Longitudinal social network analysis of perceived physical ability and peer exposure among adolescents in school-based physical activity

In school-based physical activity settings, adolescents’ perceived physical ability (PPA) and their exposure to peers’ PPA may shape physical activity behaviours and represent interrelated, modifiable factors. This study examined the longitudinal relationship between PPA and exposure to peers’ PPA among Korean adolescents in school-based physical activity. A total of 101 South Korean middle school students (all aged 14) fully participated in this longitudinal study. Participants self-reported PPA and nominated up to five peers (e.g., classmates) with whom they interacted in school-based physical activity, and reported the weekly frequency of four interaction types. Social network analysis was applied to longitudinal data collected at two time points within a year: Time 1 (T1), conducted at the end of the spring term; and Time 2 (T2), conducted at the end of the fall term. Exposure networks were analysed using UCINET and NetDraw. A cross-lagged panel model was fit in Stata 19. Exposure to peers’ PPA – reflecting peer number, interaction frequency, and peers’ PPA – was higher over time, as indicated by higher mean out-degree, a slight increase in mean exposure contribution per non-zero tie, and visual inspection of the sociogram. PPA at T1 significantly predicted exposure to peers’ PPA at T2, whereas exposure at T1 did not predict PPA at T2. An implication that can be drawn from these findings is that identifying low-PPA network isolates (i.e., adolescents who are unlikely to form natural connections in school-based activity) and using teacher-directed peer grouping may increase opportunities for peer exposure and support their physical activity.