfer (Chomat, 2004, 2005; Cohen, 2003; Harrison, 2002). Even though there is limited research on transf er of strategies in second lan guage setting, but the qualitative data in this study, despite only one in number, revealed the ev idence of the possible strategy transfer to other learning tasks from one of the interviewees, and this instance provided insights for language teachers to realize the important role of metacognition in EFL readers learning processes. Studies on both L1 and L2 learning have consequently placed increasing emphasis on developing metacognitive understanding rather than simply teaching discrete cognitive strategies (Harris & Grenfell, 2004). Because there are differenc es in learners style and preferences, it is necessary for language teachers to be aware that not all the reading strategi es fit all EFL students. Instead, promoting metacognition in strategies employment and a llowing language learners to figure out the most effective ones for them is th e best. Harrison (2002) further summarizes the research evidence that learners who have a cons cious awareness and cons cious control over their learning strategies can apply that knowledge in new learning cont exts and learn more than those who have not been taught any strategies. English teachers roles consist in not only teaching EFL students language and culture, but teaching them how to approach English reading tasks in and out of the classroom. The information and result of reading strategies in this study will prove 196

PAGE 197

very useful for languag e teachers who want to he lp their EFL students become more independent and empowered learners. When EFL students engage in reflecti ng upon their reading strategies they become better prepared to make conscious decisions about what they can do to improve their learning in terms of reading. Strong metacongitive skills empower learners. This empowerment not only improves Eng lish learning but also transf ers to other aspects of EFL students performance and lives. 2. Values for EFL students The value in making metacognitive reading st rategies known to EFL students consists in that it can help facilitate their positive Englis h reading experience and ga in control over their own learning. Reading strategy journals, part of the experiment of this study, provide a written record and give EFL students opportunities to th ink about how they perceive their own reading processes. Reflection on how they read gives them more awareness, which lead to more control over their learning. The strategi es the readers used often depend on their age, personality, preferred leaning styles or purpose for reading. By keeping read ing strategy journals, they are encouraged to think about which kinds of strategi es they use more often and to perhaps consider adding new strategies or rarely us ed strategies to their repertoire s as they become more aware of other ways to deal with a text. Through c onstant reflection can in crease EFL students responsibility for their own learning which by th e way facilitates EFL students autonomy not only in learning English but also in other subjects. In addition, Vogotskys (1986) sociocultural perspectives emphasize that peer sharing and collaborative learning in classroom can lead to more effective learni ng and learner efficacy. This point is further supported by this study. Following Confucian learning tradition, where the teacher is highly respected and expected to pe rform a teachers duty, EF L students in Taiwan may find it is difficult for them to accept the le arning activities to be dialogic among teachers 197

PAGE 198

and their peers. However, the result of this study dem onstrated that adolescent EFL students actually like this kind of learning environment, and adjust will to it. Small group discussion and sharing in MRSI has made EFL students gained different perspectives from their peers on thinking while reading together. By dialogic in teraction, they challenged one another and support one anothers thinking. Conclusion The purpose of this research study was to fill th e gap in the literature linking theories and practice in the field of adolescent EFL metac ognitive strategy instruction. The results are supportive in many ways. First, this study provides evidence that metacognitive reading strategy instruction, which has been pr oven effective for L1 students, can also improve ESL/EFL students reading comprehension and reading motivation at all levels, quantitatively and qualitatively. The prominent feature of this study is that while most studies on metacognitive reading strategy instruction measure the success of ESL/EFL students reading comprehension solely by quantitative reports, this one has tried to balance the statistical findings with the qualitative explanations of how metacognitive reading strategy instruction and EFL students learning experience impact their EFL reading ex perience and how EFL students perceptions of the learning process changed over the period of the study. Second, one major characteristic in this study lie s in that it explored the effects of reading strategy journal as a self-reflection and an ev aluation tool for EFL students to examine their reading strategy use in each reading task. The process of wring it down has made the adolescent EFL students actually see their thinking in the reading. Mean while, by talking and group sharing, it also encouraged EFL students to make their thinking process visible to their peers. The results indicated that the use of explaining out loud how they tackle the reading problems was effective not only in improving EFL students readi ng strategy awareness but also 198

PAGE 199

in lowering their reading anxiet y and reluctance. In additi on, by speaking out their thinking process, EF L students learned from each other an d learned how to modify their reading strategy when problems arose during their reading. Third, the present study also explored the effect of metacognition on EFL secondary students motivation towards readi ng in English. The results indicat ed that the integration of the metacognitive reading strategy instruction (MRSI) was effectiv e in improving EFL students motivation towards reading in Engl ish; in particular, the motivati on of struggling readers. Few studies have been conducted in an L2 context re garding this issue. This study offered evidence that after the MRSI experiment, EFL students be came confident readers a nd finally they became motivated because they began to understand the re lationship between their use of strategies and their success in reading English. That is to say, once EFL students get over the initial reluctance of reading and are given freedom to choose suitable reading materi als which they can read with joy and ease, they often become quite enthusias tic on reading. From this point, they often continue enthusiastically under their own initiative, increasing their exposure to the target language and vocabulary. Fourth, nowadays, English teachers in Taiw an are under great pressure in terms of preparing their EFL students for the yearly entr ance examinations. Some may argue that they dont have the luxury to spend time teaching their st udents how to self-direct learn and read. In order to meet school and parents expectations, most of them are diligent teachers, trying to teach whatever they know and to cover whatever their EFL students needs to know in order to perform well in the exams. They feel it would be more useful to teach their EFL students test-taking strategies, which are different from reading comp rehension strategies for effective understanding or mean-making. However, while these test-taki ng strategies could have helped them become 199

PAGE 200

test-wise, th eir ability for in-dep th understanding of th e reading materials at hand might not have improved (Cohen, 1998). Under this product-orient ed teaching method, the problem arises: many of my colleagues often have doubts as to why, af ter having taught so hard and given their EFL students so much English knowledge input without wasting class tim e in other things, their EFL students still dont read well or lik e to read. In other words, English teachers in Taiwan have been tried to cover everything in English class, but EFL students only learn a part. The reason behind this problem is now clear: if EFL students are not taught how to be independent learners, dependence on English teachers inst ruction can hardly allow them to reach a satisfactory level. If the only reason that EFL students read is to pr epare for school exams, then it is possible that they will never read after school ing, when the exams have ended. However, to master or acquire a second language, EFL students are the ones who can take control over their learning process and they are the ones who should take the initiativ e to read more to expand their exposure to language input and to build up their English pr oficiency level. This study, by integrating metacognitive reading strategy in struction into a regular EFL E nglish class, provides evidence that effective metacognitive reading strategy pr ogram can transform adolescent EFL students learning experience and improve their English readi ng achievement. This is a great and inspiring outcome for language educators in Taiwan a nd I hope that the findings of this study will contribute to the knowledge base and offer encouragement for researchers and practitioners working towards improving the English achievement of EFL students. Some issues with regard to metacognitive reading strategy instruction (MRSI) remain unclear here. For instance, long-term impact of MRSI on English reading comprehension as well as L1 literacy and L1 strategy transfer or interference on EFL read ing remain unexplored. Therefore, there is a need for more comprehensiv e research on a wild range of variables affecting 200

PAGE 201

reading strategy instruction and use. In summ ary, findings in this study can help language teachers, cu rriculum developers, and researchers to design and develop appropriate programs to meet the needs of the school, the teacher, and the students. The ultimate goal for this research is to offer suggestions to EFL or ESL language teac hers to understand that EFL/ESL learners need to explore different reading strategies, experime nt, and evaluate those reading strategies and eventually choose their own set of effective strategies. By cr eating a supportiv e and cooperative learning context, that is, inte grate MRSI along with constructi ve learning environment into regular EFL instruction, then EFL students are ab le to think more and evaluate their reading processes and further become confident and deve lop responsibility for their own learning and learner autonomy. 201

PAGE 202

202 APPENDIX A DEMOGRAPHIC DATA SURVEY: STUDENTS PROFILE Age. Student ID. Class. Years studied English. Joint high school entrance examination English test score Have you ever been taught how to read or reading strategies in English reading? YesNo. Please provide the following information. Try to be as specific as possible. This survey should take approximately 15 minutes to complete. What do you think is the most difficu lties part of learning a language? 2. How do you rate your overall prof iciency in English as compared to that of your classmates? ) 3. How important is it for you to b ecome proficient in English? ( ?) 4. Do you enjoy learning English? ( 5. Reasons for learning English ( ..Interested in English Int erested in we stern culture ..Have friends who speak English only.Need English for future career or study ..Need it for travel ..Other 6. How many hours per week you spend reading English material? What do you read? ( 7. Do you like to read in English? ( ) 8. Do you like English class? Why or why not? ( 9. What do you find most difficult about English? ..) 10. How do you learn to reading in English? ( 11. Do you think you are a good reader in English? Why or why not? () 12. What makes a person a good reader? ()

PAGE 203


PAGE 204


PAGE 205


PAGE 206


PAGE 207


PAGE 208


PAGE 209


PAGE 210

APPENDIX D READING STRAT EGY BOOKLET & JOURNAL Reading Strategies that Help to Become a Better Reader Reading Stagetegies Before you read While you are reading. After youve read. Planning Monitoring ( Evaluating Questions to Ask Questions to ask at this stage What do I already know about this topic? ( Questions to ask at this stage Do I understand what I am reading? Does this make sense in comparison to what I already know? Questions to ask at this stage How did I do? Did I accomplish my task? What could I have done differently? Strategies to Use Strategies to use: Predicting ( Previewing ( Activate your background knowledge Strategies to use Focus you attention Clarifying Predict as you go along Reread Keep going if you get the gist Ask yourself questions Summarize sections of the text Talk to yourself through problems Visualizing through semantic mapping ( Strategies to use Summarize entire text ( Ask yourself questions 210

