Download Running head: Reading Interventions and more Study Guides, Projects, Research Learning disability in PDF only on Docsity!
Running head: Reading Intervention
A Strategy Intervention to Increase the Reading Comprehension of Junior High School Students with Reading Disabilities
Trudy G. Mothus and Judith C. Lapadat University of Northern British Columbia
Author note. Trudy G. Mothus, Education Program. Judith C. Lapadat, Education Program. We thank the following colleagues for their helpful comments: Deborah L. Butler, Peter MacMillan, Judith A. Scott, and Bruno Zumbo. Correspondence concerning this article should be addressed to Judith C. Lapadat, Northwest Regional Chair and Professor of Education, University of Northern British Columbia, Northwest Regional Campus, 4741 Park Avenue, Terrace, BC, Canada, V8G 1W2. E-mail: [email protected].
Abstract A challenge facing educators is to find ways to arrest and reverse the cumulative deficit in reading experienced by many students with learning disabilities. In this study, we evaluated the effect of a strategy intervention to increase the reading comprehension of eighth grade students with reading disabilities in intact junior high school classes (N = 98). Reading comprehension gains made by students taught a paraphrasing strategy (SIM) were compared to those of students who received conventional learning assistance (LA) and to a control group who participated in no extra intervention (No-LA). Over a year, the reading comprehension gains of students taught the paraphrasing strategy were significantly higher than those of students in LA, while attrition from the No-LA group was too high to prevent a valid comparison. Findings support the efficacy of using a strategy intervention to improve the reading comprehension of students with reading disabilities, and demonstrate that the intervention can be implemented effectively within the context of daily classroom instruction in a junior high school.
Low Achievement, Dropout, and LLD Researchers have found close relationships between reading disabilities and learning disabilities, between learning disabilities and low achievement, and between low achievement and dropout. However, there is a paucity of research tracing the relationship across all of these variables, particularly at the junior high school level. For example, McCaul, Donaldson, Coladarci, and Davis, (1992), who used the tenth grade data from the U.S. High School and Beyond data base of 600 dropouts and 2000 graduates who did not continue in post-secondary education, found that dropouts had significantly lower achievement scores than the graduates who did not continue their schooling after graduation. Half of all students who drop out of school may be categorized as having LD (Bender, 1995; Blackorby, Edgar, & Kortering, 1991; Deshler, Schumaker, & Lenz, 1984; Hasazi, Johnson, Hasazi, Gordon, & Hull 1989; Levin, Zigmond, & Birch, 1985; Mellard & Hazel, 1992; Tanner, Krahn, & Hartnagel, 1995). Bender (1995) estimates that this rate is actually conservative and predicts that LD dropout rates in schools without special programs may be higher. Poor achievement of students is listed in the dropout literature as one of the major reasons of school failure (Barrington & Hendricks, 1989; Ponsford & Lapadat, 2001; Shaw, Cullen, McGuire, & Brinckerhoff, 1995; Trusty, & Dooley-Dickey, 1993). Students at risk for dropping out characteristically are identified by student-related factors such as low school achievement, poor social skills, deficient cognitive/academic skills, socio-behavioral problems, socio-economic factors, alienation from school, absenteeism, and dropout from school (Bender, 1995; Deshler, Schumaker, & Lenz, 1984; Schumaker, Alley, Warner, & Deshler, 1980; Shaw et al., 1995; Soodak & Podell, 1994; Stanovich, 1988; Tanner et al., 1995; Weber, 1994). Although Hallahan and Kauffman (1982) have called poor academic achievement the "hallmark" of learning disabilities (p. 115), little of this dropout research addresses a root achievement issue for students with learning disabilities – reading difficulties. Rather, most interventions for
dropout have centered on the attempt to change behavioral characteristics of students with learning disabilities (Blackorby et al., 1991). Learning Disabilities and Reading Disabilities In contrast to research on dropout, research on learning disabilities indicates reading disability and difficulty with reading comprehension – not just low achievement – to be among the main defining and most frequent characteristics of students with learning disabilities (Bender, 1995; Hallahan & Kauffman, 1982; Karlin, 1980; Lindsey & Kerlin, 1979; Norman & Zigmond, 1980; Rush & Vitale, 1994; Stanovich, 1988; Torgesen, 1988a, 1988b, 1989; Zigmond, Vallecorsa, & Leinhardt, 1990). Some researchers have estimated that 90% of all students classified as LD are reading disabled (Kaluger & Kolson, 1978; Levin et al., 1985). Most students who are classified as having LD are unable to