The Effects of Partner Learning During Spelling for Students with Severe Disabilities and Their Peers

Journal of Behavioral Education, Vol. 10, Nos. 2/3, 2000, pp. 107 121 The Effects of Partner Learning During Spelling for Students with Severe Disabilities and Their Peers John McDonnell, Ph.D., 1 Nadine Thorson, M. Ed., 2 Cathi Allen, M. Ed., 3 and Connie Mathot-Buckner, M. Ed. 4,5 A multiple baseline across subjects design was used to examine the effects of partner learning on the spelling performance, academic responding, and competing behavior of three students with severe disabilities and three of their classmates without disabilities. The students were enrolled in different general elementary classes. All students in these classes were assigned to partner learning triads by the general education teacher. One triad in each class included the student with severe disabilities and two classmates without disabilities. All students in the class received two, 20 min partner learning sessions each week. During partner learning, each member of the triad was asked to spell words, present words to be spelled, provide feedback to the speller, and check the spelling accuracy. These roles were rotated among the members of the triad after each trial. These roles were adapted as necessary to accommodate the academic and communication skills of the students with severe disabilities. The effects of partner learning on spelling accuracy were assessed through weekly spelling tests. The effects of partner learning on academic responding and competing behavior were assessed using the MS-CISSAR (Carta, Greenwood, Schulte, Arreaga-Mayer, Terry, 1988). Results of weekly spelling indicated that partner learning led to improved spelling accuracy for students with severe disabilities and did not negatively affect the spelling accuracy of their peers. Partner learning also led to improved rates of academic responding and reduced rates of competing behavior for 5 of 6 students. 1 Professor, Department of Special Education, University of Utah, Salt Lake City, Utah. 2 Program Specialist, School and Community Inclusion Program, Department of Special Education, University of Utah, Salt Lake City, Utah. 3 Program Specialist, School and Community Inclusion Program, Department of Special Education, University of Utah, Salt Lake City, Utah. 4 Program Specialist, School and Community Inclusion Program, Department of Special Education, University of Utah, Salt Lake City, Utah. 5 Correspondence should be directed to John McDonnell, University of Utah, Department of Special Education, 1705 E. Central Campus Dr., Rm 221, Salt Lake City, Utah 84112-9253. 107 1053-0819/00/0900-0107$18.00/0 C 2000 Human Sciences Press, Inc.

108 McDonnell, Thorson, Allen, and Mathot-Buckner The results are discussed in terms of their implications for practice and future research. KEY WORDS: partner learning; inclusion; severe disabilities; spelling. One challenge facing practitioners involved in inclusive educational programs for students with severe disabilities is designing instruction tailored to their unique educational needs and providing a sufficient number of instructional trials to promote efficient learning. If inclusive education is going to succeed, then practitioners need validated instructional practices based on these two key principles (McDonnell, 1998). In addition, these instructional practices must blend with the overall structure and organization of general education classes (Fantuzzo & Atkins, 1992). Peer mediated instruction has been suggested as one approach for addressing these critical needs (Giangreco & Putnam, 1991; Harper Meheady, & Mallet, 1994; McDonnell, 1998). This approach includes strategies such as Classwide Peer Tutoring (Delquardi, Greenwood, Whorton, Carta, & Hall, 1986, Numbered Heads Together (Kagan, 1985), and Partner Learning (McNeil, 1994), Although the procedural elements of these strategies differ slightly, they all are designed to allow students to serve as instructional agents for one another (Harper et al., 1994). Several research reviews and meta-analyses have concluded that peer mediated instructional strategies consistently lead to increased student achievement and opportunities to respond to instructional tasks for students without disabilities and those with mild disabilities (Cohen, Kulik & Kulik, 1982; Fowler, 1988; Good & Brophy, 1987; Krouse, Gerber, & Kaufman, 1981; Thousand & Villa, 1990; Topping, 1991). Studies also have shown that students without disabilities can serve as effective instructors for students with severe disabilities (Carr & Darcy, 1990; Charlop, Schreibman, & Tyron, 1983; Donder & Nietupski, 1981; Egel, Richman, & Koegel, 1981; Fenrick & McDonnell, 1980; Kamps, Locke, Delquadri, & Hall, 1989; Kohl & Stettner-Eaton, 1985; Kohl, Moses, & Stettner- Eaton, 1983; Lancioni, 1982). Students without disabilities appear to derive many social benefits as well from such programs including improved self-concept, increased acceptance of students with disabilities, and the development of friendships with peers with disabilities (Peck, Donaldson, & Pezzoli, 1990; Sasso & Rude, 1988). Although the effects of peer mediated instruction for students with severe disabilities have been positive, most studies focus on strategies in which a student with severe disabilities was the recipient of instruction provided by a peer without disabilities. The instructional roles of tutor and tutee have rarely been reversed; presumably, because students with severe disabilities do not have the academic and communication skills to provide instruction to their peers. Furthermore, in most of these studies peer tutoring was implemented in separate special education classes. One exception was a study conducted by Kamps, Barbetta, Leonard, and Delquadri