PAGE 211

211 1(Reading Strategies 1) Predicting What do we use the strategy of predicating? To help us become better readers. ) To think about what you already know and how this relates to what might happen next.( Gives you a reason for reading to see if your prediction came true. ( To look for clues that tell us what may happen next. Good prediction can be explained. They are not just wild guesses but are based on the following things: What is already known about the reading text? Activate background knowledge ) What information do you know from titles, headings, pictures, and charts? o Sample ways to predict 1. I think that. 2. I believe 3. I bet. 4. I predict. 5. Because of., I believe that.. 6. Based on what I have read or on the type of story it is, I predict that 2(Reading Strategies 2) Clarifying What do we use the strategy of clarifying? To know what to do when we dont understand something To find out a word means ( To get more information about something in the text o Word Analysis When you come to a word you dont know, dont feel panic or look it up in the dictionary immediately: Reread the sentence and think of what it might make sense Break the word into pieces (prefix and suffix) ( Are there any parts of word you know? ( Identify its part of speech Substitute a word or idea that make sense ( ) Look it up in the dictionary

PAGE 212

o hen something is confusing Reread ( ) Slow down ( ) Look at the pictures, titles, headings. ( ) Clarifying statement 1. A word I dont know the meaning of is 2. Is there anything I dont understanding? 3. I am not sure what is happening when..? 4. This is confusing to me so I need to 5. This doesnt make sense to me so I need to 3 (Reading Strategies 3) Questioning a. Why do we use the strategy of questioning? To see if we really understand what we are reading? To decide what is the main idea or mo st important about what we have read (Finding main idea Global reading strategies ) o Question words Who What ? Where Why How o Questioning (question words such as who, where, what, why, when, and how?) 1. What was the main idea? 2. I wonder why? 3. What was happening? 4. What would you do if? 5. Why do you think ? 6. How would you feel if..? 4 (Reading Strategies 4) Summarizing a. Why do we use the strategy of summarizing? To check if we understand what we read To help us remember the main ideas 212

PAGE 213

To find out what the author wants to reader to know b. Good Summarizing includes the following parts: Key people, (who) Key place, (where) Key information, Key ideas, concepts, and events. Focus on the main idea or most important information and dont tell every detail Use your own words. Trying to use semantic map to help you summarize. (visualizing) Sample summarizing statements: 1. The main point of this was.. 2. The most information thing I read was.. 3. In my own words, this is about.. 4. I understand that.. Practice One Using clues to clarifying words 1. The job requires that you are proficient at typing and using the computer. Proficiency probably means: 2. We went to Bilbos for lunch last week. Bilbos probably means: 3. Generally I dont like candy but yesterday I ate a lot. Generally probably means 4. Dont blurt out the answer. Remember to raise your hand. blurt probably means 5. I really like Neerus backpack. Neerus probably means: 213

PAGE 214

Practice Two Mammoth A mammoth was an elephant that live d during the Ice Ages. The Ice Ages were times when much of the Earth wa s covered with snow and ice. It was very hard to live during th e ice ages. It was very co ld. Mammoths could live in the cold because their bodies helped them. They had tusks that were very l ong. The tusks curled up at the ends. Mammoths used their tusks to scrape th e snow off grass and plants so they could eat them. Mammoths had long, shaggy hair to help keep them warm. The long, shaggy hair kept the cold off of the mammoth 's body. The hair also helped keep the heat the mammoth's body made in. Mammoths had a hump between their shoul ders. The hump was made of fat. When mammoths couldn't find food to eat they could use up the fat in the hump. Read this paragraph. As you are reading, are there any words or ideas you need clarified? 2. Summarize the important points. Mammoth Summary .. .. Mammoth Predict Clarify..Question..Summarize..Vocabulary.. ?Tell one place where you used one of the strategies: Write at least three questions you could as k someone else who read the story: 214

PAGE 215

Reading Strategies Self Checklist ( Think about the reading you read and answer the following questions about the strategies used. 1. predicted what the reading would be about by the title and the illustrations. Not at all Some( A lot( 2. clarified unknown words by using the context, substituting, or a good guess. Not at all Some A lot ) 3. pay attention to my own thoughts to see if I understand what I was reading. Not at all( Some( A lot( 4. went back or reread when something didnt make sense. Not at all Some( A lot( 5. skipped unknown words that I couldnt clarify if I understood the gist. Not at all Some A lot( 6. made predictions about what would happen next in the reading text. Not at all( Some( A lot 7. After reading, I summarized the main points of the reading text. Not at all( Some A lot Practice Three Animals Bodies Animals' bodies have different forms so they can live in different places. Some animals live in the water. Other animals fly through the air. Some animals live on the land. Animals' bodies are specially made so that each animal can live well in its environment. An animal that lives in the water usually has a tail an d fins, but no legs. It needs to swim, not to walk. If it does have legs, its feet are usually webbed so it can swim well. A water animal's body is built to do the job of swimming! An animal that flies through the air has wings. It also has legs because it can't fly all of the time. A bird's or insect's body is built to do the job of flying! An imals that live on the land usually have legs, except for snakes and other animals the crawl on their bellies. To move on land animals need to walk, run and climb. ========================================================= Read this paragraph. As you are reading, are there any words or ideas you need clarified? 2. Summarize the important points. .. .... . 215

PAGE 216

. .. .. .. .. Animals bodies Predict Clarify..Question..Summarize..Vocabulary.. Tell one place where you used one of the strategies: Write at least three questions you could as k someone else who read the story: Reading Strategies Self Checklist ( Think about the reading you read and answer the following questions about the strategies used. 1. predicted what the reading would be about by the title and the illustrations. Not at all( Some( A lot 2. clarified unknown words by using the context, substituting, or a good guess. Not at all( Some( A lot 3. pay attention to my own thoughts to see if I understand what I was reading. Not at all Some( A lot 4. went back or reread when something didnt make sense. Not at all Some( A lot( 5. skipped unknown words that I couldnt clarify if I understood the gist. Not at all( Some( A lot 6. made predictions about what would happen next in the reading text. Not at all( Some( A lot 7. After reading, I summarized the main points of the reading text. Not at all ) Some( A lot ) 216

PAGE 217

Reading comprehension Tsunamis The 2004 tsunamis took tens of thousands of peoples lives and the tragedy remains etched in out memory. But by December 2006, the Indi an Ocean will have a hi-tech tsunami early warning system in place. The warning system will go off at once should another tidal wave begin. There is no doubt that if the warning system had been in place, a lot of lives would have been saved from the 2004 tsunamis. However, some people dont think that is enough to defend people there against the assault of anothe r tsunami. Fishermen living along the coasts next to the Indian Ocean are often too poor to live decent lives let al one buy mobile phones. That means that a warning system might not be able to get to them in case of an emergency. 1. The main idea of this passage is: A. The 2004 tsunamis caused a lot of damage and took a lot of lives. B. The early warning system will go off before a tsunami hits. C. An early warning system will not be enough to guard fishermen against a tsunami. D. Fishermen along the coasts of India are mostly poor. 2. By when will the Indian Ocean be equipped w ith a hi-tech tsunami early warning system? A. October, 2006 B. December, 2006 C. November, 2006 D. September, 2006 3. Why is the warning system not enough to save the people along the coasts of the Indian Ocean? A. Because the warning system wi ll not go off on the beach. B. Because mobile phones are no t available in p oor countries. C. Because the coasts stretch to the horizon. D. Because the people are too poor to have access to the warning system. 4. The author expands on his main idea by A. giving reasons. B. offering arguments C. making a contrast D. giving examples Reading Strategies Self Checklist Think about the reading you read and answer the following questions about the strategies used. 1. predicted what the reading would be about by the title and the illustrations. Not at al Some A lot( 2. clarified unknown words by using the context, substituting, or a good guess. Not at all( Some( A lot 3. pay attention to my own thoughts to see if I understand what I was reading. Not at all Some( A lot( 4. went back or reread when something didnt make sense. Not at all Some A lot 5. skipped unknown words that I couldnt clarify if I understood the gist. Not at all( Some( A lot 6. made predictions about what would happen next in the reading text. Not at all( Some A lot ) 7. After reading, I summarized the main points of the reading text. Not at all Some( A lot 217

PAGE 218

. . .. ... .. .. 218

PAGE 219

219 Reading comprehension Dear Grandpa, How are things going back in Taiwan? I really miss you, but Ive found some really great people to spend time with here at the university in England. Peggy is such a funny girl. She reminds me a lot of my sister and we have become very close friends. On the 18th, we went to see a concert at Wembley Stadium. That place is huge. The stadium is brand new and very beautiful. I think it holds 100,000 people. Anyway, thanks for everything. It was hard when I first moved here, but your advice was really good. I mean, at first, I kept thinking about home, but then I remember what you said, and you were right. One year isnt that long and if I think about where I am and not where I was, I will have a lot more fun. Its beautiful and there are lots of people that I want to see and things I want to do. I am so glad I am not ho mesick anymore. Thanks again. Love, Sandy 1. What did Peggy and Sandy do no the 18th? They A. went to see a garden B. went to talk to Wembley C. went to ask for advice D. went to a concert 2. What holds 100,000 people? A. Wembley Stadium B. the concert C. England D. the university 3. What did Grandpa give the girl? A. a big hug B. a bit of fun C. a little advice D. a little homesickness Reading Strategies Self Checklist Think about the reading you read and answer the following questions about the strategies used. 1. predicted what the reading would be about by the title and the illustrations. Not at all Some A lot( 2. clarified unknown words by using the context, substituting, or a good guess. Not at all( Some( A lot( 3. pay attention to my own thoughts to see if I understand what I was reading. Not at all( Some( A lot 4. went back or reread when something didnt make sense. Not at all Some A lot( 5. I skipped unknown words that I couldnt clarify if I understood the gist. Not at all Some( A lot( )

