Printer Friendly

Classroom and playground interaction of students with and without disabilities.

ABSTRACT: The classroom and playground behaviors of students with disabilities, in an integrated classroom, are frequently cited as reasons for rejection of these students by their regular-class peers. We compared the classroom and playground behavior of 95 students with mild disabilities with that of 95 students without disabilities, age and sex matched, and enrolled in the same classes. The children, age 8-13 years, attended state government elementary schools. Behaviors were observed using a time-sampling method with nine categories of behavior in each setting. The behavior of both groups of students was in many ways similar, regardless of the presence of disabilities.

Integration in the regular school is receiving increasing recognition as a legitimate educational option for students with disabilities. Many issues surrounding the process of integration, however, require empirical verification. The mere placement of these students within a regular-school context does not automatically result in increased social interaction between students with disabilities and those without. As Gresham (1982) suggested in a recent review of the literature, little evidence supports the notion that the increased contact between the two groups of students in an integrated setting results in either more positive attitudes toward students with disabilities or social acceptance of these students by their peers without disabilities.

The social behavior of children has often been investigated because of its relationship to social acceptance and rejection by peers. Certain social behaviors, such as positive peer interaction, greeting others, asking for and giving information, and making conversation, are predictive of social acceptance (Asher & Hymel, 198 1; Asher, Oden, & Gottman, 1977). Given the finding that students with mild disabilities often hold low social-status positions in regular classrooms (Gottheb, Semmel, & Veldman, 1978), it has been hypothesized that these children display patterns of social behavior that are different from those of regular-class peers (Gresham, 1982).

The classroom behavior of students with mild disabilities has been studied using observational research methods (Espiner, Wilton, & Glynn, 1985; Forness & Esveldt, 1975; Gampel, Gottlieb, & Harrison, 1974; Gottlieb, Gampel, & Budoff, 1975; Herink & Lee, 1985; Hudson & Clunies-Ross, 1984; Kaufman, Agard, & Semmel, 1985); but the social behavior of such children in the playground setting has been a neglected area of research. In this investigation, both the classroom and playground behavior of students with mild disabilities (mildly intellectually handicapped and borderline) and of their regular-class peers were studied.

Gampel et al. (1974) compared the classroom behavior of 12 segregated students classified as educable mentally retarded (EMR) and 14 recently integrated EMR students with that of regular-class children with low IQs and children of average intellectual ability, in the primary grades of a U.S. elementary school. These researchers found that the integrated EMR students' behavior more closely resembled that of the low-IQ, regular-class children than that of the segregated EMR students, after only 4 months in an integrated setting. When observed again at the end of a year of integration (Gottlieb et al., 1975), the integrated EMR students were found to be displaying more prosocial behaviors than both the segregated EMR students and the regular-class students without retardation.

More recent research carried out in Australia (Hudson & Clunies-Ross, 1984), New Zealand (Espiner et al., 1985), and the United States (Herink & Lee, 1985; Kaufman et al., 1985) suggests similar results. A more detailed analysis of the social behavior of the two groups of children, however, reveals some differences. Hudson and Clunies-Ross (1984) systematically observed 15 students with intellectual disabilities in both the regular classroom and the playground (grades 1 3). They found no difference between overall rates of both positive and negative interactions for the students with disabilities and a randomly selected peer-group sample without disabilities. The disabled students, however, were observed to initiate only half as many positive interactions as were their regular-class peers. They were also observed to initiate twice as many interactions with the classroom teacher as were their non-disabled peers. Unfortunately, because these authors did not report classroom and playground behavior separately, the results relate to the overall behavior patterns observed.

Herink and Lee (1985) compared the social behavior of 20 preschoolers with mild and moderate intellectual disabilities with that of 20 children without disabilities and found that the children with disabilities were substantially integrated into the emotional and social life of the peer group, but not fully integrated into the verbal life of the peer group. These researchers also found an inverse relationship between teachers' initiation of social interaction with the children with disabilities and the degree to which they were integrated into the peer social group.

Kaufman et al. (1985) observed 300 children classified as EMR and regular-class children in the United States. They found that the integrated EMR learners displayed more antisocial and less friendly/cooperative behavior than did their regular-class peers; but the differences were too small to be of educational significance.

