互惠式教學法對不同英語能力學生的閱讀理解之成效

全文

(1)國立臺灣師範大學英語學系碩士論文 Master’s Thesis Graduate Institute of English National Taiwan Normal University. 互惠式教學法對不同英語能力學生的閱讀理解之成效 The Effects of Reciprocal Teaching on Reading Comprehension of Students with Different Competence in English. 指導教授：陳秋蘭博士 Advisor: Dr. Chiou-lan Chern 研究生：蔡涵如 Han-ju Tsai 中華民國一百零二年七月 July, 2013.

(2) 摘要許多研究顯示「互惠式教學法」能有效地提升學生的閱讀理解能力；然而，在台灣，尚未有研究同時囊括兩個不同能力的群組進行此法教學並比較其不同。因此，本研究旨在檢視「互惠式教學法」施行於不同英語能力的台灣國中生，探討其英語閱讀理解方面的提升成效，及他們對此教學法的回應及建議。參與本次研究的對象為台中市某國中的六十四位國三學生，學校依照他們上學期的英語平均成績將其分為兩個班級：排名前半的學生編入「成就較高班」，而另一半學生編入「成就較低班」。這兩班的英文老師（也是本次研究者）進行了為期五週，每週兩次的互惠式教學研究實驗。實驗中使用的閱讀教材為九篇敘述性文章，皆取自各版本教科書及英語學習雜誌；而研究工具則採用取自全民英檢的初級閱讀理解試題為前後測試卷，及一份具體探討學生對此教學實驗看法的回饋問卷。為分析前後測分數的變化，本研究使用成對樣本 t 檢定的方式來判斷兩組學生是否有顯著進步；而研究對象對於此教學法的回應和建議則透過描述性統計的方式來呈現。根據本次研究結果顯示，「互惠式教學法」對「成就較高班」及「成就較低班」的學生們在英語閱讀理解方面皆有顯著進步，而大多數的實驗對象對「互惠式教學法」抱持正面肯定的態度，並認為此法同時提升他們的學習興趣、引領他們反思自己的學習或思考過程……等等。其中，針對四個閱讀策略，兩班都認為「預測」最簡單，而「釐清」最有助益，但「成就較高班」認為最難的是「提問」，而「成就較低班」則認為是「摘要」。依據上述結果，本研究建議「互惠式教學法」可融入一般課室英語教學，不同英語能力的學生都將各有收穫，值得一試；此外，結論中並提出實施此法的建議，期盼能為未來研究提供參考。關鍵字:互惠式教學法、閱讀理解、不同能力. i.

(3) Abstract. A lot of research showed that reciprocal teaching could promote students’ reading comprehension effectively, but few related studies in Taiwan have been conducted to examine the effects by including students of different English competence in a real teaching scenario and comparing the differences between them. Therefore, the purpose of the present study is to explore the effects of reciprocal teaching on reading comprehension of students with different proficiency levels, and to investigate their responses and suggestions to the intervention. The participants were 64 ninth graders in a junior high school in Taichung City. Based on their average English scores in the previous semester, the top 50% of the two classes went to a higher achiever class while the rest to a lower achiever class. In this study, the former was assigned as the higher achiever group, and the latter as the lower achiever group. Meanwhile, the researcher was the teacher for both groups. The reciprocal teaching intervention comprised ten sessions during five weeks, i.e. two sessions per week. The reading materials were nine descriptive articles from either textbooks or English magazines. The research instruments were a reading comprehension test from GEPT and a perception questionnaire. To analyze the results from pretest and posttest, paired sample t-test was used to investigate whether the two groups showed significant improvement. Besides, the responses and suggestions to the intervention were presented via descriptive statistics. According to the results, reciprocal teaching is effective in promoting reading comprehension of students with different proficiency. Most of the participants were positive about the treatment and credited it for increasing learning interests, enhancing metacognition, etc. Among the four reading strategies, both groups thought predicting was the easiest while clarifying the most helpful. But, in terms of the most difficult ii.

(4) strategy, the higher achievers voted for questioning whereas the lower achievers chose summarizing. As revealed in the findings, reciprocal teaching can be incorporated in the English classroom in junior high schools because students of different competence would benefit from it in some way. Furthermore, suggestions were included in the conclusion for future research. Key Words: reciprocal teaching, reading comprehension, different competence. iii.

(5) Acknowledgements To get the master’s degree is my dream. I would like to express my appreciation to those who supported me to make my dream come true. First of all, I want to show my upmost gratitude to Dr. Chiou-lan Chern, my dear adviser. She is very kind and generous to me. When I first talked with her about my idea, she gave me some books related to my topic and showed me some possible ways to start. When I encountered problems, she offered immediate help for me. She read my writing word by word and helped me to present my thesis in a clearer way. I always tell the people around me that I am the luckiest one to have Dr. Chern as my adviser. From her, I learn about what a good teacher should be like. Without her, I would not be able to complete my thesis. Besides, I would like to express my deepest appreciation to my committee members, Dr. Chih-cheng Lin and Dr. Chin-kuei Cheng. They raised inspiring questions for me to think about. They also offered insightful comments and suggestions for me to make more exploration of the collected data. Last but not least, I want to say thanks to my family who have supported and encouraged me over the past five years. They kept me company when I needed somebody to talk to. They left me alone when I had to be busy with my study. Without their support and love, I might not have had the strength to make my dream come true. With sincere gratitude, I dedicate this study to them.. iv.

(6) Table of Contents. Abstract (in Chinese) ...................................................................................................... i Abstract..........................................................................................................................ii Acknowledgements ...................................................................................................... iv Table of Contents........................................................................................................... v List of Tables ................................................................................................................ ix Chapter One Introduction .............................................................................................. 1 Background and Motivation .................................................................................. 1 Significance of the Study....................................................................................... 3 Research Questions ............................................................................................... 4 Overview of the study ........................................................................................... 4 Chapter Two Literature Review..................................................................................... 6 Reading Theory ..................................................................................................... 6 Reading models. ............................................................................................ 6 Reading comprehension. ............................................................................... 7 Reading Strategies ................................................................................................. 8 Definitions of reading strategies.................................................................... 8 Empirical studies on reading strategies. ........................................................ 9 Explicit Instruction of Reading Strategies........................................................... 11 v.

(7) The importance of explicit strategy instruction in junior high school......... 13 Reciprocal Teaching ............................................................................................ 15 Definition of reciprocal teaching. ................................................................ 15 Four strategies of reciprocal teaching.......................................................... 15 Empirical studies on reciprocal teaching..................................................... 17 Chapter Three Method................................................................................................. 22 Setting.................................................................................................................. 22 Participants .......................................................................................................... 23 Materials .............................................................................................................. 23 Reciprocal Teaching Intervention........................................................................ 26 Stage I: Demonstration of four strategies. ................................................... 26 Stage II: Teacher-led discussion. ................................................................. 29 Stage III: Student-led discussion. ................................................................ 32 Research Instruments........................................................................................... 34 An English reading comprehension test. ..................................................... 34 A perception questionnaire .......................................................................... 37 Data-collecting Procedures.................................................................................. 38 Data Analysis....................................................................................................... 40 Chapter Four Results and Discussion.......................................................................... 41. vi.

