Until recently, little emphasis has been placed on the social functioning of students with mild disabilities (Gresham, 1992; Haring, 1993; Kopp et al., 1992; La Greca and Vaughn, 1992), and lack of social skills, like learning problems, may be a cause of placement into special education programmes (Coleman and Minnett, 1993; Gresham and Elliott, 1989a, b; Shores et al., 1992). Thus examining how the social ecology functions to support peer social behaviour within the regular education classroom is important, especially because it has been suggested that integrated, rather than segregated, settings may offer greater social opportunity and normalcy to students with mild disabilities (Putnam, 1993; Snell and Eichner, 1989). Some have argued that students with mild disabilities should be given opportunities to interact with their typical peers in regular classroom settings because positive social interaction may lead to friendships and other social relations (Haring, 1993; Snell and Eichner, 1989).
Evidence does exist that typical children on average: (1) are more socially competent than their peers with mild disabilities (Landau and Moore, 1991; McKinney et al., 1993; Stone and La Greca, 1990; Vaughn and Lancelotta, 1990; Vaughn et al., 1993), (2) prefer to interact with other typical children as opposed to children with mild disabilities (Guralnick,1980; Osguthorpe et al., 1985; Roberts et al., 1991), and (3) do not socially accept children with disabilities (Landau and Moore, 1991; Milich et al., 1992; Roberts and Zubrick, 1993). With regard to students with mild disabilities, they tend to stay alone and engage in little social play as compared with their typical peers (Kopp et al., 1992; Roberts et al., 1991; Williams and Asher, 1992).
For example, Merrell et al. (1992) examined the social competence of 135 students with learning disabilities (LD), 109 students having Educable Mental Retardation (EMR),114 students having Behavioural Disorders (BD),100 low-achieving students, and 108 average, regular education peers based on their teacher ratings. Using the Walker-McConnell Scale of Social Competence and School Adjustment (SSCSA), the results indicated that the typical peers received significantly higher sociometric ratings than the groups of students with mild disabilities and low-achievers on all sub-scales of the SSCSA. Overall, the BD group scored significantly lower than all other groups on subscale 1 of the SSCSA and the total scale.
Also, the short and long-term effects of mainstreaming on the social behaviours of the students with mild disabilities have been questionable (e.g. Roberts et al., 1991). For example, Brewer and Smith (1989) reported that the social acceptance of students with mental retardation who were educated in inclusive settings was not increased over a course of two years of integration. Compared with the typical students, the students with disabilities received significantly lower social ratings on sociometric measures of acceptance based on a test administered to the whole classroom.
In summary, the above findings favour the lower social status of students with mild disabilities, and social isolation/rejection by typical peers. However, as was the case with the prior work focused on classroom behaviour, this work on social behaviour is primarily descriptive and comparative. Other than examining the effects of students placed from one setting or service delivery format to another, few studies have experimentally manipulated instruction within these settings to affect the social behaviour of target students with disabilities or their typical peers. Thus it is imperative for instruction supporting the inclusion of children with disabilities to demonstrate its effectiveness in addressing both social and academic goals. The objectives of this study were to (1) assess the amounts and types of within-classroom social interaction of students with mild disabilities and their typical peers during the implementation of class-wide peer tutoring (CWPT) and teacher-mediated instruction, (2) examine the amounts of inappropriate behaviour during both conditions, (3) measure the effects of these procedures on the spelling performance of three students with mild disabilities and three of their typical peers, and (4) assess the social validity of both instructional procedures within an ABAB experimental single-subject design. …