With regard to playground behavior, Hudson and Clunies-Ross (1984) observed 15 students with disabilities; however, no results specific to the playground setting were reported. A recent study completed in New Zealand by Pipe, Redman, and White (1983), however, found high levels of social interaction between preschoolers with intellectual disabilities and those without, both in the classroom and on the playground.

Hence, empirical evidence from studies investigating the classroom behavior of students with and without disabilities appears to be equivocal in its support for the hypothesized difference in social behavior patterns between the groups. Most previously cited studies (except Kaufman et al., 1985), however, used relatively small samples and have predominantly investigated children in preschool or the primary grades of elementary school. Older students without disabilities in middle and upper grades have been shown to display more stable friendship choices (Ollendick, 198 1). These students also would be expected to show different patterns of interactive behavior. Although Kaufman et al. (1985) considered a large sample of students, they did not compare the social behavior of students with and without disabilities in a playground setting.

There is a need, therefore, to investigate and compare the social behavior of students with mild disabilities with that of their regular-class peers, in both classroom and playground settings. It is important also to consider older students. In this article, we seek to meet these aims by investigating the classroom and playground behavior of 95 students with mild disabilities and 95 peers without disabilities, aged 8 to 13 years.



Subjects were 190 elementary school students ranging in age from 8 to 13 years. Of these students, 95 were students with mild disabilities who were enrolled in Educational Support Centres attached to 11 regular state elementary schools within the metropolitan area of Perth (Western Australia). These students were all integrated into regular classrooms for varying periods of time during the day. The remaining 95 students, who were without disabilities, were enrolled in the same integrated classrooms.

The students with disabilities fell into the following three diagnostic categories when classified on the basis of the American Association on Mental Deficiency definition (Grossman, 1983); moderately intellectually disabled, 3% (n = 3); mildly intellectually disabled, 35% (n = 34); and those functioning intellectually at a borderline level, 61 % (n = 58). All 95 students had been referred to the Support Centres because they were performing at a level of academic achievement 2 years or more below that expected for their chronological age.

Mean percentages of time integrated and other demographic details on these students are listed in Table 1.

Students without disabilities were matched with an integrated student with disabilities by choosing a regular-class peer of the same sex, with the closest birth date to the targeted student. The sample of students without disabilities included 69 males and 26 females. In Table 2 we show mean age and number of students with and without disabilities according to their grade. As it is not uncommon for students with mild disabilities to repeat a grade, in some cases the children without disabilities were somewhat younger than their matched peers with disabilities.

Measures of Classroom and Playground Behavior

Models available in Gampel et al. (1974), Hudson and Clunies-Ross (1984), and Ollendick (1981) were used in developing two sets of behavioral categories for observing behavior in the classroom and on the playground. In both settings, the pattern of interactions and the type of activity engaged in were recorded.

Observation and recording of the students' interaction patterns were the same for both classroom and playground settings. The following details were recorded:

1 .The type of interaction-an initiation, whether successful or unsuccessful or an ongoing interaction.

2. The person involved in the interaction other than the target child-whether peer, teacher, or another adult.

3. The quality of the interaction-either positive or negative. Where an interaction initiated by a teacher or another adult was coded, a further distinction was made between a statement, request or a nonverbal initiation.

The type of student behavior in the classroom and playground required two different sets of categories.

In the classroom, the attending behaviors of the target student were recorded. These behaviors consisted of the following:

1. On task.

2. Off task-quiet.

3. Off task-disruptive.

4. Off task-aggressive.

On the playground, the student's activity was recorded in one of the following categories:

1. Unoccupied behavior.

2. Solitary play.

3. Onlooker behavior.

4. Parallel play.

5. Interactive play.


A time-sampling method was used to gather the observational data. The observational interval was 25 seconds (s) including 15 s of observational time and 10 s of recording time. The first behavior to be observed within the 15-s observational time was the behavior that was coded. Intervals were cued by a beeper mechanism, which was heard through an earpiece.

In both the classroom and playground settings, targeted students were observed in a series of 5-minute (min) blocks broken up into 12, 25-s intervals. The total length of any single observation session was up to 30 min, and each student in a matched pair was observed for three blocks of 5 min in every 30-min session. This resulted in a total of 36 observational entries for each child in a matched pair for the 30-min session. All pairs of students were observed for three sessions in the classroom and three sessions on the playground, resulting in a total of 108 classroom and 108 playground observational entries per student.