(8) Results of the Reading Comprehension Tests ..................................................... 41 Results of the Perception Questionnaire.............................................................. 44 Students’ overall perception towards learning via reciprocal teaching. ...... 44 Students’ reflection on their practice of the four reading strategies. ........... 49 Students’ belief of reciprocal teaching in promoting reading comprehension and interests. ................................................................................................ 54 Students’ attitudes toward using reciprocal teaching in the future. ............. 56 Discussion............................................................................................................ 61 Effects of reciprocal teaching instruction on reading comprehension......... 61 Perception of reciprocal reaching instruction .............................................. 63 Feasibility of reciprocal teaching in EFL classrooms in Taiwan................. 67 The issues of time and materials.................................................................. 70 Chapter Five Conclusion ............................................................................................. 72 Summary of Findings .......................................................................................... 72 Pedagogical Implications..................................................................................... 73 Suggestions for Future Research ......................................................................... 74 References ................................................................................................................... 76 Appendices .................................................................................................................. 82 Appendix A: Guides for Different Roles in Reciprocal Teaching....................... 82. vii.

(9) Appendix B: The Reciprocal Teaching Worksheet ( A shrunk size from the original) ............................................................................................................... 83 Appendix C: Sample of Instruction ..................................................................... 84 Appendix D: Lesson Plans .................................................................................. 86 Appendix E: Reading Comprehension Test......................................................... 90 Appendix F: The Perception Questionnaire ........................................................ 98 Appendix G: The Perception Questionnaire (in Chinese) ................................... 99. viii.

(10) List of Tables. Table 1 The Readability and Length of the Reading Materials ................................... 25 Table 2 Description of the Four Roles in Reciprocal Teaching................................... 32 Table 3 The Design of the Reciprocal Teaching Intervention ..................................... 33 Table 4 The Results of Pre-and Posttest in Reading Comprehension in Group A and Group B ............................................................................................................... 37 Table 5 Three Parts of the Present Study..................................................................... 39 Table 6 Results of the Pre-and Posttest in Reading Comprehension in Both Groups . 42 Table 7 Result of the Anchor Questions in Reading Comprehension in Both Groups 43 Table 8 Improvement in Anchor Questions ................................................................. 43 Table 9 Overall Perceptions towards Learning via Reciprocal Teaching in Both Groups ................................................................................................................. 45 Table 10 Reflection on Practice of the Four Reading Strategies ................................. 50 Table 11 Belief in Promoting Reading Comprehension and Interests......................... 54 Table 12 Expectation of continuing Receiving Reciprocal Teaching.......................... 56. ix.

(11) Chapter One Introduction. This introductory chapter comprises four sections. The first part starts with the background and motivation of the present study. The second section illustrates the significance of this study. Next, the research questions to be addressed are presented in the third part. Finally, an overview of the chapters in this thesis is included at the end of this chapter.. Background and Motivation Reading plays an important role in learning. In a first language context, learners have to develop the reading ability first before they can proceed to learn new information or knowledge via reading (Alvermann & Earle, 2003). Moreover, readers start to ask questions requiring deep thoughts after learning to read actively (Molden, 2007). The ability to ask good questions is often associated with the ability to think analytically (Paul & Elder, 2001). In brief, reading ability forms the basis of further learning and thinking. Reading plays a crucial role in second and foreign language learning. Language input is usually initially received via listening in a first language. However, it is often via reading that learners are exposed to language input in a second or foreign language (Schramm, 2008). Since English is a foreign language in Taiwan, reading may be the most principal source of language input for EFL learners in Taiwan. Stated another way, EFL learners in Taiwan have much more chances to read English than to listen to, speak in, or write in English. With the understanding of the importance of English, the Ministry of Education (MOE) in Taiwan implemented English instruction to Grade 3 curriculum in 2005. Besides, more and more parents have their pre-school children take English classes 1.

(12) before they begin the formal courses in elementary schools. Even with English courses at school, many students attend extra classes after school. However, with so much attention paid to English, there are still one third of the junior high school students identified as underachievers, which leads to the polarization of students’ English proficiency (Hsu & Chen, 2007). To solve the problems mentioned above, many plans and programs are implemented to help struggling English learners. For example, Focus 300 aims to promote the reading habits of elementary school children in remote areas. Remedial instruction offers lessons for underachievers to meet their individual needs. However, little attention is paid to regular classrooms. In addition to taking care of the struggling students after class, there may be some approaches that can benefit both good readers and poor readers. Reading is a multifaceted process requiring both decoding and meaning construction. Successful readers sometimes rely more on reading strategies than decoding ability to construct meaning (Adams, 1990; Goodman, 1967; Pressley, 2000). However, generally speaking, English teachers in Taiwan spend most of the class time on teaching decoding skills, ignoring reading comprehension strategies (Chen, 2003; Hsu, 2009; C. Wu, 2002). Students always learn passively, i. e. listen to lecture quietly or take notes consistently with little thinking. They seldom participate actively in class. In the long run, they become either bored with the course or overwhelmed by the ineffective learning. In fact, the effectiveness of reading strategy instruction on reading comprehension has been proved in many studies (Boulware-Gooden, Carreker, Thornhill, & Joshi, 2007; Dymock, 2007). Reciprocal teaching, a method for reading strategy instruction, was first proposed by Palincsar and Brown (1984). After they developed the framework, a number of studies were conducted to investigate how 2.

(13) reciprocal teaching help language learners (Alfassi, Weiss, & Lifshitz, 2009; Ash, 2005; Goodman, 2005; Lysynchuk, Pressley, & Vye, 1990; Myers, 2005; Takala, 2006; Williams, 2010). Reciprocal teaching benefits learners with different proficiency levels (Alfassi et al., 2009; Ash, 2005; Goodman, 2005; Lysynchuk et al., 1990), in different settings (Myers, 2005; Williams, 2010), and in language contexts other than English (Takala, 2006). In Taiwan, plenty of studies were conducted to explore the effects of reciprocal teaching on EFL learners (Chern, 2005; Chou, 2008; Shiau, 2010; Su, 2010; Wu, 2012). Some of them probed the issue in terms of either a single group of students (Chern, 2005; Shiau, 2010; Wu, 2012) or the control/ experimental homogeneous group design (Chou, 2008; Su, 2010). However, few studies have tapped into the comparison of the effects of reciprocal teaching on students of different English proficiency levels. Therefore, the present study is motivated to bridge the gap in existing literature and explore how reciprocal teaching helps EFL learners of different competence in English.. Significance of the Study The present study is significant in many ways. First, the present study explored the impact of reciprocal teaching on reading comprehension of learners of different English proficiency levels. Previous studies tended to employ the control-and-experimental-group design to investigate the effect of reciprocal teaching (Chou, 2008; Su, 2010). However, the present study, based on the previous findings of positive impacts on reading comprehension, divided students of different proficiency into either a higher achiever group or a lower achiever group, and both groups participated in the intervention. Thus, the present study provided a different point of view to explore the effects of reciprocal teaching on reading 3.

