Observing social inclusion of children with ASD

Background: Social isolation is one of the most enduring challenges facing children with autism spectrum disorders (ASD). Gathering information from multiple informants and including observations of children in natural settings provides a more complete picture of the social inclusion of individuals with ASD in schools.  Most children with ASD are not isolated at school, but vary in their social connectedness to peers in their classrooms (Chamberlain, et al., 2007).  However, observing children’s interactions with their peers on the schoolyard yields more information about the nature of their social inclusion. The Playground Observation of Peer Engagement (POPE) has been utilized in several studies accounting for approximately 100 high functioning children with ASD, fully included in general education classrooms.  

Objectives: The objective of this paper is to describe an observational measure for use on school playgrounds that measures children’s social engagement with others and that can be collected live. For this presentation, the reliability and validity of the POPE will be discussed in relation to other measures collected on children’s social relationships. 

Methods: Using the POPE, children’s peer interactions were observed during school (recess and lunch periods) for levels of engagement, initiations and responses, and affect. Observers watched children for forty seconds and coded for twenty seconds, for an average of fifteen-minutes per period. Social network, friendship self-report data, and teacher and parent report of social behavior were also examined in relation to playground engagement. 

Results: The POPE has high inter-rater agreement, κ>.80, and has been successfully used in multi-site studies with suitable inter-rater agreement (κ>.70). Data from one study found children with ASD were engaged on the playground for an average of 38.6% of observed intervals (Kasari et al., in press).  Even children with more friends and high social network scores were as unengaged as children with fewer friends and less social network centrality.  Teacher and aide ratings of social behavior in the classroom were significantly related to observer-rated playground engagement (r=.48, p<.05), total playground initiations(r=.51, p<.05), to peers and appropriate responses (r=.60, p<.05) (Locke, 2011). Parent ratings of play dates at home were also significantly associated with observations of playground engagement. Number of reciprocal friendships and social network status were non-significantly associated with playground engagement. 

Conclusions: These data suggest that the POPE is a reliable measure of peer social engagement on the playground, and is significantly associated with teacher and parent report of social behavior.  Peer and self report of friendships are less associated with playground behavior, and taken together suggest that children who are nominated by peers as friends have great difficulty engaging with their friends on school playgrounds. By utilizing the POPE, it is possible to understand the level of engagement children with ASD have with their peers and their spontaneous social behaviors, which ultimately may inform intervention, monitor progress, and measure efficacy of social skills treatments.