The effects of Classwide Peer Tutoring on the reading achievement of urban middle school students.
This study investigated the effects of Classwide Peer Tutoring (CWPT) on the reading skills of urban middle-school students using novels as the curriculum. Teacher-led instruction was compared with CWPT and CWPT plus a lottery contingency for appropriate on-task and tutoring behaviors. Three sixth-grade general education reading classes under the direction of one teacher participated. Data were collected on all students from weekly written tests of vocabulary and comprehension. Additional oral reading rate and academic engagement data were collected from three "low-achieving" target students. Overall, results demonstrated improved performance on weekly tests under CWPT conditions compared with teacher-led instruction. CWPT plus lottery resulted in further increases. Data also revealed differences in the types of academic responses made during teacher-led instruction and CWPT and increases in oral reading rates for two target students. These findings suggest that CWPT, particularly CWPT plus lottery, can improve the reading skills of urban middle school students.
Data from the National Assessment of Educational Progress (NAEP) indicate that more than one-third of our nation's fourth-grade students are performing below basic achievement levels in reading. An alarming 26% of eighth-graders, and 23% of twelfth-graders are in that same category. Further, Caucasian students consistently outperform their African-American, Hispanic, and other minority peers (National Center for Educational Statistics, 1999). These figures are disturbing, especially when one considers the fact that students who are already at the greatest risk for academic failure (i.e., economically disadvantaged students) typically receive the least amount of instruction and practice as they progress through school (Hall, Delquadri, Greenwood, & Thurston, 1982; Kamps et al., 1989). Mathes, Fuchs, Fuchs, Henley and Sanders (1994) reported, however, that when low-achieving, at-risk students do receive effective reading instruction, they tend to experience greater success in their remaining school years.
Fortunately, effective strategies exist to improve students' reading skills. One such strategy, peer tutoring, is an instructional method in which students teach one another in school settings under the direction of a teacher (Wagner, 1982). The general term "peer tutoring" is often used to describe instruction delivered by same-age students (peer tutoring) and older, more advanced students (cross-age tutoring). While this instructional strategy is not new (cf. Lancaster, 1805), experimental validation of its efficacy has only been of research interest for about the last 35 years.
Peer Tutoring and Reading
Research assessing the effects of peer tutoring on "reading" has focused on many skills. Because standardized tests of reading proficiency often include assessment of skills such as sight word reading, vocabulary, oral reading fluency, and comprehension, efforts to improve students' reading skills generally focus on these areas. Studies directed at assessing the effects of peer tutoring on sight word reading have generally obtained positive results (e.g., Barbetta, Miller, Peters, Heron, & Cochran, 1991; Butler, 1999; Chiang, Thorpe, & Darch, 1980; Cochran, Feng, Cartledge, & Hamilton, 1993; Heron, Heward, Cooke, & Hill, 1983; Giesecke, Cartledge, & Gardner, 1993).
Other studies, which focused specifically on oral reading fluency, have produced positive results when comparing peer tutoring and teacher-led instruction. Jenkins, Mayhall, Peschka, and Jenkins (1974) found that peer tutoring produced better oral reading rates on timed readings then when students participated in small group instruction. Salend and Nowak (1988) used a multiple baseline design to demonstrate the effects of a peer tutoring strategy on the oral reading fluency of three second- and third-grade students. After 15 days of tutoring, error rates decreased from 24.5, 25.6, and 18.8 during baseline to 8.2, 10.4, and 6.5 during tutoring for each of the three students respectively. The marked decrease in student error rates over such a short period of time suggests that tutoring improved reading fluency for these students.
While the research to this point is encouraging, each of these studies only examined small component pieces of the skill called "reading" (see also Sindelar, 1982; Vaughn et al., 2000), and although reading fluency is considered to be an important component reading skill, "reading words well does not constitute good reading" (Stoddard, Valcante, Sindelar, O'Shea, & Algozzine, 1993, p. 54). Fortunately, peer tutoring programs incorporating two or more of the component skills (i.e., vocabulary, comprehension, oral reading fluency) also show promise for improving student reading skills (e.g., Ezell, Kohler, & Strain, 1994; Fuchs, Fuchs, Mathes, & Simmons, 1997; Kamps, Barbetta, Leonard, & Delquadri, 1994; Kamps, Leonard, Potucek, & Garrison-Harrell, 1995; Mathes & Fuchs, 1994; Simmons, Fuchs, Fuchs, Hodge, & Mathes, 1994; Simmons, Fuchs, Fuchs, Mathes, & Hodge, 1995; Stoddard et al., 1993; Trovato & Bucher, 1980).
Another program, Classwide Peer Tutoring (CWPT), was originally developed by researchers in conjunction with a third-grade teacher who wanted to improve the spelling achievement of her students (Delquadri, Greenwood, Stretton, & Hall, 1983). The premise of the program was to allow students more "opportunities to respond" to academic material. That is, by using a reciprocal tutoring format, the CWPT program allowed all students in the classroom opportunities to practice spelling in a situation that provided constant monitoring, error correction, and consequences for correct responding. Certainly, these circumstances would be optimal for improving reading skills of children of varying ability levels, as well as those with disabilities (see Greenwood, Delquadri, & Hall, 1984; Hall et al., 1982; Kamps et al., 1994).
While the data suggest that peer tutoring can improve students' reading skills, the majority of the research documenting the effectiveness of peer tutoring on reading has been conducted with elementary school children. What is noticeably absent from this literature is the examination of the efficacy of peer tutoring on the reading achievement of students beyond primary grades. The data from NAEP show clearly that, although a smaller percentage of students are categorized as "below basic achievement" levels after fourth-grade, nearly one-third of eighth-graders and one-quarter of high school seniors demonstrate insufficient reading skills. This indicates that the problem persists beyond elementary school and, therefore, requires attention at the secondary school level.
Peer Tutoring in Secondary Schools
Although peer tutoring research has been conducted at the secondary school level (i.e., middle school and high school), the literature, especially with regard to reading, is much more sparse than that available for peer tutoring with elementary school students. Perhaps this is due to the nature of the school day at the secondary school level, which is characterized by short class periods and movement between classes, as opposed to spending the day with a single teacher for academic instruction. Another potential problem is that, when entering middle and high school, unless students are identified as having disabilities, reading is often not provided as a separate area of instruction (Fuchs, Fuchs, & Kazdan, 1999; Harris, Marchaud-Martella, Martella, 2000). Arguably, these circumstances are less than optimal for implementation of an intervention such as peer tutoring.
Despite these potential barriers, available research examining the effectiveness of peer tutoring with older students has produced positive results. Harris et al. (2000) combined peer tutoring with the Corrective Reading Program (Engelmann, Hanner, & Johnson, 1989) and repeated readings to examine the effects on reading achievement of at-risk high school students. Dependent measures included pre-post performance on vocabulary and comprehension subtests of a standardized reading test, oral reading rates, number of repeated readings, and lessons completed. Eighty-eight students participated in 50 min sessions five days per week over the course of the school year. At posttest, students' median grade equivalent on both the vocabulary and comprehension subtests had improved by at least two grade levels, an increase of a full grade level above what would normally be expected in one school year. Further, mean oral reading rates improved by at least 97 words per minute from pre- to posttest, with collateral improvements in accuracy.
