• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Effects of teacher-directed and student-interactive summarization instruction on reading comprehension and written summarization of Korean fourth graders

Dissertation
Author: Jongseong Jeong
Abstract:
The purpose of this study was to investigate how Korean fourth graders' performance on reading comprehension and written summarization changes as a function of instruction in summarization across test times. Seventy five Korean fourth graders from three classes were randomly assigned to the collaborative summarization, direct instruction, and control groups. During six sessions of 20 min instruction, the experimental students received instruction in summarization via two instructional approaches: collaborative summarization and direct instruction. The collaborative summarization model was developed as an alternative to direct instruction to incorporate scaffolding and active social interaction. Reading comprehension was assessed using three sentence verification tests. Summarization was examined in terms of identifying main ideas and including extraneous information. Writing quality was a combined rating of focus, support, and organization indices. The collaborative summarization group's performance was significantly improved on identification of main ideas and writing quality in comparison to the control group, but the benefit of collaborative summarization was not reflected in reading comprehension. All three groups tended to produce shorter summaries on posttest and follow-up test than on the pretest, which led to inclusion of significantly fewer extraneous idea units on posttest and follow-up test than on the pretest. The two students at risk for reading difficulties benefited from receiving summarization instruction on reading comprehension, identification of main ideas, and writing quality, but this benefit was not durable. Collaborative summarization needs to be further refined to establish overarching instructional procedures that incorporate scaffolding and peer interaction.

Table of Contents List of Tables..................................................................................................................viii List of Figures.....................................................................................................................x Chapter I Rationale...........................................................................................................1

Chapter II Literature Review.........................................................................................15 Summarization Across Grade Levels.................................................................16 Summarization of Younger Children................................................................22 Content of Summarization Instruction..............................................................26 Instructional Approaches to Teaching Summarization...................................29 Measures of the Effect of Summarization Training.........................................42 Summary...............................................................................................................52 Research Hypotheses...........................................................................................53 Research Question...............................................................................................55

Chapter III Method.........................................................................................................56 Setting....................................................................................................................56 Pilot Study.............................................................................................................57 Participants...........................................................................................................58 Reading Materials................................................................................................61 Measures...............................................................................................................63 Procedures............................................................................................................67 Data Analysis........................................................................................................76

Chapter IV Results..........................................................................................................77 Group Equivalence..............................................................................................79 Multivariate Analysis of Dependent Variables.................................................79 Reading Comprehension.....................................................................................80 Identification of Main Ideas................................................................................82 Writing Quality....................................................................................................85 Inclusion of Extraneous Information.................................................................88 Students at Risk for Reading Difficulties..........................................................90 Summary...............................................................................................................91

Chapter V Discussion......................................................................................................93 Interpretation of Major Findings.......................................................................94 Limitations of the Study....................................................................................101 Implications for Practice...................................................................................104 Recommendations for Future Research..........................................................104 Conclusion..........................................................................................................106

References.......................................................................................................................107

vi

Appendix A Grading Criteria for Written Summaries..............................................115

Appendix B Writing Quality Rubric............................................................................117

Appendix C Implementation Integrity Checklists......................................................119

Appendix D Levels of Prompts for the Four Summarization Rules.........................122

Appendix E Range of Scores on the Four Dependent Measures...............................127

Appendix F Examination of Group Equivalence........................................................129

Appendix G Repeated Measures Analysis of Time-by-Condition Interaction......................................................................................................................131

Appendix H Pairwise Comparisons by Condition......................................................134

Appendix I Korean Instructional Passages for the Pilot and Study.........................136

Appendix J English Instructional Passages for the Pilot and Study.........................142

Appendix K Korean Test Materials for Reading Comprehension and Written Summarization.........................................................................................148

