Observational Coding Scheme for Adolescent Peer Social Support (2012)

Undergraduates: Kori Brady, Julia Shadur, M.A.

Faculty Advisor: Andrea Hussong
Department: Biostatistics

---

Peer social support is known to impact adolescents’ use of alcohol. One of the barriers to understanding how peer social support impacts an adolescent is the lack of reliable and valid indicators of enacted social support. Currently, the most prevalent methods of measuring peer social support only measure perceived social support, which may not always correspond to the actual enacted social support. Observational measures can help to overcome this weakness. This study focused on adapting observational coding schemes previously used with adults for the purpose of assessing social support present in adolescents’ interactions. The adapted coding scheme was then used to analyze interactions between 52 rising 9th grade friendship dyads. Inter-rater reliability between two independent coders was calculated across 25% of the tapes using an aggregate percent agreement across categories of social support (mean percent agreement= 80.4%). Construct validity was assessed by comparing frequency scores of positive and negative social support from the observational measure to scores for friendship intimacy and friendship conflict as indicated by the adolescent’s self-report (Network of Relationships Inventory). The positive frequency score showed stronger correlation with friendship intimacy (r=.19) than with friendship conflict (r= -.036). The negative frequency score showed a modestly stronger correlation with friendship conflict (r= .16) than with friendship intimacy (r=.099).