Observation, Parent, and Self Report of Social Engagement in Adolescents with ASD

Background: Individuals with ASD often exhibit limited insight into their peer relationships. As a result, parent report is frequently utilized in ratings of youth social engagement. In fact, individuals with ASD display a systematic tendency to over-rate their social functioning relative to parents’ report (Lerner et al., 2012; Vickerstaff et al., 2007). These informant discrepancies suggest the need to include observer measures of social functioning, particularly when examining change in social functioning in response to intervention. Unfortunately, there is limited research examining convergence among observation, parent, and self report in ASD.  

Objectives: The main objective was to examine the relationship between observation of teen engagement and adolescent and parent report of engagement and social interactions during a school-based social skills intervention. 

Methods: Adolescents with ASD (n = 34) participated in an 8-week social skills intervention in the school setting.  The Teen Observation of Peer Interaction (TOPI) was administered during school lunches at pre-intervention and post-intervention. The Teen Observation of Peer Interaction (TOPI) is a behavioral observation measure that allows for observation of peer interactions (e.g., proximity to peer, level of complexity in social engagement with another peer). The TOPI consists of five 90-second observation intervals and rates “engagement state” from “solitary” to “joint” on a six point scale for each interval. An overall average engagement state was calculated at each time point. In addition to completion of the TOPI at each time point, parent and self report measures of adolescent engagement and social interactions were collected via the Behavior Assessment System for Children (BASC) and Social Skills Improvement Scale (SSIS). Adolescents also completed the Index of Peer Relations (IPR). 

Results: Adolescent report of engagement (SSIS), social stress (BASC), and interpersonal relatedness (IPR) were not related to engagement state (TOPI) at pre-intervention. However, at post-intervention, adolescents who were observed in more complex engagement states reported stronger peer relationships via the Index of Peer Relations (r = -.509, p = .007) (lower IPR = better peer relationships) and lower social stress via the BASC (r = -.451, p = .012). There continued to be no correlation between self-report of engagement on the SSIS and observation of engagement. Parent report of social engagement (SSIS) was significantly correlated with observation of engagement state (TOPI) at pre-intervention (r = .535, p = .001) and post-intervention (r = .459, p= .024). 

Conclusions: Adolescents who were observed in more complex engagement states reported stronger peer relationships and lower social stress at post-intervention. Observer rating of adolescent engagement was convergent with parent, but not adolescent report of engagement at both time points. Although these relationships need to be explored further, these preliminary results suggest that there is validity in using observer report of engagement in addition to self-report and parent-report when measuring impact of intervention on social engagement.