A Social Network Analysis of Middle and High School Students with ASD in an Inclusive Public Charter High School

Background: Inclusive education—or the co-participation of students with and without disabilities in general education classrooms—increases academic and social outcomes in children with and without disabilities (Abramovitch et al., 1987; Cosier et al., 2013; Gallagher, 2000; Stoneman et al., 1987; 1989). However, students with disabilities in inclusive classrooms may be at increased risk for social rejection (Ochs et al., 2001), which may hinder their social and academic development (Siperstein & Parker, 2008; Cooc & Kim, 2017). To optimize inclusive education’s outcomes, students with disabilities must be effectively integrated into peer social networks. This is particularly important for students with autism spectrum disorder (ASD), who, due to their difficulties with social communication, are at increased risk of social isolation, and resulting negative social and academic outcomes (Lasgaard et al., 2010; OECD, 2003).

Objectives: The present study examines the extent to which secondary students with ASD are socially included in peer networks at an inclusive charter school serving a high proportion (~50%) of students with disabilities. Relationships between social inclusion and social competence, self-efficacy, and depression symptoms are also investigated, in students with and without ASD.

Methods: Chronological age-matched students (10-20 years) with and without ASD (n_ASD=31; n_non-ASD = 67) completed electronic questionnaires about their friendships, self-efficacy, and depression symptoms. Each student’s social competence was also assessed using a brief teacher-report questionnaire. Social network analysis methods were used to explore ASD students’ social inclusion in peer networks at school, and within-group correlations tested for associations between measures of social inclusion and age, social competence, self-efficacy, and depression symptoms, in students with and without ASD.

Results: On average, students had equal proportions of friendship ties with students with and without ASD, regardless of their own diagnostic status (External-Internal Index_ASD = -0.199 [-0.443, 0.045]; E-I Index_non-ASD=-0.015 [-0.134, 0.103]; t_ASDvnon-ASD=1.537, p_ASDvnon-ASD=0.128). That is, students with/without ASD reported having roughly equal numbers of friends with and without ASD. However, students without ASD tended to report more friendships than students with ASD (p=0.009). For students with ASD, numbers of friendships were positively associated with teacher-reported social competence (ASD: r=0.417, p=0.020; non-ASD p=0.300). Furthermore, number of friendships was positively associated with self-efficacy (r =0.477, p=0.010), and marginally negatively associated with depression symptomatology (r=-0.388, p=0.050) in students with ASD (all non-ASD p>0.120). None of the variables examined varied significantly by age in the sample.

Conclusions: The school under study serves a population of students with disabilities four times higher than the U.S. national average, thus presenting a unique opportunity to examine the social relationships of students with and without disabilities in an inclusive educational context. Students in this context did not self-segregate into peer networks based on the presence or absence of an ASD diagnosis, although students with ASD appeared to have fewer friendships on average than those without ASD. Given the relationships between numbers of friends, self-efficacy, and depression symptoms in students with ASD, the current results emphasize the importance of strong peer networks for promoting positive mental health outcomes in secondary school students with ASD.