小老師制度在國小英語補救教學上之個案研究 - 政大學術集成

全文

(1)國立政治大學英國語文學系碩士在職專班碩士論文. 指導教授：招靜琪博士 Advisor：Dr. Chin-Chi Chao. 政治大小老師制度在國小英語補救教學上之個案研究立. ‧ 國. 學. A Remedial Peer Tutoring Case Study in an Elementary School. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 研究生：張順宜撰 Name：Shun-yi Chang 中華民國一百年七月 July, 2011.

(2) A Remedial Peer Tutoring Case Study in an Elementary School. A Master Thesis Presented to Department of English,. 立. 政治大. ‧. ‧ 國. 學. National Chengchi University. n. er. io. sit. y. Nat. al. i n U. Ch. v. hi In e Partial n g cFulfillment. Of the Requirements for the Degree of Master of Arts. by Shun-yi Chang July, 2011.

(3) To Chin-Chi Chao 獻給我的恩師招靜琪教授. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(4) Acknowledgements. These brief acknowledgments can only begin to express my genuine appreciation for all of those who helped me in this educational pursuit. I feel extremely fortunate to have been a student under Dr. Chin-chi Chao who opened my eyes and ears to the wonders of language development. She has been an inspiring source of knowledge and provider of guidance in this study. I would also like to thank Dr. Chen-kuan Chen and Dr. Chieh-yue Yeh for their time, insights and suggestions.. 政治大 pursue and complete this educational endeavor. I would like to thank my parents for 立 Above all, I wish to express deep gratitude to my family who enabled me to. emphasizing the importance of education, giving me the foundation to set high goals. ‧ 國. 學. and supporting me unconditionally. I am deeply grateful to my school colleagues for. ‧. their loving support and complete confidence in me. Finally, I would like to thank my. sit. y. Nat. church sisters for caring and constant prayers and more importantly, for making this. io. n. al. er. study personally meaningful to me.. Ch. engchi. iv. i n U. v.

(5) Table of Contents. Acknowledgments…………………………………………………………………….iv Table of Contents………………………………………………………………………v Chinese Abstract………………………………………………………………………ix English Abstract………………………………………………………………………xi Chapter 1. Introduction…………………………………………………………………..1. 政治大 Purpose of the Study…………………………………………………….4 立. Background of the Study………………………………………………..1. Significance of the Study……………………………………………….4. ‧ 國. 學. 2. Literature Review…………………………………………………………….6. ‧. Sociocultural Theory in SLA…………………………………………...6. sit. y. Nat. Peer tutoring…………………………………………………………….9. io. er. Peer Tutoring in SLA…………………………………………………..11 Research Questions…………………………………………………….16. al. n. v i n Ch Methodology………………………………………………………………...17 engchi U. 3.. Context of the Study…………………………………………………...17 Participants…………………………………………………………….20 Materials……………………………………………………………….23. Data Collection………………………………………………………...23 Data Analysis……………………………………………………..……31 Data Analysis Procedures……………………………………………...34 4. Results………………………………………………………………………35 Preliminary Overview of the Data: What was accomplished?.………..36 Section I—Basic Spelling………………………………………….36 v.

(6) Section II—Synthetic Spelling…………………………………….38 The Tutee’s Bids for Help vs The Tutor’s Developmentally Appropriate Assistance……………………………………………………………...42 Section I—Basic Spelling………………………………………….45 Section II—Synthetic Spelling…………………………………….52 Overview of the Tutee’s Gains and the Tutor’s Assistance…................65 Development of CVC Structure…………………………………...65 Development of Long Vowel Structure………………....................71. 政治大 Vowel Diagraphs Structure………………………………………...73 立 Development of Consonant Blends, Diagraphs and. The Tutee’s Internalization of Spelling Rules…………………………78. ‧ 國. 學. Perceptions of the Peer Tutoring………………………………………80. ‧. The Tutor’s Perceptions……………………………………………80. sit. y. Nat. The Tutee’s Perceptions…………………………………………...82. io. er. 5. Discussions………………………………………………………………….84 Answers to the Research Questions…………………………………...84. al. n. v i n Ch Conclusion and Implications………………………………..........................94 engchi U. 6.. Summary for the Findings……………………………………………..94 Findings of the Present Study vs. Findings of Previous Studies………95 Theoretical and Pedagogical Implications……………………………..97 Limitations of the Study……………………………………………….99. References……………………………………………………………………..101 Appendixes A.. Basic and Synthetic Spelling Word Lists……………………..110. B.. Screen Test……………………………………………………114. C.. Tests for the Tutee…………………………………………….117 vi.

(7) D.. Interview Questions…………………………………………..124. E.. Transcription conventions…………………………………….126. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. vii. i n U. v.

(8) List of Tables and Figures Table 1 Assistance From Implicit to Explicit Level (Five levels)…...……………….28 Table 2 Examining Sociocultural Constructs Related to Assistance and Internalization…………………………………….31 Table 3 Signaling “continuing” or “not continuing”…………………………………32 Table 4 Assistance From Implicit to Explicit Level (Nine levels)………………...…44 Figure 1 Procedures and Data Collection of the Study……………………………....25 Figure 2 Timeline of the Study………………………………………........................26. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. viii. i n U. v.

(9) 國立政治大學英國語文學系碩士在職專班碩士論文提要. 論文名稱：小老師制度在國小英語補救教學上之個案研究指導教授：招靜琪博士研究生：張順宜論文提要內容：以往傳統研究探討同儕相互幫助的機制多在於量化性的陳述其有效結果，然. 政治大. 而，在小老師的教學與被幫助者的學習動機上卻較少提出在兩位小朋友的言談互. 立. 動中，何以促使此同儕相互幫助的機制有正面的效果。本個案研究旨在探討英語. ‧ 國. 學. 為外語的學習環境中，在國小階段的英語補救教學裡，同儕相互幫助的動機與影響。本文採用維高斯基 (Vygotsky) 社會文化學習理論來探討在台灣的國小補救. ‧. 教學環境中，教室兩人一組的同儕學習中，小老師如何使被幫助者語言知識建構. sit. y. Nat. 起。. n. al. er. io. 本研究對象針對 30 位五年級學生，分為兩人一組，共 15 組，實施為期 20. i n U. v. 週小老師制度教學，並且針對其中一組的互動方式，作深入的研究。研究方法. Ch. engchi. 採用質化的言談分析及晤談訪問。資料分析重於探討被幫助者如何藉由小老師的協助，完成該年級應學會的字母、單字認讀與基礎閱讀的過程。目的是希望呈現小老師在與被幫助者言談互動之中，小老師如何一步一步地提供合適的協助以建立被幫助者的語言知識概念。本研究主要的發現有： 1. 這一名受過小老師制度訓練的小學生能夠很快掌握學習者的學習困難處並提供協助。 2. 這一名小老師用程度分級的協助方式來探測學習者的能力，且能夠發揮創意，給予適當的協助。學習者也表示小老師用多種問答的方式來幫助自己釐清學 ix.

(10) 習盲點，這樣有助於了解自己的學習潛力。 3. 這一名小老師與學習者之間的朋友關係在學習過程中扮演重要的媒介角色。本研究根據以上議題討論結果，在文末進一步提供建議，作為教育學者們參考。. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. x. i n U. v.

