Title: The Social Engagement of Girls with ASD At School: Comparisons with Boys and Girls with and without ASD

Background:  Most of what we know about school-age children with autism (ASD comes from studies with largely male samples (Chamberlain, Kasseri, & Rot Herman-Fuller, 2007; Kasseri, Chamberlain, & Bauminger, 2001; Rotheram-Fuller, Kasari, Chamberlain, & Locke, 2010). Given that girls with ASD show impairment in the same core deficits as boys with ASD (Lord, 1997), they are bound to share similar characteristics. However, girls with ASD are similar in many respects to typically developing girls. Yet, it is unclear if the social behaviors of school-age girls are similar to boys with ASD or typically developing girls. More research is needed to examine the social engagement of girls with ASD. 

Objectives:  The purpose of this research was to examine the social engagement of girls with high functioning autism during unstructured social periods at school. 

Methods:  This study is a secondary analysis of two large multi-site randomized control trials (AIR-B, in progress). Because there were fewer girls with ASD in the original studies, all female subjects that met the criteria for ASD were used in this study. Typically developing children attended the same school as the subjects with ASD. The sample (N=288) included girls with ASD (n=33) boys with ASD (n=78), and typically developing girls and boys (girls: 44; n=; boys:  n=78).

The children (grades 1-5) and adolescents (grades 6-12) with ASD had a confirmed diagnosis of ASD (Autism Diagnostic Schedule, Lord, 1983) without intellectual disability (IQ ≥ 70; confirmed by the Stanford-Binet Fifth Edition), and were educated in the general education classroom for a minimum of 80% of the school day. Social engagement was measured using the Playground Observation of Peer Engagement (POPE; Kasari, et al.  2010), which was collected on school campuses during unstructured social periods at baseline.

Results:  Children and adolescents with ASD were significantly more likely to be solitary during unstructured social periods than typically developing populations (f, (4,288)  = 61.26, p = .00).  An interaction effect indicated that girls with ASD (μ =  33. 53; sd = 34.19), were significantly more likely than boys with and without ASD (μ = 24.27; sd =31.60;  ASD:  μ = 19.65; sd = 23.48), and typically developing girls (μ =14.43; sd = 23.48) to be in a parallel engagement state (f, (4,288)  = 61.26, p = .00).  Likewise, an interaction effect indicated that girls with ASD (μ =20.43; sd = 26.80) were significantly less socially engaged during unstructured social periods at school compared to typically developing girls (μ =34.55; sd = 34.39) and boys with and without ASD (μ =29.59; sd =31.25;  ASD:  μ =37.69; sd = 34.33); (f, (4,288)  = 5.61, p = .02. 

Conclusions:  Girls with ASD have social experiences that are different from boys with ASD and their typically developing peers. While both boys and girls with ASD are more likely to be solitary, boys with ASD have higher levels of social engagement than girls. More research is needed to identify qualitative behaviors that promote or interfere with the social engagement of girls with ASD.