Peer Relationships of Children with ASD in General Education Settings

Background: Many children with autism are included in the general education classroom. Little is known about the social relationships that exist between children with autism and their typical peers in the classroom.   

Objectives:   This study examined the social connections between peers in general education classrooms when one of the peers had an autism spectrum disorder.  Children’s acceptance, reciprocity and self-reports of friendship quality were examined for first to fifth graders. 

Methods: Sixty, first-to-fifth grade children with autism participated across 56 classrooms in 30 different schools in a large urban city.  These children were compared to a matched sample of typical children from the same classroom, same gender, and same age.  Of the children with autism, 16 children were in first grade, 17 children in second grade, eight children in third grade, 11 children in fourth grade, and eight children in fifth grade. Children with autism were from diverse ethnic backgrounds (46.7% Caucasian, 5% African American, 21.7% Latino, 16.7% Asian, and 10% Other) and were predominantly male (90%). All were fully included in regular education classrooms and were an average of 8.14 years old (sd=1.56), with an average IQ of 90.97 (sd=16.33). Typically developing children were an average of 7.87 years old (sd=1.42). Measures included a sociometric measure of children’s friendships based on peer nominations (Cairns & Cairns, 1994), a self-report questionnaire of friendship quality (FQS; Bukowski, Hoza & Boivin,1994) and playground behavioral observations.   

Results: Outcome measures of social network status and friendship quality were analyzed using mixed-effects linear models.  Results indicated that the children with autism had significantly fewer social connections in the classroom (1.36 ± 0.1) as compared to their typically developing matched peers (2.16 ± 0.1), F(1,4) = 33, p<.01.  Thus, children with autism nominated significantly fewer children (3.9 ± 0.36) from their classroom as friends than typically developing children (5.4 ± 0.37), F(1,4) = 9.06, p<.05. In addition, children with autism received significantly fewer friendship nominations from classmates (1.4 ± 0.25) than their matched peers (3.1 ± 0.25), F(1,4) = 21.90, p<.01. Also the proportion of children’s reciprocal friendships was significantly lower in children with autism (18.1% ± 5.58) in comparison to their typically developing matched peers (64.1% ± 5.39) F(1,4) = 35.12, p<.01. When examining children’s number of rejections, there was no significant difference between the number of rejections received in children with autism and their matched peers.

Children with autism reported significantly poorer friendship quality in the domains of companionship (F(1,4)=8.00, p<.05), help (F(1,4)=12.94, p<.05) and closeness (F(1,4)=15.52, p<.05). They did not have significantly different ratings in the domains of security or conflict; P>.05).

Conclusions: This study shows that children with autism in general education settings are self-reporting poorer friendship quality than their typical classmates and these findings are validated by the typical children’s lower ratings of social functioning for children with autism on the sociometric questionnaire.  These results provide additional information about the social development and competence of children with high functioning autism and demonstrate that systematic social skills interventions are needed in the public schools.