Appendix L English Test Materials for Reading Comprehension and Written Summarization.........................................................................................179

vii

List of Tables Table Page

1 Distribution of Participants by Instructional Condition and Gender....................................................................................................................60 2 Title of Instructional Passages, Relevant Domains, and Number of Sentences (Paragraphs) by Session.....................................................61 3 Title, Readability Level, and Source of Testing Passages by Test Time...........................................................................................................62 4 Teacher Responses to Examples of Student Moves................................................72 5 Mean (SD) of Scores on a Diagnostic Test and Four Dependent Measures .............................................................................................78 6 Pearson Correlation Coefficients Among Dependent Variables at Pretest................................................................................................................80 7 Repeated Measures Analysis for Time-by-Condition Interaction Effect on Reading Comprehension Between Two Groups.....................................82 8 Repeated Measures Analysis for Main Effect of Time on Identification of Main Ideas Involving Two Test Times....................................84 9 Repeated Measures Analysis for Time-by-Condition Interaction Effect on Identification of Main Ideas Between CS and Control Groups Involving Two Test Times.........................................................................85 10 Repeated Measures Analysis for Main Effect of Time on Writing Quality Involving Two Test Times.......................................................87 11 Repeated Measures Analysis for Time-by-Condition Interaction Effect on Writing Quality Between Two Groups....................................................88 12 Mean (SD) of Extraneous Idea Units for Three Groups Across Three Test Times........................................................................................89 13 Mean (SD) of Sentences in Written Summaries for Three Groups Across Three Test Times ..........................................................................90 14 Scores on a Diagnostic Test and Four Dependent Measures for Students at Risk for Reading Difficulties..........................................................92

viii

Table Page

B1 Three Writing Quality Indices Rubric..................................................................118 C1 Collaborative Summarization Integrity Checklist ..............................................120 C2 Direct Instruction Fidelity Checklist...................................................................121 D1 Levels of Prompts for Get Rid of Unnecessary Detail.........................................123 D2 Levels of Prompts for Collapse Lists and Paragraphs........................................124 D3 Levels of Prompts for Use Topic Sentences ........................................................125 D4 Levels of Prompts for Polishing the Summary.....................................................126 E1 Range of the Scores on Reading Comprehension, Identification of Main Ideas, Writing Quality, and Inclusion of Extraneous Information...................................................................................128 F1 Analysis of Variance for Group Equivalence......................................................130 G1 Repeated Measures Analysis for Time-by-Condition Interaction Effect Between Two Test Times.........................................................132 G2 Repeated Measures Analysis for Time-by-Condition Interaction Effect on Writing Quality Involving Two Test Times Between Two Groups.........................................................................133 H1 Pairwise Comparisons of Extraneous Information by Condition Using Tukey HSD...........................................................................135

ix

List of Figures Figure Page

1 Mean scores on reading comprehension by condition across three test times.......................................................................................................81

2 Means of identified main ideas by condition across three test times................................................................................................................83

3 Means of writing quality by instructional condition..............................................86

4 Average numbers of extraneous idea units by instructional condition................................................................................................................90

x

Chapter I Rationale Reading is integral for academic achievement during schooling and is critical for successful employment (National Reading Panel [NRP], 2000). The essential goal of reading is to comprehend a variety of materials written for diverse purposes. Reading comprehension is defined as “the process of simultaneously extracting and constructing meaning through interaction and involvement with written language” (NRP, p. 11). In the definition, three dimensions of reading comprehension are involved: reader, text, and activity (NRP, 2000). Reader refers to readers’ cognitive abilities, motivation, and knowledge of vocabulary, domains, and comprehension skills. Text involves genre, reading level, and content presented in the text. The more familiar a text is to a reader’s topical knowledge and experience with genre, the better readers will understand the text. Activity includes purposes for reading, operations employed to process the text, and consequences of reading. Prior to reading, readers have one or more purposes, which are internally generated or externally imposed. During reading, readers use cognitive operations to process the text. As a result, readers are expected to acquire vocabulary and domain knowledge, critical thinking, and new ideas. Currently, the importance of text genre, as a feature of text dimension, has received attention in relation to reading comprehension. With a rapid increase of informational texts at third and fourth grades in comparison to primary grades, students are more likely to encounter reading difficulties (Chall, 1983; Duke, 2000; Jeong, Gaffney, & Choi, in press). Gersten, Fuchs, Williams, and Baker (2001) argued that 1