read competently at grade level. Torgesen (1989) points out that a child who fails in reading will respond with inappropriate social behaviors in the classroom, truancy, lowered self-esteem, and school-leaving before grade twelve. Behavioral difficulties and absenteeism decrease opportunities to learn, initiating a downward spiral of behaviors, such as lesson evasion, leading to failure. Failure, in turn, may lead to decreased self esteem and even less productive achievement behaviors (Ciborowski, 1995; Hallahan & Kauffman, 1982; Lapadat, 1998; Torgesen, 1989; Vauras, Lehtinen, Olkinuora, & Salonen, 1993). It is often a reading and writing disability that is at the nucleus of poor achievement and, ultimately, school failure. One reason that poor reading ability is likely to result in low achievement and academic failure is that children who have difficulty with reading also have difficulty acquiring domain- specific information via the reading process (Perfetti, 1984, 1986; Shaw et al., 1995; Torgesen, 1989). Bender’s label for this is “cumulative deficit” (1995, p. 179; see also Chall, Jacobs, & Baldwin, 1990; Lapadat, 1991; Rose, Medway, Cantrell, & Marus, 1983). Students with learning or reading disabilities fall further behind each year by mastering perhaps only three-fourths of each year's content of instruction. At the entry to junior high school, students with LD tend to
Traditional Learning Assistance Interventions Most high schools use a combination of tutorial, basic remediation, and compensatory learning assistance (LA) approaches to address the consequences of learning, reading, and writing disabilities (Bender, 1995; Alley & Deshler, 1979; Deshler, Schumaker, & Lenz, 1984). In the tutorial approach, students with LD obtain help from a learning assistance teacher to complete class assignments and prepare for course examinations in order to meet the requirements of core subjects. In the basic skills remediation approach, students obtain extra work from the LA teacher in order to remedy deficiencies in foundational skills deemed necessary for grade level learning. The compensatory approach involves the use of modifications and diverse formats to present information to the students. This may include use of audiotaped lessons and books, oral testing, charts of simplified text data, vocabulary sheets, and controlled reading levels on tests. The emphasis in all three of these approaches is to help students complete the assignments of the regular curriculum. Interventions that explicitly address reading deficits in order to improve reading competence rarely are attempted (Soodak & Podell, 1994). Instead, content mastery is the focus. Deshler, Schumaker, Lenz, and Ellis (1984) and Clark (1993) found little empirical evidence that these three LA approaches were effective in increasing reading comprehension ability. Schumaker, Deshler, Alley, and Warner (1983), Clark (1993), and Gottesman (1979) found that students in LA programs demonstrated minimal achievement gains of 0.2 to 0.3, 0.6, and 0.4 grade levels a year respectively while in remedial instruction. Any achievement gain of less than one grade level per year is not enough for students in these programs to keep pace with their non-LD peers (Bender, 1995; Zigmond et al., 1990). Instead, these findings show that LD students continued to fall further behind in their level of reading comprehension. Strategy Interventions Strategy instruction is an alternative to traditional LA approaches. This intervention approach directly addresses students’ difficulties in social skills, communication, behavior, study
skills, writing, and reading comprehension by teaching them to use strategies (Palincsar & Brown, 1987; Deshler, Schumaker, Lenz, & Ellis, 1984). Strategies are a set of skills and principles that enable students to solve problems independently, along with the decision-making rules that guide their selection and use. Lenz et al. (1996) define strategies as follows: “An individual’s approach to a task is called a strategy when it includes how a person thinks and acts when planning, executing, and evaluating performance on a task and its outcomes” (p. 5). Thus, a strategy approach includes both cognitive and metacognitive elements. The Strategies Intervention Model (SIM), developed by researchers at the University of Kansas, is based on the theory that students with LD have information processing difficulties, are strategy deficient, and are inactive learners. That is, they do not create or use appropriate cognitive and metacognitive strategies spontaneously to process information, to cope with problems they encounter, or to learn new material (Alley & Deshler, 1979; Bender, 1995; Clark, 1993; Deshler, Schumaker, Lenz, & Ellis, 1984; Ellis, Deshler, & Schumaker, 1989; Shaw et al., 1995; Palincsar & Brown, 1987; Torgesen, 1988a, 1988b). Rabren and Darch (1996) found that instead of using text-based strategies to comprehend a story, students with LD used less effective approaches such as getting