(11) Abstract Recognizing the important role of peers in the learning process, this study investigated the effect of peer tutoring in Taiwan elementary remedial instruction. Findings of prior studies had demonstrated the quantitative effectiveness of peer tutoring in enhancing educational outcomes and generating various benefits in psychological domain, yet what had contributed to learning in the peer interaction was little discussed. Drawing on Vygotskian sociocultural perspectives, this study aimed to explore the “inside stories” of how slow learners’ knowledge is constructed through peer tutoring in an EFL remedial class. What and how those aspects yet to be found. 政治大. contributing to slow learner’s performance and in addition, how the tutor-tutee. 立. interaction affecting the perceptions of the tutor, the tutee and the teacher about the. ‧ 國. 學. peer-tutoring model were examined.. Conducted through a class-wide peer-tutoring program, this study arranged 30. ‧. participants into 15 dyads in a remedial class. The combination was of high-low level. y. Nat. io. sit. of English proficiency. In order to obtain detailed, in-depth observation and accounts. n. al. er. of the one-semester long peer interaction, focus was on one dyad in this study.. i n U. v. The major findings are summarized. First, with tutor training ahead of the peer. Ch. engchi. tutoring, this particular tutor was able to get the gist of helping the tutee in solving learning difficulties. Second, the levels of assistance provided by the tutor clearly diagnosed the tutee’s learning difficulties so as to help the tutor creatively devised appropriate assistance. Third, “friendship” between the tutor and the tutee played an important role in the process. The tutor was a significant other in the tutee’s learning. In light of these findings, pedagogical implications and suggestions are presented at the end of the paper.. xi.

(12) CHAPTER ONE INTRODUCTION Background of the Study Peer tutoring is considered a discovery of professional development in education that students can participate in the delivery of educational services. (Bergen, Mi, 2002) Students, traditionally viewed as recipients of these services, are now seen as a potential source of help and assistance. Peer tutoring, a context in which students work with one another and support each other’s efforts, has been noted as holding an. 政治大 belief that students perform differently when studying independently compared to 立. important role in the learning process. The theoretical aspect of this concept lies in the. studying with help of others (Vygotsky, 1978).. ‧ 國. 學. Peer tutoring, in recent years, has also been regarded as a solution to the problem. ‧. of polarization of English proficiency observed in Taiwanese elementary schools.. sit. y. Nat. With the implementation of Grades 1 to 9 Curriculum, all public elementary schools. io. er. in Taiwan have started English courses in the fifth grade since 2001 and further extended the courses to grade three in 2005. However, more and more students. al. n. v i n receiving English instructionCmuch earlier than theU h e n g c h i formal English courses in schools. A serious problem in English education thus has arisen---the widening gap among students’ prior learning experiences and knowledge of English (Tsou, 2002 ). According to Lin (2003), 76% of the elementary school English teachers in Taiwan perceived a wide knowledge gap among students’ English competency (Lin, 2003), which has been making English teaching difficult and ineffective (Chan & Chiang, 2002; Liu, 2001; Lu, 1999). As a consequence, students receiving English instruction prior to the formal courses are less likely to have problems catching what the teacher says; on the contrary, they might find the lessons too easy and become bored. By comparison, those who have not had any English learning experiences prior to school 1.

(13) English lessons, as argued by Chang (2002), tend to find themselves in inferior positions and are very likely labeled as “slower learners,” “disadvantaged students,” or “underachievers” on their very first day of the English class. To ensure that none of these students are left behind or even abandoned on the road to English learning, it is necessary for teachers to identify these students’ learning difficulties and needs in order to enhance the effectiveness of instruction (Chang, Chou, Chen, Yeh, Lin & Hsu, 2003; Chou, 2002) and remedial instruction is recognized as an important and necessary strategy to bring every student up to the. 政治大 the Taipei City English teachers were willing to provide or have already provided 立. grade level (Chang, 1997; Chang, 2001; Li, 1996). According to Chen (2004), most of. remedial instructions for the underachievers. Among many teaching strategies, the. ‧ 國. 學. most adopted remedial instructions were peer-tutoring and cooperative learning.. ‧. Because of time limit and heavy workload, teachers tend to have peers play the role of. sit. y. Nat. “a teacher” to provide slow learners with extra and remedial instruction or for. io. er. reinforcement. In peer-tutoring, the flexible, friendly environment has been found to reduce anxiety and facilitate learning (Cohen, 1986; Nevi, 1983). Tutees would have. al. n. v i n Cpersonally more individual concerns and assistance on their learning U h e n gappropriate i h c difficulties according to their learning needs. On the other hand, tutors usually. become better comprehend the learning contents during the peer-tutoring sessions which subsequently benefited their own learning.. In a Vygotskian sense, children. are working within each others’ zone of proximal development, providing feedback to each other and developing argument skills (Slavin, 1996). This perspective is useful in explaining why children learn from being taught by peers. Although peer tutoring has been employed in meeting students’ academic and psychological needs in different education levels and courses as well as in Taiwan elementary English remedial classes (Fantuzzo, King, & Heller, 1992; Fantuzzo, 2.

(14) Polite, & Grayson, 1990; Jia, 2004), few have attempted to explore the “inside stories” of the peer tutoring. Among those studies mentioned above, it is obvious that the emphasis was highly placed on quantitative evaluation of learning outcomes, cognitive and affective changes in tutees and tutors, while little attention has been paid on investigating the detailed interactional process between individuals or inside individuals which leads to cognitive processing (Lantolf, 2000). The inside story of how linguistic knowledge has been scaffold and acquired in higher proficiency learners’ assistance to lower proficiency learners in peer tutoring process in. 政治大 of how more experienced peers assist slow learners in regard to language knowledge 立. elementary English language learning has been little explored. They left the question. construction/internalization in the peer tutoring process and what exactly the more. ‧ 國. 學. experienced peers provided the slow learners in the process which led to language. ‧. development.. sit. y. Nat. In the researcher’s teaching experience, a twenty-week peer-tutoring model was. io. er. implemented in the English remedial class in the previous year. Thirty-six tutors and thirty-six tutees participated in the remedial class. The study began with a pre-tutoring. al. n. v i n C h tutor training provided preparatory stage for a two-week by the researcher using engchi U. supplementary materials based on the textbook. Then the peer tutoring sessions were conducted for 18 weeks. Each dyad is of high-low proficiency combination. Every three weeks, the researcher and the tutors went over reflection reviews to ensure that. the tutoring process proceeded without serious problems. At the end of the semester, according to their formative tests scores, significant improvement was made by the tutees in basic letter/word recognition, spelling and reading abilities in the English remedial class. From cognitive/interactionist perspectives, students’ language ability improved might have been due to the aid of the peer tutoring sessions, but from sociocultural perspectives, the reasons why and how tutor-tutee interactions leading to 3.