Fuchs et al. (1999) compared the effects of peer tutoring (Peer-Assisted Learning Strategies; PALS) and teacher-led instruction on the reading skills of high school students enrolled in 15 special education or remedial reading classes. Comparison classes used traditional teacher-led instruction, which included no peer-mediated instruction. PALS classes supplemented reading instruction with five tutoring sessions biweekly. The total amount of instruction was held constant across the two conditions. Components of the PALS program included partner reading, paragraph shrinking (i.e., main idea and summary), prediction, and earning points for correct responses. A lottery system was added to augment the points and team competition component usually used with elementary school students in the PALS program. Each day, both students in the highest scoring dyad posted their names on a chart in the classroom and entered a monthly "drawing" that resulted in one student winning $10. Results indicated that students who participated in PALS outperformed comparison students on standardized measures of reading comprehension, however, generalized reading fluency was similar for the two groups (see also Mastropieri et al., 2001). These findings suggest that peer tutoring can effectively improve students' reading comprehension; effects on fluency, however, warrant further research. Recognizing that the novelty of peer mediation was insufficient to maintain student motivation resulted in a potentially cost-effective reinforcement system that also merits further investigation.
The research with elementary aged students indicates that peer tutoring can be more effective than teacher-led instruction at improving reading skills; thus, one might assume that extensions to secondary school would produce similar changes in students' reading. However, preliminary research, although encouraging, has produced mixed results, making it difficult to ascertain the efficacy of peer tutoring as a strategy for improving the reading skills of secondary school students. These deficits necessitate that more research be conducted to determine the viability of peer tutoring as a reading intervention for older students.
The data from NAEP indicate that unless something is done to dramatically improve the reading skills of our nation's students, nearly one-quarter of them will graduate from high school with insufficient reading skills. The prospects grow even dimmer for students who begin school at high risk for academic failure due to challenges such as low socioeconomic status. Because reading is a fundamental skill, students who are inefficient readers suffer cumulative deficits across many content areas that require reading as a prerequisite skill (e.g., social studies, science). Furthermore, data indicate that insufficient reading skills increase the likelihood of dropping out of school (Greenwood et al., 1989), out-of-wedlock pregnancies, dependence on welfare, and arrests (Smith & Lincoln, 1988). Thus, finding effective strategies to improve students' reading skills is imperative.
While the body of research documenting the effects of peer tutoring on reading skills for elementary school children is abundant, there exists little evidence to support the use of peer tutoring for improving reading with secondary school students. This is problematic, because the crisis of insufficient reading instruction and practice persists beyond elementary school. Therefore, studies involving more secondary school students are needed to determine the extent to which peer tutoring can affect students' reading. Further, research directed at reading with different content (e.g., novels, other texts) and simultaneous intervention on multiple skills (i.e., vocabulary, comprehension, oral reading fluency) is needed. Finally, research to determine the feasibility of implementing peer tutoring with secondary school students deserves attention.
Purpose and Research Questions
Given what is known about the efficacy of peer tutoring, and the gaps in the research base, the purpose of this study was to advance what is known about the use of CWPT in middle school reading and the contribution of an additional reinforcement contingency on student academic performance.
The first objective of this study was to examine the effects of CWPT on the reading achievement of sixth-grade middle school students in an urban, economically disadvantaged area. The following questions were addressed:
1. What are the differential effects of teacher-led instruction, CWPT, and CWPT plus lottery on weekly reading and vocabulary tests as measured by class means?
2. What are the differential effects of teacher-led instruction, CWPT, and CWPT plus lottery on weekly reading vocabulary and comprehension tests for individual "low-achieving" students?
3. Does student engagement change as a function of method of instruction?
4. Do teacher-led instruction and CWPT differentially affect the oral reading rates of "low-achieving" students?
A second objective of this study was to assess the acceptability of the procedures (Schwartz & Baer, 1991; Wolf, 1978). The following questions were addressed through questionnaires:
1. What is the level of student satisfaction with the CWPT procedures?
2. Is CWPT acceptable to the teacher in terms of workload, effects on student academic outcomes, and overall management of students?
Setting and Participants
Seventy-one sixth-grade students enrolled in three general education reading classes in an urban middle school in the Midwest participated in this study. The school qualified as a Title I school, due to the fact that approximately 75% of the student population was eligible for free or reduced lunch. Ethnicity data for the entire school indicated that the student population was comprised of 69% African-Americans, 24% Caucasians, 5% Asian Pacific Islanders, and 2% Hispanics. Class A (n=25) contained 12 females and 13 males comprised of 80% African-American students, 16% Caucasian students, and 4% Hispanic students. Class B (n=20) contained 5 females and 15 males, with 80% African-Americans and 20% Caucasians. Eleven females and 15 males comprised Class C (n=26), with 61% African-Americans, 27% Caucasians, and 12% Asian Pacific Islanders. The same teacher, who had been teaching in the district for four years and at this school for three of those years, taught all three classes. All sessions were conducted in the classroom in which the teacher regularly taught. Originally, the teacher used CWPT in only one class (Class A), but later began implementation in Classes B and C after viewing the data from the first seven tests with Class A.
Target students. In order to assess the effects of different methods of instruction on the weekly test performance, academic engagement and oral reading rates of individual lower achieving students, the experimenter asked the teacher to identify three or four students in Class A who were either receiving special education services or were "low-achieving", from whom these measures could be obtained. Because no students in the participating class were receiving special education services at that time, the teacher chose four "low-achieving" students, based on the results of a recently administered Informal Reading Inventory (IRI; Burns & Roe, 1999), used to assess student reading skills school-wide.
Thomas was a 12-year-old African-American male who performed at a second-grade level on the IRI prior to the start of the study. Cathy, an 11-year 3-month-old African-American female, and Michelle, an 11-year 3-month-old Caucasian female, both scored in the second- to third-grade reading level on the IRI. These three students served as the target students from whom individual measures of academic achievement (i.e., weekly test scores), academic engagement, and oral reading rates were obtained.
Experimental Design and Conditions
A reversal or ABAB design (Baer, Wolf, & Risley, 1968), modified to accommodate a third "C" condition, was used to assess the effects of teacher-led instruction (A), CWPT (B), and CWPT plus lottery (C) on student academic achievement as measured by weekly tests. The sequence of manipulation of the independent variables for Class A was ABABCACACA, and ABCACACA for Classes B and C. See Table 1 for novels, content coverage, and experimental conditions.
Teacher-led instruction (A). The teacher met with each class four times weekly for 43 min (Monday, Tuesday, Thursday, and Friday) and one time weekly (Wednesday) for 25 min. The curriculum was composed mainly of reading from novels chosen by the teacher interspersed with other reading related activities. The three novels read tionary and writing definitions in a notebook, students taking turns reading aloud from the novel as the teacher called on them, following along silently in the novel while listening to selections on an audiotape, and completing activities designed to coincide with novels from Novel Units[c] student packets or other teacher or publisher developed materials (e.g., character mapping, summarizing story events in journals, comprehension worksheets). Introduction of vocabulary occurred on the first day of new material coverage. Generally, one full class period (usually Monday) was devoted to the vocabulary activity and the other four days were spent reading from the novel (either taking turns or listening to the audiotape) and completing worksheets corresponding to the selection. Prior to participating in the study, assessment of student progress was based solely on completion of these activities. Usually, students worked on activities either independently or in whole group fashion; however, on rare occasion they assisted each other in looking up words in the dictionary.
