29709
Discrepancies between Parent- and Self-Reported Social Skills in Autism Spectrum Disorder: Magnitude and Treatment Effects
While most people usually rate themselves as “above average” on objective measures of performance (e.g., test performance; Kreuger & Dunning, 1999), for certain populations the discrepancy between self- and other-report ratings is particularly pronounced (i.e., the positive illusory bias; Hoza et al., 2000; Gresham et. al., 2000). This bias is particularly pronounced among those with developmental disorders. For instance, youth with ADHD persistently over-estimate their own skills relative to others’ reports of their skills (Owens et al., 2007). A growing body of literature suggests that youth with autism spectrum disorder (ASD) exhibit a similar – perhaps even greater – discrepancy (Lerner et. al., 2012; Vickerstaff et. al., 2007; Kalyva, 2010). While social skills interventions have been shown to affect parent- and self-reported social skills independently (Gates et. al., 2017; Lerner et. al., 2011), the impact they may have on this informant discrepancy remains unknown. Similarly, it is unclear what role common treatment moderators (e.g., age, sex, comorbidity) may play in this relationship.
Objectives:
To investigate the impact of a social skills intervention program on the discrepancy between parent- and child-rated social skills in adolescents with ASD.
Methods:
Adolescents and their parents (Table 1) completed the Social Skills Rating System (Gresham & Elliot, 1990) before and after participating in an empirically-supported intervention (McMahon et al., 2013) known to increase both self- and parent-reported social skills for youth with ASD (Lerner, Mikami, & Levine, 2011). These activities provided the participants with additional opportunities to practice social skills in an engaging and reinforcing group environment (Lerner, Mikami, & Levine, 2011).
Results:
Adolescents rated their own social skills higher than their parents rated them both pre-intervention (t(44) = 7.84, p < .001) and post-intervention (t(44) = 6.53, p < .001). However, the discrepancy between child- and parent-reports was significantly smaller post-intervention than pre-intervention (t(44) = 2.16, p = .04); this effect was driven by an increase in parent-reported scores rather than a reduction in child reported scores. Hierarchical multiple regression analyses revealed that older adolescents experienced a relatively greater reduction in discrepancy than their younger peers (Table 2).
Conclusions:
This is the first study to assess the impact of a social skills intervention program on the discrepancy between parent- and child-rated social skills for adolescents with ASD. Results indicated that while adolescents with ASD overestimate their abilities relative to parent-report even after undergoing social skills treatment, the magnitude of discrepancy reduces following a social skills intervention program. This indicates two potential explanations for our findings: either the adolescents are showing real improvements and their ratings that reflect increased skill and awareness of the skills, or we are observing parents’ tendency to report changes due to expectancy effects that are common within intervention research (Jones et al., 2017). In order to tease this apart and discover the cause of the discrepancy decrease, further research should employ the use of peer-report and observational data of social skills. This will help us to understand if social skills interventions are truly increasing social skills and/or self-awareness.