PAGE 220

6. made predictions about what would happen next in the reading text. Not at all Some( A lot( 7. After reading, I summarized the main points of the reading text. Not at all( Some( A lot( . .. .. .. .. .. ... Reading comprehension Chinese Recipe Here is a great idea for some delicious Chinese food that you can make yourself. Chinese noodles with chicken are a big favorite at my house, so I thought I wo uld share the recipe with you. First, you will need the ingredients. You need chicken breast, onion, broccoli, carrots, noodles and sauce. Once you have the ingredients, you need to cook the chicken in a little oil until it is just about fully cooked. Then, add the onion, broccoli, and carrots. Once you have added the vegetables, cover the pan and heat for 3 or 4 minutes. The vegetables will steam inside the covered pan and cool quic kly. Lastly, add the noodles and sauce mix and stir. Keep stirring for about 5 minutes until the noodles are soft and its time to serve. This is a great little Chinese recipe that is sure to please your family. 1. What do you need first? A. The recipe. B. The chicken. C. The onion. D. The ingredients. 2. What do you do first? A. Cook the chicken. B. Cook the vegetables. C. Find the recipe. D. Cook the noodles. 3. When are the noodles done? A. When they are crispy. B. When they are soft. C. When mom says so. D. When the chicken is cooked. 220

PAGE 221

Readin g Str a Think about the reading you read and answer the following questions about the strategies used. 12. predicted what the reading would be about by the title and the illustrations. Not at all Some( A lot 13. clarified unknown words by using the context, substituting, o r Not at all Some( A lot( 14. pay attention to my own thoughts to see if I understand what I was reading. Not at all( Some( A lot( 15. went back or reread when something didnt make sense. Not at all Some( A lot 16. skipped unknown words that I couldnt clarify if Iunderstood the gis Not at all( Some( A lot( 17. made predictions about what would happen next in the reading text. Not at all Some( A lot( 18. After reading, I summarized the main points of the reading text. Not at all Some( A lot( .. .. .. ... 221

PAGE 222

222 APPENDIX E READING STRATEGIC JOURNAL GUIDELINE Assigned Reading Article Date (1) For this assigned reading passage, what st rategies did you use before reading, during reading, and after reading? (2) How easy or difficult this assigned te xt is for you to read and why? ( ) (3) What you have learned from this reading? ( ) (4) Personal response: do you like this assigned reading passage and why? ( ) (5) During the past week, what strategies did you use while reading? ( (6) Write about your experience reading this week (e ither in class or outsi de). Here are some questions to help you. You dont have to answer all the questi ons, just choose one or two. Write in the box below. ( Write about a difficult experience reading Eng lish that you had this week. What strategy could you use to improve in this area? ) Write about a successful experience reading English that you had this week. What strategy did you use that helped you compre hend? If you did it again, would you do it differently? ( ) Did any of your classmate use strategies fo r reading that you would like to try? What strategies did they use? ( )

PAGE 223

APPENDIX F SURVEY OF READING STR ATEGY 05/2007 Statement Neve r Al wa ys 1 I have a purpose in mind when I read. 1 2 3 4 5 2 I take notes while reading to help me understand what I read. 1 2 3 4 5 3 I think about what I know to help me understand what I read. 1 2 3 4 5 4 I take an overall view of the text to see what it is about before reading it. 1 2 3 4 5 5 When text becomes difficult, I read aloud to help me understand what I read. 1 2 3 4 5 6 I think about whether the content of the text fits my reading purpose. 1 2 3 4 5 7 I read slowly and carefully to make sure I understand what I am reading. 1 2 3 4 5 8 I review the text first by noting its characteristics like length and organization. 1 2 3 4 5 9 I try to get back on track when I lose concentration. 1 2 3 4 5 10 I underline or circle information in the text to help me remember it. 1 2 3 4 5 11 I adjust my reading speed according to what I am reading. 1 2 3 4 5 12 When reading, I decide what to read closely and what to ignore. 1 2 3 4 5 13 I use reference materials (e.g., a di ctionary) to help me understand what I read. 1 2 3 4 5 223

PAGE 224

14 When text becomes difficult, I pay closer attention to what I am reading. 1 2 3 4 5 15 I use tables, figures, and pictures in text to increase my understanding. 1 2 3 4 5 16 I stop from time to time and think about what I am reading. 1 2 3 4 5 17 I use context clues to help me better understand what I am reading. 1 2 3 4 5 18 I paraphrase (restate ideas in my own words) to better understand what I read. 1 2 3 4 5 19 I try to picture or visualize information to help remember what I read. 1 2 3 4 5 20 I use typographical features like bo ld face and italics to identify key information. 1 2 3 4 5 21 I critically analyze and evaluate the information presented in the text. 1 2 3 4 5 22 I go back and forth in the text to find relationships among ideas in it. 1 2 3 4 5 23 I check my understanding when I come across new information. 1 2 3 4 5 24 I try to guess what the content of the text is about when I read 1 2 3 4 5 25 When text becomes difficult, I re-read it to increase my understanding. 1 2 3 4 5 26 I ask myself questions I like to have answered in the text. 1 2 3 4 5 27 I check to see if my guesses about the text are right or wrong. 1 2 3 4 5 28 When I read, I guess the meaning of unknown words or phrases. 1 2 3 4 5 29 When reading, I translate from English into my native language. 1 2 3 4 5 30 When reading, I think about information in both English and my mother tongue. 1 2 3 4 5 Source: Mokhtari K., & Sheorkey, R., (2002), Measuring ESL Students Awaren ess of Reading Strategies, Journal of Developmental Educ ation, 25(3), 2-8. 224

PAGE 225

225 APPENDIX G ENGLISH READING MOTIVATION QUESTIONNAIRE 05/2007 Statement 1 By learning to read in English, I hope I will be able to read English novels. 1 2 3 4 2 I get immersed in interesting stories even if they are written in English. 1 2 3 4 3 Learning to read in English is important in that we need to cope with internationalization. 1 2 3 4 4 I am learning to read in English because I might study abroad in the future. 1 2 3 4 5 By being able to read in English, I hope to understand more deeply about lifestyles and cultures of Eng lish speaking countries (such as America and England). 1 2 3 4 6 Even if reading were not a required subject, I would take a reading class anyway. 1 2 3 4 7 *I am learning to read in Englis h merely because I would like to get good grades. 1 2 3 4 8 Long and difficult English passages put me off. 1 2 3 4 9 *I am taking a reading class merely because it is a required subject. 1 2 3 4 10 I would like to get a job that uses what I studied in English reading class. 1 2 3 4 11 I am good at reading in English. 1 2 3 4 12 I like reading English novels. 1 2 3 4 13 I liked reading classes at junior and senior high schools. 1 2 3 4 14 By learning to read in English, I hope to be able to read English newspapers and/or magazines. 1 2 3 4

PAGE 226

15 It is fun to read in English. 1 2 3 4 16 I like reading English newspapers and/or magazines. 1 2 3 4 17 English reading is my weak subject. 1 2 3 4 18 Learning to reading in English is important because it will be conducive to my general education. 1 2 3 4 19 By learning to read in English, I hope to learn about various opinions in the world 1 2 3 4 20 *I think learning to speak and/or listening is more important than learning to read in English. 1 2 3 4 21 My grades for English reading classes at junior and senior high schools were not very good. 1 2 3 4 22 I enjoy the challenge of difficult English passages. 1 2 3 4 23 I do not have any desire to read in English even if the content is interesting. 1 2 3 4 24 Learning to reading in English is important because it will broaden my view. 1 2 3 4 25 *By learning to read in English, I hope to search information on the Internet. 1 2 3 4 26 Reading in English is important because it will make me a more knowledgeable person. 1 2 3 4 27 It is a waste of time to learn to read in English. 1 2 3 4 28 I would not voluntarily read in English unless it is required as homework or assignment. 1 2 3 4 29 I tend to get deeply engaged when I read in English. 1 2 3 4 30 It is a pain to read in English. 1 2 3 4 Source: Mori, S. (2002). Redefining Motivation to Read in a Foreign Language, Reading in a foreign language Vol. 14, No. 2, p. 91-110. 226

PAGE 227

APPENDIX H READING COMPRE HENSION TEST Questions 1-3 Book Editor Wanted Job Responsibilities: The editor is responsible for coor dinating and overseeing book projects by working with authors, artists, designers, and printers to produce quality books in a timely manner. Job Requirements: The applicants must hold a bachelors degree in Chinese literature or journalism and have five years of experien ce as a book editor or a comparable position. The editor must be able to research and analyze information with accuracy and work closely with team members. A ccurate typing skills of at le ast 45 w.p.m. are preferable. The applicants must demonstrate know ledge of the book production and editorial processes, and proficiency in word-processi ng and publishing-related computer programs is a definite advantage. Please fax your resume to Era Publisher at (02) 1234-5678. 1. What type of educational backgroun d must the job applicant possess? A. A masters degree in journalism. B. A bachelors degree in Chinese literature. C. A Ph.D. in Chinese or journalism. D. A bachelors degree in English literature. 2. What will the primary job responsibility of the book editor? A. Writing books. B. Printing books. C. Coordinating various book projects. D. Correcting peoples articles. 3. Which of the following is NOT the required skill for this job? A. Good typing skills. B. Knowledge about book production. C. Book writing experience. D. Skill in work-related computer programs. 227