(15) language learning were unknown. Purpose of the Study In an attempt to address the unknown gap as well as to provide insight into this under-explored area, this study investigated how slow learners’ knowledge is constructed through peer tutoring through their dialogues. What and how those aspects yet to be found contributing to slow learner’s performance and in addition, how the tutor-tutee interaction affecting the perceptions of the tutor and the tutee about the peer-tutoring model were examined.. 政治大 To explore aspects of tutor-tutee interactions contributing to tutee’s 立. Drawing on Vygotskian sociocultural perspectives, this study aimed to 1.. performance. ‧ 國. 學. 2. To examine how those aspects of tutor-tutee interactions scaffold tutee’s. ‧. language development. io. er. research were presented in next few chapters.. sit. y. Nat. Literature review, methodology, results, discussions and conclusions of this. Significance of the Study. al. n. v i n The study is significantC to the researcher as a U h e n g c h i practitioner and the research community at large.. For the researcher as a practitioner, the study leads to a deeper understanding of the uniqueness of the tutoring context, learning atmosphere and styles of learners’ interaction and to be able to become more sensitive to learners’ thoughts and perceptions (Ernest, 1994). Bogdan and Biklen (2003) note that the goal of this kind of research helps practitioners to ‘understand the learners’ world and to determine how and with what they judge it. For the research community, the study offers a glimpse into how teachers may need to adjust their teaching perspectives towards students’ learning. It raises 4.

(16) questions for further research and offers more depth to our understanding of teaching.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 5. i n U. v.

(17) CHAPTER TWO LITERATURE REVIEW This literature review chapter has two parts. The first part discusses the meanings and interpretations of sociocultural theory in second language acquisition. The second part provides a discussion of meanings and applications of peer tutoring in second language acquisition. Sociocultural Theory in SLA According to Fawcett and Garton (2005), sociocultural and cognitive approaches. 政治大 second language acquisition is essentially a mental process of acquiring systems of 立. to second language acquisition (SLA) differ considerably. For cognitive approaches,. knowledge which make up the target language. Researchers working in this approach. ‧ 國. 學. are primarily interested in how the brain processes, stores and retrieves information.. ‧. The main focus is on the cognitive abilities of the learner and the way these abilities. sit. y. Nat. interact with the task of processing and, hopefully, acquiring a second language.. io. er. Progress in acquiring the second language (L2) system is seen as manifested by increased fluency and accuracy and a wider range of syntactic structures, as these. al. n. v i n C h that the learnerUcan draw on automatically. reflect an expanding of knowledge engchi. For sociocultural approaches, on the other hand, language development is. essentially a social process. These approaches view mind as distributed and learning as something inter-mental, embedded in social interaction. This means that individuals and environments mutually constitute one another and the persons are not considered to be separate from the environment and interaction through which language development occurs. In this view, knowledge is not owned solely by the learner, but it is also a property of social settings and the interface between person and social context. Language development can be studied by examining distributed cognition—how a learner makes use of the L2 in interaction with other people and 6.

(18) artifacts. Development is visible through analyses of episodes of interaction, as the learner demonstrates increased independence (Hall & Verplaeste 2000; Lantolf 2000; Ohta 2001). Mediation. As Vygotsky once said, “A colt is already a horse; a human baby is only a candidate to become a human being” (Vygotsky, 1962, p.68). Vygotsky believed human beings share lower mental functions with other animals. What differentiates human beings so that we go beyond other animals is the mental or psychological tools we acquire to help us think. The way human beings acquire. 政治大 responses from an individual. However, when we acquire a psychological tool, such 立. psychological tools is from culture. As with behaviorism, at first, outside stimuli elicit. as language, the tool mediates between the outside stimuli and the responses. Human. ‧ 國. 學. beings’ psychological tools create intentionality, comparisons, and higher-order. ‧. planning. Culture is handed down to us through our society, which is handed down to. sit. y. Nat. us through adults in the society (such as through our parents). What was in the culture. io. er. is incorporated into our own cognitive processed as the psychological tools that we use. Vygotsky said, “what was inter-mental becomes intra-mental.” (Vygotsky, 1978,. al. n. v i n C h human beingsUcan not function on an adult level p.68) From Vygotsky’s perspective, engchi without the culture of which we are a part bringing us along and providing what is. necessary. This conceptualization acknowledges a deeper level of social interaction than the simple social influence to guide our thinking, planning and actions. The tools that mediate human beings’ psychological processing in interaction are symbolic: language, symbolic play, art and writing. In a classroom, the mediation can take form of the textbook, visual material, discourse pattern, opportunities for second language interaction, types of direct instruction, or various kinds of teacher assistance. All forms of mediation are embedded in some context that makes them sociocultural processes. 7.

(19) Language as a mediating tool in learning. The sociocultural perspective, deriving in part from the concepts of Vygotsky (Vygotsky, 1962, 1978), illuminates the role of social interaction in creating an environment to learn language, learn about language, and learn “through” language. This perspective examines interaction within a broad social and cultural context. In Vygotsky’s view, human learning and development are bound up in activity, that is, purposeful action mediated by various tools (Vygotsky, 1978; Wertsch, 1979). The most important of these tools is language, the semiotic system that is the basis of human intellect (Halliday, 1993; Vygotsky,. 政治大 he stated: “Every function in the child’s cultural development appears twice: first, on 立. 1978). All higher-order functions develop from language-based, social interaction, as. the social level, and later, on the individual level; the first, between people. ‧ 國. 學. (interpsychological), and then inside the child (intrapsychological)” (Vygotsky, 1981,. ‧. p. 183).. sit. y. Nat. Sociocutural theory underscores the importance of conceptualizing language. io. er. learning as a developmental process mediated by semiotic resources appropriated from the classroom (Wertch 1991, 1998). These semiotic resources include print. al. n. v i n C h gestures, andUmost notably, classroom discourse materials, the physical environment, engchi which is manifested in peer interactions. It has been recognized that classroom discourse plays an important social role as a semiotic mediator of knowledge construction with respect to learning content in peer interaction. As Bakhtin (1986) emphasized that language is not encountered or learned as an abstract system of decontexualised rules and definitions. Rather, language occurs as dialogue. Vygotskian perspectives underpin the peer tutoring approach in learning—that is, it is assumed that learners gain mastery and develop cognitive skills through interaction with others and their environment (Hogan & Tudge, 1999; Kozulin, 1998). Sociocultural theory underscores the importance of conceptualizing language learning 8.

(20) as a developmental process mediated by semiotic resources appropriated in the classroom (Wertsch 1991, 1998). This perspective contrasts sharply with cognitive approaches based solely on the acquisition metaphor of development, which rigidly ascribes language learning to various internal mental processes. Within the cognitive-acquisition perspective, the individual is seen as the sole channel through which knowledge is gained. Within a sociocultural perspective, according to Lantolf (2000), however, learning, including the learning of the second languages, is seen in a more holistic perspective and is a semiotic process attributable to participation in. 政治大 learning is considered an important socially-mediated activity that fosters language 立. socially-mediated activities. And peers interaction frequently employed in language. acquisition.. ‧ 國. 學. The nature of peer assistance is closely related to Vygotsky’s (1978) notion of. ‧. zone of proximal development (ZPD). Vygotsky (1978) defined ZPD as “the distance. sit. y. Nat. between the actual development level as determined by independent problem solving. io. er. and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers” (p.86). In other words,. al. n. v i n Ch with the help of a more knowledgeable person, a novice is able to master concepts. engchi U. Although Vygotsky’s idea of the ZPD was originally restricted to the novice-expert relationship such as adult-child and teacher-student, recent studies that applied his notion to peer interaction have shown that peers can simultaneously be experts and novices ( Anton & DiCamilla, 1998; Donato, 1994; Kowal & Swain, 1997; Ohta, 2001; Swain & Lapkin, 1998). Peer tutoring Collaborative learning is a commonly used teaching strategy in many classrooms and learners participating in collaborative learning mostly show significant learning improvement (Fawcett & Garton, 2005). Among collaborative learning, peer tutoring 9.