Classwide peer tutoring (CWPT) (B). The tutoring procedure used during reading was similar to that described by Delquadri, Greenwood, Whorton, Carta, and Hall (1986) and Duvall, Delquadri, Elliott, and Hall (1992). The procedures implemented are summarized below:
1. Tutoring occurred for the entire 43 min class period on Mondays, Tuesdays, and Thursdays. Oral reading procedures, described below, were in place on these days. On Wednesdays and Fridays, CWPT sessions lasted approximately 15 min and consisted of vocabulary procedures only (described below).
2. The experimenter instructed the teacher to pair students based on "ability" and compatibility. In most cases, the teacher paired students of approximately equal "ability" to work together, however, students judged by the teacher to be low-achieving were paired with higher achieving students who might provide a better model of correct performance. Once paired, dyads were instructed to sit together and then assigned to one of two competing teams. Student dyads remained the same for the duration of each novel, however, the teacher reassigned dyads to different teams more frequently so that students always had the chance to be on the winning team. For instance, dyad 1 might be on team A for the first couple of weeks but then at the start of the third week be reassigned to team B. Dyads were only reassigned at the start of a new week, and not all dyads were reassigned at the same time in order to keep the teams as heterogeneous as possible.
3. Folders containing materials for tutoring (e.g., point sheets, vocabulary flashcards, "help" card, comprehension questions, and crayons for marking point sheet) were pre-assembled and located in plastic tubs in the classroom. Daily, the teacher wrote the materials needed for that class period on a dry erase board outside the classroom door (e.g., tutoring folder, a copy of the novel Freak the Mighty), so that students could obtain materials upon entry to the classroom and be prepared to begin when the bell rang.
4. During the oral reading procedures, each student in the dyad performed each role (i.e., tutor and tutee) for approximately 13-15 minutes. The teacher began each session by reading aloud the entire selection, typically six to eight pages in length, to be covered that day. Students were instructed to follow along silently in their books. After the teacher finished reading, she told the students the stopping point for the first tutee, wrote the page numbers on the chalkboard, and set the timer for 8 to 10 min. The tutee read the selection and earned two points for each correctly read sentence. Any errors (i.e., mispronounced words, omissions, additional words, substitutions, mixed word order, or hesitations longer than 4 s) resulted in correction by the tutor. A correction consisted of the tutor saying, "stop", pointing out the error, requiring the tutee to repeat the word correctly and then reread the sentence correctly. Sentences read correctly after an error correction resulted in one point. If students reached the end of the passage before the timer sounded, they returned to the beginning of the selection and continued reading. When the timer sounded, the teacher reset the timer for 5 min and instructed students to begin asking comprehension questions. The teacher provided scripted comprehension questions for the first 7 sessions for class A and the first two sessions for classes B and C. After that point, students used only "question starter" prompt cards (i.e., Who, What, Where, When, How printed on them) for asking comprehension questions. When using the scripted comprehension questions, tutors asked tutees questions about the story and compared the responses against an answer key. Later, when only the prompt cards were used, students determined correctness of answers without the aid of an answer key, based on their own comprehension of the reading selection. Correct responses resulted in two points. If the tutee could not answer a question, the tutor directed the tutee to the page in the book on which the answer could be found. Questions answered correctly after prompts to refer back to the book resulted in one point. If the students answered all questions before the timer sounded, they were instructed to go back through the list or make up questions of their own from the reading until the timer sounded. At the end of the 5 min question session, students switched roles and resumed the oral reading procedure. The second tutee began reading at the point where the first tutee had been instructed to stop, and continued reading to the end of the passage read by the teacher (For the first novel read, The Watson's Go To Birmingham-1963, both students read the same selection orally).
5. The teacher introduced new vocabulary words (12-15 words weekly) via the overhead projector on Wednesdays, immediately preceding vocabulary tutoring. The teacher read each word and definition while students echoed chorally. As soon as vocabulary introduction was complete, the students took the vocabulary flashcards out of their folders and the teacher set the timer for 5-6 min. The tutor held up one vocabulary flashcard at a time and required the tutee to read the word and attempt to give the definition. If the tutee was unable to read the word, the tutor prompted the correct response or requested assistance from the teacher by raising a "help" card. If after reading the word the tutee could not give the correct definition, the tutor presented the correct definition, either orally or visually, and the tutee said the word and definition aloud three times. Initially correct responses earned two points, while responses requiring error correction earned one point. Students cycled through the vocabulary flashcards as many times as possible before the timer sounded. When the timer sounded, students switched roles and followed the same procedure.
6. During CWPT, the teacher circulated among the students monitoring and assisting when needed and awarding bonus points for good "tutoring behavior" (e.g., praising and giving points to tutees, following procedures, giving correct responses).
7. After completion of CWPT, students totaled their points and reported them to the student assistant appointed by the teacher. The assistant recorded and totaled the points on a CWPT chart at the back of the classroom. Due to district concern about promoting competition between students, points were posted but announcement of the "winning team" was downplayed.
CWPT plus lottery (C). All procedures described above for CWPT remained intact during the CWPT plus lottery condition. The lottery differed from CWPT in the following way:
When the teacher circulated among the students, she passed out small "tickets" (e.g., slips of paper, or "raffle tickets") to students for good tutoring and on-task behavior. These tickets replaced the bonus points used in the CWPT condition. At the end of each tutoring session, the lottery tickets, on which the students had written their names, were collected and placed in a bucket. On Fridays, the teacher or experimenter drew between 5 and 8 names from the tickets collected that week. Additionally, occasional surprise drawings were done throughout the study by the experimenter. Students whose names were drawn were allowed to choose an item from a box of assorted inexpensive novelties (e.g., pop-a-point pencils, gel pens, high bounce balls, potato chips, candy bars). Students could only "win" once during each drawing, and a name pulled more than once was returned to the bucket. Each Monday, the ticket bucket was emptied so that everyone began the week with no tickets.
CWPT training. The teacher in this study was trained during a single two-hour staff in-service training at the school. Teacher training included verbal explanation, demonstration by the project staff, and role-playing. Additionally, the experimenter provided the teacher with a copy of the CWPT training manual, Together We Can! (Greenwood, Delquadri, & Carta, 1997).
Students were trained during the first post baseline week (i.e., five class sessions) by the experimenter and another project staff person. Training procedures consisted of (a) explanation of the tutoring procedures, (b) modeling of each component (i.e., oral reading, error correction, point recording, flashcard presentation) by the experimenter and project staff person, (c) practice by the teacher and a student in front of the class, and (d) students role-playing while project staff and teacher circulated among the students praising correct tutoring behavior and assisting or correcting when necessary. Additional training for the lottery involved explaining to the students behaviors that would earn "tickets" and making explicit that the more tickets earned, the greater the chances of "winning" the drawing on Fridays.