PAGE 228

Questions 4-6 Dear Sales Manager: This letter is a complaint regarding the Baker TX 2200 bread Machine that I purchased on April 1st at Sams Electronics in Taichung. Unfortunately, this product did not perf orm as well as I had expected. After making two successful loaves of whole-wh eat bread, the kneading paddle began to function badly and the belt kept coming off. The quality of the product is simply disappointing. To resolve the problem, I would appreciate a full refund. Enclos ed are copies of the receipts, warranty, product model, and serial number. I hope that this problem can be solved in a mutually agreeable way. Please contact me within 2 weeks. If I dont receive response by then, I will start seeking assistance from third parties, such as a c onsumer right organization of an attorney. Thank you for your anticipated assistan ce in resolving my problem. Please contact me at 02-23458910 if you have any questions. Sincerely, Joyce Ma 4. What was the m ain purpose of this letter? A. To purchase a bread machine. B. To exchange a product. C. To complain about customer service at Sams Electronics D. To request for money back for a product. 5. What was the problem of the product? A. It burned the bread. B. The kneading paddle was defective. C. The belt was shattered. D. the motor overheated. 6. What did Ms. Ma request in the letter? A. A brand new bread machine. B. her money back. C. an attorney. D. Gift certificate. 228

PAGE 229

Questions 7-9 The Dead Sea, Located in the Middle East, be tween Jordan and Israel, has the saltiest water on earth. It is almost six times as salty as the ocean. It is called dead because almost nothing could live in it. The high concentration of salt prevents fi sh, seaweeds, plants and other visible aquatic organism from living in or near th e water. The water in the Dead Sea is fatal to living creatures. If fish accidentally swim into water from a freshwater stream, they will quickly die. Their bodies will be quickly coated with a layer of salt crystals and then tossed onto a shore by wind of waves. However, humans are adaptable to the salty water in the Dead Sea. The high density of the water makes it impossible for people to sink. Instead, almost everyone can actually easily float in the Dead Sea. 7. What is the main topic of the passage? A. The ecology of the Dead Sea. B. The formation of the Dead Sea. C. The location of the Dead Sea. D. The reason why it is called the Dead Sea. 8. What is the location of the Dead Sea? A. Between Asia and Europe. B. Above sea level. C. In the Middle East. D. Between Israel and Iraq. 9. Which of the following statement is true about the Dead Sea? A. The Dead Sea is almost as salty as the ocean. B. The water in the Dead Sea is too salty for fish to survive in it. C. The Dead Sea is the only salty lake in the world. D. The Dead Sea is deadly to humans b ecause of the high density of the water. 229

PAGE 230

Questions 1 0-11 Public speaking is a common fear for many people. Even experienced speakers have anxiety when speaking in front of a group of peopl e. Being nervous is perfectly normal. The important thing is how to reduce the stress. There are some tips which can help you promote successful public speaking. First, you need to know your audience and match your content to their interests. Then, know the material and arrange what you want to say in a seque ntial and logical order. If you have the time, use a tape recorder or video camera to record your presentation, and then analyze your strengths and weaknesses. Before the presentation, try to relax by doing relaxation exercise to ease your tension. Visualize yourself as being co nfident and giving a successful speech. Finally, remember that your audience will expect you to succeed. The want you to be simulative, informative, encouraging, and interesting. 10. What is the main idea of this passage? A. Why public speaking is stressful. B. How to write a good speech. C. The importance of reducing stress before speaking in public. D. How to be a successful public speaker. 11. According to the passage, which of the following is NOT listed as one of the tips to promote successful public speaking? A. Visualize yourself as a successful, confident speaker. B. Remember that the audience is hypercritical. C. Know the material thoroughly. D. Rehearse the speech in advance. 230

PAGE 231

Questions 1 2-15 In Japan geisha are professional entertainers who entertain guests through different types of performing arts in teahouses. They are trained in a wide range of traditional arts such as Japanese classi cal dancing, singing, and playing mu sical instruments. Geisha women are also trained to serve dr inks and talk with the guests. Geisha usually continue to excel in those skills throughout their career. Geisha usually wear their hair in a bu n or a uniform style of a single comb and two pins. They also wear eleg ant kimonos and white makeup with red lipstick. Outside of Japan, there still remains some confusi on about the nature of the geisha profession. They are different from prostitutes. A geisha is not paid for sex. However, a geisha may choose to have patron, with whom she is invo lved economically and sexually. Ge isha do not marry throughout their career. If they choose to get married, th en they must retire from the profession. 12. According to the passage, what do geisha mostly do? A. Serving tea at teahouses. B. Dancing in a theater. C. Performing religious ceremonies. D. Entertaining guests. 13. According to the passage, which of the following in NOT mentioned as geishas skills? A. Playing musical instruments. B. Planting flowers. C. Dancing. D. Serving drinks. 14. Which of the following best descri bes the appearance of geisha? A. Geisha always dress in white kimonos. B. Geisha usually wear hair straight. C. Geisha wear white makeup with red lipstick. D. Modern geisha sometimes wear western gowns when performing. 15. According to the passage, which of the following statements is true about geisha? A. People usually think geisha ar e the same as prostitutes. B. A geisha can continue to work as a geisha after she marries her patron, C. Geisha are trained in a vari ety of traditional skills. D. Geisha are only found in Kyoto, Japan. 231

PAGE 232

232 APPENDIX I INTERVIEW QUESTIONS In order to understand in depth the students experience of the implementation of metacognitive reading strategies awareness in the English reading course, a face-to-face interview with the students will be conducted as a further data colle ction method after the experiment is complete. In conducting the inte rviews, the research herself served as the interviewer. The language used in the interview is participants first language, Mandarin. The consent to the students will be obtained. The interview questions listed as following. Before the interview, I will remind the students th at there is no right answ er or wrong answer to the questions. 1. Please state your previous experi ences in reading in English. 2. Please state the reading difficulties you have had. 3. Please state the reading methods you have used. 4. Please state your attitudes towards reading in English. 5. Please provide your view points of metacognitive reading strategies. 6. Please give you comments on the training.

PAGE 233


PAGE 234

LIST OF REFERE NCES Abbott, M. L. (2006). ESL reading strategi es: Differences in Arabic and Mandarin speaker test performance. Language Learning, 56(4), 633-670. Adams, M.J. (1990). Beginning to read: Thinki ng and learning about print. Cambridge, MA: Massachusetts Institute of Technology. Aebersold, J.A. & Field, M.L. (1997). From reader to reading teacher: Issues and strategies for second language classrooms Cambridge University Press. Alderson, J.C. (1984). Reading in a foreign language (Ed.) Longman. Alderson, J.C. (2000). Assessing reading. Cambridge, UK: Cambridge University. Almasi, J.F. (2003). Teaching strategic proces ses in reading. New York: Guilford Press. Anderson, N.J. (1991). Individual differences in strategy use in sec ond language reading and testing. Modern Language Journal 75, 460-472. Anderson, N.J.(1985). Role of the readers sche ma in comprehension, learning, and memory. In Singer & R.B. Ruddell (Eds.), Theoretical models and processes of reading (pp. 372384). Newark, NJ: Internati onal Reading Association. Anderson, R.C., & Freebody, P. (1981). Voca bulary Knowledge. In J. Guthrie (Ed.), Comprehension and teaching: research reviews (pp.77-117). Newark, DE: International reading association. Anderson, R.C., & Pearson, P.D. (1984). A schema -theoretic view of reading comprehension. In P.D. Pearson(Eds.), Handbook of readin g research (pp.225-291) New York: Longman. Anderson, R.C., & Pearson, P.D. (1993). A schema-theoretic view of basic process in reading comprehension. In Carrell, P.L., Devine, J. & Eskey, D.E. (Eds.), Interactive approached to second language reading (pp. 37-55). Cambridge: Cambridge University Press. Anderson, N.J. (2002). The role of metacong ition in second language teaching and learning. ERIC Digest April, 2002, 3-4. Anderson, N.J. (2003). Metacognitive readi ng strategies increase L2 performance. The Language Teacher Online, 27(7). Anderson, N.J. (2003). Scrolling, clicking, and re ading English: online reading strategies in a second/ foreign language. Reading in a Foreign Language, Vol. 3, No. 3. Auerbach, E. R., & Paxton, D. (1997). "It's not the English Thing:" Bri nging reading research into the ESL classroom. TESOL Quarterly, 31 237-261. 234

PAGE 235

Autrey, J. H. (1999). Effects of direct inst ruction and precisio n teaching on achievement and persistence of adult learners. Dissertation Abstracts International, 60 (06), 1863A. (UMI No. 9932955) Baker, L (2002). Metacognition in comprehension instruction. In C. C. Block & M. Pressley (Eds.), Comprehension Instruction: Re search based best practices 77-95, New York: Guilford. Baker, L. (2004). Reading comprehension and sc ience inquiry: Metacognitiv e connections. In W. Saul (Ed.), Crossing borders in literacy and science instruction: Perspectives on theory and practice (pp. 239-257). Newark, DE: Inte rnational Readin g Association. Baker, L. (2005). Developmental differences in metacognition: Implications for metacognitively oriented reading instruction. In S. Israel, C. C. Block, K. L. Bauserman, & K. KinnucanWelsch (Eds.), Metacognition in literacy learning: Theory, assessm ent, instruction, and professional development (pp. 61-79). Mahwah, NJ: Erlbaum. Baker, L. & Brown, A. (1984). Mrtacognitive skills and reading. In P.D. Pearson, R. Barr. M.L. Kamil, & P. Modrnthal, (Eds.). Handbook of Reading Research (pp. 353-394). New York: Longman. Barber, Theodore Xenophon, (1973). Pitfalls in rese arch: nine investigator and experimenter effects", in Robert M. W. Travers (ed.), Second handbook of research on teaching (Chicago: Rand McNally), pp. 382-404. Bartlett, F.C. (1932). Remembering: A study in expe rimental and social psychology Cambridge: Cambridge University press. Barnett, M. A. (1988). Reading th rough context: How real and perc eived strategies use affects L2 comprehension. Modern Language Journal, 72, 150-162. Baumann, J.F. and Kameenui, E.J. (1991). Research on vocabulary instruction: Ode To Voltaire. In J. Flood, J.J. Lapp, and J.R. Squire (Eds.), Handbook of research on teaching the English language arts (pp. 604-632). New York: MacMillan. Baumann, J., Jones, L., Seifert-Kessell, N. ( 1993). Using think-alouds to enhance childrens comprehension monitoring abilities. The Reading Teacher, 47 (3), p. 184-193. Bernhardt, E.B. (1991). Reading development in a second language Norwood, NJ: Ablex. Bernhardt, E.B. (2003). Challenges to reading research from a multilingual world. Reading Research Quarterly, 38, 112-117. Bernhardt, E.B. (2005). Progress and proc rastination in seco nd language reading. Annual Review of Applied Linguistics 25, 133-150. 235