students with learning disabilities (LD) have limited understanding of different structures Heading Levels This literacy achievement gap was reflected in Annual Reports to Congress from 1997-1998 through the 2001-2002 school years (Office of Special Education and Rehabilitative Services, 1999, 2000, 2001, 2002, 2003). Approximately 52% of students with LD were identified at grades 3 and 4. According to the 2007 reports of the annual reading assessment by the National Center for Education Statistics (NCES), the proportion of fourth graders whose reading scores were below the basic reading level is 33% (NCES, 2007). Performance below the basic reading level means that one third of fourth graders cannot read well enough to effectively complete grade-level work. Leach, Scarborough, and Rescorla (2003) examined the types of reading difficulties that students identified with LD at grades K-3 and grades 4-5 exhibit. They revealed that the percentage of students struggling with word recognition decreased from 95% at grades K-3 to 67% at grades 4-5 but students who exhibited difficulties with comprehension increased from 52% to 64%. Based on the Annual Reports, NCES, and Leach at al., students are more likely to have difficulties with reading comprehension around fourth grade and these reading difficulties correspond to the increase of informational text. Saènz and Fuchs (2002) found that high school students’ performance was significantly better in reading fluency and comprehension with narrative than expository text. In Korea, informational text first appears in first grade textbooks such as Korean Language, Right Life, and Wise Life. Right Life combines morality and social studies. Wise Life corresponds to science. Most information texts are presented in Wise Life at 2

first grade, although Korean Language and Right Life contain a few informational passages. The amount of informational text, however, remarkably increases as social studies is taught as an individual subject beginning in third grade. Social studies deals with a broad range of knowledge and information regarding economy, geography, history, and politics (Kim & Kim, 2000). Accordingly, students encounter many technical words and unfamiliar information related to each area. Thus, students struggling with reading are likely to experience overload on understanding social studies text. Korean elementary students report that social studies is their most difficult subject (Cited in Min & Kim, 2002). According to 2007 educational statistics reported by the Korea Ministry of Education, Science, and Technology, approximately 0.8% (65,940) of students (8,299,228) enrolled in schools receive special education services and 10.6 % (6,982) of them are identified with learning disabilities (Ministry of Education, Science, & Technology, n.d., 2008). Since 2002, the Korea Ministry of Education, Science, and Technology has conducted annual nationwide assessment to ensure that every third grader reaches basic scholastic abilities in reading, writing, and math. On average, 3.14 % of third graders performed below basic reading ability from 2002 to 2005. About 87% of the students that did not attain the basic level had difficulties with reading comprehension and 20% and 40% of them struggled with decoding and vocabulary knowledge, respectively (Korea Institute of Curriculum and Evaluation [KICE], 2004). Korean students ranked in the first place in the reading test of Programme for International Student Assessment (PISA) 2006, which was conducted by the Organisation for Economic Cooperation and Development (OECD) (KICE, 2007). The reading test 3