In the classroom setting, a matched pair of students was observed in an alternating sequence of 5-min blocks over the 30-min session. The order of observation of the pair of students was counterbalanced to reduce the effect of variance in behavior due to the beginning or end of classroom lessons. Classroom observations were completed at varying times of the day, coinciding with those periods where the students with disabilities were integrated into the regular classrooms. Because these students were more likely to be integrated into the regular classrooms for the two afternoon sessions, 81% of the classroom observations were completed during these time periods.

Several lesson types were observed. In Table 3 we show a breakdown of the types of lessons observed and the percentage of the total number of observation entries for each lesson type. Because students were observed in pairs, the percentages are the same for both groups of students. Students spent the majority of their time in individual seatwork or listening to the teacher giving a lesson. Small group activities were also observed.

In the playground setting, students were observed during free-play sessions. Students within a target pair were observed one after the other, the first child being observed for up to three consecutive 5-min blocks, then the second child. The change in procedure from the classroom to playground settings was related to the practical difficulties of observing children in large playground areas. Observing target pairs in alternating 5-min blocks was impossible because of the amount of time needed to search for children.

The playground observations were completed during recess time when the students were involved in free-play activities. A variety of playground games and activities was observed during these periods, for example, structured games such as cricket and basketball, involvement with playground equipment, or talking.

Five psychology students and the chief investigator acted as observers. An initial training of three observers was carried out in eight 1-hour (hr) sessions over a 3-week period. Both videotaped sequences of behavior and in-vivo practice were used to train observers. A further five sessions were completed to train the remaining two observers who were employed at a later date to complete the total number of observations. Observations were carried out over a 6-month period in the last two terms of the school year to ensure that the students' behaviors would not be influenced by a new environment.

Reliability checks were carried out before observation began and at four points during the 6-month period of observations, with all six observers present. The reliability coefficients were calculated by dividing the number of exact agreements for all observers in each observational interval, by total agreements plus disagreements (Anastasi, 1976). Initial interrater reliability coefficients were .94 for classroom observations and .81 for the playground observations, respectively. Interrater reliability coefficients for the four reliability checks carried out throughout the data collection phase ranged from .80 to .89 for classroom behaviors and .81 to .83 for playground behaviors.


To obtain a subset of behaviors for further statistical analysis, two stages of data reduction were carried out with both classroom and playground observational data.

The first stage involved eliminating and collapsing categories that exhibited very low rates of occurrence. Frequency scores were calculated for the data characterizing the type of activity, four categories of classroom attending behavior and five categories of playground behavior. These frequencies were then transformed into proportions of the total number of observations for each subject in both settings (108 in each case). Scores represented as percentages for students with and without disabilities are shown in Figures 1 and 2.

Figures 1 and 2 show that aggressive behavior in the classroom and unoccupied behavior on the playground occurred with frequencies of less than 3% for both groups of students. These two categories were excluded from further data analysis.

Frequency scores for the large number of categories of interactional behavior were also calculated for the classroom and playground settings. As would be expected, both groups of students engaged in less interactive behavior in the classroom than on the playground, (students with disabilities: 10.36%-classroom, 67.13%-playground; students without disabilities: 1 1.64%classroom, 77.89%, playground).

A number of interaction categories occurred with very low frequency in both the classroom and playground settings; hence, the data were collapsed to obtain a subset of variables. These combined categories consisted of the following variables, with scores calculated separately for both the classroom and playground settings:

1. Positive adult interactions.

2. Negative adult interactions.

3. Positive peer interactions.

4. Negative peer interactions.

5. Successful initiations from a child.

6. Unsuccessful initiations from a child.

7. Successful initiations from an adult.

8. Unsuccessful initiations from an adult.

9. Successful initiations from a peer.

10. Unsuccessful initiations from a peer.

All positive and negative, adult and peer interactions were then transformed to proportions of the total number of interactions for each subject. Figure 3 presents the mean percentage of these interactions for students with and without disabilities in both settings. Positive and negative interactions were mutually exclusive; therefore, only positive interactions were included in further analyses.