Partner Learning in General Education Classes 109 (1994) in which they examined the impact of the Classwide Peer Tutoring program on reading skills and social interactions of students with autism enrolled in general education classes. The peer tutoring program consisted of approximately 30 min of mutual instruction between the students with autism and peers without disabilities on reading fluency and comprehension. The results showed that the students and their peers improved both in reading fluency and comprehension skills. In addition, the tutoring program resulted in general increases in the total duration of social interactions between students during free time. The purpose of the present study was to extend the research base on peer mediated instruction as an approach for supporting students with severe disabilities in general education classes. We were also interested in examining a strategy in which students with severe disabilities would serve in both the tutor and tutee roles in general education classes. Consequently, the present study examines a partner learning process in which one student with severe disabilities and two peers without disabilities provided each other with spelling practice and review. The process was designed to ensure that students received instruction tailored to their own educational needs, implemented within the typical structure of students general education classes, and to maximize their interaction with their peers without disabilities. We evaluated partner learning in terms of its effects on the student s academic engagement and competing behavior, and spelling test performance. METHOD Settings and Participants Students participating in the study attended two elementary schools and were members of three different grade-level classes. Each school had participated in state and/or federal outreach projects administered by the senior author. The teachers and administrators in each building had made a strong commitment to including all children with disabilities in the general education program. In addition, staff in these schools had received training and technical assistance on strategies for supporting the participation of children with disabilities in the regular curriculum. All of the schools were in suburban and middle class neighborhoods. Participants in the study included three students with severe disabilities who were placed in general education classes and three students without disabilities enrolled in the same general education class. Special education teachers in each school that had agreed to take part in the study identify a pool of potential study participants with severe disabilities. This group included all students in the school who were (1) classified as having moderate to profound mental retardation or multiple disabilities and (2) were included in the general education class for most of their instruction. Table I provides basic demographic information about the students selected for participation in the study.