parents to help, relying on existing prior knowledge, and rote memorization. They concluded that students with learning disabilities were not passive learners; rather, they actively used poor strategies for coping with text which led to less effective comprehension of text. Reynolds (2000) attributed the lower reading comprehension of poor readers to insufficient automatization of both basic and higher level strategic comprehension processes, which he in turn linked to having limited attentional resources and allocating those attentional resources inefficiently. In contrast to tutorial, remedial, and compensatory approaches, the main focus of strategy intervention is not increased content knowledge, but rather, knowing how to learn. Through instruction in the Paraphrasing Strategy, a SIM strategy designed to improve reading comprehension, University of Kansas researchers found that students learn to acquire, retrieve,
approach provides valuable guidance to teachers working with adolescents with LD; however, more efficacy data is needed (Wong, 1993, 1996). Purpose We think that, for many adolescents with LD, reading failure is a crucial variable in achievement failure and, therefore, needs to be addressed in the junior high school setting. Students must be able to read to learn in order to have successful educational experiences. Greater academic success may lead to greater satisfaction at school; improved motivation, behavior, and attendance; and ultimately, better career opportunities. Also, as the majority of students who drop out do so between grades seven and ten, intervention at this stage of schooling may be key in keeping students in school. Strategy interventions show promise as an approach that may enhance reading comprehension along with other aspects of achievement. In a climate of scarce resources, educators will be more likely to adopt strategy interventions as an alternative to LA if there is evidence that they both efficient and effective. Therefore, more research is needed to evaluate strategy interventions as applied in actual junior high school classrooms. Our study compares the effectiveness of classroom-based strategy instruction and traditional learning assistance for low achieving eighth grade junior high school students with reading comprehension deficits. We hypothesized that a Strategies Intervention Model, employing the Paraphrasing Strategy based on the theoretical constructs and experimental work of Schumaker et al. (1984), would be more effective in ameliorating a reading disability than other traditional LA interventions. We theorized that if students learn techniques to cope with mainstream classroom text-based instruction, they will be more able to keep pace with their peers academically and will demonstrate fewer symptoms of discouragement and alienation from school. Our intent was to address what we believe is an important underlying cause of poor achievement for many students – reading disability.
Method This study used information from the databases of two junior secondary schools. Data reported here include student pre- and post reading comprehension subtest scores as assessed by the Stanford Diagnostic Reading Test (SDRT) (Karlsen, Madden, & Gardner, 1976), attendance records, and course grades (both academic and work habit grades). We compared the reading comprehension gains of eighth grade poor readers who were taught the SIM Paraphrasing Strategy in intact classrooms for a full school year, to the reading comprehension gains of poor readers receiving resource room based learning assistance utilizing tutorial, compensatory, and remedial methods (the customary approach to intervention used in the school district). Reading gains of a control group of students whose elementary teachers had not recommended intervention despite low reading scores, or whose parents had opted for them to receive no intervention, also were used in the comparison. In addition to comparing the reading gains of the three groups, we also compared them on measures of absenteeism, behavior, and subject failure. Reading scores of the SIM intervention group were supplemented by qualitative observations compiled by the classroom teacher. Participants The students in this study were from a school district in the province of British Columbia, Canada. The district has a student population of approximately 20,000 students. Two junior secondary schools, enrolling grades eight to ten, and matched on socioeconomic factors, school size ( n = 500), and student ethnicity, were used in the study. Both schools are located in the same small city, and serve mainly working-class neighbourhoods. In data collected by the provincial Ministry of Education, approximately ten percent of the students self-identified as Aboriginal, approximately ten percent were of East-Asian or Indo-Canadian extraction, and most of the rest were Caucasian. The number of students in eighth grade are approximately the same in each school, ranging from 150 to 180 students in any year.