The successful and unsuccessful initiations were transformed to proportions of the total number of initiations for each subject. Figure 4 presents the mean percentage of child initiations, and

Figure 5 presents the mean percentages of adult and peer initiations for both groups of students in the two settings. Successful and unsuccessful initiations were mutually exclusive categories. They were therefore excluded from further analysis.

After the completion of the first stage of data reduction, each subject had a score profile of 17 proportional scores representing their behavior in the classroom and playground.

Factor 2 represented classroom attending behavior; and it accounted for 13.52% of the total variance. This factor contained the categories of on task and off task-quiet. The latter category loaded negatively on this factor.

Factor 3 accounted for 11.01% of the total variance and grouped together all child/peer initiation patterns. This factor was composed of successful child initiations on the playground (negative loading), successful peer initiations on the playground, and successful peer initiations in the classroom.

Factor 4 represented both types of positive interaction in the classroom-peer interaction (negative loading) and adult interaction. This factor, which defined classroom interaction, accounted for 8.09% of the total variance.

Finally, Factor 5 consisted of two classroom variables, successful initiations from a child and from an adult (negative loading). This factor accounted for 7.42% of the total variance and represented child/adult initiation patterns in the classroom.

Variables loading over .40 on each of Factors 6-8 showed no meaningful groupings of interaction or behavior; therefore, the five remaining variables were included separately in all further analysis. These included disruptive behavior in the classroom, positive adult interactions on the playground, successful initiation from an adult on the playground, solitary play, and onlooker behavior. The results of the varimax rotation for Factors 1-8 are shown in Table 4.

Following the second stage of data reduction, factor scores on the first five factors were obtained for all subjects by inversing the scale of variables that loaded negatively and calculating an average of those variables with a factor loading of greater than .40 on an individual factor.

These five factor scores plus scores on the remaining five categories not represented in the factor analysis were then used as dependent measures in a multivariate analysis of variance to determine differences in behavior between the two groups of students. The means and standard deviations of these 10 variables are shown in Table 5.

A Hotellings t[.sup.]2 test for repeated measures (Dixon, 1981) revealed a significant difference between groups, f (10,85) = 7.80, p < .0001. Further paired comparisons using matched sample t tests revealed significant differences on 4 of the 10 behavioral categories. The students with disabilities were observed to engage in less peer-play interaction, t(94) = 4.07, p < .00 1, and more solitary play, t(94) = 5.7 1, p < .000 1, than were their peers. The students with disabilities also engaged in less classroom attending behavior, t(94) = 6.02, p < 0001, and more positive adult interaction on the playground, t (94) = 2.14, p < .05, than did their peers. No significant differences were found in the following interaction variables; child/peer initiations, positive class interactions, child/adult initiations and successful adult playground initiations. The two groups of students also did not differ in the amount of disruptive behavior displayed in the classroom and onlooker behavior on the playground.


These results indicate that there are many similarities in behavior patterns between students with mild disabilities and those without disabilities attending integrated classes. Both groups of students showed low levels of negative, disruptive, and aggressive behavior in the classroom and playground settings. The two groups of students also did not differ significantly in their amounts of interaction with adults or peers in the classroom. Finally, the patterns of initiation with peers did not differ across groups in either setting.

There were differences, however, between the groups in the type of behavior engaged in, both in the classroom and on the playground. On the playground, students with disabilities interacted and played less with their peers than did the other students. Although the results show that the students with disabilities were not totally isolated, this finding has implications for the social acceptance of these students in an integrated setting. When compared with results of recent studies of social acceptance and social behavior in children without disabilities (Dodge, 1983), the behaviors of the students with disabilities resemble closely those of neglected children.

When not interacting with their peers, the students with disabilities were observed to engage in significantly more solitary play and more positive interactions with adults. These results are similar to both Herink and Lee's i 985) findings with preschoolers and the findings of Hudson and Clunies-Ross (1984), who found that elementary school children with disabilities initiated twice as many positive interactions with adults as did their peers. Because the current study found no significant differences between the groups in the proportion of adult initiations on the playground, this finding suggests that students with disabilities seek out and initiate interactions with adults on the playground.