PAGE 236

Benedetto, R. A. (1984). A psychol inguistic investigation of the use of top-level organizational strategies in first and second language reading: Five case studies. Dissertation Abstracts International, 45. 2049-A. Black, D. J. (1995). An investigation into the knowing and regul ating comprehension behaviors of skilled and less skilled seventh grade readers. Masters Abstracts International, 33 (03), 717. (UMI No. MM92150) Block, E. (1986). The Comprehension stra tegies of second language readers. TESOL Quarterly 20, 463-494. Block, E.L. (1992). See how they read: Comprehension monitoring of L1 and L2 readers. TESOL Quarterly 26, 2. Brantmeier, C. (2002). Second language read ing strategy research at the secondary and university levels: Variations, di sparities and generalizability. The Reading Matrix: An International Online Journal 3 (2), 1-14. Brooks, J. G., & Brooks, M. G. (1993 ) In search of understanding: The case for constructivist classrooms Alexandria, VA: Association for Supe rvision and Curriculum Development. Brown, T.S. (1991). A comparison of three lear ning strategies for ES L vocabulary acquisition. TESOL Quarterly Vol.25, No.4. Brown, A.L., Campione, J.C., & Day, J.D. (1981) Learning to learn: On training students to learn from texts. Educational Researcher 10, 2, 14-21. Brunning, H. R. Schraw, G. J. & Ronning, R. R. (1999). Cognitive psychology and instruction (3rd Ed.). NJ: Prentice-Hall. Canagarajah, A.S. (1993). Critical ethnography of a Sri Lankan classroom: ambiguities in opposition to reproduction through ESOL. TESOL Quarterly 27 4, 319-343. Carriedo, N., & Alonso-Tapia, J. (1995). Co mprehension strategy training in content areas. European Journal of Psyc hology of Education, 10 (4), 411-431. Calero-Breckheimer, A., & Goetz, E.T. (1993). Reading strategies of biliterate children for English and Spanish texts. Reading Psychology, 14, 177-204. Casanave, C.P. (1988). Comprehension monitori ng in ESL reading: a neglected essential. TESOL Quarterly 22, 283-302. Carrell, P.L. (1983). Three components of b ackground knowledge in reading comprehension. Language Learning, 33, 183-207. 236

PAGE 237

Carrell, P. L. (1985). Facilitating ESL reading by teaching text structure. TESOL Quarterly, 19 727-752. Carrell, P. L. (1989). Metacognitive aw areness and second language reading. Modern Language Journal, 73, 120-134. Carrell, P.L. (1991). Second language reading: reading ability or language proficiency? Applied Linguistics 12, 159-179. Carrell, P.L.,& Eisterhold, J.C.(1993). Schema theory and ESL reading pedagogy. In P.L. carrel, J. Devine, & D.E. Eskey(Eds.), Interactive approaches to second language reading (pp. 73-92). Cambridge: Cambridge university press. Carrell, P.L., Devine, J. & Eskey, D.E. (Eds.) (1993). Interactive approached to second language reading. Cambridge: Cambridge University Press. Carrell, P. L., Pharis, B. G., & Liberto, J. C. (1989). Metacognitive strategy training for ESL reading. TESOL Quarterly, 23, 647-678. Carrell, P.L., Gajdusek, L., & Wise, T. ( 1998). Metacognition and EFL/ESL reading. Instructional Science 26, 97-112. Chaplin, M (1982). Rosenblatt revisited: Th e transaction between reader and text. Journal of Reading November, 150-154. Clark, M. (1979). Reading in Spanish and English. Language Learning, 29, 121-150. Clay, M. (1991). Becoming literate: The cons truction of inner control Portsmouth, NH: Heinemann Chamot, A.U. (2004). Issues in language learning strategy research and teaching. Electronic Journal of Foreign Language Teaching, 1(1). 14-26. Retrieved September 27, 2007, from Chamot, A.U. (2005). Language le arning strategy instruction: Current issues and research. Annual Review of Applied Linguistics 25, 112-130. Chamot, A.U., & Keatley, C.W. (2003) Learning strategies of adol escent low-literacy Hispanic ESL students. Paper presented at the 2003 Annual Me eting of the Am erican Educational Research Association, Chicago, IL. Chamot, A.U., & OMallety, J.M. (1994). Childrens learning strategies in language immersion classrooms. Modern Language Journal 83, 319-338. Chia, H.L., & Chia, H.U.(2001). English readi ng concepts held by Taiwanese EFL readers. English Teaching and Learning, 26, 23-44. 237

PAGE 238

Cheng, C. K. (1998) A descriptive study of reading strategi es used by Chinese ESL students from Taiwan, R. O. C. Unpublished doctoral dissertation, University of Kansas. Chern, C.L. (1993). Chinese students word-solving strategies in read ing in English in T. Huckin, M. Haynes, & C.Coady (Eds.), Second language reading and vocabulary learning (pp.67-85). Norwood, NJ: Ablex. Chern, C.L. (2003). English language teaching in Taiwan today. In H.W. Kam & Y. L.Wong (Eds.), English language teaching in East Asia today: Changing polic ies and practices (pp. 427-439). Singapore: Times Academic Press. Choochom, O. (1995). Text comprehension as a f unction of motivational orientation and reading strategy. Dissertation Abstracts International, 55 (07), 1874A. (UMI No. 9431605) Clarke, M.A. (1980). The short circuit hypothesis of ESL reading: or when language competence interferes with reading performance. Modern Language Journal 64, 203-209. Coady, J. (1979). A psycholinguistic model of the ESL reader. In R. Mackay, B. Barkman, & R.R. Jordan (Eds.). Reading in a Second Language (pp.5-12). Rowley, MA: Newbury House. Cohen, A.D. (1990). Strategies in second language le arning: Insight from research In R. Phillipson. E. Kellerman, L. Selinker, M.S. Smith, and M. Swain (eds,): Foreign/Second Language Pedagogy Research Cleve: Multilingual matters. Cohen, A.D. (2003). The learners side of forei gn language learning: Wher e do style, strategies, and tasks meet? International Review of Applied Linguistics 41, 279-291. Cohen, A.D. (2007). Coming to terms with language learner strategies: Surveying the experts. In A.D. Cohen & E. Macaro (Eds.), Language learner strategies: Th irty years of research and practice (pp. 29-46). Oxford, England: Ox ford University Press. Cornford, I.R. (2002). Learning-to-leran strate gies as a basis for e ffective lifelong learning. International Journal of Lifelong Education 21, 4, p.357-368. Cook, V. (2001). Using the first language in the classroom. Canadian Modern Language Review 57, 402-423. Cooper, J. M. (1998). An exploratory study of the metacognition of verbally gifted/learning disabled learners with and without reading difficulties. Dissertation Abstracts International, 58 (12), 4559A. (UMI No. 98183223) Cordero-Ponce, W.L. (2000). Summarization in struction on foreign language comprehension, Reading Research and Instruction 39 (4), 329-350. 238

PAGE 239

Cziko, G.A. (1980). Language com petence and reading strategies: A comparison of first and second language oral reading errors. Language Learning, 30, 101-114. Day, R.R., & Bamford, J. (1998). Extensive reading in the second language classroom Cambridge University press. Dewey, J. (1911). How we think Boston: D.C. Heath & Company. Devine, J. (1984). Readers inte rnalized models of reading proc esses. In J. Handscombe, R. Orem and B.P. Taylor (eds). On TESOL Washington, DC: TESOL. Devine, J. Carrell, P.L., & Eskey, D.E. (1987). Research in reading in English as a second language. Washington, D.C.: Teachers of E nglish to Speakers of Other English. Devine, J. (1993). The role of metacognition in second language reading and writing. In G. Joan and L. I. Carson (Eds.), Reading in the composition classroom, second language perspective (pp. 105-130). Boston: Heinle and Heinle Publishers. Drnyei, Z. (2001). Teaching and researching motivation. New York: Longman. Drnyei, Z. (2003). Questionnaires in second language rese arch: construction, administration, and processing. Mahwah, NJ: Erblaum. Drnyei, Z. (2005). The psychology of the language learne r: Individual differences in second language acquisition Mahwah, NJ: Lawrence Erlbaum. Droop, M., & Verhoeven, L. (2003). Language proficie ncy and reading ability in first and second language learners. Reading Research Quarterly, 38, 78-103. Druitt, E. E. (2002). Investigating students achievement before and after a reading intervention program. Masters Abstracts International, 40 (06), 1341. (UMI No. 1409408) Duffy, G. (2005). Metacognition and the development of reading teachers. In C.Block, S. Israel, K. Kinnucan-Welsch & K. Bauserman (Eds.) (pp. 299-314). Metacognition and literacy learning. Mahwah, NJ: Erlbaum. El-Dib, M.A.B. (2004). Language learning strategies in Kuwait: Links to gender, language level, and culture in a hybrid context. Foreign Language Annals, 37, 85-95. Ellis, R. (1994). The Study of Second Language Acquisition. Oxford: Oxford University Press. Eskey, D. (1973). A model program for teaching adva nced reading to students of English as a foreign language. Language Learning 23(2), 169-184. 239