assessed abilities to use, understand, and reflect on written information for various purposes and test items were developed considering three dimensions: structure of the text, reading process, and situation of reading. Particularly, structure of the text is divided into continuous and non-continuous texts. Continuous texts are composed of narration, exposition, description, argumentation, instruction, and document. Non-continuous texts included charts, graphs, diagrams, tables, maps, and so on. Although PISA did not specify the genres of texts, the majority of texts seem to be informational. Based on the 2003 nationwide assessment of basic scholastic abilities and PISA 2006, the proportion of Korean students struggling with reading is low but the majority who do exhibit difficulties have problems with reading comprehension. As a consequence, reading comprehension should be the main focus of literacy education to prevent students from lagging behind in school. Since no description of classroom reading instruction is available in Korea, one cannot assume how teachers deal with reading comprehension of informational texts with students at third grade and above. In the U.S., Durkin (1978-79) conducted classroom observations of reading and social studies at grades 3-6 and found that instruction in reading comprehension is rarely provided in classrooms. In comparison to the lack of comprehension instruction in classrooms, the effectiveness of various comprehension skills have been studied. Summarization is a method that has been effective in improving comprehension of informational texts (Armbruster, Anderson, & Ostertag, 1987; Bean & Steenwyk, 1984; Cordero-Ponce, 2000; Day, 1986; Friend, 2001; Malone & Mastropieri, 1991; Rogevich & Perin, 2008; Rosenshine, 1983). Summarization is defined as finding or extracting main ideas and 4

supporting elements, and integrating the ideas and elements into a coherent form (Brown & Day, 1983). Brown and Day identified basic rules of summarization, which are cognitive operations that take place while writing a summary. The rules are deletion of unnecessary and redundant information, superordination of lists, selection of a topic sentence, and invention of a topic sentence. In an attempt to extract commonality from different terminology that researchers use to describe cognitive operations involved in summarization, Hidi and Anderson (1986) reduced cognitive operations into the selection process, condensation of information, and recognition that a summary should be concise and needs complex integration and transformation of the original text. In the process of selection, in which the importance of text segments is evaluated relative to the whole passage, segments to be judged important are selected but those to be evaluated less important are deleted. In the condensation process, details are replaced with more general concepts. Major topics are concisely represented through complex integration and combination. Summarization is often confused with either identification of main ideas or retelling. Finding main ideas is an important element of summarization but summarizers must organize main ideas in a hierarchical order based on importance of idea units (Thistlethwaite, 1991). Finding main ideas does not necessarily lead to making a quality summary. Retelling is a task wherein readers reproduce, either orally or in writing, a holistic understanding of what they read. While retelling, readers may become aware of text organization and the relationship among pieces of information they recall (Bromley, 1998). Although texts are reconstructed in retelling and summarization, summarization requires higher cognitive operations than retelling. Summarization requires the reader to 5

determine what should be included or excluded and integrate idea units in a coherent format, whereas the intent of retelling is to reproduce whatever readers can recall. The effectiveness of summarization has been assessed through comprehension tests such as researcher-developed multiple choice tests (Gajria & Salvia, 1992; Hare & Borchardt, 1984; Nelson & Smith, 1992; Rogevich & Perin, 2008; Wood, Winne, & Carney, 1995) and recall assessments (Armbruster et al., 1987; Cordero-Ponce, 2000; Rinehart, Stahl, & Erickson, 1986; Wood et al., 1995). Instruction in summarization was also found to enhance identification of important information, application of diverse summarization rules, and quality of a written summary (Armbruster et al., 1987; Cordero- Ponce, 2000; Day, 1986; Friend, 2001; Gajria & Salvia, 1992; Hare & Borchardt, 1984; Nelson & Smith, 1992; Rogevich & Perin, 2008; Wood et al., 1995). Except for Gajria and Salvia (1992)’s study, in which students produced oral summaries of expository text, written summaries were evaluated in the other summarization studies. Summarization studies, however, have been limited to upper grade levels (Hidi & Anderson, 1986; Stahl, 2004). In a search of Eric, PsycINFO, and Wilson Web for studies in which summarization was an independent variable, a total of 13 studies were found. Six studies were conducted with college students and five studies were implemented with middle and high school students. Only two studies involved fourth graders as participants. The lack of studies that employed summarization as an intervention for elementary grades is also obvious. In a search through ProQuest, 16 dissertations were located that used summarization as an independent variable. One study was conducted with third graders and three studies were done with fifth graders. The remaining 12 studies were implemented with students at middle school or above. The 6