(21) is one effective teaching strategy employed in different ways in different situations (Beasley, 1997; Boudouris, 2005). Peer tutoring is a system whereby learners help each other and learn by teaching. The most common approach is when an experienced student assists one or more learners (tutees) out of class time in content or strategy training (Beasley, 1997; Rafoth, 1998). Peer Tutoring Defined. The main difference between teaching and tutoring is that teachers must focus an entire curriculum to an entire class and tutors focus on specific areas of learning, the problem areas experienced by their tutees. Tutoring. 政治大 ways that classroom teaching cannot (Pugh, 2005). 立. complements and supplements classroom teaching, reaching the struggling students in. Peer Teaching Approaches. In general, according to Puchner (2003), there are. ‧ 國. 學. four kinds of peer teaching approaches: A. cross-age and same-age peer tutoring, B.. ‧. cooperative learning, C. reciprocal peer tutoring, D. reciprocal teaching.. sit. y. Nat. First of all, cross-age and same-age peer tutoring is tutoring of learners by other. io. er. learners with the same or more advanced ability, which is considered beneficial for both the tutor and the tutee. Compared to conventional classroom teaching in. al. n. v i n C h peer tutoring brings elementary and secondary schools, more significant effects. engchi U. Learners’ achievement, attitudes, self-concept were positively escalated (Cohen, Kulik & Kulik, 1982). The second kind of peer teaching approach is cooperative learning. By 1990 there was a strong research base on cooperative learning. Studies comparing cooperative learning to conventional instruction demonstrated that cooperative learning was effective, and benefits to cooperative learning have been found at grade levels two to twelve, in many major subjects, with higher and lower order tasks, and with urban and rural populations in the U.S. Areas of benefit included self-esteem, intergroup relations, achievement, attitudes towards school, and acceptance of 10.

(22) learners with difficulties (Cohen, 1994; Slavin, 1991, 1996). The third kind is reciprocal peer tutoring, which involves students being paired and in a structured manner taking turns to act as tutor and tutee. These programs have found to be effective in terms of academic achievement in subject areas of math, reading, spelling, and science (Greenwood, Arreaga-Mayer, & Utley, 2001). They have been researched mostly at the elementary level, but appear to be effective at both elementary and secondary levels. There are studies of the effects of these programs on a variety of student populations, including high-, average-, and low-achievers,. 政治大 positive results (Du Paul, Ervin, & Hook, 1998). Studies of these programs generally 立. students with disabilities, low-income students, and minority students, generally with. do not separate effects on tutor from effects on tutee because all students play both. ‧ 國. 學. roles.. ‧. Finally, reciprocal teaching is an approach to improving reading comprehension. sit. y. Nat. which involves instruction and practice of the following four instructional support. io. er. strategies: generating questions, summarizing, attempting to clarify confusing words or text meaning, and predicting what might happen next. The strategies are taught and. al. n. v i n C hread a passage, paragraph modeled by the teacher as they by paragraph. The students engchi U gradually take over the role of facilitator and provide instructional support to each other. Students are encouraged to initiate discussion and comment on each other’s contributions, with the goal of achieving a dialogue about the meaning of the text (Rosenshine & Meister, 1994). Peer Tutoring in SLA Peer tutoring has been helpful for consistently academically unsuccessful or. challenged students (Cohen, Kulik & Kulik, 1982). There are a number of benefits to students participating in a peer tutoring program for both tutors and tutees. These benefits include developing networking opportunities; making friends; building 11.

(23) confidence and self-esteem; enhancing team-working skills; and developing leadership skills (Beasley, 1997; Kalkowski, 1995). In addition to these benefits, some challenges of peer tutoring have also been reported. One challenge is the ability of tutors to teach tutees. Previous studies suggested that the cognitive benefits of peer collaboration may depend on a complex set of factors such as age, comparative ability level of partners, motivation, confidence, gender and the task (Garton & Pratt, 2001; Hoagn & Tudge, 1999; Strough, Berg, & Meegan, 2001). Some researchers (King, 1999; Kruger, 1992; Light & Littleton, 1994; Rogoff, 1990; Samaha & De Lisi, 2000). 政治大 through verbal communication. Exchange of ideas through verbal communication is 立. argue that a key element of effective peer collaboration is the active exchange of ideas. not about easy talk or casual chats but about the tutors providing some clear. ‧ 國. 學. explanations to the tutees and the responsive actions elicited from the tutees as well. ‧. which leads to development. This occurs usually in high/low combination of students. sit. y. Nat. due to high ability students tend to provide more clearer explanations or instructions. io. 1992).. er. which help the low ability tutees understand and respond to them (Fuchs, 1996; Webb,. al. n. v i n C h behaviors thatUinfluence learning, research In terms of individual students engchi. indicates that students who learn most in peer tutoring are those who provide detailed and complex explanations to the peers (Cohen, 1994; Fuchs, 1996; Webb, 1992). In addition, the students who provide these explanations are more likely to be high achieving students (Cohen, 1994; Fuchs, 1996). Incidentally, explainees only appear to learn when the explanation is of high quality, and tutees do learn more when tutored by a high ability tutor as opposed to a medium ability tutor (Fuchs, 1996; Webb, 1992). However, students placed in collaborative contexts do not spontaneously use higher order thinking and ask good questions, nor do they use appropriate social skills 12.

(24) (Cohen, 1994). Without training, explanations in collaborative groups or dyads are often confused. (Fuchs, Fuchs, Bentz, Phillips, & Hamlett, 1994; Fuchs, Fuchs, Kazdan & Allen, 1999). Some positive effects have been found for training in peer tutoring (Cohen, 1994). Thus, Cohen suggested that training students in skills of working together and peer tutoring increases the effectiveness of peer teaching. Other than the training of students’ ability to elicit meaningful dialogues, another challenge that might have impacts on peers’ constant dialogic activities is the relationship between the tutor and the tutee. According to Pugh (2005), tutees often. 政治大 “can’t learn.” A peer tutor, whether the same age or older, should be seen more as a 立 express (and sometimes mean it) that they don’t “like” their teacher and that they. friend than an authority figure. The tutor can have an edge by being “liked” and. ‧ 國. 學. encourage a “can do” attitude. Pugh suggested the first thing a tutor must do is. ‧. establish a friendship with the tutee. Having a good relationship with one another is. sit. y. Nat. more likely to trigger more dialogues and responses. This is most teachers cannot and. io. er. should not do because there will always be a friend/authority figure conflict. On the other hand, peers should be friends. Thus, friendship is a natural and productive. al. n. v i n C h relationship. Pugh starting point for the peer tutoring said students should not be engchi U. placed in a tutoring relationship if they express a dislike for their tutor or tutee. Both students should be given the opportunity to say what and who would make them feel most comfortable in order to set effective conditions for learning. She pointed out that matching tutors and tutees should be on the basis of relationships more than skills. Personalities and personal preferences should be taken into consideration. Previous researches placed much emphasis on tutors’ ability to teach and teaching quality, overlooking that friendship might have impacts on the teaching quality and tutees’ responses too. How friendships play a role in peer interaction is an underexplored area which inspired the present study to do more in-depth investigation. 13.