110 McDonnell, Thorson, Allen, and Mathot-Buckner Table I. Student Demographic Data Number of students Time in general without disabilities Student IQ Classification education class in class 1 37 Severe intellectual disabilities 3 hours 26 2 45 Severe intellectual disabilities 3 hours 24 3 40 Severe intellectual disabilities 4.5 hours 21 Student 1 participated in a typical fourth grade class for 3 hrs per day. She was pulled out of the general education class for speech, reading, math instruction for the remainder of the school day. In her grade level class, she completed the same work as the other students or was provided parallel instruction on the goals and objectives included in her Individualized Education Program (IEP). She was provided no additional curriculum modifications or paraprofessional support in her grade level class. Student 2 participated in a typical fifth grade class for 3 hrs per day. She also received instruction on her IEP goals and objectives in a resource program for the rest of the school day. While she participated to some extent in the activities of the general education class, the bulk of the instruction she received was done through parallel instruction or independent seat work. During time in her general education class she worked independently on individual goals. She received no paraprofessional support when she was in her grade level class. Student 3 was enrolled in a fifth grade class for 4.5 hrs per day. During the remainder of the school day she received instruction on IEP goals and objectives in a resource program. During the time she spent in her general education class she was provided parallel instruction or completed independent seat work tied to tied to her IEP goals and objectives. She received no paraprofessional support while in the grade level class. The general education teachers were asked to form heterogeneous partner learning triads that included students of differing spelling performance levels (i.e., above average, average, or below average spelling performance) and personal characteristics (i.e., gender, ethnicity). Once the triads were established, we identified the potential pool of students without disabilities in the class who matched the gender of the target student with severe disabilities and were performing at or above average in spelling. From this pool, one peer in a different triad from the student with severe disabilities was randomly selected for participation in the study. Dependent Measures The effectiveness of partner learning was assessed in terms of its impact on the level of academic responding and competing behaviors during spelling lessons, and weekly spelling test scores.

Partner Learning in General Education Classes 111 Academic Responding and Competing Behavior The Code for Instructional Structure and Academic Response Mainstream Version (MS-CISSAR) (Carta, Greenwood, Schulte, Arreaga-Mayer, Terry, 1988) was used to collect data on these two variables. This instrument has undergone extensive validation and has been used to address a number of research questions focusing on the academic performance of students who are at-risk of school failure (Greenwood, Carta, & Atwater, 1991). Greenwood and his colleagues (1993) describe academic responses as those student behaviors made directly in response to academic tasks, commands, or prompts. The specific student behaviors included in this subcategory are (a) writing, (b) manipulating objects that are relevant to completion of an academic task such as a computer, (c) reading aloud, (d) reading silently, and (e) engaging in verbal behaviors related to the academic task such as talking with a peer about subject matter as part of a collaborative learning group. Competing behaviors are defined as those responses that are unacceptable because they are against commonly accepted social conventions, classroom rules, or teacher directions. The behaviors in this subcategory of the MS-CISSARS include (a) aggression toward others, (b) disrupting the academic task, (c) talking with peers or the teacher about subjects not directly related to the academic task, (d) looking around the classroom and not attending to the academic task, (e) noncompliance with teacher directions or commands, (f) self-stimulatory behavior, and (g) self-abuse. Data on academic responding and competing behavior were summarized as the percent of observation intervals in which the behaviors occurred. Scores on Weekly Spelling Tests Each Friday students were given a spelling test on the words introduced during the week. For students with severe disabilities, these words were selected by their general and special education teachers consistent with their IEP goals and objectives. The spelling words for Students 1 and 2 were drawn directly from the general education spelling curriculum. The spelling words for Student 3 were drawn from the Edmark reading program. The number of words included in each student s weekly spelling lists ranged from 5 to 20 words, and was adjusted by their teachers to based on their previous weekly spelling performance. The weekly spelling lists for peers without disabilities was guided by the approved spelling curriculum for the school and were selected by the classroom teacher. A total of 20 words was introduced each week to peers throughout the study. For both students and peers, the performance on weekly spelling tests was summarized as the percent of words on the list spelled accurately.