restriction in participants’ grade levels was also obvious in Korea. Using KERIS, a comprehensive search engine that catalogs all kinds of studies produced in Korea, including journal articles and master’s theses and doctoral dissertations, only one of the 10 identified studies involved fifth graders. Participants in the other studies were in junior and senior high or college. Hidi and Anderson (1986) attributed the scarcity of summarization studies conducted with elementary participants to developmental differences in judgment of importance, integration of idea units, and selection or construction of topic sentences. Identification of important information is critical to successful comprehension (Williams, 1988) and is an essential component of summarization (Hidi & Anderson, 1986). In examination of the developmental trend in sensitivity to structural importance with third, fifth, seventh, and college students, Brown and Smiley (1977) revealed that seventh graders and college students were more sensitive to identifying structural importance in comparison to third and fifth graders. Taylor (1986) also found that fourth and fifth graders had difficulties finding and expressing main ideas. The fourth and fifth graders based their selection of content on how interesting the information was to them, rather than weighing the importance of the content relative to the whole text (Taylor, 1986). In Brown, Day, and Jones’ (1983) study, eleventh graders and college students performed better than fifth and seventh graders in examining importance of sentences in a text, and integrating idea units into a summary. The ability to select or construct topic sentences develops very slowly so that even college students exhibit difficulties inventing topic sentences when they are not explicit (Brown & Day, 1983; Garner & McCaleb, 7

1985). Hidi and Anderson (1986) argued that selection and construction of topic sentences can be completed by adolescents and older students. Despite young children’s difficulty identifying important information and applying summarization rules, developmental differences across grade levels do not necessarily indicate that elementary students cannot learn to summarize. In studies conducted by Brown and Smiley (1977), Brown and Day (1983), and Taylor (1986), students did not receive training for writing a summary. The difference in summarization across different grade levels reflects the extent to which the students already acquired knowledge and skills needed for summarization up to the time when their summarization performance was assessed. For this reason, one cannot be sure how elementary students would perform in writing summaries once they had received instruction. The difference across age levels is likely to be accounted for by an intervening variable, prior knowledge of content and text structure, for which the testing materials used were not controlled. In Brown and Day’s (1983) and Brown and Smiley’s (1978) studies, readability of the texts was at a fifth grade level, which was the lowest level of participants. By using the same passages across different age levels, older students are likely to have more prior knowledge of relevant content and text structures and more cognitive resources to allocate to comprehension because of the ease of the text, which likely contributed to better performance in identifying important information. Jenkins, Heliotis, Stein, and Haynes (1987) and Nelson and Smith (1992) provided third, fourth, and fifth graders instruction in summarization and, as a result, their reading comprehension and ability to identify main ideas improved. Comparing summaries written by students at different age levels, Brown et al. (1983) discovered that 8

age was not the best predictor for writing quality summaries. Rather, whether students did or did not plan before writing summaries had a greater impact on summarization than age level. Providing instruction in summarization and planning prior to writing a summary are two major factors that contribute to improvement in summary production and reading comprehension for elementary students. Thus, once adequate training is given and students plan beforehand, grade level is unlikely to be an obstacle to learning summarization. Instructional procedures employed to teach summarization vary. Most typically, a set of summarization rules are taught. Otherwise, summarization guidelines or steps that students should follow to write summaries are offered (Armbruster et al., 1987). Instructional procedures in teaching a set of summarization rules and offering guidelines or steps take the form of direct instruction. Rosenshine (1983) identified six instructional steps that appear in direct instruction: (a) reviewing previous day’s work, (b) presenting content and skills to teach, (c) guided student practice, (d) providing feedback and corrections, (e) independent practice, and (f) weekly and monthly review. These steps became a standard instructional process and are followed in the summarization studies wherein direct instruction is a major instructional approach. Summarization is also implicitly taught without addressing summarization rules and guidelines or steps (Bean & Steenwyk, 1984; Jenkins et al., 1987). Students were allocated space following each paragraph and were asked to write a summary of each paragraph. Finally, they combined the paragraph summaries into a cohesive summary of the passage (Bean & Steenwyk, 1984). Bean and Steenwyk referred to the implicit teaching of summarization as an intuitive approach. In the intuitive approach, students 9