A number of alternative explanations for this behavior can be suggested. Gresham (1982) believed that students with disabilities were socially rejected and excluded from interactions with their regular class peers because of their antisocial behavior and lack of social interactional skills. However, the results of this study show that the frequencies of negative interactional behavior in classroom and playground settings were low and did not differ across the groups of students. Moreover, no significant difference was found between students with and without disabilities in their child/peer initiation patterns (Factor 3), evidence that appears to contradict Gresham's thesis.

Though the students with disabilities did not engage in more disruptive or negative behavior than their peers, they may have been viewed by their peers as cognitively less competent and therefore less desirable as playmates. Disabled students were observed to engage in less on-task behavior and more quiet, off-task behavior than were their peers. Although these behaviors did not appear to be associated with differences in adult attention between the two groups in the classroom, they may have been noticed by other class members. The integrated students with disabilities could have been perceived by their peers as unable to meet the cognitive demands of the games played outside on the playground.

An alternative explanation is that the students with disabilities chose to engage in more solitary play and interaction with adults on the playground. This may have been seen as a less threatening option than attempting to join already established groups. The majority of disabled students in this study were not full-time members of their regular classes; and they often had joined the class within the past I to 2 years. Hence, they had not had the advantage of mixing with the same peers consistently all day from the beginning of their school careers. Their behavior could be viewed as similar to that of students without disabilities who are new to a class or school.

Finally, a note of caution needs to be addressed regarding the initiation categories. The frequencies of occurrence of these six behaviors were low in both groups of students compared with those of other categories investigated. The use of factor scores to summarize the data improved the stability of these ratings. However, categories such as successful play initiations from an adult should be interpreted conservatively.

One of the major aims of integration is to provide students with disabilities with opportunities for interaction with and social acceptance by students without disabilities. The present data suggest that integration is achieving this aim. The students with disabilities did not engage in any more disruptive or negative behavior than did their peers. They were shown to be interacting with peers on the playground approximately 50% of the time sampled, and their patterns of initiation and response with peers did not differ from that of the group without disabilities. Differences in behavior patterns, however, were observed, particularly in peer interactions on the playground and attending behavior in the classroom.

The results of the current study suggest that students with disabilities interacted less with peers and more with adults than did the students without disabilities. An explanation that considers only the strengths and weaknesses of the individual child could suggest that the students with mild disabilities need coaching in the area of social skills (Gresham, 1982; Gresham & Elliott, 1987). To promote generalization, the integrated setting is the logical place in which to learn and practice these social skills. However, aspects of the regular classroom/school environment can also be modified to promote more interaction and acceptance across groups.

Attending behavior in the classroom also needs to be addressed. Because of their learning difficulties, students with mild disabilities need to spend more time on task to maintain and improve their academic performances. If this can be achieved through changes in classroom organization and alternative instructional strategies that allow for more individual difference, students with disabilities will be perceived by their peers as being better able to cope with the demands of the integrated classroom. Structured opportunities for successful interaction between both groups of students could lead to better interaction patterns on the playground.

The results of this investigation support Gottlieb's (1981) suggestion that it is not sufficient to provide contact between students with and without disabilities to build intergroup social interaction. If intergroup social interaction is to be an aim of integration programs, opportunities for interaction should be carefully planned and all influential factors investigated, such as the roles of teachers and regular-class peers.


Anastasi, A. 1976) Psychological testing (4th ed.). New York: Macmillan.

Asher, S. R., & Hymel, S. (1981). Children's social competence in peer relationships: Sociometric and behavioral assessment. In D. J. Wine & M. D. Smye (Eds.), Social competence (pp. 125-157). New York: Guilford Press.

Asher, S. R., Oden, S. L., & Gottman, J. M. 1977). Children's friendships in school settings. In L. G. Katz (Ed.), Curremt topics in early childhood education. (pp. 33-61). Norwood, NJ: Ablex.

Dixon, W. J. (1981). BMDP statistical software. Berkeley: University of California Press.

Dodge, K. A. (1983). Behavioral antecedents of peer social status. Child Development, 54, 1386-1399. Espiner, D., Wilton, K., & Glynn, T. 1985). Social interaction and acceptance of mildly retarded children in a mainstream special education setting. Australasian Journal of Special Education, 9, 8-15.

