What Are They Doing at Recess? Examining Playground Behavior Between Elementary School Children with and without Autism Spectrum Disorder

Background:  The social communication and interaction deficits associated with ASD can dramatically impair peer engagement and social relationships at school. The majority of children with ASD experience poorer social outcomes in schools than those of children without ASD (Kasari et al., 2011; Locke et al., 2013). Studies have reported that children with ASD spend approximately a third of the recess period solitary or unengaged with peers (Frankel et al., 2011; Kasari et al., 2011); less is known about the playground engagement and peer interaction behaviors of children withoutASD during recess. This comparison is necessary to determine an appropriate benchmark of peer engagement during recess and reasonable expectations for improvement for children with ASD who undergo interventions designed to improve social ability in schools. 

Objectives:  The purpose of this study was to document the recess engagement and peer interaction behaviors of children with and without ASD in inclusive school settings to provide a benchmark of the range of engagement and peer interaction during recess of typical classmates.

Methods: Participants included 51 children with ASD (9 females and 42 males; M_age = 8.1, SD = 1.6 years old; M_IQ = 86.9, SD = 12.6) and 51 typically developing children (20 females and 31 males; M_age = 8.1, SD =1.5 years old) matched on gender, classroom, grade, age, and ethnicity (wherever possible) from 42 classrooms in seven public schools in a large urban school district (AIR-B Network). The Playground Observation of Peer Engagement (POPE), a timed interval behavior coding system, was collected twice for both children with ASD and the matched sample during two separate recess periods within one week.

Results: Classification and Regression Tree and Receiver Operating Characteristic curve analysis indicated that the optimal cut-point for discriminating between children with ASD and typically developing peers was .58 on the POPE (sensitivity, 0.72; specificity, 0.79). There was a statistically significant difference in the percentage of time spent in solitary (M_diff = 19.6 after adjusting for age) and joint engagement (M_diff = -28.0 after adjusting for age) between children with ASD and their matched peers (p<0.001 for both). Children with ASD also had significantly fewer successful initiations (p<0.001), total initiations (p=0.02), lower initiation rate (p<0.001), positive responses (p=0.01), total opportunities to respond (p=0.03), and lower response rate (p=0.01) than did the matched sample.

Conclusions: Comparing the engagement of children with ASD with that of typically developing classmates in the same context may determine whether children with ASD need intervention and what the expectation for improvement should be. A goal of 58% (the cut point that differentiates children with ASD from their typically developing peers) engagement may be a reasonable intervention objective.