112 McDonnell, Thorson, Allen, and Mathot-Buckner Experimental Design and Conditions A multiple baseline across subjects design was used to assess the effectiveness of partner learning on rates of academic responding and competing behavior, and spelling test scores (Tawney & Gast, 1984). The experimental conditions consisted of Baseline and Partner Learning. Students with severe disabilities (1 through 3) and their peers (4 through 6) were yoked for the initiation of the experimental conditions. That is, Students 1 and 4, Students 2 and 5, and Students 3 and 6 had the same number of baseline probe sessions and initiated partner learning at the same time. Baseline Participating students were observed weekly during the first 20 min of regularly scheduled spelling lessons. For 5 of 6 participating students, these lessons typically included whole class presentations by the general education teacher and independent practice on the targeted spelling words or other related language concepts. For Student 3, the typical instructional activities included independent practice on spelling or other academic skills. The special education teachers did not provide any direct instruction to the students with severe disabilities or the peers without disabilities during the spelling lessons. However, the teachers did observe the students with severe disabilities in the general education classroom twice weekly to monitor their behavior. Data for baseline probe sessions 4 and 5 for Student 3 are not reported because of unanticipated changes in regularly scheduled instructional activities. Partner Learning Partner learning was scheduled for all students in the class for twenty min twice per week during the regularly scheduled spelling lessons. Before each partner learning session, the general education teacher developed two word lists. The first list included the words that the student with severe disabilities would spell. The words on this list were selected jointly by the general and special education teacher using the procedures described above. The second list was for the students without disabilities in the class and included all of the spelling words introduced during the week. During an instructional trial, each student in a triad served in one of three partner learning roles. The first role was the word wizard. The word wizard wrote and verbally spelled a word. The second role was the word conjurer. The word conjurer, selected a word from the appropriate word list (the individually designed list for the student with severe disabilities or the general word list for students without disabilities), presented the word to the speller, and provided feedback to the speller. The word keeper held the word lists, checked the written and verbal

Partner Learning in General Education Classes 113 spelling of the word with the word conjurer, and showed the written word from the list to the speller as part of the standard error correction procedure if he or she made a mistake. For example, the word conjurer randomly selected a word from the spelling list and presented it (i.e., Spell.) to the word wizard. The word wizard wasgiven5stowrite the word and spell it verbally to the other members of the triad. Once the word wizard wrote and verbally spelled the word, the word conjurer and the word keeper jointly checked the spelling accuracy. The word conjurer then provided feedback to the speller (i.e., I agree or I disagree). If the word wizard did not spell the word correctly, the word keeper would show him/her the word on the list and the steps would be repeated until the word wizard spelled the word correctly. These roles were rotated after each instructional trial. Students received at least one trial on each word on the list during the partner learning sessions. Partner learning lessons were designed to accommodate the academic and communication skills of students with severe disabilities and to maximize their participation in the tutoring process. For example if a student could not read the word to be spelled, the word keeper would read and model the word for her and then she would present it to the word conjurer. Similarly, once the word conjurer had spelled the word the student and the word keeper would jointly check the accuracy. If necessary the word keeper would prompt the student to reinforce the word conjurer or to implement the error correction procedure. The roles of the general and special education teachers during partner learning included organizing the word lists for the triads, initiating and monitoring transition activities, monitoring the activities of the group during partner learning, monitoring the designated learning roles, providing feedback to triads about their performance, and providing assistance to triads when questions arose about the spelling tasks. Procedures Implementation Training for the Classroom Teachers and Students Two of the authors provided 1 hr of training to participating classroom teachers on the partner learning process. Training was done one-to-one with each participating teacher prior to the initiation of partner learning in their classrooms. During training, the elements of the partner learning process were discussed and examples how the students would implement the process were presented. Following the completion of baseline probe sessions for each student and peer, the classroom teacher provided instruction to all students in her class on the partner learning process. The teachers described the steps of the process and modeled the roles of the word conjurer, word wizard, and work keeper for students. The general education teacher and special education staff gave the triads additional training and assistance on the process and their roles during partner learning sessions as needed.