(25) In Ohta’s research (1995), reciprocal peer tutoring was conducted. Two Japanese learners engaged in form-focused collaborative activity that revolved around concerns for grammatical accuracy. In their interaction, the learners negotiated language form and the giving/receiving of form-related assistance. Analysis proceeded using procedures for discourse analysis, including attention to the two learners’ dialogic interaction. Their interaction was closely examined for evidence of assisted performance, with the focus of analysis on episodes in which help was or was not provided to each other and how help was negotiated in between.. 政治大 interaction happens in the zone of proximal development (ZPD) (Lantolf 2000, p54), 立 According to Vygotsky (1978), internalization of the language of social. the interactional space within which a learner is enabled to perform a task beyond his. ‧ 國. 學. or her own current level of competence, through assisted performance. Ohta (1995). ‧. examined the collaborative pair-work task and found that emergence of a ZPD. sit. y. Nat. resulted in learners performing at a higher level of competence (Lantolf 2000, p55). In. io. er. Lantolf and Aljaafreh’s peer tutoring research (1995), they examined interaction between adult ESL learners and more expert tutors. Learners were shown to progress. al. n. v i n C h sensitive assistance in the ZPD through developmentally in tutoring sessions. engchi U According to Lantolf and Aljaafreh,. Determining a learners’ ZPD is an act of negotiated discovery that is realized through dialogic interaction between learner and expert. In other words, the learner and expert engage each other in an attempt to discover precisely what the learner is able to achieve without help and what the learner can accomplish with assistance … . Importantly, the help negotiated between the novice and expert is graduated and contingent in the sense that it moves from more explicit to more implicit, or strategic, levels, and is offered only when needed and is withdrawn once the novice shows signs of self-control and ability to function independently (Aljaafreh and Lantolf 1994:468) or even rejects help when it is offered (Wertsch and Hickmann 1987). (Lantolf and Aljaafreh 1995:620) 14.

(26) Their research carefully examined their dialogic activities in which the expert discovered the potential developmental level of the novice and provided appropriate help accordingly. Results of analysis showed that the help led development did not occur in a random way, but in an orderly and developmentally sensitive manner. A list of levels of assistance from explicit to implicit was identified in the analysis of the interactions that occurred during the tutorial sessions. The assistance was negotiated between the novice and the expert. The idea was to offer just enough assistance from the expert to encourage and guide the learner to participate in the activity and to. 政治大 combination of the peers and whether some training was given before the peer 立. assume increased responsibility for arriving at the appropriate performance. However,. tutoring were not known in their researches. Domestic peer tutoring studies have. ‧ 國. 學. focused more on the types of assistance offered by tutors in isolation without. ‧. considering the relation of assistance to the development of language learning. They. sit. y. Nat. overlooked the tutors’ reaction to their tutees and the tutees’ involvement,. io. er. responsiveness and appropriation. Little was discussed about whether tutor training given ahead of peer tutoring has impacts on tutees’ learning or tutors’ teaching. n. al. perceptions.. Ch. engchi. i n U. v. This researcher found similar positive results in her remedial class resulted from employing peer-tutoring program in assisting low achieving learners in English basic reading in past years’ experiences. The result echoes studies mentioned above that peer interaction facilitates improved competence. In the past, the researcher did have tutor training before the peer tutoring. The tutor training was, at most, guiding tutors to give oral encouragement and praises to tutees. Teaching skills were not included. Combination was based on peers’ good relationships for it was easier and faster to get each other involved. In the present study, instead of letting the 5th-grade tutors randomly help their tutees in solving the tasks, the researcher trained the tutors with 15.

(27) some methods ahead of the peer tutoring. One of the methods was to provide roughly several levels of help from explicit to implicit kinds for the tutors to employ which kind would suit their tutees’ current needs when their tutees asked for help. The tutors needed to pay attention to their tutees’ learning difficulties so as to offer appropriate assistance. Since the orderly and developmental levels of help can lead development in language learning in peer interaction (Lantolf & Aljaafreh, 1994), the present study was curious about whether any similar or particular phenomenon could be found in analyzing and explaining how the 5th-grade tutors, if given some trainings ahead of. 政治大. peer tutoring, helped the peer tutees in their dialogic interactions which led to language development.. 立. Therefore, the study employed Aljaafreh, Lantolf and Ohta’s research framework. ‧ 國. 學. to carefully examine how the trained 5th-grade tutors and tutees engage in tasks.. ‧. How assistance negotiated and offered in the peer interaction was discussed and. sit. y. Nat. whether good relationships play a role in the interaction.. io. er. Research Questions. Drawing on Vygotskian sociocultural perspectives, this study aims to investigate. n. al. Ch. the following three research questions:. engchi. i n U. v. 1. What role does the tutor training play in the peer interaction? 2. How does the tutor training have impacts on the tutor’s teaching and the tutee’s learning? 3. How do relationships between the tutor and the tutee influence the tutor’s teaching and the tutee’s language development? What are their perceptions?. 16.

(28) CHAPTER THREE METHOD Drawing on Vygotskian sociocultural perspectives, this study aims to explore aspects of a dyad of tutor-tutee interactions and to understand how their interactions scaffold and contribute to the tutee’s performance in an English remedial class. This chapter is comprised of three sections. Section one describes background of the participants. Section two illustrates the designed materials and tasks employed in the peer interaction. Section three presents the procedure, data collection and analysis of the study. Context of the Study. 立. 政治大. The elementary school of this study is located in Shu-lin city, Taipei county,. ‧ 國. 學. which is a central school in charge of administratrative affairs in San-xia and Ying-ge. ‧. school districts. It is a large-size school with a total of seventy-two classes and. sit. y. Nat. thirty-five students in each class. This elementary school is similar to most of other. io. er. Taipei county elementary schools. First, each class is of academically normal distribution. English is a required subject with two hours of English class for all. al. n. rd. students from 3 to 6. th. v i n C hand one hour forU1 and 2 grades engchi st. nd. grades in every week.. Second, two county-wide formal English proficiency tests are administered in the 4th and 6th grades every year to make sure that students have attained the required basic English proficiency of their grades. Third, English remedial classes are subsidized by the government and conducted in extra time for slow learners whose English proficiency falls behind their peers. However, this elementary school is dissimilar from many others in that it is located in the center of San-xia and Ying-ge schools district, in which houses a number of families of high socioeconomic status. Therefore, the students in this school are normally considered good students who receive more attention as well as better familial education at home from parents who are usually of 17.