SIM group in School 1. Normal SIM initial class size ranges from 24 to 27 students. Thirty-three SIM students for whom we had both pretest and posttest SDRT scores over the two years were included in the analysis. Despite some heterogeneity, all students had an identified reading and/or writing disability as measured by the SDRT and a curriculum-based narrative writing sample. Most students participating in the strategy intervention demonstrated word recognition and comprehension ability between the grade four to grade six level, indicating that, although limited, their basic reading vocabulary was adequate to learn the Paraphrasing Strategy. These students took the SIM class instead of French 8. The students’ remaining coursework consisted of the regular eighth grade curriculum in inclusive classrooms. The SIM classes were taught by two special education teachers in a team teaching model. One was the first author of this study and the other was the school LA teacher. Neither teacher knew that data from these students would be analysed for the purpose of this study at the time of the strategy intervention. Both teachers took equal responsibility for class management, lesson planning, and instruction. The first author took a principal role in managing and teaching the SIM curriculum, whereas the LA teacher had the major responsibility for the writing component of the curriculum. LA group in Schools 1 and 2. During the 1995-1996 school year, a group of students ( n =
- from the two schools were identified by the same criteria as used for identifying the SIM groups. In place of French 8, they participated in an LA class that combined tutorial, basic skills remediation, and compensatory approaches in a resource room model. The LA classes were taught by four LA teachers, two in each school. Class size in the LA room did not exceed nine students per class. The LA teachers used adapted materials and tests to help these adolescents attain competency in content-based materials and prepare for tests. This included provision of simplified vocabulary sheets of subject content information as a tutoring device. Tests, in social studies and science for example, were rewritten with adjusted vocabulary levels or provided orally. Students received remedial instruction in comprehension, spelling, grammar, punctuation,
and narrative and descriptive writing, using worksheets from publishing companies. They were given tutorial assistance with their homework assignments. Also, study skills, organizational skills, and test taking skills were addressed. No-LA group in Schools 1 and 2. The No-LA group consisted of students in the two schools who were not given LA during grade 8 (1995-1996; n = 31). Although these students demonstrated similar deficits in reading comprehension ability according to the SDRT as the SIM and LA groups, the No-LA students received no extra intervention. The membership of the No-LA group was determined post hoc solely through standardized reading test scores in the database; it included those eighth grade students in the regular program who obtained a SDRT pretest score between the 5th and the 37th percentile, but whose elementary teachers did not recommend them for LA, or who chose not to enroll in LA despite an LA recommendation. Their program consisted of inclusive regular class placements and curricula. Data excluded from analysis. In all three groups (SIM, LA, and No-LA), we included all data from all students for whom we had both pre- and posttest SDRT data, except for 1 SIM student who was diagnosed as intellectually challenged subsequent to placement in the SIM class, and 2 LA and 6 No-LA students in School 2 whose SDRT test scores were unreliable or invalid (test forms were spoiled or not complete). However, the study n s are smaller than the number of students who actually received SIM or LA intervention, or who were identified for the No-LA control group, because of absenteeism on the posttest date, mid-year transfers to other schools, and drop-out. In particular, there was a high rate of attrition from the No-LA group. Supplementary data from subsequent years. We report pre- and posttest scores schoolwide and for SIM students in School 1 for three years subsequent to the comparison data analysed in this study. Equivalent data were not available for LA and No-LA comparison groups in these years. Because of curricular and administrative changes^1 , these cohorts are not directly comparable to the SIM, LA, and No-LA groups that we analyzed so we have not incorporated
First, students' listening vocabulary is greater than their reading vocabulary. Therefore, while students were first learning to RAP information by looking for main ideas and details, their processing capacity was not being overloaded by the requirement of also decoding words (Chall et al., 1990). Second, students had to learn to make one- or two- keyword notations in the margins of the RAP sheet to remember details and what the paragraph was about. This encouraged one-word notetaking strategies using keywords. Outlining and notetaking, common requirements expected of students by high school teachers (Bianco & McCormick, 1989), was taught incidentally in this manner. Third, this method ensured that students paraphrased but did not copy the text since they could not memorize paragraphs verbatim. Cognitive interaction with text was encouraged by this means; the students could not remain passive in their relationship with the text. Students also had opportunities to discuss the content of paragraphs with a partner. The next five stories were shown to the students using an overhead projector, also an adaptation, so that students could follow along with the reading while the teacher drew their attention to paragraphs in the stories. This was intended to teach the students to transcribe ideas without copying or plagiarizing. Also, the students were taught paragraph structure using this process. Finally, the students could refer to projected paragraphs to check spelling. The second ten expository texts and stories were written at the students' grade level (grade 8) rather than at their below-grade reading level. These also were presented on the overhead projector. Reading materials for these 20 lessons included selections from Science Research Associates: Individualized Reading Skills Program (1970) - Orange (grade 4) and Blue (grade 8) Level reading series (now out of print), as well as texts from the students' core classes, other approved textbooks, and library books such as encyclopaedias and reference books. Students also were taught to RAP informational and documentary videos (see Appendices A and B for an example).^2 Generalization of the Paraphrasing Strategy learned in the intervention class was promoted by the Humanities teacher, who encouraged students to use RAP to make notes in English and Social Studies.