114 McDonnell, Thorson, Allen, and Mathot-Buckner Observation Procedures The observations of the student s level of academic responding and competing behavior were conducted weekly during one of the regularly scheduled partner learning sessions. At the beginning of the session, two of the researchers positioned themselves in discrete locations of the classroom but close enough to the triads that included each of the target students to record data. One researcher observed the triad that included the student with severe disabilities, and the second researcher observed the triad that included the peer without disabilities. Data were collected on a portable lap top computer using the MS-CISSAR software. Each 1 min interval is divided into three, 20 s segments in which the observer records the ecological context of the lesson, data on the teacher, and the students response in that order. The software prompts the observer to record data every 20 seconds through a low tone. At the end of each 20 s interval, the observer looks at the events, records the information on the lap top computer, and then rests briefly before the next interval begins. Reliability of Observation Data Prior to the initiation of the study, three of the authors learned to implement the MS-CISSAR using video tapes of children with low incidence disabilities enrolled in inclusive classrooms but who were not participating in the study. Training continued until all of the observers met a criterion of 90% agreement across observation intervals for 2 consecutive sessions. Interobserver agreement on the rates of academic responding and competing behavior by students was conducted during three sessions. One reliability probe was conducted with each student with severe disabilities. During the sessions, two of the researchers independently observed the target student. Interrater reliability was assessed for all behaviors coded during each observation interval. The level of agreement between the observers was calculated by dividing the number of agreements by the number of agreements plus disagreements, and multiplying by 100. Interrater reliability ranged from 90.2% to 93.6% with an average of 91.7% across the sessions. RESULTS Academic Responding and Competing Behavior Figure 1 presents the rates of academic responding and competing behavior for Students 1 through 3. The rates of academic responding for Student 1 increased from baseline (an average of 39% of observation intervals across probes) to Partner Learning (an average of 75% of observation intervals across probes). Student 1 s rates of competing behaviors also decreased slightly during Partner Learning

Partner Learning in General Education Classes 115 Fig. 1. Percent of observation intervals of academic responding and competing behavior for students with severe disabilities.

116 McDonnell, Thorson, Allen, and Mathot-Buckner (M = 9% of observation intervals across probes) from the baseline condition (M = 19% of observation intervals across probes). Partner Learning also led to substantial increases in academic responding for Student 2 during Partner Learning (M = 83% of observation intervals across probes) from baseline (M = 32% of observation intervals across probes). Student 2 also decreased their rates of competing behavior during Partner Learning (M = 11% of observation intervals across probes) from baseline sessions (M = 32% of observation intervals across probes). Although similar patterns were seen with Student 3, the changes from Baseline to Partner Learning in both academic responding (M = 62% of observation intervals across probes to 72%) and competing behavior (M = 19% of observation intervals across probes to 17%) was less appreciable. Figure 2 presents the rates of academic responding and competing behavior for students without disabilities. Student 4 had moderate increases in their rates of academic responding during Partner Learning (M = 58%) from the baseline condition (36%). However, their rate of competing behavior remained essentially the same during Baseline (M = 17%) and Partner Learning (M = 14%). There was a substantial amount of overlap of data points for Subject 5 between Baseline and Partner Learning for both academic responding (M = 53% to 70%) and competing behavior (M = 13% to 17%). Finally, Student 6 had substantial increases in their rates of academic responding during Partner Learning (M = 73%) over Baseline (M = 24%). Their rates of competing behavior decreased slightly during Partner Learning (M = 10%) from Baseline (M = 17%). Scores on Spelling Tests Table II presents the mean percentage correct on spelling tests for all students during Baseline and Partner Learning conditions. All of the students with severe disabilities (Students 1 3) increased the percentage of words they spelled correctly during Partner Learning from Baseline. The mean percent of words spelled correctly increased by 11%, 40%, and 62% for Students 1, 2, and 3 respectively. The mean spelling performance of peers remained very high and stable from the baseline condition to Partner Learning. Table II. Mean Post-Test Scores by Phase Baseline Partner Learning Student Mean Range Mean Range Student 1 85 80 90 96 65 100 Student 2 40 20 60 80 20 100 Student 3 28 15 35 90 80 100 Student 4 89 98 100 95 75 100 Student 5 96 95 100 96 95 100 Student 6 99 95 100 100

Partner Learning in General Education Classes 117 Fig. 2. Percent of observation intervals of academic responding and competing behavior for students without disabilities.