(14) comprehension and the application of reciprocal teaching in a real teaching scenario in which students are at different proficiency levels. Moreover, the present study evaluated the effects of reciprocal teaching from the perspective of Taiwanese EFL junior high school students. It explored students’ attitudes and preference towards each of the four strategies.. Research Questions The research questions to be addressed in the present study are as follows: 1. Do both higher achievers and lower achievers improve their reading comprehension after reciprocal teaching? 2. How do higher achievers and lower achievers perceive reciprocal teaching procedures? The first research question will be answered by the results of the pretest and the posttest of the reading comprehension. Paired sample t-test will be employed to measure the scores of the pretest and posttest in reading comprehension in each class. The result of the paired sample t-test showed whether there is a significant effect of reciprocal teaching in each group. It will also show if there is any significant difference between the two groups. The second research question will be explored by examining students’ responses to the perception questionnaire. The results will reveal how students perceive reciprocal teaching and what suggestions they have in terms of future application of reciprocal teaching.. Overview of the study This thesis is composed of five chapters. The first chapter provides a brief introduction to the research background and motivation, the significance of the study, 4.

(15) and the research questions. After the introductory section, literature about reading theory, reading strategies, explicit instruction of reading strategies and reciprocal teaching are reviewed in the second chapter. Next, the research method, including setting, participants, reciprocal teaching intervention, materials, research instruments, data collecting procedures, and data analysis as well as the pilot test, are included in Chapter Three. Then, the results of data and findings of the analysis are discussed in Chapter Four. Finally, summary of the major findings, pedagogical implications, limitations and suggestions for future research are presented in Chapter Five.. 5.

(16) Chapter Two Literature Review. This chapter provides a theoretical background of the present study. Reading theories are presented first. Then, reading strategies are reviewed and followed by explicit instruction of reading strategies. Finally, reciprocal teaching is introduced to promote reading comprehension with the four reading strategies.. Reading Theory Reading models. Reading is a psycholinguistic process in which it starts with a linguistic surface representation encoded by a writer and ends with meaning which the reader constructs (Goodman, 1967). Among various reading models, bottom-up reading, top-down reading and the interactive reading are the most widely cited ones (Brown, 2001). A bulk of research documents how reading theory has progressed from bottom-up, top-down to the current interactive model. In the past, reading was considered as a passive, receptive and decoding process. In the bottom-up model, readers extract the meaning of the text by recognizing different kinds of linguistic signals, such as morphemes, words, phrases, and grammatical rules (Brown, 2001). However, the over-simplified processing was challenged by Goodman (1988), who pointed out this difference between bottom-up and top-down models. Different from bottom-up model, top-down model explains that the process of reading is driven by a more general concept, rather than linguistic segments. In other words, construction of meaning starts from the reader’s application of prior knowledge and experience to infer from the context. In the top-down model, readers make predictions based on the context or their background knowledge. To understand a text, readers confirm or deny the prediction by the clues they attain 6.

(17) during the reading process. After the debate between the bottom-up and the top-down models, however, researchers on reading have recently reached consensus on interactive model, the combination of bottom-up and top-down models (Brown, 2001). In the interactive model, readers shift from bottom-up to top-down and vice versa to compensate for the comprehending failure and to strive for meaning. To sum up, this model balances the insufficiency of either the bottom-up model or the top-down one. Instead of passive decoding, the interactive model includes the interaction between the reader and the text. Among the substantial body of research on reading theory, most reading models are L1-based. L1 models are helpful to explain the reading process in general. However, they are not specific enough to indicate the unique characteristics of L2 reading. Therefore, Bernhardt (2005) presented an interactive compensatory version of the reading model. The three-dimensional reading model illustrates the L2 reading process in terms of not only textual processes and intratextual variables but also L2-specific factors. First, textual processes include word recognition, syntactic processes and graphophonic ones. Second, intratextual variables contain prior knowledge and strategies. Last but not least, the L2-specific factors indicate the important roles that L2-proficiency and difference between L1 and L2 play in the reading process. Though different from each other, Bernhardt’s model and its previous counterparts share a goal: to explain the interactive nature of the process of achieving comprehension.. Reading comprehension. Snow (2002) defined reading comprehension as the process of simultaneously extracting and constructing meaning through interaction and involvement with written 7.

(18) language. Based on the involvement of literal and inferential understanding of what is read, comprehension was explained in terms of three elements: the reader, the text, and the activity or purpose for reading (Samuels, 1993). Besides, comprehension occurs within a larger sociocultural context that shapes and is shaped by the reader. In other words, readers can be both passively shaped and actively participating in the process of comprehending. Since readers can participate actively in the reading process, they may employ some methods to approach comprehension more effectively. One of the good methods to facilitate comprehension is to use reading strategies.. Reading Strategies The National Reading Panel (2000) indicates that reading strategies, such as comprehension monitoring, question generation and summarization, have accumulated sufficient empirical evidence for effectiveness. In the following section, existing definitions of reading strategies will be presented first and then some empirical studies supporting the effects of reading strategies will be discussed.. Definitions of reading strategies. Reading strategies were early viewed as an action employed by the reader to construct meaning (Garner, 1987). To be more specific, Gambrell and Dromsky (2000) defined reading strategies as plans that engaged the reader in gathering, monitoring, evaluating ,and using text information to solve problems encountered during the reading process. In terms of interactive reading model, reading strategies are considered a variety of behaviors that interact with each other so as to help readers tackle a text, construct meaning, and develop language proficiency (Devine & Kania, 2003; Walczyk & Griffith-Ross, 2007). However, Cohen (1990) did not focus on the 8.

(19) interaction between readers and texts, but emphasized the characteristics of readers’ awareness. He indicated that reading strategies were mental processes which readers consciously chose to take to achieve the reading tasks. Instead of coping with the issue of definitions, some researchers have moved a step further to discuss what a strategic reader does while reading. Readers were considered to possess a strategy repertoire from which they were able to retrieve appropriate reading strategy to deal with their reading difficulties and to accomplish their reading tasks. Though they had adequate knowledge to solve the problems, they might not be able to activate it constructively (Schoenfeld, 1987). Once they had the ability to employ proper reading strategies flexibly and successfully, they showed the characteristics of strategic readers (Hsiao & Oxford, 2002). According to Block (1986), readers were categorized as integrators and non-integrators. The former focused on getting the main ideas during reading while the latter on recalling the details. Furthermore, integrators related the text information to their prior knowledge, became aware of text structure, monitored their comprehension process consistently and made efforts to solve the problems they encountered. Since the characteristics mentioned above contributed to their reading comprehension, integrators could be viewed as strategic readers. Moreover, Long and Long (1987) analyzed the differences between successful readers and unsuccessful readers. Successful readers were found to focus their attention on the key points and to relate the text to their background knowledge. This finding echoed Block’s (1986) conclusion and indicated some characteristics strategic readers shared in common.. Empirical studies on reading strategies. A growing number of studies are now available to shed light on the effects of specific reading strategies: analyzing grammar (Layton, Robinson, & Lawson, 1998), 9.

