Time-lagged effects of social support on exercise adherence among Korean college students: serial mediation of self-efficacy, athletic identity, and intrinsic motivation

Introduction Physical activity levels decline substantially during the college years, yet the psychological mechanisms linking social support to sustained exercise adherence remain insufficiently understood through longitudinal research. Grounded in Social Cognitive Theory, this study examined the time-lagged mediating mechanisms through which social support influences exercise adherence among Korean college students. Methods A three-wave longitudinal design was employed with 950 Korean college students (47.1% male; M age = 20.3 years). Social support and baseline exercise adherence were assessed at T1 (baseline); exercise self-efficacy and athletic identity at T2 (2 weeks later); and intrinsic motivation and exercise adherence at T3 (4 to 6 weeks later). Time-lagged mediation was tested using structural equation modeling with bias-corrected bootstrap confidence intervals (5,000 resamples). Results After controlling for baseline exercise adherence, the total effect of social support on T3 exercise adherence was small (β = 0.06, p = 0.061), while the direct effect was negligible (β = 0.00, p = 0.924). The total indirect effect was significant (β = 0.05, 95% CI [0.030, 0.073]), consistent with an indirect-only mediation pattern. Three significant indirect pathways were identified: (1) social support → self-efficacy → exercise adherence (effect = 0.029, 95% CI [0.013, 0.049]); (2) social support → self-efficacy → intrinsic motivation → exercise adherence (effect = 0.013, 95% CI [0.007, 0.021]); and (3) social support → athletic identity → intrinsic motivation → exercise adherence (effect = 0.004, 95% CI [0.002, 0.008]). Discussion These findings provide time-lagged evidence that social support is associated with exercise adherence primarily through enhancing self-efficacy, which influences adherence both directly and through intrinsic motivation. Athletic identity contributed indirectly via intrinsic motivation rather than directly predicting adherence. The results support Social Cognitive Theory's emphasis on self-efficacy as a key mediator linking environmental factors to health behaviors, and suggest that interventions combining social support enhancement with self-efficacy building may be most effective for promoting sustained physical activity among college students.