118 McDonnell, Thorson, Allen, and Mathot-Buckner DISCUSSION The results of this study suggest that the Partner Learning strategy was an effective approach for supporting students with severe disabilities during spelling lessons in general education classes. The data showed that partner learning was more effective than the traditional lecture instructional formats used with Students 1 and 2, and to the parallel instruction provided to Student 3 in increasing their spelling accuracy. Our findings are consistent with previous research showing that peer tutoring can be an effective strategy for promoting the acquisition of academic and other developmental skills by students with severe disabilities (Carr & Darcy, 1990; Charlop et al, 1983; Donder & Nietupski, 1981; Egel et al, 1981; Fenrick & McDonnell, 1980; Kamps et al, 1989; Kohl & Stettner-Eaton, 1985; Kohl et al, 1983; Lancioni, 1982). In addition, the partner learning strategy was superior to traditional lecture formats and parallel instruction in promoting these students rates of academic responding during spelling lessons. Interestingly, the rates of academic responding of students with severe disabilities during partner learning were comparable to their peers without disabilities. This finding is consistent with other studies that have found that when provided appropriate accommodations and supports that the academic engagement of students with severe disabilities in general education classes is similar to peers without disabilities (Hollowood, Salisbury, Rainforth, & Palombaro, 1995; Logan, Bakeman, & Keefe, 1997; McDonnell, Thorson, McQuivey, & Kiefer-O Donnell, 1997). An important advantage of the partner learning strategy for students with severe disabilities enrolled in general education classes is that it can be tailored to their individual needs. For example, the spelling words presented to the students in this study were based on their own IEP objectives and their previous performance in spelling instruction. Furthermore, the structure of partner learning allows teachers to use a wide range of adaptations and accommodations to meet the unique learning needs of students with severe disabilities. A second advantage is that partner learning allows students with severe disabilities to receive the type of direct instruction teaching format shown to be associated with their achievement (Snell & Brown, 2000; Westling & Fox, 2000). For example, this group of students received repeated opportunities to practice targeted spelling words during partner learning lessons. In addition, they were provided with clear and succinct response prompts, reinforcement, and error corrections. Partner learning did not result in substantial increases in spelling performance of the students without disabilities who participated in the study. However, this finding is not surprising given that the average test scores of these students during baseline ranged between 89 and 99% correct. Two of the 3 students without disabilities had higher rates of academic responding during partner learning when compared to the traditional lecture and independent seat work instruction they received during baseline sessions. These findings are important because many general and

Partner Learning in General Education Classes 119 special educators are concerned that the implementation of instructional strategies designed to accommodate the needs of students with severe disabilities in general education classes may negatively impact the quality of the education services receive by students without disabilities. These data suggest that such concerns may not be warranted. It is not clear from the study what impact partner learning had on the performance of students without disabilities who were in triads with students with severe disabilities. We decided to focus the study on the overall utility of partner learning in inclusive classrooms because previous research suggests that there are a number of positive educational and social benefits for individuals who tutor students with severe disabilities (Peck et al, 1990; Sasso & Rude, 1998). Our observations, and the anecdotal reports of the general education teachers, suggest that there were no negative effects of partner learning for these students. However, these are issues that should be addressed in future research studies. Although it was not a primary focus of the study, we also attempted to assess the perspectives of general education teachers about the relative utility of the partner learning strategy in meeting the spelling needs of students with severe disabilities and their peers. Using a post-hoc written survey, we asked the general education teachers participating in the study to rate the overall effectiveness of partner learning as an instructional strategy for all students on a five point Likert scale (Poor = 1; 5 = Excellent), two of the teachers rated the strategy as Excellent and one rated it as Very Good. One teacher commented that...the time spent in this process was much more effective than the worksheets from the spelling program. Another wrote that partner learning had led to an increase in confidence for low students it s beneficial for them. In addition, all three of the participating teachers indicated that they planned to continue to use partner learning their classrooms. Several weaknesses of this study should be noted. First, the small number of subjects restricts the external validity of the study and thus limits the generalizations that may be made beyond the study sample. Second, because of unforseen changes in the instructional activities of Student 3 we were unable to follow the convention of collecting baseline data points immediately before the implementation of the intervention (Tawney & Gast, 1994). This hampers our ability to demonstrate clear experimental control of partner learning for this student. Despite of these limitations, the present study suggests that partner learning can be an effective tool for supporting effective instruction for students with severe disabilities in general education classes. Equally important, this strategy was easily infused within the typical structure of the general education classes in which these students were enrolled. Additional research on partner learning is needed including (a) replication across other subject areas, (b) replication at the secondary level, (c) examination of the impact of partner learning on the social interactions between students with severe disabilities and their peers without disabilities,