By the completion of student training the teacher was conducting sessions independently. Monitoring of procedures was ongoing during tutoring sessions, with prompts given as necessary. A single, classwide refresher training was done by the teacher after the students returned to school from the Winter Break.
Fidelity of Treatment
Fidelity of the oral reading procedure was verified by the experimenter or other project staff member using a 38-item fidelity checklist, which included questions regarding materials (e.g., Team point charts posted, All tutors have score sheets), teacher procedures (e.g., Teacher moves among students during tutoring, Teacher awards bonus points for tutoring correctly), students procedures (e.g., Tutee reads orally, Tutor scores properly for correct sentences read), and comprehension (e.g., Tutee answers questions, Tutor corrects responses quickly). Fidelity checklists were completed twice during CWPT in Classes A and B, and once in Class C while the teacher implemented procedures independent of the experimenter. Because the experimenter was frequently in the classroom observing and collecting various measures, the teacher was unaware when fidelity was being assessed. Following fidelity observations, the experimenter provided a copy of the checklist and feedback based on the checklist to the teacher. Percentage of procedures properly implemented during oral reading CWPT was 88% and 82% for class A, 86% and 92% for class B, and 93% for class C. CWPT procedures did not occur during teacher-led instruction.
Class mean scores from weekly posttests and periodic pretests were collected in each of the three classrooms, and additional measures of individual test scores, oral reading rates and academic engagement were collected for three target students.
Weekly tests. Weekly tests, developed by the experimenter and edited and approved by the classroom teacher, were administered on Fridays immediately following vocabulary CWPT. Pretests contained the same questions as the weekly posttests with the order of questions and answer choices changed. Administration of pretests occurred on eight occasions for Class A and on three occasions for Classes B and C following completion of the weekly test on Fridays. Tests were composed of 20 questions, consisting of five vocabulary word and definition matching, five vocabulary fill-in-the-blank questions, seven short answer questions, and three multiple-mark questions (i.e., similar to multiple choice, but students choose all correct answers). The teacher scored weekly tests, with spot checks done by the experimenter, and the experimenter scored all pretests at the teacher's request. Once during baseline and once during tutoring, the teacher and experimenter scored the tests independently and another project staff member figured point-by-point reliability on each of the 20 test items. An agreement was recorded if both the experimenter and the teacher marked an item the same way (i.e., correct or incorrect). Reliability equaled the number of agreements divided by the number of disagreements multiplied by 100. Reliability of student test scores equaled 96% (range 78%-100%).
Academic engagement. The behavior of the three target students was measured twice during teacher-led instruction (A) and twice during CWPT conditions (i.e., CWPT [B] or CWPT plus lottery [C]). Observations were conducted using the Code for Instructional Structure and Student Academic Response (CISSAR: Greenwood, Carta, Kamps, & Delquadri, 1997), a momentary time-sampling computerized data collection program. Categories observed included activity (e.g., reading, math), task (e.g., worksheet, readers, pen and paper), structure (i.e., whole group, small group, individual), teacher behavior (e.g., teaching, no response, disapproval), and student response (e.g., reading aloud, raising hand) (see Greenwood, Delquadri, & Hall, 1984).
Observers were trained via a combination of written definitions and classroom practice sessions to a criterion of 90% overall reliability and 85% across each individual category (e.g., student response) before data collection began. Interobserver agreement was calculated for 10% of the CISSAR observations by dividing the number of agreements by the number of agreements plus disagreements and multiplying by 100. Overall reliability was 91.73% agreement, with a range of 80% to 100% across each individual category. Reliability for student response was 89.33%
Oral reading rates. To determine the effects of oral reading procedures on student reading fluency, oral reading rate data were obtained from each target student. At least one check during teacher-led instruction (A) and one check during CWPT (B or C) occurred for each novel. Two-min timed readings were done with each target student over material that had been covered in either baseline (A) or CWPT (B or C). After completion of the timed reading, the assessor asked five comprehension questions about the material read. Errors made during oral reading, as well as correct and incorrect responses to comprehension questions were recorded. Oral reading rates (i.e., correct number of words read) were calculated by taking the total number of words read, subtracting the number of errors made, and dividing by two.
Social validity. To ascertain consumer satisfaction with CWPT, the students and teacher completed questionnaires at the end of the school year. Student surveys were completed anonymously in the absence of the experimenter. Fourteen of the 16-items on the student questionnaire used a 4-point Likert scale to solicit information from students regarding activities (e.g., How did you like doing the vocabulary lists?), peer interaction (e.g., Do you think your classmates are friendlier to you when you do peer tutoring?), and learning (e.g., How much did peer tutoring help you with reading?). It also contained two open-ended questions that required comment on what students liked least and most about CWPT. The teacher questionnaire, which was also completed in the absence of the experimenter, solicited responses to 24 items using a 5-point Likert scale. Again, two additional questions required written responses to questions of least and most preferred aspects of CWPT.
Differential effects of instructional method on student quiz performance. As shown in Figure 1, student performance on weekly tests improved during CWPT and CWPT plus lottery compared with teacher-led instruction. Classes B and C began participation in the study at point 8, thus there are no data for the first seven tests for these two classes. Notations at 1, 8, and 15 (8 and 15 only for Classes B and C) on the graph indicate when novels changed. The notation at 13 indicates when scripted comprehension questions were withdrawn. Also at test 13, vocabulary words that had been written on the chalkboard earlier in the week were inadvertently left on the chalkboard during the test. The line on each graph running horizontally from the 60% point on the abscissa indicates the criterion for "passing" grades established by the classroom teacher.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Weekly data for Class A are displayed in the top panel of Figure 1. Mean percent correct for Class A across conditions equaled 32% during pretest, 63% during teacher-led instruction, 75% during CWPT, and 81% during CWPT plus lottery. Weekly data for Class B are shown in the middle panel of Figure 1. Mean percent correct for Class B equaled 32%, 58%, 61%, and 74% during pretest, teacher-led, CWPT, and CWPT plus lottery, respectively. Similar results were obtained for Class C, shown in the lower panel of Figure 1. Class means by condition for Class C were 32%, 62%, 71%, and 82% for pretest, teacher-led, CWPT, and CWPT plus lottery, respectively.
Differential effects of instructional method on target students' quiz performance. Percentage correct on weekly tests for each target student is shown in Figure 2. As with the class mean figures, notations at 1, 8, and 15 indicate when novels changed, and the notation at 13 shows when scripted comprehension questions were withdrawn. Data for post test 16 are absent for all three target students because they participated in a field trip that day and missed the test.
For Thomas (top panel), mean percentage correct on weekly tests was 16%, 33%, 61%, and 72% across pretest, baseline, CWPT, and CWPT plus lottery, respectively. With the exception test 9, percentage correct on weekly tests during CWPT and CWPT plus lottery exceeded performance on tests during teacher-led instruction. Pre- to posttest gains for Thomas were minimal under teacher-led instruction, but improved substantially during CWPT and CWPT plus lottery.