(29) high socioeconomic status in the conventional sense. In this school, there are 10 classes in each grade. Each class is comprised of approximate 35 English mixed ability students (high, moderate, and low levels). Because students’ in-class English proficiency levels are getting polarized, English teachers of this school spend extra 3-4 hours per month conducting remedial classes in order to help slow learners catch up with the progress of their grades. Each English teacher has his/her own unique ways of conducting remedial classes. The remedial class conducted by the researcher was comprised of 15 dyads of tutor-tutee during. 政治大 learning tasks. According to the researcher’s past successful experience of grouping 立. data collection. Each dyad had a tutor assisting a tutee to complete basic, graded. remedial class students, each dyad was composed by one high and one low level of. ‧ 國. 學. English proficiency, with the high-level students assisting the low-level students to. ‧. complete collaborative tasks which gradually led to the low-level students’ language. sit. y. Nat. development. According to Pugh (2005), besides considering the tutor’s academic. io. er. performance, the researcher would match the tutee with the one whom the tutee was acquainted with as his/her tutor. During data collection, the researcher conducted the. al. n. v i n remedial class by followingC the same method of U h e n g c h i matching tutor-tutee dyads: That is, according to their English academic performance in the previous year, their willingness to learn/help and their acquaintanceship with each other. The tutees were learners whose English performance was far behind the normal learning progress according to his/her English academic scores /performance over the past year, having difficulties in learning current 5th-grade contents and need to improve in basic letter and sound correspondence. Their overall English academic scores were below the average scores of their grade. They expressed needs and willingness to attend the remedial class to improve their basic English ability when inquired by the research. 18.

(30) The tutors were of high level of English proficiency based on his/her English academic scores /performance over the past year. Their overall English academic scores were above the average scores of their grade. They volunteered to help their peers when inquired by the researcher/their English teacher. Not only was their English academic performance the reason why they were involved in the selection but also the strong willingness that both the tutee and the tutor demonstrated to learn/help before they were admitted in the remedial sessions. Therefore, in the beginning of this semester, according to the new 5th graders’ overall scores/performance of the previous. 政治大 English remedial class to improve their basic ability. Fifteen high-level proficiency 立. year (i.e., 4th grade), there were fifteen students being selected as needing to attend the. students were also selected to tutor the 15 lower peers in the remedial class. The tutor. ‧ 國. 學. and the tutee matched in each dyad were of the same class, gender and were familiar. ‧. with each other. Materials used were basic spelling reinforcement tasks based on the. sit. y. Nat. textbook progress. Tasks were of gradation and designed to strengthen the tutees’. io. er. basic letter/sound correspondence and spelling ability which assist them in keeping up with reading ability of normal progress of their grade. The tutees were required to. al. n. v i n taskCand the tutors accompanied hengchi U. complete each graded. them, offering assistance. throughout the whole tasks. The tutees were given two proficiency tests in the middle and end of the peer tutoring sessions to examine whether or not the tutees made progress and how much progress the tutees have made. Before the start of tutorial sessions, the tutors received a week or two tutor training programs to help equip themselves with some basic coaching hints on how to deal with matters in the tutorial sessions. In order to provide detailed, in-depth observation and accounts of the one-semester long tutor-tutee interaction, one dyad was selected among the 15 dyads for this study. The selection of the case was primarily based on opportunistic 19.

(31) convenience sampling. The chosen dyad was much closer to the researcher in terms of teacher-student relationship. According to Duff (2008), the advantage of studying people with whom the researcher was already familiar with is that access and informed consent are easier to obtain. In addition, it may be possible to observe or interact with familiar participants for a more extended or intensive period, and as a result, the researcher may obtain more useful data about the case. Participants The tutee. Since the tutee’s basic English ability needed to be improved in order. 政治大 same level of low English proficiency were selected to attend the remedial class in 立. to catch up with the normal learning progress, fifteen students of approximately the. this semester. The fifteen students were those who had difficulties in learning current. ‧ 國. 學. 5th-grade contents and need to improve the basic skill of letter and sound. ‧. correspondence. Each tutee was assigned to a familiar tutor. The tutee informant of. sit. y. Nat. the study, Mary (a pseudonym), who the researcher was acquainted with was selected. io. er. for this study and was assigned to be in the dyad assisted by her familiar classmate, Kate. Kate and Mary spent a lot of time together at school. Since the homeroom. al. n. v i n teacher sometimes designedC in-class collaborationU h e n g c h i tasks in order to encourage students. to learn from each other, Kate and Mary usually teamed themselves up for group projects. Most of the time, Kate was ready and willing to provide solutions in group projects when Mary turned to her for help. In the English class, Kate also actively helped Mary in some spelling or reading problems. Mary and Kate were good friends, so Mary felt very comfortable about grouping in the same dyad. Mary thus agreed to participate in this study. Mary was a 5th-grade girl from a middle-class, aboriginal family with two younger sisters and a younger brother. Both of her parents worked in factories. Her parents seldom participated in her learning at home. As the eldest child in the family, 20.

(32) Mary had to take care of her sisters and brother right after school. Looking after her siblings usually took up most of the evening so that little time was spared for her studying and doing homework. Sometimes Mary would even be late for school. At school, Mary got along well with her classmates and she liked to be helpful in class. However, in terms of academic performance, she didn’t show very high motivation for learning. Her overall academic performance was below the average level according to her previous two years’ grades. She felt inferior to others in terms of academic learning which results in her tendency to get nervous and frustrated in. 政治大 English teacher’s extra individual instructions and concerns for her difficulties, Mary 立 learning. However, in English class, due to the relaxing learning atmosphere and the. showed interest. She has been attending the English remedial class since the fourth. ‧ 國. 學. grade and did make improvement in basic letter and sound correspondence (according. ‧. to her 4th-grade overall academic performance). Since there was still a lot for Mary to. y. Nat. improve, including basic spelling and reading skills which a normal 5th grader should. er. io. sit. obtain before starting her 5th grade, Mary was arranged to be in the remedial class again this year. She did not have any outside-school English learning experience and. n. al. Ch. was willing to participate in this study.. engchi. i n U. v. The tutor. The fifteen tutors were approximately of the same level of high English proficiency according to their overall past-year English academic scores/performance. 10 of them had experiences of tutoring peers in English remedial class; the rest of 5 had none. Whether or not the tutor had the experience of tutoring peers, they showed willingness to help and receive tutor training before the remedial class. They were all very friendly with their assigned tutees. Among the 15 tutors, Kate (a pseudonym), who was teamed with Mary, was selected as the tutor informant for this study. The researcher/teacher also had a good relationship with Kate. Kate had the experiences of peer tutoring other students in the English remedial class in the 21.