PAGE 240

Eskey, D. (1986). Theoretical foundations in Dubin, F., Esdey, D, & Grabe, W (Eds.). Interactive approaches to second language reading New York: Cambridge University Press. Eskey, D. & Grabe, W. (1988). Interactive models for second language re ading: perspective on instruction. In Carrell, P.L., Devi ne, J. & Eskey, D.E. (Eds.) (1993). Interactive approached to second language reading Cambridge: Cambridge University Press. Eskey, D.E. (2002). Reading and the teaching of L2 reading. TESOL Journal 11, 1, pp.5-9. Eskey, D. E. (2005). Reading in a second language. In E. Hinkel (Ed.), Handbook on Second Language Learning and Teaching (pp. 563-579). Mahwah,NJ: Erlbaum. Ethrman, M & Oxford, R. (1989). Effects of sex differences, career choice, and psychological type on adult language learning strategies, The Modern Language Journals 73, 1, 1-13. Ethrman, M & Oxford, R. (1990). Adult language lear ning styles and strategies in an intensive training setting. The Modern Language Journal 74, 3, 311-327. Ethrman, M., Leaver, B. & Oxford, R.(2003). A brie f review of individual differences in second language learning, System, 31 (3), 313-330. Farrell, T.S.C. (2001). Teaching read ing strategies: It takes time! Reading in a Foreign Language 13 (2), 631-646. Farrell, T.S.C (2009). Teaching Reading to English Language Learners: A reflective Guide. CA: Corwin Press. Faul, F., Erdfelder, E., Lang, A.-G. & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioural, and biomedical sciences. Behaviour Research Methods, 39(2), 175-191. Flavell, J. H. (1978). Metacognitive development. In J. M. Scandura & C. J. Brainerd (Eds.), Structural/process theories of complex human behaviour (pp. 213-245). The Netherlands: Sijthoff and Noordhoff. Flavell, J. H. (1992). Metacognition and cogni tive monitoring: A new area of cognitivedevelopmental inquiry. In T. Nelson (Ed.), On TESOL (pp.171-183). Washington, DC: TESOL. Fosnot, C. T. (1996). Constructivism: A psychologi cal theory of learning. In C. T. Fosnot (Ed.) Constructivism: Theory, perspectives, and practice (pp. 8-33). New York: Teachers College Press. Fu, D. (1995). My Trouble is My English: Asian Students and Their American Dreams. Portsmouth, NH: Heinemann. 240

PAGE 241

Gall, M. D., Borg, W .R., & Gall, J.P. (2003). Educational research: An introduction (7th Ed.). White Plains, NY: Longman. Garcia, G.E (1991). Factors influencing the E nglish reading test performance of Spanishspeaking Hispanic children. Reading Research Quarterly 26, 371-393. Garner, Hare, Alexander, Haynes, & Winogra d, (1984). Inducing Use of a text look back strategy among unsuccessful readers American Educational Research Journal Vol. 21, No. 4 (Winter, 1984), pp. 789-798. Garner, R. (1980). Monitoring of understanding: An investigation of good and poor readers' awareness of induced miscomprehension of text. Journal of Reading Behavior, 12, 55-63. Garner, R. (1987). Metacognition and reading comprehension. Norwood, NJ: Ablex Publishing. Garner, R. (1994). Metacognition and executive c ontrol. In R. Ruddell, M. Ruddell & H. Singer (Eds.), Theoretical models and processes of reading (pp. 715-732). Newark, Delaware: International Reading Association. Good, T., & Brophy, J. (1991). Looking in classrooms. (5th ed.). New York: HarperCollins. Goodman, K. (1976). Reading: A psycholinguistic guessing game. Journal of the Reading Specialist 6, 126-135. Goodman, K. S. (1992). Reading, writing, and written texts: A transactional sociopsycholinguistic view Tucson, AZ: Literacy and Learning Center, University of Arizona. Goodman, K. (1996). On reading. Heinemann. Portsmouth, NH. Gough, P. B. (1972). One second of reading. In J. F. Kavanaugh & I.C. Maltingly (Eds.), Language by ear and eye (pp. 331-358). Cambridge, MA: MIT Press. Grabe, W. (1988). Reassessing the term interactiv e. In P.L.Carrell, J. Devine, & D. E.Eskey (Eds.), Interactive approaches to second language reading (pp. 56-70). New York: Cambridge University Press. Grabe, W. (1991). Current developments in second language reading research. TESOL Quarterly, 25, 3, 375-406. Grabe, W. (2002). Foundations for L2 reading instruction. The Language Teacher online, 26, Retrieved December 3, 2003, from articles/2002/07/grabe 241

PAGE 242

Grabe, W (2004). Research on teaching reading. Annual Review of applied Linguistics 24, 4469. Grabe, W., & Stoller, F.L. (2002). Research on teaching reading. Annual Review of Applied Linguistics 24, 44-69. Green, J. M. & Oxford, R. (1995). A closer look at learning strategi es, L2 proficiency and gender. TESOL Quarterly 29, 2, 261-297. Griffiths, C. (2002). Using reading as a strategy for teaching and learning language. ERIC Database of Educational Documents Griffiths, C. (2003a). Patterns of language learning strategy use. System, 31, 367-383. Griffiths, C. (2003b). Language learning strate gy use and proficiency. Retrieved October 15, 2009, from Griffiths, C. (2004). Language learning strategies: Theory and Research. Research Paper Series Centre for Research in International Education. Retrieved October 5, 2009, from Griffiths, C. & Parr, J. (2000). Language lear ning strategies, nationa lity, Independence and Proficiency. Independence, 28, 7-10. Griffiths, C. & Parr, J (2001). Strategies for success: How language learning strategies relate to proficiency in language learning. Many Voices 17, 27-31. Griffiths, C. (2004). Language Learni ng Strategies: Theory and Research. Occasional Paper No.1. Auckland, New Zealand: Center for Re search in Intern ational Education. Griffiths, C. (2008). Lessons from good language learners Cambridge University Press. Gu, P.Y. (1996). Robin Hood in SLA: What has the learning strategy re searcher taught us? Asian Journal of English Language Teaching, 6, 1-29. Guthrie, J.T., & Alao, S. (1997). Designing contexts to increase motivation for reading. Educational Psychologist, 32, 95-105. Guthrie, J. T., & Alao, S. (1997). Engagement in reading for young adolescents. Journal of Adolescent & Adult Literacy, 40 (6), 438-446. Guthrie, J. T. (2001). Contexts for engagement and motivation in reading. Reading Online, 4 (8). Retrieved May 6, 2004, from ticles/handbook/guthrie/index.htm l 242

PAGE 243

Ha mp-Lyons, L. (1985). Two approaches to teaching reading: A classroom-based study. Reading in a Foreign Language, 3 363-373. Hansen, J. (1981). The effects of inference training and practice on young childrens reading comprehension. Reading Research Quarterly, Vol. 16, No. 3 (1981), pp. 391-417. Hansen, J., & Pearson, P.D. (1983). An inst ructional study: Impr oving the inferential comprehension of good and poor fourth grade readers. Journal of Educational Psychology, 75(6), 821-829. Hassan, X. et al. (2005). Strategy tr aining in language learning a systematic review of available research. In: Research Evidence in Education Library London: EOOI-Centre, Social Science Research Unit, Institute of Education, University of London. Hare, V., & Borchardt, K.M. (1984). Dir ect instruction of summarization skills. Reading Research Quarterly, 21, 62-78. Harris, V. (2003). Adapting classroom-based strate gy instruction to distance learning context. In N.J. Anderson (Ed.), TESL-EJ: Special issue: St rategy research and training 7 (2), A-1. Harris, V. & Grenfell, M. (2004). Language learning strategies: A case for cross-curricular collaboration. Language awareness 13, 2, 116-130. Harrison, C. (2002). The National Strategy for English at Key Stage 3: Roots and Research. Department for Education and Skills, London: HMSO. Hauptman, P. C. (1979). A comparison of first and second language reading strategies among English-speaking university students. Interlanguage Studies Bulletin 4, 173-201. Hosenfeld, C. (1977). A preliminary investigation of the reading strategi es of successful and nonsuccessful second language learners. System 5, 110-123. Hosenfeld, C. (1984). Case studies of ninth grad e readers. In J.C. Alderson and A.H. Urquhart (Eds.): Reading in a foreign language London: Longman. Hosenfeld, C. (1995). Reading in English as a second language: some current issues, applications, and directions for future research. College ESL 5(2), 21-34. Hua, T. (1997). The relationship between read ing comprehension processes in L1 and L2. Reading Psychology, 18, 249-301. Huang, J. (2006). Learner resistance in metacogni tion training? An exploration of mismatches between the learner and teacher agendas. Language Teaching Research 10, 1, 95-117. Huang, Shu-chen (2006). Reading English for academic purpose what situational factors may motivate learners to read? System 34, p. 371-383. 243