Mean percent correct on weekly tests for Cathy (middle panel) was 28%, 65%, 69%, and 87% during pretest, baseline, CWPT, and CWPT plus lottery, respectively. Responding during baseline was unstable for this student, and percent correct on weekly tests during CWPT only slightly exceeded baseline performance. The CWPT plus lottery intervention produced more stable, correct responding on weekly tests when compared with both teacher-led instruction and CWPT.
Mean percent correct on weekly tests for Michelle (lower panel) equaled 27%, 59%, 72%, and 82% during pretest, baseline, CWPT, and CWPT plus lottery, respectively. Performance on weekly tests during baseline and CWPT was variable; however, implementation of CWPT plus lottery resulted in more stable, consistently higher correct responding on weekly tests. Although the discrepancy between means may seem small, it should be noted that these numbers, when associated with letter grades typically given by teachers, indicate substantial changes in student achievement. For instance, for Thomas, a mean of 33% during teacher-led instruction equals "F" in letter grades, but 61% equals "D", a difference of one letter grade (every 10 percentage points equals one letter grade from 60% up). Further, the mean score of 72% obtained by Thomas during CWPT plus lottery equals "C" in letter grades, a difference of two letter grades from baseline performance. Letter grade equivalents for Cathy were equal to "D" for both baseline and CWPT and "B" for CWPT plus lottery. Similarly, letter grade equivalents for Michelle equaled "F" during baseline, "C" during CWPT, and "B" during CWPT plus lottery.
Changes in student engagement as a function of instructional method. Figure 3 displays the distribution of student responding across the four most frequently recorded codes: writing, reading aloud, reading silently, and attending to task. These four codes comprised between 76% and 93% of all responding during both baseline and tutoring. The most substantial change in academic responding for all three target students was the marked increase in reading aloud during tutoring when compared with baseline levels. As a direct function of increased reading aloud, reading silently decreased for Thomas and Cathy and remained the same for Michelle.
Overall, academic responding remained fairly consistent across baseline and CWPT respectively for both Thomas (67%, 70%) and Cathy (76%, 75%), but increased for Michelle (54%, 66%). Task management increased during CWPT for Thomas (15%, 28%) and Cathy (16%, 23%), but decreased for Michelle (42%, 30%). Further, competing responses decreased for all three students, with the most substantial difference evidenced by data obtained from Thomas (17%, 1%).
Oral reading rates of "low-achieving" students. As displayed in Figure 4, the number of words read correctly per minute by Thomas during baseline ranged from 43 to 59.5, with corresponding ranges from 61-78.5 during CWPT.
The data for Cathy were more variable and less conclusive than those obtained from Thomas. The number of correctly read words per minute ranged from 86 to 120 during baseline and from 87.5 to 132 during CWPT. Similarly, Michelle's number of correctly read words per minute ranged from 79 to 106.5 during baseline, and 87 to 114 during CWPT.
Error rates for all three students decreased from baseline to CWPT. Thomas's error words per minute ranged from 3.5 to 6 (86%-93% accuracy) during baseline and 2 to 5 (93%-97% accuracy) during CWPT. Error rates for Cathy ranged from 1.5 to 5 (95%-99% accuracy) during baseline and 0 to 3 (97%-100% accuracy) during tutoring. Similarly, Michelle's error rates ranged from 1 to 6.5 (92%-99% accuracy) during baseline and .5 to 2 (98%-99% accuracy) during CWPT. Data from student responses to comprehension questions revealed no differences across conditions, with ranges of 60% to 100%.
[FIGURE 4 OMITTED]
Student satisfaction with the CWPT procedures. Sixty one of the seventy one participating students responded to the consumer satisfaction questionnaire. Overall, students rated CWPT positively. When asked, "How much do you like CWPT?" the majority of students (59%) responded "quite a bit" (n=22) or "very much" (n=14). Forty-five students responded that CWPT helped them learn "quite a bit" (n=26) or "very much" (n=19), and most of the students (n=49) replied that they would like to do peer tutoring next year. With regard to individual components of the program, the majority of students responded that they enjoyed doing flashcards (n=40), earning points (n=51), and reading from the book (n=34). Fifty-six students replied that they liked getting prizes in the lottery "a lot". Replies to whether or not students liked being the tutee were the least positive, with a total of 32 student responses of "not at all" or "very little."
Teacher satisfaction with CWPT. Satisfaction data obtained from the teacher indicated that the teacher "strongly agreed" that CWPT was beneficial to students of all ability levels, easy to implement, and an enhancement to traditional teaching approaches. Only two questions, both of which related to student cooperative and supportive behaviors, received responses of "3" (i.e., not sure), and all other questions were rated "agree" or "strongly agree" by the teacher. In response to "What did you like most?" the teacher wrote, "I didn't have to pull teeth to get them to read. My discipline problems were greatly decreased because the students were always on task. Easy program to implement."
Findings from this study of CWPT in urban middle-school reading classes were quite favorable. The data from weekly tests indicate that, overall, method of instruction was functionally related to student performance on tests. In all three classrooms, CWPT was more effective than teacher-led instruction for improving student test scores, including scores of lower ability students. In addition, implementation of CWPT plus lottery resulted in further improvement in student test scores. Means for Classes A and B improved from 62% and 58% respectively during baseline to 81% and 74% respectively during CWPT plus lottery. Mean performance for Class C improved from 58% during baseline to 74% during CWPT plus lottery. In terms of letter grades, Class A as a whole improved from "D" performance during baseline to "B" levels under CWPT plus lottery conditions. Class B's scores increased from "F" levels under baseline conditions to "C" when CWPT plus lottery was in effect, a gain of two letter grades. Because Class B convened during the first hour of the day (7:50 A.M.), students often arrived late to class or missed it altogether if they missed the bus. According to the teacher, this may have contributed to Class B's lower performance on weekly tests. Class C's performance, as a whole, was similar to that of Class A, with "D" performance under baseline condition that improved to "B" when CWPT plus lottery was in place. As mentioned previously, letter grade improvements for each of the three target students were also substantial.
These findings are similar to results obtained by Trovato and Bucher (1980) who found that elementary school students in a reading peer tutoring program outperformed students receiving traditional instruction on several dependent measures, but students who received peer tutoring plus back-up reinforcement made the greatest gains. In the present study, students earned points for correct responding in both the CWPT and the CWPT plus lottery conditions, however, changing the teacher bonus points to a lottery system in which students could "win" small tangible items had a more profound effect on academic outcomes. The lottery system seemed to solve some of the motivation challenges presented by the prohibition of student competition imposed by the district. Although the lottery was not directly tied to academic performance on weekly tests, it apparently had enough strength as a motivation system to improve that performance. These findings contribute to the research base on the effectiveness of CWPT with middle school reading and the addition of a cost-effective individual contingency (i.e., lottery), however, further research with more participants should be conducted to confirm these results. This extends prior research (e.g., Fuchs et al., 1999; Fuchs et al., 2001) that found improvements in student academic performance when tutoring plus a lottery was tied directly to desired terminal performance.