(33) previous year when she was a 4th grader and has won reputation among her peers for being very helpful. According to the students she tutored, Kate really had her ways of teaching and assisting them in basic spelling and reading. Kate was patient with slow learners in clearly guiding them every word spelling. In order to create relax learning atmosphere, she liked to crack jokes in the beginning of the tutoring. Kate liked to make friends with the tutees. By doing so, Kate thought it would be much easier for the tutee to be open to learning. They liked her very much. Apart from their personal viewpoints towards Kate, based on the final proficiency tests of these tutees. 政治大 and might have worked in improving 立. administered at the end of the remedial class, those ways that Kate used to help seemed effective. the tutees’ language. development. In this semester, Kate volunteered to assist other fellow students in the. ‧ 國. 學. remedial class as in the past. Kate’s ways of tutoring peers won her good reputation of. ‧. effectively assisting low level peers in language learning and problem solving. The. sit. y. Nat. study was highly interested in exploring how and in what ways such a high English. io. er. proficiency level student help a low English proficiency level peer attain basic language development in the tutor-tutee communicative interaction.. al. n. v i n C h family according Kate came from a well-fixed to her personal information from engchi U. her homeroom teacher. She was an outgoing girl. She has learnt English ahead of. formal elementary English education for five years. She went to a private language school for extra 4-6 hours of English language per week. Her overall academic performance was above the average level in her class. Her English proficiency was as good as an 8th graders’ level in junior high according to Kate’s junior-high English mock exams taken in school. Though Kate’s English proficiency was far above her class’ level, she still actively participated in her English class activities. Apart from Kate’s English academic performance and strong willingness to help, her good relationships with Mary made the researcher finally decided to assign her to 22.

(34) assist Mary in the English remedial class. Kate agreed to participate. Materials As a qualitative study, this study’s sources of data are the pre-tutoring training program and peer tutoring materials. The pre-tutoring training program was provided by the researcher as supplementary materials based on the school textbooks to guide the tutors how to assist the tutees, while the peer tutoring materials were designed based on the school textbooks by the researcher (see Appendix A). Structure of the peer tutoring. 政治大 basic word recognition, synthetic word spelling and reading. Basic letter/sound 立. materials was of gradation from fundamental phonemic awareness, CVC spelling to. correspondence, letters writing, CVC spelling, basic readings were the key elements. ‧ 國. 學. designed in the teaching materials. There were two phases of peer tutoring: first,. ‧. from week 3 to 11, the emphasis was placed on building up the tutee’s ability of. sit. y. Nat. basic word recognition and reading. The review order was according to the. io. er. textbook’s designed arrangement: CVC spelling with the five short vowels (a,e,i,o,u) and five long vowels spelling (a_e, e_e, i_e, o_e, u_e). Second, from week 12 to18,. al. n. v i n C hto generalize her U focus was on helping the tutee acquired rules to more varied words engchi reading. The review order was from systematic words to synthetic words. Those. words were varied and more difficult. Sources came from: Basic & Intermediate Phonics by Melody publisher and East & West publisher, and self-developed worksheets; multimedia-based materials such as tutorial websites, computer software were also employed; reading materials were mainly selected from basic Scholastic Readers. Data Collection The procedure and time frame of this research are shown in Figure 1 & 2. The study lasted for the whole semester (twenty weeks). It began with a pre-tutoring 23.

(35) preparatory stage, which was a two-week tutor training provided by the researcher using supplementary materials before the peer tutoring sessions started. The tutor was trained with basic teaching knowledge and teaching techniques by the researcher according to the supplementary materials. Then the peer tutoring session was conducted for 18 weeks: thirty minutes per session, two sessions per week. It was conducted in mornings as the tutee’s extra learning hours. All of the sessions were videotaped and tape recorded. Video recording was used so that the participants’ paralinguistic expressions such as gestures and facial expressions during their. 政治大 the tutee’s linguistic as well as affective outcomes, close examination of their 立. interaction could be incorporated into the analysis. Since this study examined not only. paralinguistic expressions during their interaction was worthwhile. Every three weeks. ‧ 國. 學. during the peer tutoring sessions, the researcher and the tutor went through a. ‧. reflection review to ensure that the tutoring process proceeds without any serious. sit. y. Nat. problems. In the last week, interviews were conducted after the peer tutoring sessions.. io. al. n. timeline.. er. Figure 1 below depicts the data collection process in this study, while Figure 2 is the. Ch. engchi. 24. i n U. v.

(36) Selection of tutees and tutors. Tutor training (week 1-2). Peer tutoring (weeks 3-19). ‧. ‧ 國. Tutoring reviews & improvement. 學. Videotaping & observation. 立. 政治大. Nat. n. al. er. io. sit. y. Interviews (week 20). Ch. engchi. i n U. v. Data analysis. Figure 1：Procedures and Data Collection of the Study. 25.

(37) Timeline of the study ---A total of twenty weeks Stage. Week. Data Collection Activities. Stage 1. 1st & 2nd. The tutors received tutor training programs provided by the researcher.. Stage 2. 3rd. Tutorial sessions began: Basic 26 letters & sounds correspondence review/quiz. Tutoring Activities. Providing tutor training. Observation th. 4. Basic CVC spelling task: short vowel a & e Session 1, 2. 政spelling治 Basic CVC task: short 大vowel i & o立. th. 5. Observation Tutor reflection reviews. Session 3, 4. ‧ 國. 學. th. 6. th. Observation. y. sit. 8. Basic CVC spelling task: long vowel a_e, e_e Session 7, 8. io. Basic CVC spelling task: long vowel i_e, o_e Session 9, 10. er. Nat th. Observation. ‧. 7. Basic CVC spelling task: short vowel u Session 5, 6. al. v i n BasicC CVC task: long h espelling i U vowel u_e h n c g Session 11, 12. Observation. Review & Proficiency Test I Session 13, 14. Observation. n 9th th. 10. th. Tutor reflection reviews. 11. Reinforcement: Basic CVC spelling, with blends and diagraphs added. Session 15, 16. 12th. Synthetic spelling task: letter A-D initial word set Session 17, 18. Observation. Synthetic spelling task: letter E-H initial word set Session19, 20. Observation. Synthetic spelling task: letter I-L initial word set Session 21, 22. Tutor reflection reviews. th. 13. th. 14. 26. Tutor reflection reviews.

(38) 15th. Synthetic spelling task: letter M-P initial word set Session 23, 24. Observation. Synthetic spelling task: letter Q-T initial word set Session 25, 26. Observation. 17. Synthetic spelling task: letter U-Z initial word set Session 27, 28. Tutor reflection reviews. 18th. Review Session 29, 30. Observation. Proficiency Test II Session 31, 32. Observation. th. 16. th. 19th Stage 3. th. 20. Individual interviews & group interviews. 政治大 Stage 1 Pre-tutoring 立preparatory stage. An English test with basic letters writing, Figure 2 : Timeline of the study. ‧ 國. 學. letter/sound correspondence, word matching and spelling was to be conducted in the class. Those who failed to get 70% correct were required to take the remedial class.. ‧. Tutors were of high English proficiency according to their past English academic. sit. y. Nat. performance and were recommended by their fourth-grade English teachers. These. n. al. er. io. tutors should pass a screening test (see Appendix B) designed by the researcher before. v. the tutoring to make sure they had enough language ability to assist their peers.. Ch. engchi. i n U. A two-week tutor training was then implemented before the peer tutoring sessions during the semester. In the training, the tutors were trained with basic teaching knowledge and teaching techniques based on the peer tutoring materials. The researcher led the training and the tutors followed the guidance for each different task in the peer tutoring materials. Most of the time, the tutors were given courage and encouragement by the researcher as to how to get along with their peers and how to proceed the tutorial sessions. As to how to proceed the tutorial sessions, a list of rules were given to the tutors. The rules were: first, the tutees did the tasks alone and the tutors observed their tutees’ reaction to the tasks. Second, if their tutees were able to. 27.