(20) identifying text structure (Carrell, 1985; S. Dymock, 2005; Fisher, Frey, & Lapp, 2008), metacognitive strategies (Fitzgerald, 1995; Kletzien, 1992; Long & Long, 1987; Mokhtari & Reichard, 2002), and semantic mapping (Chuang, 2010; Sinatra, Stahl-Gemake, & Burg, 1984). These studies on reading strategies have been proved to help students enhance reading comprehension, build confidence in reading and raise awareness of reading strategies. Since many reading strategies were proved helpful to enhance reading comprehension, researchers went further to examine the reading process and to identify the characteristics strategic readers had in common. Hosefeld (1977) compared reading strategies of successful L2 readers with those of unsuccessful L2 readers. Successful readers were found to employ more strategies, such as reading in broad phrases and skipping unimportant words, to get the main idea. In contrast, unsuccessful readers focused on decoding and read in short phrases. They were not flexible enough to skip unimportant words/information and were unable to use their prior knowledge to decode unknown words. The result revealed that the successful readers used more top-down strategies whereas the unsuccessful readers used more bottom-up ones. Carrell (1989) conducted a study on the strategy use of 75 native English speakers learning Spanish and 45 native Spanish speakers learning English. The results showed that proficient English learners used more top-down strategies, while less-proficient Spanish learners used more bottom-up strategies in reading. Interestingly, it seemed that the studies reviewed above showed that the difference between good readers and poor readers is their tendency to choose the strategies. In other words, good readers tended to use effective strategies consistently and comprehend successfully while poor reader tended to choose less effective strategies and fail to reach comprehension frequently. However, recent research challenged the 10.

(21) simplified relationship between reading strategies and reading comprehension. Carrell (1991) indicated that “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.” Anderson (1991) conducted a qualitative and quantitative study to examine individual difference in strategy use by L2 learners. The participants were engaged in two reading tasks: taking a reading comprehension test and reading academic texts. As revealed in the results, both good readers and poor readers used the same kinds of strategies in either task. Their reading comprehension depended not on how many reading strategies were used but on how the strategies were applied. Moreover, the poor readers in the case study appeared to know what strategies to use but might not know how to make sure whether she applied them successfully. These phenomena suggested that strategic readers not only know what strategies to apply but also when, why, and how to apply them appropriately. To sum up, many studies evidence the value of reading strategy. However, assessment of comprehension is common in the classroom while strategy instruction is rare (Kelly, Moore, & Tuck, 1994; Stricklin, 2011). Therefore, explicit instruction of reading strategies is highly needed.. Explicit Instruction of Reading Strategies King and Johnson (1999) claimed that the implementation of reciprocal teaching is explicitly beneficial for the students in acquiring and utilizing strategies while reading. When implementing reciprocal teaching in their classrooms, Pilonieta and Medina (2009) mentioned that explicit instruction of strategies was one of the key elements found in effective comprehension strategy training. The following section will first discuss the value of explicit instruction of reading strategies, and then move 11.

(22) on to the importance of urging explicit instruction of reading comprehension in junior high school classrooms. The value of explicit instruction of reading strategies. Though the value of reading comprehension strategies has been well-established, some teachers undervalued the necessity of reading strategies (Denton & Fletcher, 2003). Denton and Fletcher (2003) indicated the misunderstanding of the nature of reading strategies. Those who assumed that reading comprehension would develop naturally without any formal instruction placed reading in the same developmental pattern as speaking. In fact, children’s ability to acquire speech naturally relies on the exposure to it. However, reading and speaking do not share the same progression in common. Besides, people have been communicating via speech for a much longer time than via the print (Boulware-Gooden et al., 2007; Wren, 2002). Therefore, reading strategies are prerequisites for fluent reading and need to be taught through formal education. In fact, considerable research has documented the effectiveness of reading strategy instruction and proved that reading strategies could be taught and trained via explicit instructions (Boulware-Gooden et al., 2007; Dymock, 2007; Herrmann, 1988; Onofrey & Theurer, 2007; Paris, Wasik, & Turner, 1991; Pressley, 2000; Snow, 2002). Snow (2002) indicated that teachers played a vital role in helping students develop reading strategies. Reading strategies should be taught explicitly and modeled for sufficient time. Thus, with appropriate instruction, students at all grade levels would learn to promote their comprehension development and prevent reading comprehension problems. Snow’s (2002) validation of explicit strategy instruction was not new. Herrmann (1988) suggested that teachers use direct explanation to help students 12.

(23) understand how reading process worked and how to learn from text. She first encouraged teachers to model reasoning process experienced by successful readers. This mental modeling made invisible reasoning process become visible to students. Next, the teacher asked students to articulate how they construct meaning from the text. Additional explanation was provided by the teacher to make students reason like experienced readers. With the scaffolding from the teacher, students gradually understand the rationale of reading and using the strategies. Pressley (2002) echoed the idea with contention for acquainting students with a strategic process and to teaching them a small repertoire of strategies. Boulware-Gooden et al. (2007) conducted a study to enhance reading comprehension and vocabulary achievement of 119 third graders via explicit strategy instruction. They divided students into the intervention group and the comparison group. Both groups participated in the same instruction activities, but only the intervention group was supplemented with direct strategy instruction. After a five-week intervention, the intervention group outperformed the comparison group by a 40% difference in gains in vocabulary and a 20% difference in gains in reading comprehension.. The importance of explicit strategy instruction in junior high school. Though being documented to contribute to students’ reading comprehension, reading comprehension strategy instruction has made little progress in classrooms (Drukin, 1978/1979; Kelly et al., 1994; Pressley, Wharton-McDonald, Mistretta-Hampston, & Echevarria, 1998; Sporer, Brunstein, & Kieschke, 2009). An early observation of the lack of reading strategy instruction was made by Drukin (1978/1979), who found that many assessments of reading comprehension existed in elementary school classrooms but little explicit comprehension instruction was 13.