Other findings, consistent with those of CWPT in elementary settings (Greenwood et al., 1984), suggest that student academic engagement is functionally related to method of instruction. For example, Thomas's competing behaviors (e.g., looking around, talking inappropriately) decreased markedly from a baseline occurrence of 17% of intervals to just 1% of intervals during CWPT. This translates to a decrease from approximately 7.5 min off-task during baseline, compared with less than 1 min during CWPT. Hypothetically, this student gained 5.5 min of reading instruction per day during CWPT, potentially translating to more than 16 min per week and almost 12 hours per school year for just one class. If these changes could be affected across all academic periods of the day (e.g., reading, social studies, science, language, math), the potential exists to increase the time this student spends making active academic responses an additional 60 hours per school year (cf. Greenwood, 1991).
Although the target students' academic responding was high across both baseline and CWPT, the types of responses differed. The difference between reading silently and reading aloud from baseline to CWPT was the most substantial change noted in student academic responding. Reading aloud for all three students ranged from 0% to 4% (0 to approximately 2 min) of intervals during baseline, with increases ranging from 27% to 29% (approximately 12 min) of intervals during CWPT. Reading silently ranged from 31% to 62% (approximately 13 to 27 min) of intervals during baseline, and 28% to 40% (approximately 12 to 17 min) during CWPT. Although both of these behaviors are desirable in that they are academic responses, it can be argued that these are qualitatively different responses. First, when coding student behavior "reading silently", the observer assumes that the student is actually reading silently. This is a judgment call based on other observable behaviors, such as eyes oriented to the book and eye movement indicative of scanning. Reading aloud leaves no such margin for error; when a student is reading aloud, the actual behavior is observable. In addition, reading aloud is a well-documented strategy for improving fluency, while reading silently has not been shown to do so (Fuchs, Fuchs, Hosp, & Jenkins, 2001; National Reading Panel, 2000)
Second, reading aloud allows opportunities for error correction and consequences for correct responding that are unavailable when the behavior of interest is not directly observable. As noted by Skinner (1989), students' academic achievement is improved when opportunities for immediate consequences are available. It is likely that the changes in academic responding, particularly reading aloud, produced by CWPT were functionally related to students' reading achievement, as evidenced by improved test scores under CWPT conditions. These findings support the research (e.g., Hall et al., 1982; Kamps et al., 1995) claiming that students gain academically when they have opportunities to respond actively to academic material.
Another benefit of CWPT was improved reading fluency for two target students. For Thomas and Michelle, correctly read words per minute and accuracy increased during CWPT conditions. For Thomas, this was true for all three novels. This is perhaps due to the fact that reading rates for this student were so low to begin with, and the practice provided during CWPT was enough to significantly improve fluency. For Michelle, actual correct words per minute were higher for two of the three novels (i.e., The Watson's Go To Birmingham-1963 and Freak the Mighty) and accuracy was higher for all three novels during CWPT. While oral reading rates did not reach the benchmark rate of 125 words per minute (Fuchs, Fuchs, Hamlett, Walz, & Germann, 1993), these data support previous research findings that increased opportunities to respond to academic material result in improved performance. It is probable that the change in student responding from reading silently to reading aloud contributed to the change in oral reading rates, because students had more opportunities for practice, error correction, and consequences for correct responding when reading aloud. It is unclear why oral reading rates for Cathy were so variable. It could be speculated that because this student's reading rates were approaching the benchmark rates for sixth-grade students, somewhat of a ceiling effect occurred.
Social validation measures indicated that both the students and the teacher were satisfied with the results of CWPT. Students enjoyed participating and often complained that they did not like baseline weeks because they "did not do as well on the tests." Further, the teacher was reluctant to remain in baseline conditions for longer than one week at a time after the initial baseline period, because student scores were so greatly improved with tutoring. These anecdotal reports are consistent with the data obtained from questionnaires, suggesting that CWPT was preferred over the teacher's traditional instruction by both the students and the teacher.
Although performance on the IRI was not a dependent measure in this study, it should be noted that all three target students' scores on the IRI improved beyond the normally expected one grade level per school year gains. Thomas's performance on the IRI improved from a second-grade level on the fall administration to a fourth-grade level on the spring administration. Cathy's score improved from a second-to third-grade performance to sixth-grade performance. Similarly, Michelle's score increased from second- to third-grade performance, to fourth-grade performance. While it is not clear that these improvements resulted from the students' participation in CWPT, it is unlikely that students who were performing well below their grade level would have made such substantial gains without intervention.
Limitations and Suggestions for Future Research
Because there is still sparse evidence of the efficacy of CWPT with older students, much more research is warranted on peer tutoring as an intervention to improve combinations of skills, such as vocabulary, oral reading fluency, and comprehension.
While the data demonstrate that CWPT conditions were more effective than teacher-led instruction for improving students' reading achievement, there are limitations of this study. First, because the lottery was implemented to enhance a particular aspect of the CWPT package (i.e., teacher bonus points), examining its effects separately was not possible. While it is unlikely that the lottery system combined with teacher-led instruction would have produced results like those obtained from CWPT with lottery, further systematic investigation is necessary to make such a statement definitively. Another limitation is the failure to return to CWPT once CWPT plus lottery was implemented. As is typical in applied research, the teacher and experimenter chose to keep the "best procedure" in place, in order to provide the best intervention to the students. Pragmatically, this was the proper course to pursue, however, in terms of demonstrating experimental control further comparison of CWPT to CWPT plus lottery would have been more desirable. While it is unlikely, it is certainly possible that the same outcome may have resulted had the CWPT plus lottery condition been run prior to the CWPT condition.
The limited number of academic engagement observations leaves some room for speculation about the adequacy of those data. In addition, these data were collected only twice during baseline and twice over the entire CWPT condition (both B and C). Even though there were clear differences across conditions, there is potential confounding of the data that merits investigation in future research.
Although the CWPT conditions overall improved student test performance, long-term maintenance of skills was not assessed. Thus, conclusions about the long-term effects of this intervention cannot be drawn. Research by Greenwood and colleagues (1989; 1993; 1995) demonstrated long-term gains for students who received intensive CWPT intervention in the primary grades (i.e., first- through fourth-grade), however no such evidence exists to suggest that the same long-term gains are possible when intervening with older students who already suffer cumulative deficits. Such evidence would substantially improve the case for structured tutoring programs as effective reading interventions.
Another shortcoming of this study was the failure to assess generalization. It would have been interesting to examine if students showed collateral gains in other classes as a result of the increased reading practice, or if changes in oral reading fluency would have resulted in other novels or content areas. Future research should focus on specifically programming for generalization of such skills so that students can be maximally effective when working independently (see Hock, Deschler, & Schumaker, 2000).
Other issues that may have affected outcomes include material difficulty and content coverage. It is possible that some vocabulary words, tests, or reading passages may have been more difficult than others. Further, the number of pages read varied from week to week, although the number of vocabulary words, comprehension questions, and test questions remained constant. These problems are, perhaps, inherent when teachers must attempt to deliver instruction to students performing at varying instructional levels. Thus, no attempt was made to match materials to students' instructional levels. However, because this study was conducted using the curriculum already in place, the results provide a more realistic view of how this program might fare in other classrooms with similar curricula.
The use of a comparison group may have provided additional information about how the participating students measured up against peers attending the same school but not receiving the intervention. Although the reversal design provided a clear picture of student performance across conditions, a comparison group might have allowed for more conclusive statements about the effects of CWPT on reading outcomes.