120 McDonnell, Thorson, Allen, and Mathot-Buckner (d) the effects of partner learning on the acceptance of students with severe disabilities by their peers without disabilities, (e) long-term effects of partner learning on the academic performance of both students with and without disabilities, and (f) an in depth analysis of the perceived acceptability and utility of partner learning by general education teachers. ACKNOWLEDGMENTS This study was supported in part by the Division of Services for At-Risk Students, Utah State Office of Education. The opinions expressed herein do not necessarily reflect the position of the Utah State Office of Education. The authors would like to thank Andrea McDonnell for her feedback on earlier drafts of this manuscript. REFERENCES Carr, E. G., & Darcy, M. (1990). Setting generality of peer modeling in children with autism. Journal of Autism and Developmental Disorders, 20, 45 60. Carta, J. J., Greenwood, C. R., Schwartz, I. S., & Miller, P. A. (1990). Code for instructional structure and student academic response: Mainstream Version (MS-CISSAR). Kansas City, KS: Juniper Gardens Children s Project, Bureau of Child Research, University of Kansas. Charlop, M. H., Schreibman, L., & Tyron, A. (1983). Learning through observation: The effects of peer modeling on acquisition and generalization in autistic children. Journal of Abnormal Child Psychology, 11, 355 365. Cohen, P. A., Kulik, J. A., & Kulik, C. C. (1982). Educational outcomes of tutoring. American Educational Research Journal, 19, 237 248. Delquardi, J. C., Greenwood, C. R., Whorton, D., Carta, J. J., & Hall, R. V. (1986). Classwide peer tutoring. Exceptional Children, 52, 535 542. Donder, D., & Nietupski, J. (1981). Nonhandicapped adolescents teaching playground skills to their mentally retarded peers: Toward a less restrictive middle school environment. Education and Training of the Mentally Retarded, 16, 270 276. Egel, A. L., Richman, G. S., & Koegel, R. L. (1981). Normal peer models and autistic children s learning. Journal of Applied Behavior Analysis, 14, 3 12. Fantuzzo, J., & Atkins, M. (1992). Applied behavior analysis for educators: Teacher centered and classroom based. Journal of Applied Behavior Analysis, 25, 37 42. Fenrick, N., & McDonnell, J. (1980). Junior high school students as teachers of the severely retarded: Training and generalization. Education and Training of the Mentally Retarded, 15, 187 194. Fowler, S. A. (1988). The effects of peer-mediated interventions on establishing, maintaining, and generalizing children s behavior changes. In R. H. Horner, G. Dunlap, & R. L. Koegel (Eds.), Generalization and maintenance: Life-style changes in applied setting (pp. 143 171). Baltimore: Paul H. Brookes. Giangreco, M. F., & Putnam, J. (1991). Supporting the education of students with severe disabilities in regular education environments. In L. H. Meyer, C. A. Peck, & L. Brown (Eds)., Critical issues in the lives of people with severe disabilities (pp. 245 270). Baltimore: Paul H. Brookes. Good, T. L., & Brophy, J. E. (1987). Looking into classrooms (4th ed.). New York: Harper & Row. Greenwood, C. R., & Delquadri, J., (1991). Code for instructional structure and student academic response (CISSAR). In M. Hersen & A. S. Bellack (Eds.), Dictionary of behavioral assessment (pp. 120 122). New York: Pergamon. Harper, G. F., Maheady, L., & Mallette, B. (1994). The power of peer-mediated instruction: How and why it promotes academic success for all students. In J. S. Thousand, R. A. Villa, & A. I.