(24) implemented to help students tackle the text. To be more accurate, less than 1% of the reading period was spent on teaching reading strategies. Almost three decades later, Pressley et al. (1998) investigated instructional practice in ten fourth-grade and fifth-grade classrooms and found that direct teaching of reading strategies was minimal. Some of the teachers mentioned strategy use, but they did so in a passive manner. They taught the strategies neither actively nor explicitly. Some teachers taught students to use strategies to find keys to reading comprehension questions on tests, but their students were not observed using the strategies during reading. While some teachers tried to incorporate strategy use in their instruction, most teachers just ignored teaching the use of strategies. The absence of explicit instruction about reading strategy has been noticed recently (Pressley, 2006; Sporer et al., 2009; Taylor, Peterson, Pearson, & Rodriguez, 2002). Although strategy instruction should help students at all grade levels (Sweet & Snow, 2003), Paris et al. (1991) proposed a special need for adolescent learners to receive explicit strategy instruction. In terms of reading, by adolescence, children began to show tendency to use strategies, which made them able to respond actively and adequately while reading. Moreover, even though most readers should have already grasped basic reading process by age 12, they still lacked well-articulated concepts about effective strategy use and needed explicit instruction to tell them what strategy to apply, how they functioned, and why they enhanced comprehension. Via explicit instruction, students can learn the effective strategy use and be reminded to enforce the strategy. In conclusion, explicit instruction on reading strategies promotes the development of comprehension. Not only L1 learners, but also EFL learners need to receive strategy instruction through formal education. When thinking of reading strategy instruction, many teachers and researchers think of reciprocal teaching. 14.

(25) Reciprocal Teaching Definition of reciprocal teaching. Reciprocal Teaching was an instructional method first described by Palincsar (1982) and by Palincsar and Brown (1984). The details of this method were elaborated in their later articles. Characterized as a dialogue between the teacher and students, reciprocal teaching was an instructional procedure to foster and to monitor reading comprehension. To achieve the goals of comprehension-fostering and comprehension-monitoring, the dialogue between the teacher and students should be carried on with four strategies: predicting, questioning, clarifying, and summarizing (Palincsar, David, & Brown, 1989).. Four strategies of reciprocal teaching. In reciprocal teaching, four metacognitive strategies, predicting, questioning, clarifying, and summarizing are utilized to construct meaning of the text. To start with, predicting involves activating students’ schemata to make a reasonable guess about the topic, the following passages, and the author’s purpose (Palincsar & Brown, 1984). When making a prediction, students can read the title and the pictures to activate the prior knowledge before reading. During reading, students may confirm, reject, or revise their predictions (Hashey & Connors, 2003). Students can also look back and forth to collect more information for further prediction. To sum up, predicting motivates students to read and provides them with a purpose for reading (Stricklin, 2011). Second, questioning requires students to recognize the main idea and specific features, so they can promote their comprehension and create their own questions (Palincsar & Brown, 1984). Students may generate a variety of questions. Hashey and 15.

(26) Connors (2003) categorized them into “skinny” questions and “fat” questions. The former can be answered simply with yes, no, or any short responses based on the text while the latter call for deeper thoughts and complete answers. Fat questions are similar to what a teacher may ask on a test because students need sufficient comprehension to answer them. In other words, fat questions are indicators of reading comprehension. Therefore, students are encouraged to generate questions of this kind. In brief, questioning enhances students’ comprehension because students must understand what they have read in order to propose questions (Stricklin, 2011). Next, clarifying helps students to tackle the unknown words and unclear ideas by monitoring their reading comprehension (Palincsar & Brown, 1984). When examining their comprehension by themselves, students can utilize the context clues and employ the idea of prefix and suffix to construct meaning. When sharing ideas in group discussions, students can provide their peers with their different background knowledge and strategy use. More specifically, while using the strategy of clarifying, students utilize their metacognition process to monitor their reading comprehension. Accordingly, clarifying can also promote comprehension and encourage students to become independent readers who can repair comprehension breakdown by themselves (King & Johnson, 1999). Finally, summarizing requires students to separate the main idea from the supporting details. It requires highly integration of different skills which students need to make a clear and concise summary (Palincsar & Brown, 1984). According to Hashey and Connors (2003), summarizing helps to enhance students’ comprehension since it requires them to concentrate on the key points and to exclude all the details. Besides, summarizing was viewed as the strategy which indicated whether students comprehended the text after they finished the summaries of each paragraph (King & Johnson, 1999). 16.

(27) The four strategies above can be arranged in different order to accommodate individual needs. In fact, Palincsar and Brown (1984) did not arrange the four strategies in a fixed order. They only emphasized that the four strategies should be instructed one after another. Moreover, before introducing a new strategy, a review of the previously taught strategies is important.. Empirical studies on reciprocal teaching. Originally designed for adequate decoders but poor comprehenders in junior high schools, reciprocal teaching has been proved to be an effective teaching method in a variety of settings (Alfassi, Weiss, & Lifshitz, 2009; Ash, 2005; Chern, 2005; Goodman, 2005; Lysynchuk, Pressley, & Vye, 1990; Myers, 2005; Palincsar & Brown, 1984; Takala, 2006; Williams, 2010). Similar to the experiment in Palincsar and Brown’s (1984), poor readers were recruited in the studies of Goodman’s (2005) and Lysynchuk et al.’s (1990). However, gifted and advanced students were found in Chern’s (2005) and Ash’s (2005) studies. Besides, research was done with average students in a classroom setting (Myers, 2005; Williams, 2010), as well as with students who have intellectual disabilities (Alfassi et al., 2009; Takala, 2006). Beside Takala (2006), Chou (2008) and Wu (2012) explored the effects of reciprocal teaching on students’ reading comprehension in EFL contexts. Except in Takala’s (2006) study, however, there have been few attempts to include students of different proficiency levels in a study and explore whether they perceive and benefit from reciprocal teaching in the same way. Besides developing the framework of reciprocal teaching, Palincsar and Brown (1984) provided empirical studies to evaluate this method. Palincsar and Brown (1984) recruited 12 seventh graders all classified as “adequate decoders, but poor comprehenders.” The participants were equally divided into the experimental group 17.

(28) and the control group. The former received reciprocal teaching intervention for 20 sessions, thirty minutes each, while the latter was instructed in the traditional way. Both groups were assessed by a pretest and a posttest of standardized comprehension tests. In terms of the results, the reciprocal teaching group improved significantly in reading comprehension tests. However, this salient finding seemed to be controversial to some researchers (Carver, 1987; Pearson, 1985). To verify the effects of reciprocal teaching on reading comprehension, Lysynchuk et al. (1990) made a constructive replication (Lykken, 1968) of Palincsar and Brown’s (1984). In the study, 72 participants, 36 fourth graders and 36 seventh graders, participated in 13 sessions of reading experiment. Participants of the identical grade were grouped into the reciprocal group and the control group. The procedure and the type of comprehension measures were similar to those in Palincsar and Brown’s (1984). This design yielded a great increase of scores from the pretest to the posttest of reading comprehension. Even recently, reciprocal teaching has been verified with statistically significant results to show the effect on reading comprehension improvement of poor readers (Goodman, 2005). Besides struggling readers, proficient readers could also benefit from reciprocal teaching beyond standardized comprehension measurements (Ash, 2005; Chern, 2005). In Chern’s (2005) study, 34 English-gifted junior high school students participated in an intensive course, five days a week, three hours a day. After the reciprocal teaching strategy instruction, students not only finished reading an English book but also developed their thinking. Judging from their worksheets requiring students to generate questions, Chern (2005) found that the questions shifted from those started with what, where and when to those with why and how. Stated another way, students paid more attention to logical interpretation of the text rather than to literal understanding with reciprocal teaching. 18.