The writing strategy. A teacher-created remedial writing program based on the Paraphrasing Strategy model and procedures was used to teach paragraph and essay writing during the second 20 weeks of the school year. The writing strategy, PAR, is essentially the RAP strategy in reverse. It follows upon a brainstorming session of ideas on a topic of interest to the students or a curricular topic: Put all the ideas into categories; Ask yourself, what is the main idea and what are the details of the category; and Record the main idea and details in a paragraph using your own words. In this way, the techniques of the Paraphrasing Strategy were used as an intervention to improve both reading and writing (Mothus, 2001). The strategy is kept simple, as recommended by Ellis et al. (1989), as the less complex the approach, the better students will learn and use it (Anderson & Roit, 1993). Therefore, one basic strategy was taught and students were shown how to adapt it to promote reading comprehension, take notes, and compose essays (See example in Appendix C). Using both RAP and PAR, the students were taught to paraphrase the stories and curricular materials that they read, rewriting them into essays of varying lengths. They were encouraged to add their own prior knowledge to each paragraph, as well as to RAP increasingly complex and difficult articles from encyclopaedias and reference books and add the information from these sources to their essays. Essay structure was explicitly taught and practiced. Students were taught to write an introductory paragraph outlining what the essay was about and to compose an interesting topic sentence or thesis statement. The RAP content constituted the body of the essay. They were taught to add a concluding paragraph in which they summarized main points, expressed their opinion about the topic, and added a question of interest to which they would like to know the answer. These questions led to further research and essay writing using library books. The students eventually wrote an essay for their Social Studies or Science class using library information which was accepted for credit by their teacher.
0.19, p < .85 respectively) also revealed no significant differences in initial reading levels. Therefore these groups of students could be assumed to come from a similar population.
Insert Table 2 about here
Comparative Analysis of Interventions as Measured by the SDRT Gain (change) scores in reading comprehension achievement on the SDRT were used to compare the effectiveness of the SIM and LA interventions in improving reading comprehension, as contrasted with the No-LA control group. Gain scores were calculated by subtracting the raw pretest score from the raw posttest score of each student. Therefore, a positive gain score reflects an increase in reading comprehension^3. Table 3 shows the mean gain scores of each group of students in the study, as well as the range of gain scores, and the mean difference scores in grade equivalents. A 3X1 ANOVA (group by gain score) on the gain scores of the SIM, LA, and no-LA groups indicated that a significant difference existed in the three treatment groups' reading comprehension gain scores ( F (2, 95) = 3.86, p < .02). The SIM group was found to be different from the LA group ( t = 2.63, p < .01); they made significantly greater gains in reading comprehension. The SIM group was not significantly different from the No-LA group ( t = 1.42, p < .16). Nor were the LA and No-LA groups found to differ significantly in gain scores ( t = 1.40, p < .17). However, the No- LA gain score may be affected by the selective attrition of 10 No-LA students in School 1 who had especially low pretest scores (Mean Raw Score = 31, Mean GE = 4.9), thus inflating the average No-LA gain score, and limiting the conclusions we can draw about the no-intervention control group.
_______________
Insert Table 3 about here
To test the magnitude of the difference in gain scores for the SIM, LA, and No-LA groups, effect size of the raw SDRT pretest to posttest scores using Cohen’s d was calculated for the SIM group ( d = 1.07 ), the LA group ( d = 0.43), and the No-LA group ( d = 0.87). Reading comprehension improved by more than one standard deviation for the SIM group but not for the LA and No-LA groups. In comparison, effect sizes for the grade 8 school populations were: 1993-94 School 1 ( d = 0.28), 1994-95 School 1 ( d = 0.34), 1995-96 School 1 ( d = 0.33), and 1995-96 School 2 ( d = 0.43). Measures of Alienation and Academic Failure In order to examine our secondary hypotheses that reading disability is related to alienation from school, as indicated by absenteeism and negative behaviors, and to wider academic failure, we calculated rate of absenteeism, teacher-assigned behavior scores, and failure rate in school subjects for each of the three groups (see Table 4). A 3X1 ANOVA (group by days absent) of the SIM, LA, and No-LA groups showed a significant difference in absenteeism between the three groups ( F (2, 95) = 3.74, p < .03). Pairwise t-tests showed a significant difference in school attendance between the LA and No-LA groups, with the LA students absent more ( t = 2.65, p < .01). There were no significant differences in attendance between the SIM and LA groups ( t = 1.86, p < .07), or the SIM and No-LA groups ( t = 0.70, p < 0.49). The mean number of days missed by SIM and No-LA students ( M = 9.7 and 8.4 days respectively) is comparable with the grade 8 schoolwide average absenteeism calculated for School 1 in 1995-96 ( M = 9.3; SD = 8.2), whereas the mean number of days missed by LA students was 14.0.