PAGE 244

Hung, C.H. (2000). Understand ing the culture of EFL in Taiwan: A qualitative inquiry. The Proceedings of the Fifth International Symposium on English Teaching, 19-26. Taipei: Crane. Hurst, J. L. (2004). Transacting with the word: At-risk adolescents a nd informational texts. Dissertation Abstracts International, 64(10), 3581A. (UMI No. 3108703) Janzen, J., & Stoller, F. (1998). Integrating strategic reading in l2 instruction. Reading in a Foreign Language 12(2), 251-269. Jimenez, R.T., Garcia, G.E., Pearson, P.D. (1995). Three children, two languages, and strategic reading: case studies in b ilingual/monolingual reading. American Educational Research Journal 32, 31-61. Jimenez, R.T., Garcia, G.E., Pearson, P.D. (1996). The reading strategies of bilingual Latino students who are successful English readers: Opportunities and obstacles. Reading Research Quarterly, 31, 90-112. Jimnez, R. T. (1997). The strategic reading abi lities and potential of five low-literacy Latina/o readers in middle school. Reading Research Quarterly, 32, (3), 224-243. Jimnez, R. T. & Gersten, R. M. (1999). Le ssons and dilemmas derived from the literacy instruction of two Latina/o teachers. American Educational Research Journal 36 (2) 265-301. Kelly, K. & Maxwell, S.E. (2003). Sample size for multiple regression: Obtaining regression coefficient that are accurate not simply significant, Psychological Methods 8, 305-321. Kern, R. (1989). Second language reading strategy instruction: its effects on comprehension and word inference ability Modern Language Journal, 73, 135-149. Knight, S. L., Padron, Y. N., & Waxman, H. C. (1985). The cognitive reading strategies of ESL students. TESOL Quarterly, 19 789-792. Knoll, C. L. (2000). The relationship betw een motivation and reading comprehension. Masters Abstracts International, 38 (05), 1156. (UMI No. 1398853) Koda, K. (1987). Cognitive strate gy transfer in second language reading. In J. Devine, P.L.Carrell, and D.E. Eskey (Eds.), Research in reading in English as a Second language (pp. 125-144). Washington, D.C.: Teachers of English to Speakers of Other Languages. Koda, K. (1995). Cognitive consequences of L1 and L2 orthographies. In I. Taylor and D.R. Olson (eds.) Scripts and Literacy: Reading and Lear ning to Read Alphabets, Syllabaries and Characters (pp. 311-326). Dordrecht : Kluwer Academic Press 244

PAGE 245

Koda, K. (1996). L2 word recogni tion research: A critical review. Modern Language Journal 80, 4, 450-460. Koda, K. (2005). Insights into second language re ading: A cross-linguistic approach. Ca mbridge: Cambridge University Press. Koda, K. (2007). Reading and language lear ning: Cross-linguistic constraints on second language reading development. Language Learning, 57 (suppl. 1), 1-44. Kohn, J. (1992). Literacy strategies for Chinese un iversity learners. In F. Dubin & N. Kulman (Eds.), Cross-cultural literacy (pp. 113-125). Englewood Cliffs, NJ: Prentice Hall. Kleitzien, S.B. (1992). Proficient and less profic ient comprehenders strategy use for different top-level structures. Journal of Reading Behavior. 24(2), 191-215. Krashen, S. (1989). We acquire vocabulary and spelling by readi ng: Additional evidence for the input hypothesis. Modern Language Journal 73, 446-447. Lan R., & Oxford, R. L. (2003). Language le arning strategy profiles of elementary school students in Taiwan. International Review of Applie d Linguistics in Language Teaching 41, 339-379. Lin, T. F. (2000). The teaching of vocabulary in Taiwan: A Review. The proceedings of the Ninth International Symposium on English Teaching, 460-469. Taipei: Crane. Lindsey, K. A., Manis, F. R., & Bailey, C. E. (2003). Prediction of first-grade reading in Spanishspeaking English-language learners. Journal of Educational Psychology, 95, 482. Loranger, A.L. (1997). Comprehension strategies instruction: Does it make a difference? Reading Psychology, 18 (1), 31-68. Macaro, E. (2001). Learning strategies in foreign and second language classrooms London: Continuum. Macaro, E. (2006). Strategies for language learning and for language use: Revising the theoretical framework. Modern Language Journal 90, 320-337. Merkelnach, C. (2006). The Negative Influence of ESL-Methodology on the acquisition of an L3 in Taiwan. Retrieve May 30, 2007, from m an/ejournal/Merkelbach1.htm Miller, G. E., Giovenco, A., & Rentiers, K. A. (1987). Fostering com prehension monitoring in below average readers through self-instruction training. Journal of Reading Behavior, 19, 379-394. 245

PAGE 246

Mokhtari, K., & Reichard, C. A. (2002). Assessing students m etacognitive awareness of reading strategies. Journal of Educational Psychology 94, 249-259. Mokhtari, K. & Sheorey, R. (2002). Measuring ES L Students' Awareness of Reading Strategies. Journal of Developmental Education 25, no. 3, 2-10. Mori, S. (2001). The effects of proficiency and group membership on motivation to learn and read in English. The Proceedings of the Third Temple University Japan Applied Linguistics Colloquium 55. Mori, S. (2002). Redefining motivati on to read in a foreign language. Reading in a Foreign Language, 14(2). Retrieved October 30, 2002, from Muniz-Swicegood, M. (1994). The effects of metacognitive st rategy training on the reading performance and fluent reading analysis stra tegies of third grade bilingual students. Bilingual Research Journal 18, 83-97. Nassaji, H. (2003) Higher Level and Lower lev el Text processing Skills in Advanced ESL Reading Comprehension. The Modern Language Journal. 87, 261-276. Nation, I.S.P. (2002). Vocabul ary. In D. Nunan (Ed.). Practical English language education (pp.129-152). New York: McGraw-Hill. National Reading Panel. (2000). Teaching children to read: An evi dence-based assessment of the scientific research litera ture on reading and its implications for reading instruction Washington, D.C.: U.S. Department of Health and Human Services. Nagy, W.E., & Scott, J.A. (2000). Vocabulary pr ocesses. In M.L.Kamil, P.B. Mothenthal, P.D.Pearson, & R. Barr (Eds.). Handbook of Reading Research (Vol.3; pp269-284). Nist, S., & Mealey, D. (1991). T eacher-directed comprehension stra tegies. In R. Flippo & D. Caverly (Eds.), Teaching Reading and Study Strategies (pp. 42-85). Newark, DE: International Reading Association. Nuttal, C. (1983). Teaching reading skills in a foreign language London: Heinemann. Olejnik, S. F. (1984). Planning educational research: Determining the necessary sample size. Journal of Experimental Education 53, 40-48. O'Malley, M. & Chamot, A. (1990). Learning strategies in second language acquisition New York: Cambridge University Press. Oxford, R. L. (1990). Language learning strategies: W hat every teacher should know Boston, MA: Heinle & Heinle Publishers. 246

PAGE 247

Oxford, R.L. Crookall, D. Cohen, A., Lavin, R., Nyiko, M.& Sutter, W (1990). Strategy training for language learners: Six situa tional case studies an d a training model. Foreign Language Annals, 22, 197-216. Oxford, R. & Crookall, D. (1989). Research on six situational language learning strategies: Methods, findings, and instructional issues. Modern Language Journal 73, (4), 404-419. Ordoez, C., Carlo, M., Snow, C. McLaughlin, B. (2002). Depth and brea dth of vocabulary in two languages: which skills transfer? Journal of Educational Psychology, 94(4), 719-728 Palincsar, A.S., & Brown, A.L.(1985). Reciprocal teaching of comprehension-fostering and monitoring activities. Cognition and Instruction, 1, 117-115. Pardon, Y., (1992). The effect of strategy instruction on bilingual students cognitive strategy use in reading. Bilingual Research Journal 16, 35-52. Pardon, Y., & Waxman, H. (1988). The effect of ESL students perceptions of their cognitive strategies on r eading achievement, TESOL Quarterly 22(1), 146-150. Paris, S. G., Cross, D. R., & Lipson, M. Y. ( 1984). Informed strategies for learning: A program to improve children's reading awareness and comprehension. Journal of Educational Psychology, 6 1239-1252. Paris, S. G., Lipson, M. Y., & Wixson, K. K. (1983). Becoming a strategic reader. Contemporary Educational Psychology, 8, 293-316. Paris, S. G., Wasik, B. A., & Turner, J. C. (1991) The development of strategic readers. In R. Barr, M. L. Kamil, P. B. Mosenthal, & P. D. Pearson (Eds.), Handbook of Reading Research, Volume II (pp. 609-640). New York: Longman. Pressley, M. (1998). Reading instruction that works: the case for balanced teaching New York: Guilford. Pressley, M., & Afflerbach, P. (1995). Verbal protocols of reading: The nature of constructively responsive reading Hillsdale, NJ: Lawrence Erlbaum. Pressley, M., Snyder, B., & Cariglia-Bull, B. (1987). How can good strategy use be taught to children? Evaluation of six al ternative approaches. In S. Cormier & J. Hagman (Eds.), Transfer of learning: Contempor ary research and application (pp. 81-120). Orlando, FL: Academic Press. Pressley, M. (2000). What should comprehension instruction be th e instruction of? In M. L. kamli, P.B. Mosenthal, P.D. Pearson, & R. Barr (eds.) Handbook of Reading Research Vol. III Lawrence Erlbaum Associates, Publishers Mahwah, New Jersey. 247

PAGE 248

Pressley, M. (2002). Metacognition and self-regulated com prehension. In A. E. Farstrup & S. J. Samules (Eds.), What research has to sa y about reading instruction (3rd ed., p. 291-309). Newark, DE: International Reading Association. Park, H. (2005). Second language reading: the interrelationships among text adjuncts, students proficiency levels and reading strategies Dissertation, University of Florida. Pearson, P.D. (1979). Basic Processes and Inst ructional Practices in Teaching Reading. New York: University Education Quarterly 10 (2). Pearson, P.D., & Gallagher, M.C. (1983). Th e instruction of read ing comprehension. Contemporary Education Quarterly 10(2). Phakiti, A. (2003). A closer look at gender and strategy use in L2 reading. Language Learning, 53, 649702. Pritchard, R. (1990). The effects of cultural schemata on reading processing strategies. Reading Research Quarterly, 25 (4), 273-295. Razi, S. (2004). The effects of cultural schema and reading activities on reading comprehension. In M.Singhal (Ed.), Proceedings of the 1st internati onal online conference on second and foreign language teaching and research (pp. 276-293). USA: The Reading Matrix Retrieved September 13, 2007 from onfe rence/pp/proceedings/razi.pdf Raymond, P. M. (1993). The effects of structur e strategy training on th e recall of expository prose for university students read ing French as a second language. Modern Language Journal 77, 445-458. Redmann, J. (2005). An interactive Reading Journal for all levels of the foreign language curriculum, Foreign Language Annals, Vol.38, 4, p 484-493. Readence, J. E., Bean, T. W., & Baldwin, R. S. (1989). Content area reading: An integrated approach. Dubuque, IA: Kendall/Hunt Publishing Company. Riley, L. & Harsch, K. (1999). "Using strategy journals to compare ESL and EFL students' awareness, development, and use of language le arning strategies." In J. E. Purpua and R. L. Oxford, Language Learning Strategies in the Context of Autonomy (compendium compiled for invitational symposium). New York: Teachers College, Columbia University. Roehler, L. R., & Duffy, G. G. (1984). Direct explanation of comprehension processes. In G. G. Duffy, L. R. Roehler, & J. Mason (Eds.), Comprehension Instruction: Perspectives and Suggestions (pp. 265-280). New York: Longman. Rubin, J. (2003). Diary writing as a pr ocess: simple, useful, powerful. Guidelines 25, 2, 10-14. 248