(29) Similarly, Ash (2005) described a class in which a group of advanced students were instructed with reciprocal teaching to read a series of letters exchanged by a couple. It was definitely easy for those good readers to comprehend the texts. However, by reciprocal reading, they not merely achieve literal and inferential understanding but also identify their purposes for reading and think critically about the author’s point of view. In addition to studies conducted in groups of similar proficiency levels, a substantial body of research documents the effects of reciprocal teaching for average students of mixed levels in a classroom setting (Myers, 2005; Williams, 2010). Williams (2010) implemented reciprocal teaching in an intact class for more than three months in order to enhance English learners’ reading comprehension of expository texts. Based on the teacher’s observation, students’ improvement showed that reciprocal teaching not only prepared them to elicit the literal meaning of a text but also provided them with a better understanding of the text. Kindergarten students used to be considered too young to learn a sophisticated reading strategy like reciprocal teaching. However, Myers (2005) conducted a classroom action-research project and adapted reciprocal teaching for kindergarten students via read-aloud. Since the kindergarteners were unable to read text by themselves, the materials were books that the students were familiar with. During the three-month instruction, students were assessed by three instruments: interviews, anecdotal records and rubric assessment. The findings showed that the kindergarteners were capable of thinking metacognitively and asking questions which arose from critical thinking. This project offered evidence that even students with limited language proficiency, like kindergarteners, ESL and EFL learners, could benefit from reciprocal teaching. Different from the studies mentioned above, a body of research was done to 19.

(30) further investigate whether reciprocal teaching could help students with proficiency lower than “poor readers, ” i.e. intellectually disabled students (Alfassi et al., 2009; Takala, 2006). Alfassi et al. (2009) investigated the effects of reciprocal teaching on students with intellectual disability. Seventy-five students aged 15 to 21 participated in 24 sessions of either reciprocal teaching strategy instruction or traditional remedial methods of skill acquisition. Two kinds of reading comprehension tests and a reading literacy assessment were administered. Results from all measures provided support for the claim that reciprocal teaching was superior to the traditional remedial methods. Takala (2006) examined the effects of reciprocal teaching on reading comprehension of elementary school students of different proficiency levels: pupils with typical development and those with moderate or severe specific language impairment. The study contained six similar interventions, three in the mainstream course and three in specific language impairment course. The three interventions in each course include fourth-grade and sixth-grade science classes and fourth-grade history classes. The interventions were implemented in Finnish and lasted five weeks each, two lessons per week, and a total of ten to fifteen sessions. The instruction and discussion took place with the whole class, and no small groups were arranged. Based on pretests, posttests and delayed posttests, the best results were achieved by the mainstream students. However, some positive development could also be noticed in the special education groups. Similar to Alfassi et al. (2009), Takala (2006) challenged the common perception that intellectually disabled people could hardly improve their literacy via reciprocal teaching. Besides Takala’s (2006) research mentioned above, Chou (2008) and Wu (2012) investigated the effects of reciprocal teaching in EFL contexts. Chou (2008) recruited 70 ninth-grade junior high school students from two mixed-level classes and implemented a ten-week reading course. The experimental group received the 20.

(31) reciprocal teaching intervention while the control group received the traditional reading instruction. After analyzing the data collected by a reading comprehension test, think-aloud protocols and a response questionnaire, Chou (2008) concluded that reciprocal teaching was more beneficial in promoting subjects’ reading comprehension and the subjects reported positive attitudes towards this approach. Wu (2012) conducted another study to explore the effect of reciprocal teaching on four Taiwanese junior high school students who were considered to be struggling readers. After fourteen sessions, they read six story books via reciprocal teaching and responded to a perception questionnaire. Besides, the researcher’s teaching journal and the subjects’ weekly worksheets were utilized to collect the data. The results showed that reciprocal teaching could benefit low achievers’ learning and most of them were positive about the effect of reciprocal teaching. To sum up, the effect of reciprocal teaching on reading comprehension has been widely investigated with students in different settings. Positive results have been obtained in many studies. However, most researchers looked at learners of a specific proficiency level. Little research includes students of different proficiency levels in one study and implements the intervention with groups, especially in an EFL junior high school context. The current study aims to explore the acceptability of reciprocal teaching from the perspective of Taiwanese EFL junior high school. With hopes to offer all students the foundation they need in their independent reading, the present study implemented explicit strategy instruction via reciprocal teaching to help students acquire and use strategies while reading.. 21.

(32) Chapter Three Method. This chapter is composed of seven parts. The first part briefly illustrates the setting of the present study. The second section described the background of the participants. The procedure of the intervention is introduced in the third part. The reading materials used in the intervention are illustrated in the fourth section. The fifth part presents the research instruments. Finally, data-collecting procedures and data analysis are provided in the end of the chapter.. Setting The present study was conducted in School A, a junior high school in central Taiwan. School A is a large school with a student population of 2,100 in 63 classes. The social and economic status of parents is above average. Based on the current MOE regulation, junior high school students in Taiwan should be arranged into a class by means of normal class grouping. That meant the distribution of students’ general competence should be similar among classes. However, English was one of the three subjects that could be instructed by academic competence-based grouping.1 To be more specific, students in every two or three classes could be grouped into a new class based on their English proficiency. School A adopted this policy of placing students according to their proficiency level in English. The study was conducted in two of the classes in School A.. 1. According to current MOE regulation, only English, mathematics and natural science classes can be instructed by academic competence-based grouping to avoid the difficulties arising from bimodal distribution and to facilitate teaching and learning. 22.

(33) Participants The participants of this study were 64 ninth-graders from two classes in School A. All of them had been receiving compulsory English education for at least six years. Based on the English proficiency, i.e. their average English scores in the previous semester, students were divided into two new classes for English instruction. In other words, the top 50% of the two classes went to a higher achiever class while the rest to a lower achiever class. In the present study, the former was assigned as the higher achiever group, and the latter as the lower achiever group. The researcher was the teacher for both groups. According to Palincsar et al. (1989), students participating in reciprocal teaching should be taught in small and heterogeneous groups, so every one of them can practice sufficiently while receiving feedback about his or her own performance. Before starting reciprocal teaching intervention, the teacher divided the participants into groups of four based on the average of their English grades last semester. To make the groups formed heterogeneously, the top eight participants were the first members in each group, and the next eight participants were put in groups in a reverse order. For example, in terms of the second member, number nine was put in the same group with number eight while number ten with number seven. The same steps were repeated until there were four participants in every group. Throughout the teaching intervention, the demonstration and discussion were carried out in groups. The group members stayed the same through the experiment.. Materials There were nine articles for the participants to read during the treatment. The materials were partly, seven out of nine articles, selected from the textbooks and partly, two articles, from an English magazine called ABC Interactive English. 23.