were expected to sense that summarization is to concisely integrate important information. The intuitive approach group significantly outperformed the control group on a paragraph summary writing task and a standardized test of paragraph comprehension, Nelson Reading Test, Form B. Direct instruction proceeds in small steps and detailed instructions and explanations are provided. To enhance students’ understanding, many examples are offered and immediate feedback is given. An assignment is broken down into small pieces and students practice each skill until they are automatic (Rosenshine, 1983). Teachers structure lessons by planning instruction in advance and students learn skills by rote. Although the teacher’s role fades as students practice independently, the entire lesson is run according to the teacher’s preplanning. Direct instruction is criticized for making students passive recipients. Preplanned lessons and overlearning naturally include rote learning and memorization. In contrast to direct instruction, an alternative instructional approach focuses on inducing students’ involvement through scaffolding and facilitating social interaction with and among learners. Instead of letting students learn by themselves, teachers provide responsive assistance. Rooted in constructivism, this perspective views students as constructors of knowledge and teachers as facilitators who enhance interactions with and among students and provide learning environments in which students take on active roles (Vygotsky, 1978). The concept of the zone of proximal development (ZPD) that Vygotsky (1978) introduced provides practical implications for alternative instruction. The ZPD is the gap between the actual developmental level, in which learners can independently resolve 10

tasks or problems, and the potential developmental level, where learners can figure out problems with adult guidance or in collaboration with more capable peers (Vygotsky, 1978). Learning takes place in the ZPD through scaffolding, which is a process of assisting novices in solving problems that are beyond their present capabilities (Wood, Bruner, & Ross, 1976). Students are motivated for learning by setting up instructional objectives that are just above students’ current skills and knowledge level (Jaramillo, 1996). As students complete the given task with teacher’s responsive assistance, they experience a sense of achievement and progress as confident learners. Accordingly, the teacher fades her or his assistance. In this alternative approach, prompting is the major tool that a teacher employs to enhance students’ engagement. Prompts are adjusted in response to the learner’s current understanding and need for support, which is called contingency (Stone, 2002). Rodgers (2004) categorized teacher’s responsive assistance by the degree of explicitness into four levels: questioning, directing, demonstrating, and telling. Questioning is the least specific prompt and, on the other hand, telling is the most explicit one. The teacher flexibly adjusts the level of assistance in response to students’ performance. As students come to succeed in performing the task, the level of a prompt becomes less specific and vice versa. Scaffolded instruction is represented in Collaborative Reasoning (Waggoner, Chinn, Yi, & Anderson, 1995) and reciprocal teaching (Palincsar & Brown, 1984). Collaborative Reasoning (CR) is a discussion format in which students are encouraged to engage in discussions and think in a reasoned manner about text. Students discuss the question that a teacher poses in small groups. Students state their positions and offer 11

reasons to support the positions. Students also challenge other students with whom they disagree. The teacher promotes students to develop reasoned argumentation by modeling, prompting, clarifying, summarizing, and encouraging (Clark et al., 2003). In reciprocal teaching, comprehension-fostering skills are taught in an interactive dialogue between a teacher and students (Palincsar & Brown, 1984). In the beginning stages, the teacher initiates the dialogue but as students become more capable of applying the skills, the teacher invites students to initiate discussion and to assume the teacher’s role. (Rosenshine & Meister, 1994). Most studies in which scaffolding was used were conducted with normally developing children but some studies were carried out with students with special needs, specifically those with learning disabilities (Stone, 1998). Palincsar and Brown (1984) applied reciprocal teaching to students who struggled with reading comprehension and found that their reading comprehension significantly improved and their interactions with peers and the teacher became more active over instruction sessions. Bos and Anders (1990) emphasized the importance of scaffolded instruction in teaching students with learning disabilities. The participants in their study were engaged in strategic questioning, organizing, and use of prior knowledge through ongoing dialogue between the teacher and small groups of students. Students in the experimental condition significantly outperformed those in the control group on reading comprehension. Englert, Tarrant, Mariage, and Oxer (1994) implemented a didactic instructional program and a scaffolded instructional program to teach reading comprehension strategies to students with mild disabilities. The majority of the participants were identified with learning disabilities. The didactic program was confirmatory and factually oriented, while the scaffolded 12