Partner Learning in General Education Classes 121 Nevin (Eds.), Creativity and collaborative learning: A practical guide to empowering students and teachers (pp. 229 242). Baltimore: Paul H. Brookes. Kagan, S. (1985). Cooperative Learning. Riverside, CA: University of California-Riverside. Kamps, D. M., Barbetta, P. M., Leonard, B. R., & Delquardri, J. (1994). Classwide peer tutoring: An integration strategy to improve reading skills and promote peer interactions among students with autism and general education peers. Journal of Applied Behavior Analysis, 27, 49 62. Kamps, D. M., Locke. P., Delquadri, J., & Hall, R. V. (1989). Increasing academic skills of students with autism using fifth grade peers as tutors. Education and Treatment of Children, 12, 38 51. Kohl, F. L., Moses, L. G., & Stettner-Eaton, B. A. (1983). The results of teaching fifth and sixth graders to instructional trainers with students who are severe handicapped. Journal of the Association for Persons with Severe Handicaps, 8, 32 40. Kohl, F. L., & Stettner-Eaton, B. A. (1985). Fourth graders as trainers of cafeteria skills to severe handicapped students. Education and Training of the Mental Retarded, 20, 60 68. Krouse, J., Gerber, M., & Kaufman, J. (1981). Peer tutoring: Procedures, promises, and unresolved issues. Exceptional Education Quarterly, 1, 107 115. Lancioni, G. E. (1982). Normal children as tutors to teach social responses to withdrawn mentally retarded school mates: Training, maintenance, and generalization. Journal of Applied Behavior Analysis, 15, 17 40. Logan, K. R., Bakeman, R., & Keefe, E. B. (1997). Effects of instructional variables on engaged behavior of students with disabilities in general education classrooms. Exceptional Children, 63, 481 498. McDonnell, J. (1998). Instruction for students with severe disabilities in general education settings. Education and Training in Mental Retardation and Developmental Disabilities, 33, 199 215. McDonnell, J., Thorson, N., McQuivey, C., & Kiefer-O Donnell, R. (1997). The academic engaged time of students with low incidence disabilities in general education classes. Mental Retardation, 35, 18 26. McNeil, M. (1994). Creating powerful partnerships through partner learning. In J. S. Thousand, R. A. Villa, & A. I. Nevin (Eds.), Creativity and collaborative learning: A practical guide to empowering students and teachers (pp. 243 261). Baltimore: Paul H. Brookes. Peck, C. A., Donaldson, J., & Pezzoli, M. (1990). Some benefits nonhandicapped adolescents perceive for themselves form their social relationships with peers who have severe handicaps. The Journal of The Association for Persons with Severe Handicaps, 15, 241 249 Sasso, G., & Rude, H. (1988). The social effects of integration on nonhandicapped children. Education and Training in Mental Retardation, 23, 18 23. Snell, M. E., & Brown, F. (2000). Instruction of students with severe disabilities (5th Edition). Upper Saddle River, NJ: Merrill. Tawney, J. W., & Gast, D. L. (1984). Single Subject Research in Special Education. New York: MacMillan Publishing Company. Thousand, J. & Villa, R. (1990). Sharing expertise and responsibilities through teaching teams. In W. Stainback & S. Stainback (Eds.), Support networks for inclusive schooling: Interdependent integration (pp. 151 166). Baltimore: Paul H. Brookes. Topping, K. (1991). Achieving more with less: Raising reading standards via parental involvement and peer tutoring. Support for Learning, 6, 112 115. Westling, D. L., & Fox, L. (2000). Teaching students with severe disabilities (Second Edition). Upper Saddle River, NJ: Merrill.