(34) The rationale for selecting material from textbooks was to make sure the articles were suitable for participants’ English proficiency.. However, to ensure the variety,. the articles were chosen from textbooks published by three publishers, Nani (南一), Kang Hsuan (康軒) and Han Lin (翰林) publishing companies. The articles all came from Book Five and Book Six, which suited the participants’ current English language level. Some vocabulary and sentence patterns in the articles may be new to the participants. However, the gap between what the participants had learned and what they had not gave the participants a chance to negotiate with the text for meaning. After the researcher surveyed Book Five and Book Six of junior high school English textbooks, seven articles were collected: 2 from Nani (南一), 2 from Kang Hsuan (康軒), and 3 from Han Lin (翰林). According to Risko, Walker-Dalhouse, Bridges, and Wilson (2011), providing appropriate reading material beyond textbooks holds possibilities of deepening students’ interest, engagement, and comprehension. The reason for choosing articles from the magazine was to provide materials of different genres and styles. Textbook articles tend to be confined to the sentence structure and vocabulary based on a prescribed level of proficiency. Therefore, articles in textbooks tend to present stories or articles with little content diversity and limited genre. In terms of the diversity of genres and topics, magazines serve a complementary function to textbooks. Among various kinds of magazines, a language learning magazine, ABC Interactive English, were selected for the present study for two reasons. First, every issue of the magazine contains dialogues, features, short stories, comic strips, etc. Each falls into different genres and each is shown with its optimal writing style for students of the elementary level in the General English Proficiency Test (GEPT), published by the Language Training and Testing Center. Second, though different from a textbook, the magazine offers articles relevant to the students’ lives. Sometimes a magazine article is just an 24.

(35) extension from a textbook topic because the magazine is designed for English learners of the basic proficiency level, where most junior high school students are. Therefore, familiarity to the topics and interests in the issues may motivate the participants to read. In short, the combination of articles from textbooks and a language learning magazine just provides a good balance of varieties of topics, genres, and styles. Table 1 below summarizes the readability and length of the nine articles.. Table 1 The Readability and Length of the Reading Materials Articles. Readability. No. of. No. of. No. of. Words. New Words. Sentence patterns. 1. Solar Power. 6.1. 164. 14. 2. 2. Modern Cellphones. 6.4. 197. 15. 2. 3. The Astronaut’s Life in. 6.8. 167. 16. 2. 4. Honest Abe. 6.7. 152. 11. 1. 5. Guinness World Records. 7.1. 188. 15. 1. 6. Sleeping boxes. 6.5. 151. 10. 1. 7. Ryan’s Well. 6.4. 159. 10. 2. 8. A Tomato a Day Keeps. 7.2. 155. 14. 1. 7.9. 163. 15. 1. Outer Space. the Doctor Away 9. Walt Disney’s Magic. 25.

(36) Reciprocal Teaching Intervention Reciprocal teaching intervention was incorporated into the English reading course for the two groups. The intervention lasted for five weeks, with two 45-minute sessions per week. The teaching intervention consisted of three stages: demonstration of four strategies, teacher-led discussion, and student-led discussion. During the intervention, the teacher first made the reasoning process visible and comprehensible through the demonstration of four strategies in Stage I. Next, the teacher led the discussion in Stage II and transferring the reading responsibility to students gradually by guiding them to apply the four strategies. Finally, students applied four strategies independently to tackle the text and solve reading problems during discussion in Stage III.. Stage I: Demonstration of four strategies. Stage I contained two sessions. Two strategies were introduced in each session respectively. Since explicit instruction was strongly recommended for reading strategy instruction (Snow, 2002), the teacher defined and explained the target strategies directly in the beginning of each session in order to help students monitor and regulate their comprehension process. After the explanation, the students understood how reciprocal teaching worked. What should be noted is that the demonstration aimed to familiarize subjects with the four metacognitive skills: predicting, clarifying, summarizing, and questioning. Among them, no particular order was required. However, it is recommended that the strategies should be taught one at a time with a review of the strategies taught earlier. Besides, repeated practice and strategy application should be sufficient to ensure the effects (Hashey & Connors, 2003; Palincsar & Klenk, 1992). In Session 1, predicting and clarifying strategies were covered. In the first five 26.

(37) minutes, a handout of guides for roles in reciprocal teaching (See Appendix A) and a worksheet with the assigned article (Appendix B), both adapted from Oczkus (2003), were distributed to the students. The handout consisted of 11 procedural prompts often employed in four strategies while the worksheet comprised an article for demonstration. The procedural prompts on the handout functioned as clues to specify the strategies and to guide beginners to think in the logic of the strategies. Eventually, participants would become familiar with them. The worksheet was for students to make a record of their reasoning process. This recording helped students become familiar with the content by writing down what they had in mind. After the distribution, the teacher explained what predicting was and demonstrated how to predict by reading the title and by looking at the picture along with the article. Next, students were asked to write down their prediction(s) on the worksheet. Then, the teacher invited the students to read the first paragraph aloud to confirm the previous prediction. After the participants got a clear idea of the first paragraph, the teacher made another prediction about what would happen in the second paragraph. Meanwhile, students were also encouraged to predict with the teacher. Still, they had to write down the prediction on the worksheet. Then, students read the second paragraph together to check the prediction. If the prediction did not match the content, they could revise it. If it was right, they could move on to the next paragraph. The same steps were repeated until the reading was completed. During the demonstration, new vocabulary was not explained. New words and questions were saved for the next part of the session, which was the demonstration of the strategy of clarifying. In the second part of the session, the instructor introduced the strategy of clarifying. The instructor first bold-faced and underlined the unknown words and confusing ideas in the article (See Appendix B). Then, the teacher modeled to solve 27.

(38) the problems by clarifying, including using the context to get the possible meaning or employing the prior knowledge to make meaningful and reasonable guesses. Students also needed to finish the clarifying part on the worksheet. At the end of Session 1, students were asked to think about what the main idea of the article was. This technique connected their present knowledge of the article to the next strategy, summarizing. The strategy of summarizing was introduced in the first part of Session 2. The instructor started with introducing the difference between the main ideas and supporting details. Students needed to know that a summary could only include the essence of an article, not the details of it. The idea of topic sentences might help the student locate the key point illustrated in a paragraph in some descriptive articles. Besides, topic sentences might serve to facilitate the use of predicting strategy. Experienced readers can begin with the topic sentence, make a possible guess about the development, and predict the future events. The double benefits provided by topic sentences proved again that the four strategies are reciprocal and can work together. After the brief introduction of the difference between key points and supporting sentences, the instructor modeled how to make a summary in one or two sentences for the two paragraphs. Finally, the teacher introduced the strategy of questioning in the second part of Session 2. At first, the instructor invited students to generate questions about the assigned article. With the gist in mind, students generated questions of all kinds quickly. The instructor wrote down the questions which were raised by the volunteers on the blackboard and put them into two categories: skinny questions and fat questions (Hashey & Connors, 2003). The former could be answered simply with yes, no, or any short responses based on the text while the latter call for deeper thoughts and complete answers. The instructor asked students what kinds of questions would 28.