instruction program was open-ended and collaborative. During scaffolded instruction, students talked directly to peers, rather than interacting primarily with the teacher. Students also shared their ideas and extended them by negotiating with peers’ ideas. The teacher facilitated students to elaborate on their ideas by asking guiding questions. The students who received the scaffolded instruction program outperformed those in the didactic program on written recall and strategy knowledge, which was evaluated on generating a main idea, asking a question about paragraphs, predicting, and identifying strategies to use. Scaffolding-oriented instruction includes factors that promote the learning of students with learning disabilities. During scaffolded instruction, the teacher continually assesses students’ performance and detects needs for support (Stone, 1998). According to the results from the formative assessment, the teacher may adjust the level of assistance, so students with learning disabilities are less likely to experience cumulative deficits in learning that may result from repeated failure to complete prerequisites. Furthermore, in small groups of heterogeneous members, the likelihood of learning from peers increases (Anderson et al., 2001). Peers are effective contributors who play a role in enhancing the reading performance of students with learning disabilities (Smith, 2004). Generally, students with learning disabilities tend to underachieve compared to their peers at the same ability (Smith, 2004). With fear of failure, students with learning disabilities tend to avoid learning tasks or be passive learners (Pearl, 1982; Torgesen & Licht, 1983). Through scaffolded instruction, students with learning disabilities are invited to get involved in a task that is a little beyond the students’ performance level with teacher’s 13

responsive assistance. Hopefully, participants in this study, whether or not they have learning disabilities, will benefit from scaffolded instruction. The current study examines a new approach to enhance students’ active involvement in the learning process and social interaction. In addition to evaluating an attempt to develop an alternative instructional approach, the current study intends to extend the teaching of summarization to lower grade levels. The participants in the current study are fourth graders while the majority of summarization studies have been conducted with students in middle school or beyond. Developing a new instructional approach and extending instruction in summarization to fourth grade will be explored. 14

Full document contains 220 pages
Abstract: The purpose of this study was to investigate how Korean fourth graders' performance on reading comprehension and written summarization changes as a function of instruction in summarization across test times. Seventy five Korean fourth graders from three classes were randomly assigned to the collaborative summarization, direct instruction, and control groups. During six sessions of 20 min instruction, the experimental students received instruction in summarization via two instructional approaches: collaborative summarization and direct instruction. The collaborative summarization model was developed as an alternative to direct instruction to incorporate scaffolding and active social interaction. Reading comprehension was assessed using three sentence verification tests. Summarization was examined in terms of identifying main ideas and including extraneous information. Writing quality was a combined rating of focus, support, and organization indices. The collaborative summarization group's performance was significantly improved on identification of main ideas and writing quality in comparison to the control group, but the benefit of collaborative summarization was not reflected in reading comprehension. All three groups tended to produce shorter summaries on posttest and follow-up test than on the pretest, which led to inclusion of significantly fewer extraneous idea units on posttest and follow-up test than on the pretest. The two students at risk for reading difficulties benefited from receiving summarization instruction on reading comprehension, identification of main ideas, and writing quality, but this benefit was not durable. Collaborative summarization needs to be further refined to establish overarching instructional procedures that incorporate scaffolding and peer interaction.