Finally, research specifically addressing adoption of effective programs needs to be conducted. Although programs exist which effectively improve students' reading skills, the fact remains that many students continue to fall below basic proficiency levels in reading. Without determining how to get educators to implement effective programs, the best research results will go unused. Ultimately, student achievement will not be affected if research continues to ignore how to get educators to adopt effective strategies (Abbott, Walton, Tapia, & Greenwood, 1999).
Summary and Conclusions
In summary, the results of this study contributed positive findings regarding the use of CWPT as a method for improving the reading skills of middle school students. Further, the addition of a lottery contingency to CWPT for on-task and good tutoring behavior extended previous research findings that demonstrated improvements in student achievement when the lottery was directly related to academic performance (Fuchs et al., 1999).
In addition, findings indicated that CWPT dramatically increased students' reading aloud. Both reading aloud and reading silently are desirable in that they are academic responses; however, reading aloud appears to be functionally related to improved test performance and oral reading rates. This is likely a result of increased practice, opportunity for error correction, and increased consequences for correct responding that are not available when reading silently. These strategies are generally necessary to ensure the success of low performing students (Hall et al., 1982; Skinner, 1984).
Finally, consumer satisfaction data indicated that students and teacher found the program acceptable with respect to workload and academic effects. Further, anecdotal data suggested that students preferred CWPT conditions to teacher-led instruction with respect to outcomes on their weekly tests. The teacher also seemed to prefer CWPT to her own traditional instruction as evidenced by written comments on the satisfaction questionnaire and her reluctance to keep the students in baseline conditions after seeing improvement in test scores during CWPT.
Future research should continue to address the efficacy of peer tutoring as a reading intervention for secondary school students. Further, investigation of the lottery component, separate from tutoring, is also warranted. It is possible, though unlikely, that the addition of a simple contingency such as the lottery could result in improvements in academic performance commensurate with those obtained in this study. Investigation of long-term effects and generalization of skills learned while tutoring would provide valuable information about the feasibility of tutoring as a reading intervention. Finally, research focused on adoption of effective strategies is needed so that the knowledge gained from empirical research can be put to practical use.
Abbott, M., Walton, C., Tapia, Y., & Greenwood, C. (1999). Research to practice: A "blueprint" for closing the gap in local schools. Exceptional Children, 65, 339-352.
Baer, D. M., Wolf, M. M., & Risley, T. R. (1968). Some current dimensions of applied behavior analysis. Journal of Applied Behavior Analysis, 1, 91-97.
Barbetta, P. M., Miller, A. D., Peters, M. T., Heron, T. E., & Cochran, L. L. (1991). Tugmate: A cross-age tutoring program to teach sight vocabulary. Education and Treatment of Children, 14, 19-37.
Burns, P., & Roe, B. (1999). Informal reading inventory. Boston, MA: Houghton-Mifflin.
Butler, F. M. (1999). Reading partners: Students can help each other learn to read! Education and Treatment of Children, 22, 415-426.
Byers, B. (1977). The pinballs. New York: Harper Collins.
Chiang, B., Thorpe, H. W., & Darch, C. B. (1980). Effects of cross-age tutoring on word-recognition performance of learning disabled students. Learning Disability Quarterly, 3(4), 11-19.
Cochran, L., Feng, H., Cartledge, G., & Hamilton, S. (1993). The effects of cross-age tutoring on the academic achievement, social behaviors, and self-perceptions of low-achieving African-American males with behavioral disorders. Behavioral Disorders, 18, 292-302.
Curtis, C. P. (1995). The Watson's go to Birmingham-1963. New York: Dell Laurel-Leaf.
Delquadri, J., Greenwood, C. R., Stretton, K., & Hall, R. V. (1983). The peer tutoring spelling game: A classroom procedure for increasing opportunity to respond and spelling performance. Education and Treatment of Children, 6, 225-239.
Delquadri, J., Greenwood, C. R., Whorton, D., Carta, J. J., and Hall, R. V. (1986). Classwide peer tutoring. Exceptional Children, 52, 535-542.
Duvall, S. F., Delquadri, J. C., Elliott, M., & Hall, R. V. (1992). Parent-tutoring procedures: Experimental analysis and validation of generalization in oral reading across passages, settings, and time. Journal of Behavioral Education, 2, 281-303.
Engelmann, S., Hanner, S., & Johnson, G. (1989). Corrective reading series guide. New York: Macmillan/McGraw-Hill.
Ezell, H. K., Kohler, F. W., & Strain, P. S. (1994). A program description and evaluation of academic peer tutoring for reading skills of children with special needs. Education and Treatment of Children, 17, 52-67.
Fuchs, L. S., Fuchs, D., Hamlett, C. L., Walz, L., & Germann, G. (1993). Formative evaluation of academic progress: How much growth can we expect? School Psychology Review, 22, 27-48.
Fuchs, L. S., Fuchs, D., Hosp, M. K., & Jenkins, J. R. (2001). Oral reading fluency as an indicator of reading competence: A theoretical, empirical, and historical analysis. Scientific Studies of Reading, 5, 239-256.
Fuchs, L. S., Fuchs, D., & Kazdan, S. (1999). Effects of peer-assisted learning strategies on high school students with serious reading problems. Remedial and Special Education, 20, 309-318.
Fuchs, D., Fuchs, L. S., Mathes, P. G., & Simmons, D. C. (1997). Peer-assisted learning strategies: Making classrooms more responsive to diversity. American Educational Research Journal, 34, 174-206.
Fuchs, D., Fuchs, L. S., Thompson, A., Svenson, E., Yen, L., Otaiba, S. A., Yang, N., McMaster, K. N., Prentice, K., Kazdan, S., & Saenz, L. (2001). Peer-assisted learning strategies in reading: Extensions for kindergarten, first grade, and high school. Remedial and Special Education, 22, 15-21.
Giesecke, D., Cartledge, G., & Gardner III, R. (1993). Low-achieving students as successful cross-age tutors. Preventing School Failure, 37(3), 34-43.
Greenwood, C. R. (1991). Classwide peer tutoring: Longitudinal effects on the reading, language, and mathematics achievement of at-risk students. Reading, Writing, and Learning Disabilities, 7, 105-123.
Greenwood, C. R., Carta, J. J., Kamps, D., & Delquadri, J. (1997). Ecobehavioral assessment systems software (EBASS): Practioner's manual (Version 3.0). Kansas City, KS: Juniper Gardens Children's Project.
Greenwood, C. R., & Delquadri, J. (1995). Classwide peer tutoring and the prevention of school failure. Preventing School Failure, 39(4), 21-25.
Greenwood, C. R., Delquadri, J. C., & Carta, J. J. (1997). Together we can! Longmont, CO: Sopris-West.
Greenwood, C. R., Delquadri, J. C., & Hall, R. V. (1984). Opportunity to respond and student academic performance. In W. L. Heward, T.E. Heron, D. S. Hill, & J. Trap-Porter (Eds.), Focus on behavior analysis in education (pp. 58-88). Columbus, OH: Merrill.
Greenwood, C. R., Delquadri, J. C., & Hall, R. V. (1989). Longitudinal effects of classwide peer tutoring. Journal of Educational Psychology, 81, 371-383.