(39) be proposed by the teacher more frequently and showed them how to get the correct answers to the questions. Then, the students would understand that this technique will bring them closer to the kinds of questions that teachers have in mind. A sample is provided in Appendix C to demonstrate the use of four strategies to read the first paragraph of an article named Solar Power in Book Five from Kang Hsuan (康軒). One article, i. e. Solar Power, was used in Stage I to demonstrate four strategies. Using the same article benefited the demonstration in two ways. One advantage was Session 2 could start without delays caused by unfamiliarity with a newly-introduced article. In other words, students could make use of the third and fourth strategies immediately in Session 2 since they were already familiar with the content of the article. It could ensure the teaching procedure to be carried out efficiently within a period of limited time. The other advantage of using the same article was to show students that the four strategies of reciprocal teaching can be integrated into the reading of one article. Stated another way, participants could apply all four strategies to one article and gained a clearer comprehension of it.. Stage II: Teacher-led discussion. According to Wood, Bruner, and Ross (1976), a novice is guided by an expert to solve a problem or to achieve a goal beyond his actual developmental level in the process of scaffolding. In reciprocal teaching, scaffolding is used in the context of group discussion as a method to support and guide students in the zone of proximal development (Paris & Winograd, 1990). After learning the four reading strategies, students started to take more responsibility to complete the following tasks. With teacher’s guidance, they learned to practice the four strategies in the second stage, Teacher-led discussion. 29.

(40) Teacher-led discussion contained four sessions, Session 3 to Session 6. Only one target strategy was covered in each session. The teacher started a session with a brief review of the target strategy. After the review, the teacher led the students to read the first paragraph of the article using the target strategy and invited them to practice it with the remaining paragraphs. Meanwhile, the students took time observing the reasoning process of an experienced reader or their capable peers. They were also encouraged to apply the strategies. Lesson plans of these four sessions are provided in Appendix D. In order to reduce students’ anxiety with speaking in English and to make them feel free to express their ideas, students were allowed to speak in Chinese in Stage Two and Stage Three. However, speaking in English during the discussion was also encouraged. If some students talked in English, they might make progress in not only the reading comprehension but also the other three skills. Besides, to keep track of the learning process, students were asked to complete worksheets for each session. In terms of predicting and clarifying, they could write in Chinese. As for summarizing and questioning, they should finish these sections in English. In Session 3, the strategy of predicting was practiced. In the beginning, the instructor reviewed the definition of predicting and demonstrated when and how to use it with the title of Article 2. Then, the teacher asked for volunteers to make predictions about the first paragraph. Next, both the teacher and other students gave comments on the predictions. If they didn’t think the prediction is correct, they could revise it. After everyone expressed their opinions, the teacher guided students to read the first paragraph aloud to check the prediction. The same steps were repeated until they finished the last paragraph. In the last five minutes, the teacher wrapped up the session by either reviewing the target strategy or inviting students to share their feedbacks about the strategy. 30.

(41) In Session 4, the strategy of clarifying was included. After reviewing the definition of clarifying, the teacher showed the students when and how to use it by clarifying the new word in the title. Next, the teacher asked students to read the first paragraph with clarifying strategy by themselves. Then, the teacher asked for volunteers to share the problems they encountered while reading and how to solve them. The teacher praised the volunteers when they got the correct answers. On the other hand, she provided guidance when the clarification went wrong. The same steps were repeated for the next two paragraphs. To wrap up this session, the teacher invited one or two students to make a summary of what they had learned that day. This summarizing technique would be related to the target strategy in the next session. In Session 5, the teacher first reminded the students with what they did in the end of the previous session, i.e. making a summary. After giving a brief review of summarizing, she talked about the ideas of a topic sentence and supporting details. This would help the students to tell the major idea from the minor ones. Then, she guided students to read the first paragraph of Article 4 and demonstrated how to summarize it. Next, all of the students read the second paragraph aloud, and some of them were chosen to share their summaries with the class. The teacher and students discussed which summary best captured the gist of the article. Some feedbacks were given both by the teacher and by their peers. The same steps were repeated for the last paragraph. Because clarifying strategy were more complex than the previous ones, the teacher ended this session with a review of what were covered that day. In Session 6, after reviewing the definition of good and bad questions, the teacher guided the students to read the first paragraph together and demonstrated to generate two fat (good) questions and two thin (bad) ones. Next, the teacher invited students to discuss the difference between the two kinds of questions. Then, she classified questions into four categories and offered examples for students to imitate 31.

(42) later. After that, the students read the second paragraph and shared their own questions with the class. The peers could provide comments to modify the questions. The same steps were repeated for the remaining paragraphs until they finished reading. During the four sessions, the time for teacher guidance was reduced as the students took more and more responsibility for reciprocal teaching.. Stage III: Student-led discussion. Student-led discussion contained four sessions, Session 7 to Session 10, and was conducted in the form of group discussions. In this stage, all the members in a group had to take turns playing different roles, i.e. predictor, clarifier, summarizer, and questioner. Besides, the predictor has to be the group leader, too. The group leaders should be responsible for cuing other group members to do their jobs. In the beginning of each session, the teacher distributed a worksheet to each group for them to make a record. Next, the teacher briefly reminded students what each role should do in the discussion. Table 2 lists the tasks of each of the four roles in reciprocal teaching. Then, the student-led group discussion started. The group leader cued each role. When one person was talking, the other group members either listened carefully or provided their comments. Instead of dominating the discussion, the teacher walked around as a facilitator. The teacher offered help if needed.. Table 2 Description of the Four Roles in Reciprocal Teaching Roles Predictor. Tasks Read the paragraph first and make predictions about the target paragraph. Other group members will comment on the predictions. 32.

(43) or provide their own. Clarifier. Read the target paragraph and report the problems he/ she encounter. Then, share his/ her solutions to cope with the problems. Other group members will comment on the clarifications or provide their own solution. Moreover, they may share other problems.. Summarizer. Summarize the target paragraph. Other group members will comment on the summary or provide their own. Questioner. Generate questions about the target paragraph. Other group members will share the answers to the questions, comment on the questions or generate their own questions.. The design of the reciprocal teaching intervention mentioned above is summarized in Table 3 below.. Table 3 The Design of the Reciprocal Teaching Intervention Session. Strategy introduced Stage I.. 1. Instruments/ materials Demonstration of four strategies. Prediction and clarification. A handout, Worksheet 1, Article 1. 33.

(44) 2. Summarizing and questioning. A handout, Worksheet 1, Article 1. Stage II.. Teacher-led discussion. 3. Prediction. Article 2. 4. Clarification. Article 3. 5. Summarizing. Article 4. 6. Questioning. Article 5 Stage III.. 7. Student-led discussion Worksheet 2, Article 6. Integrated strategy use. 8. Worksheet 3, Article 7. 9. Worksheet 4, Article 8. 10. Worksheet 5, Article 9. Research Instruments Two research instruments were adopted in the present study: an English reading comprehension test and a perception questionnaire.. An English reading comprehension test. Similar to the study conducted by Miller, Miller, and Rosen (1988), a ten-item English reading comprehension test is used to measure the students’ performance. The 34.