Forness, S. R., & Esveldt, K. C. (1975). Classroom observation of children with learning and behavioral problems. Journal of Learning Disabilities, 8, 49-52. Gampel, D. H., Gottlieb, J., & Harrison, R. H. (1974). Comparison of classroom behavior of special class ENW, integrated EMR, low IQ and non-retarded children. American Journal of Mental Deficiency 79, 16-21. Gottlieb, J. 1981). Mainstreaming: Fulfilling the promise? American Journal of Mental Deficiency, 86, 115-126.

Gottlieb, J., Gampel, D. H., & Budoff, M. (1975). Classroom behavior of retarded children before and after integration into regular classes. Journal of Educational Psychology 73, 307-315.

Gottlieb, J., Semmel, M. I., & Veldman, D. J. (1978). Correlates of social status among mainstreamed mentally retarded children. Journal of Educational Psychology, 76, 396-405.

Gresham, F. M. (1982). Misguided mainstreaming: The case for social skills training with handicapped children. Exceptional Children, 48, 422-433.

Gresham, F. M., & Elliott, S. N. (1987). The relationship between adaptive behavior and social skills: Issues in definition and assessment. The Journal of Special Education, 21, 167-181.

Grossman, H. J. 1983). Classification in mental retardation. Washington, DC: American Association on Mental Deficiency.

Herink, N., & Lee, P. C. 1985). Patterns of social interaction of mainstreamed pre-school children: Hopeful news from the field. The Exceptional Child, 32, 191-199.

Hudson, A., & Clunies-Ross, G. (1984). A study of the integration of children with intellectual handicaps into regular schools. Australian and New Zealand Journal of Development Disabilities, 10, 165-177. Kaufman, M., Agard, J. A., & Semmel, M. I. (1985). Mainstreaming: Learners and their environments. Cambridge, MA: Brookline.

Ollendick, T. H. ( 1981). Assessment of social interaction skills in school children. American Behavioral Counseling Quarterly 1. 227-243.

Pipe, M., Redman, S., & White, K. G. (1983). Social interactions of retarded children: Generalizations from mainstream to special school. The Exceptional Child. 30, 15-22.

Thorndike, R. M. (1978). Correlational procedures for research. New York: Gardner Press.


CLARE ROBERTS is a Lecturer and CHRIS PRATT is a Senior Lecturer in the Department of Psychology at The University of Western Australia, Nedlands. DAVID LEACH is a Senior Lecturer at the School of Social Inquiry, Murdoch University, Murdoch, Western Australia.

The research reported in this article was supported in part by a grant from the Apex Foundation for Research into Mental Retardation. We are indebted to the Western Australian Ministry of Education for providing permission to conduct this research in state government schools, and we thank the principals and school staff for their cooperation. We gratefully acknowledge Dan Milech and Steve Zubrick for their advice on statistical analysis procedures.

Requests for reprints should be sent to Clare Roberts, Clinical Psychology Unit, Department of Psychology, The University of Western Australia, Nedlands, Western Australia, 6009. Manuscript received November 1988; revision accepted May 1989.

Exceptional Children, Vol. 57, No. 3, pp. 212-224. [c]1990 The Council for Exceptional Children.
COPYRIGHT 1990 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1990 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Roberts, Clare; Pratt, Chris; Leach, David
Publication:Exceptional Children
Date:Dec 1, 1990
Previous Article:Early intervention for infants with deaf-blindness.
Next Article:Apprenticeship and intensive training of consulting teachers: a naturalistic study.

Related Articles
Effects of preschool integration for children with disabilities.
Use of instructional time in classrooms serving students with and without severe disabilities.
Educating young children with disabilities using responsible inclusion.
The One-to-One Survey: Students with Disabilities Versus Students without Disabilities Satisfaction with Professors During One-on-One Contacts.
Preschoolers' Play Behaviors With Peers in Classroom and Playground Settings. (Research Into Practice).
Everyone plays! Increasing participation of special needs children.
Increasing peer interactions for students with severe disabilities via paraprofessional training.
Boy in the Bubble: Effects of Paraprofessional Proximity and Other Pedagogical Decisions on the Interactions of a Student With Behavioral Disorders.
Differential effects of peer tutoring in co-taught and non-co-taught classes: results for content learning and student- teacher interactions.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters