In the spring of 1997, Lisa Sharon Cushing and Craig H. Kennedy conducted an experiment to study the academic effects of providing peer support in general education classrooms on students without disabilities. In other words, students were paired with other students and their behavior observed. The study was undertaken to better understand the effects of peer support stratigies of participating students. Three non-disabled students were observed and a baseline measure of academic engagement was taken. Each student was paired to be a peer supporter with a disabled student and that level of behavior was observed. The experimental question states: Does serving as a peer support have a positive or negative effect on academic engagement and associated measures on peers without disabilities?
The independent variable is the peer support of the disabled students. The dependant variable is time acadmeic engagement behavior during a fifty-minute class peroid. To sum it up, three non-disabled student who were judged to have poor class room attention and academic engagement during class were selcted in this study. A baseline measuer was obtained for a comparison point. The intervention, which is the paring up with a disabled student as a peer supporter, was introduced and a behavioral measure was again taken. The results were compared to that of the baseline measure.
Three non-disabled students were selected for this experiment; Cindy, Kealoha, and Louie. First, Cindy is a 13-year-old quiet girl who draws little attention to herself and often has difficulty following directions. Second, Kealoha is a 12-year-old boy who has problems paying attention during class lectures and is often late in turning in assignments in on time or at all. Third, Louie is an 11-year-old boy who often interupts class by blurting out and talking to other students during lecture. Louie has not been turing in his assignments and was receiving low grades.
Each of these students was paired with a disabled student to serve as a peer support during class. First, Cindy was paired with Cathy in English class. Cathy is a 13-year-old girl with severe multiple disabilities including Rett syndrome. She tracks people and objects by gazing, expresses herself by smiling or crying, and blinks her eyes to communicate to yes/no questions. She requires assistance with all activities. Second, Kealoha was matched with Karl, a 12-year-old boy, in a health class of 35 students. Karl has severe intellectual disabilities and communicates using one to three word utterances as well as gestures, touching, and pointing. Karl is invloved in self-injurious behavior, and occasionally kicks or punches other classmates. If he is left unattended, he often roams throughout the classroom. Third, Louie was paired with Leila, an 11-year-old girl. Louie and Leila attend three separate classes together; English, Social Studies, and Science. Leila has moderate intellectual disabilities including poor articulation and a limited vocabulary, which cause her problems to express herself. She frequently needs
assistance to stay focused and to follow directions. In all, three subjects were studied but six subjects were involved in the experiment.
In addition to direct observation, there were other methods of social validation employeed. Adults perceptions of the classroom performance for Cindy and Louie were also assessed. They used the CPC, Classroom Participation Checklist, which contains a set of six questions relating to the engagement in classroom activities. Special education personnel who did not know the experimental question observed the student and rated his or her performance. The article does not address whether or not informed consent was obtained but is does discuss something similar. University graduate students took agreement measures by observing them in the naturalistic setting of their classroom. The mean agreement for Cindy and Kealoha were 88% and 93%. Louies mean agreement for English, Social Studies, and Science, were 89%, 82% and 88% respectively.
For Cindy and Kealoha, a withdrawl design was selected to best assess their academic engagement alone and while serving as peer supporters. By using an ABAB design, the experimenters can compare the data between the two phases. Baseline measures were gathered for all three participants to assess their normal behavior as a comparison point. Due to the fact the Louie will be observed in three different settings, a more approiate design was selceted. A multiple baseline design will allow easy comparison between phases in the three classes and also be effective in keeping the three settings from influencing each other.
To interpret data, both statistical and graphical analyses were used. The mean percent engagement for Cindy rose from 38% to 86%. Kealoha rose from 51% to 88%. Louies mean percent engagements for English, Socical Studies and Science rose from 66% to 90%, 63% to 96%, and 72% to 92% respectively. The data was graphically displayed in a typical line graph, which made it much easier to see the increse in academic engagement, participation, and homework grades.
The results from this studt definetly conclude that the three subjects benefited from the intervention introduced. Serving as a peer support of a person with moderate to severe disabilities can have positive effects on the non-disabled student serving as that supporter. The follow-up data concludes that this experiment had a carry-over effect on their general academic life. When the adolescences served as peer supporters it caused them to become more aware of what was going on around them. As a result, in general, the academic engagement, participation, assignment completion and resulting grades increased. It is easy to see these results by viewing the graphs that corespond to the study. The study needs to be replicated to ensure that any extraneous variables that may have existed did not play a role in the benefits these children received.
Cushing, L.S. and Kennedy, C.H. Academic effects of providing peer support in general education classrooms on students without disabilities. Journal of Applied Behavior Analysis, 30, 139-151.