(39) continue the tasks, then the tutors kept on observing their tutees’ learning progress without intervening. Third, if their tutees paused at some sound or word that caused the reading break down, the tutors intervened and offered assistance according to the current language level and needs of the tutees. To help the tutors to understand where their tutees’ current levels were in order to provide suitable assistance, roughly five levels of help (see Table 1) from explicit to implicit kinds listed below were suggested to the tutors as they were to offer proper assistance to suit their tutees’ needs. The number of each help indicated the amount of assistance. The bigger the number was,. 政治大 when the tutees were still in the basic level of reading words and needed a great 立. the more explicit help the tutors offered/the tutees needed. Level 4 and 3 were used. amount of other regulation to help them to read. Level 2 and 1 were used when the. ‧ 國. 學. tutees started being able to read independently and assuming more responsibility.. ‧. Level 0 indicated that the tutees did not need the tutors’ help and could do the task. sit. y. Nat. alone. The levels of help were general guidelines for the tutors to help their tutees and. io. er. just for their reference. The tutors did not need to rigidly follow every exact level. The tutors had to carefully observe where their tutees’ learning levels were so as to offer. al. n. v i n C hwas 30 minutes eachUperiod, twice per week. The proper assistance. The training engchi sessions were carried out in Chinese.. Table 1 Assistance--- From Implicit to Explicit Assistance offered from implicit to explicit level 0. Tutor asks the tutee to finish the spelling tasks independently. 1. Tutor provides clues to help the tutee elicit correct spellings (e.g., “What’s the sound of the letter?”). 2. Tutor provides spelling rules. 3. Tutor provides the spelling answers. 4. Tutor provides some explanation for the use of the correct spelling.. 28.

(40) Stage 2 Peer tutoring sessions. From week 3 to 19, peer tutoring sessions were in progress ---there were thirty minutes per session, two sessions per week. It was conducted in mornings as the tutee’s extra learning hours. The sessions were carried out in Chinese. Session 1 to session 14 were basic spelling tasks and they were to reinforce the tutee’s basic spelling ability. Session 17 to session 30 were synthetic spelling tasks and they were to examine how the tutee was able to incorporate her basic learning in previous sessions into the more varied, synthetic spelling tasks. The tutee engaged in the graded tasks with the tutor offering guidance and assistance aside.. 政治大 tutee gradually obtain basic letter/sound correspondence and spelling ability so as to 立. The tutee was required to complete the graded tasks which were designed to help the. keep up with reading ability of normal progress of their grade. The tutor offered. ‧ 國. 學. assistance when seeing the tutee discontinued the spelling/reading, pausing at some. ‧. line or whenever questions were raised. In the study, the tutors were asked to start. sit. y. Nat. assistance from the highest level of help when their tutees discontinued since the. io. er. tutees’ language ability were of very low level and needed a great amount of other regulation to help them to read first. The tutee was given two tests in the middle and. al. n. v i n C hto examine whether end of the peer tutoring sessions or not the tutee made progress engchi U and how much progress the tutee has made. The researcher sat at the back of the. classroom, observing the tutoring process and taking notes. The researcher would not intervene in the process unless the tutor had difficulties explaining and turned to the researcher for help. If the tutor really turned to the researcher for help, the researcher would provide assistance instantly and let the tutoring continue. Every three weeks, the researcher and the tutor went through a reflection review for 20 minutes to ensure that the tutoring process proceeded without any serious problems. The process was videotaped and tape recorded.. 29.

(41) Stage 3 Researcher-tutor, researcher-tutee, researcher-tutor and tutee interviews. In week 20, interviews with the tutor and the tutee were conducted. There were individual interviews and group interviews. Each interview lasted for about 20 minutes. First, individual interviews were conducted. The tutor and the tutee were interviewed individually two times (i.e., 1st –2nd interview with the tutee, 3rd -- 4th interview with the tutor) by the researcher. Individual interviews aimed to elicit their background information, past English learning experience, past peer tutoring experience, current peer tutoring experience and perceptions of peer tutoring.. 政治大 appropriate when discussing sensitive issues or topics that require self-disclosure 立. According to Borkan and Miller (1997), the use of individual interviews may be more. because participants in a group may be reluctant to share issues of a deep and personal. ‧ 國. 學. nature due to concerns of confidentiality and privacy (Frasier, Slatt, Kowlowitz,. ‧. Kollisch & Mintzer, 1997). Second, in the 5th and 6th interviews, the researcher. sit. y. Nat. conducted group interviews with the tutor and the tutee. As Morgan (1998) notes, the. io. er. individual interview provides a detailed account of the person’s unique and personal experience, whereas the focus group generates a discussion of similarities and. al. n. v i n C h Also, in a focusUgroup, contrary opinions can be differences among the participants. engchi explored and may generate new areas of inquiry (Schattner, Shmerling & Murphy, 1993). Questions developed for the focus group interviews were based on the participants’ behavior and reactions observed in their interaction. Group interviews were to obtain an in-depth understanding of the tutor-tutee interaction: how and why the tutor offered assistance to the tutee, how and why the tutee responded to the tutor. Their feedback and reflection for each other were elicited as well. The full set of interviews are shown in Appendix D.. 30.

(42) Data Analysis Data Analysis Framework. Drawing on Ohta’s (1995) research, her procedures of analysis were used in the study to illustrate the nature of the language development processes in the data set. The analysis made use of Vygotsky’s sociocultural framework which considers the transformation of cognitive constructs from the interpsychological to intrapsychological space, and the ZPD, which proposes how this transformation occurs through a process of developmentally sensitive assisted performance. These constructs illuminate the role of collaborative interaction in L2. 政治大 processes. Table 2 below describes features of interaction which allow the analysis to 立. development, in allowing examination of assistance and the learners’ internalization. Focus of analysis. Analysis examined the sequential structure of episodes of assistance, examining what triggered suppliance of assistance. It was not assumed that all assistance is helpful. ‘Appropritate assistance’ was defined as assistance which leads to language development, with language development defined as gains in learner performance on the word-spelling task, and maintained in the subsequent reading task. Analysis focused on changes in performance and how these changes related to the assistance provided. This is examined about how language structure is appropriated for individual use. Internalization of linguistic structure was visible through increasing independence of appropriate task performance.. n. al. er. io. sit. y. ‧. Nat. Mechanism of Assistance Appropriate Assistance. 學. Construct. ‧ 國. study these processes.. Internalization processes. Ch. engchi. i n U. v. Table2 : Examining sociocutural constructs related to assistance and internalization (Lantolf, 2000. p60 ) Analysis of the data focused on at what (timing) point, in what ways the tutor offered appropriate help and how the tutee attended to and learned from the peer-assistance. In the tutor-tutee interaction, help was expectantly offered when it 31.