Greenwood, C. R., Terry, B., Utley, C. A., Montagna, D., & Walker, D. (1993). Achievement, placement, and services: Middle school benefits of classwide peer tutoring used at the elementary school. School Psychology Review, 22, 497-516.
Hall, R. V., Delquadri, J., Greenwood, C. R., & Thurston, L. (1982). The importance of opportunity to respond in children's academic success. In E. B. Edgar, N. C. Haring, J. R. Jenkins, & C. G. Pious (Eds.), Mentally handicapped children: Education and training (pp. 107-140). Baltimore, MD: University Park Press.
Harris, R. E., Marchaund-Martella, N., & Martella, R. C. (2000). Effects of a peer-delivered corrective reading program. Journal of Behavioral Education, 10, 21-36.
Heron, T. E., Heward, W. L., Cooke, N. L., & Hill, D. S. (1983). Evaluation of a classwide peer tutoring system: First graders teach each other sight words. Education and Treatment of Children, 6, 137-152.
Hock, M. F., Deschler, D. D., & Schumaker, J. B. (2000). Strategic tutoring. Lawrence, KS: Edge Enterprises.
Jenkins, J. R., Mayhall, W. F., Peschka, C. M., & Jenkins, L. M. (1974). Comparing small group and tutorial instruction in resource rooms. Exceptional Children, 40, 245-250.
Kamps, D. M., Barbetta, P. M., Leonard, B. R., & Delquadri, 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, 49-61.
Kamps, D. M., Carta, J. J., Delquadri, J. C., Arreaga-Mayer, C., Terry, B., & Greenwood, C. R. (1989). School-based research and intervention. Education and Treatment of Children, 12, 359-390.
Kamps, D. M., Leonard, B., Potucek, J., & Garrison-Harrell, L. (1995). Cooperative learning groups in reading: An integration strategy for students with autism and general education peers. Behavioral Disorders, 21, 89-109.
Lancaster, J. (1805). Improvements in education as it respects the Industrious Classes. London: Darton & Harvey.
Mastropieri, M. A., Scruggs, T., Mohler, L., Beranek, M., Spencer, V., Boon, R. T., & Talbott, E. (2001). Can middle school students with serious reading difficulties help each other and learn anything? Learning Disabilities Research & Practice, 16(1), 18-27.
Mathes, P. G., Fuchs, D., Fuchs, L. S., Henley, A. M., & Sanders, A. (1994). Increasing strategic reading practice with Peabody classwide peer tutoring. Learning Disabilities Research and Practice, 9 (1), 44-48.
National Center for Educational Statistics. U. S. Department of Education. (1999). National Assessment of Educational Progress (NAEP) 1998 reading report card for the nation (NCES 1999-500). Washington, DC: Author.
National Reading Panel (2000). Report of the subgroups: National Reading Panel. Washington, DC: National Institute of Child Health and Development.
Novel Units (2001-2002). Grades K-8 2001-2002 catalog. San Antonio, TX: ECS Learning.
Philbrick, R. (1993). Freak the mighty. New York: Scholastic.
Salend, S. J., & Nowak, M. R. (1988). Effects of peer-previewing on LD students' oral reading skills. Learning Disability Quarterly, 11, 47-53.
Schwartz, I. S., & Baer, D. M. (1991). Social validity assessments: Is current practice state of the art? Journal of Applied Behavior Analysis, 24, 189-204.
Simmons, D. C., Fuchs, D., Fuchs, L. S., Hodge, J. P., & Mathes, P. G. (1994). Importance of instructional complexity and role reciprocity to classwide peer tutoring. Learning Disabilities Research and Practice, 9(4), 203-212.
Simmons, D. C., Fuchs, L. S., Fuchs, D., Mathes, P., & Hodge, J. P. (1995). Effects of explicit teaching and peer tutoring on the reading achievement of learning-disabled and low-performing students in regular classrooms. The Elementary School Journal, 95, 387-408.
Sindelar, P. T. (1982). The effects of cross-aged tutoring on the comprehension skills of remedial reading students. Journal of Special Education, 16(2), 199-206.
Skinner, B. F. (1984). The shame of American education. American Psychologist, 39, 947-954.
Skinner, B. F. (1989). The school of the future. In B. F. Skinner (Ed.), Recent issues in the analysis of behavior (pp. 85-96). Columbus, OH: Merrill.
Smith, R. C., & Lincoln, C. A. (1988). America's shame, America's hope, twelve million youth at risk. Chapel Hill, NC: MDC, Inc.
Stoddard, K., Valcante, G., Sindelar, P., O'Shea, L., & Algozzine, B. (1993). Increasing reading rate and comprehension: The effects of repeated readings, sentence segmentation, and intonation training. Reading Research and Instruction, 34(4), 53-65.
Trovato, J., & Bucher, B. (1980). Peer tutoring with or without home-based reinforcement for reading remediation. Journal of Applied Behavior Analysis, 13, 129-141.
Vaughn, S., Chard, D., Bryant, D. P., Coleman, M., Tyler, B., Linan-Thompson, S., & Kouzekanani, K. (2000). Fluency and comprehension interventions for third-grade students. Remedial and Special Education, 21, 325-335.
Wagner, L. (1982). Peer teaching. Westport, CT: Greenwood Press.
Wolf, M. M. (1978). Social validity: The case for subjective measurement. How applied behavior analysis is finding its heart. Journal of Applied Behavior Analysis, 11, 203-214.
Mary Baldwin Veerkamp and Debra M. Kamps
Juniper Gardens Children's Project
Kansas City, Kansas Public Schools
Correspondence to Debra Kamps, Ph.D., Senior Scientist, Juniper Gardens Children's Project, University of Kansas, 650 Minnesota Avenue, 2nd Floor, Kansas City, KS 66101; e-mail: email@example.com.
Table 1 Novels, Content Coverage, and Experimental Conditions The Watsons Go To Birmingham--1963 Test Chapters Pages Condition 1 5 64-74 A 2 6 75-85 A 3 7 86-99 A 4 8 100-120 B 5 10,11 138-161 B 6 12,13 162-179 B 7 14,15 180-206 B The Pinballs 8 1-6 3-31 A 9 7-9 33-47 B 10 10-12 49-62 C 11 13-18 63-97 A 12 19-21 99-111 C 13 22-24 113-127 C 14 25,26 129-136 C Freak the Mighty 15 6-8 28-47 A 16 9-12 48-79 C 17 13-16 80-107 C 18 17-20 108-134 C 19 21-25 135-160 A Note. Condition: A = Baseline (Teacher-led instruction); B = Classwide Peer Tutoring (CWPT); C = CWPT plus Lottery Percentage of Intervals Thomas Cathy Michelle Baseline CWPT Baseline CWPT Baseline CWPT Writing 15 6 11 1 23 0 Read Aloud 4 28 4 29 0 27 Read Silently 48 28 62 40 31 31 Attend to Task 9 25 13 23 36 29 Figure 3. Percentage of intervals coded writing, reading aloud, reading Silently and attending to task for each target student during baseline and CWPT conditions. Note: Table made from bar graph.