PAGE 249

Rum elhart, D.E. (1977). Toward an interact ive model of reading in Dornic, S. (Ed.), Attention and Performance VI, 573-603. New York: Academic Press. Rumelhart, D.E.(1980). Schemata: the building blocks of cognition in R.J. Spiro, B. C. Bruce, & W.F. Brewer (Eds.), Theoretical issues in reading co mprehension: perspectives from cognitive psychology, linguistics, artificial intelligence, and education (pp.33-58). Hillsadle, NJ:Erlbuam. Samuels, S.J. & Kamil, M.L. (1988). Models of th e reading process. In Carrell, P. Devine, J. & Eskey, D.(Eds.) Interactive Approaches to Second Language Reading (pp.22-36). New York: Cambridge University Press. Salataci, R. & Akyel, A. (2002). Possible effect s of strategy instruction on L1 and L2 reading. Reading in a Foreign Language, 14, 1. Sarig, G. (1987). High-level reading in the first and in the foreign language: Some comparative process data. In J. Devine, P. L. Carrell, & D. E. Eskey (Eds.), Research in reading in English as a second language (pp. 105-120). Washington, DC: TESOL. Sarig, G., & Folman, S. (1987). Metacognitive awareness and th eoretical knowledge in coherence production Unpublished paper presented at the Communication and Cognition International Congress, Ghent, Blegium. Sarig, G. (1987). High-level reading in the first and foreign language: some comparative process data. In J. Devine, P.L. Carrell, & D. E. Eskey (Eds.), Research in Reading in English as a second language (pp.105-120). Washington, D.C.: Teachers of English to Speakers of Other Languages. Schiefele, U. (1999). Intere st and learning from text. Scientific Studies of Reading 3, 257-279. Schunk, D.H., & Zimmerman, B.J. (Eds.). (1994). Self-regulation of learning and performance issues and educati onal applications Hillsdale, N.Y.: Lawrence Erlbaum Associates, Inc. Schulz, R. A. (1983). From word to meaning: Foreign language reading instruction after the elementary course. The Modern Language Journal 67(2), 127-134. Sharp, A. (2002). Chinese L1 schoolchildren readi ng in English: The effects of rhetorical patterns. Reading in a Foreign Language, 14, 111-135. Shih, Y.H., Lin, M.S. Joe, S.G., Hung, C.H. (1999). : Shih, Y.H., Lin, M.S. Huang, S.S., & Yeh, J.F. (2000). 589-596, Taipei: Crane. 249

PAGE 250

Sheorey, R., & Mokhtari., K. (2001). Differences in the m etacognitive awareness of reading strategies among native and non-native readers. System 29, 431-449. Singhal, M. (2001). Reading prof iciency, reading stra tegies, metacognitive awareness and L2 readers. Reading Matrix 1-9. Retrieved March 10, 2005, from: Skehan, P. (1989). Individual differences in second language learning London: Edward Arnold. Smith, F. (2004). Understanding reading (6th ed.) Mahwah, NJ: Lawrence Erlbaum. Snow, C. (2002). Reading for Understanding: Toward an R&D Program in Reading Comprehension Santa Monica, CA: The RAND Corporation. Song, M. (1998). Teaching reading strategies in an ongoing EFL university reading classroom. Asian Journal of English Language Teaching, 8, 41-54. Stoller, F.L., & Grabe, W. (1993). Implications for L2 vocabulary acquisition and instruction from L1 vocabulary research. In T. Huckin, M. Haynes, & J. Coady (Eds.), Second language reading and vocabulary learning (pp.23-44). Norwood, NJ: Ablex. SPSS. (2003). SPSS 12.0 for Windows. Stanovich, K. E. (1986). Matthew effects in reading: Some consequences of individual differences in the acquisition of literacy. Reading Research Quarterly, 21 380-407. Stanovich, K. (1980). Toward an interactive-compen satory model of individual differences in the development of reading fluency. Reading research Quarterly 16, 32-71. Stanovich, K. (2000). Progress in understanding reading: Scientific foundations and new frontiers. New York: Guilford Press. Takeuchi, O., C. Griffiths, and D. Coyle (2007). Applying strategies to context: The role of individual, situational, and group differe nces. In A. Cohen & E. Macaro (Eds.) Language Learner Strategies: Thirty Years of Research and Practice. (Ch4. pp.69-92) Oxford: Oxford University Press. Tang, W.N., & Moore, D. W. (1992). Effects of cognitive and met acognitive pre-reading activities on the reading comp rehension of ESL learners. Educational Psychology 12, 3 and 4, 315-331. Taylor, A., Stevens, J.R., & Asher, J.W. (2006). The effects of explicit reading strategy training on L2 reading comprehension: A meta-analysis. In J.M. Norris & L. Ortega, Synthesizing research on language learning and teaching (pp.213-244). Amsterdam: John Benjamins Publishing. 250

PAGE 251

Urquhart, S., & W eir, C. (1998). Reading in a second language: Process product and practice. London: Longman. Vygotsky, L.S. (1978). Mind in Society. Cambridge, MA: Harvard University Press. van Duzer, C. (1999). Reading and th e Adult English Language learner. ERIC Digest ED433729. Vann, R.J. and Abraham, R.G. (1990). Strategies of unsuccessful language learners. TESOL Quarterly 24 (2), 177. Wagner, R.K. and Sternberg, R.J. (1984). Alterna tive conceptions of intelligence and their implications for education. Review of Educational Research 54(2): 179. Wallace, C. (1992). Reading. Oxford: Oxford University Press. Wallace, C. (1993). Language teaching: Reading Oxford: Oxford University Press. Wen, Q., & Johnson, R.K. (1997). L2 learner va riables and English achievement: A study of tertiary-level English majors in China. Applied Linguistics, 18 (1). Wharton, G. (2000). Language lear ning strategy use of bilingual foreign language learners in Singapore. Language Learning, 50, 2, 203-244. Wigfield, A., & Guthrie, J. T. (1995). Dimensions of children's motivations for reading: An initial study (Research Rep. No. 34). Athens, GA: National Reading Research Center. Wigfield, A., & Guthrie J. T. (1997). Relations of children's motivation for reading to the amount and breadth of their reading. Journal of Educational Psychology, 89, 420-432. Winograd, P., & Hare, V. C. (1988). Direct instru ction of reading comprehension strategies: The nature of teacher explanation. In C. E. Weinst ein, E. T. Goetz, & P. A. Alexander (Eds.), Learning and study strategies: Issues in assessment instruction and evaluation (pp. 121139). San Diego: Academic Press. Wong, B.Y.L., & Jones, W. (1982). Increasing comprehension in le arning disabled and normally achieving students through self-questioning training. Learning Disability Quarterly 5 228-239. Yang, Y.-F. (2002). Reassessing read ers comprehension monitoring. Reading in a Foreign Language, 14(1), 18-42. Yang, Y.-F. (2006). Reading strategies or comprehension monitoring strategies? Reading Psychology, 27(4), 313-343. Yorio, C.A. (1971). Some sources of reading problems for foreign language learners. Language Learning, 21, 107-115. 251

PAGE 252

Yoshim ura, F. (2000). Automaticity Theory and EF L in Japan: With Some Specific Applications for Reading. Literacy across Cultures, 4, 1-10. You, Y.L. (2004). Curriculum and Research Development center for Commerce Education. Retrieved January 26, 2004, from /e_news_12.htm Young, D.J. (1993). Processing strategi es for foreign language readers. Foreign Language Annals, 26, 451-468. Young, D.J., & Oxford, R. (1997). A gender-related anal ysis of strategies us ed to process written input in the native language and a foreign language. Applied Language Learning 8, 4373. Zheng, L. (2002). Discovering EFL learners pe rception of prior knowle dge and its roles in reading comprehension. Journal of Research in Reading, 25, 172-190. Zheng, L. J. (2008). A cognitive perspective on Si ngaporean bilingual childrens use of reading strategies in learning to read in English. British Journal of Educational Psychology 78, 245-271. 252

PAGE 253

253 BIOGRAPHICAL SKETCH Min-Tzu Wang was born and raised in Ka ohsiung City, Taiwan. She obtained her undergraduate education at the National Chunghwa University of Education in Taiwan, where she majored in special educati on and minored in English. After receiving her Bachelor of Education, Min-Tzu worked as a se condary teacher for five years. Then she was admitted as a graduate student in the School of Teaching and Learning at the University of Florida in 2001. She earned her Masters of Education from the Un iversity of Florida in December of 2003. Then she was admitted as a doctoral student at the University of Florida in 2004. Upon graduation, Min-Tzu plans to resume her sec ondary teaching career in Taiwan.