Physical Activity and Depressive Symptoms in Guangdong University Students: A Cross-Sectional Moderated Serial Indirect-Association Model Involving Self-Esteem, Positive Psychological Capital, and Perceived Social Support

Background: This cross-sectional study examined whether physical activity was indirectly associated with depressive symptoms through self-esteem and positive psychological capital and whether these associations varied by perceived social support. Methods: Data from 473 students at three Guangdong universities were used to assess physical activity, depressive symptoms, self-esteem, positive psychological capital, and perceived social support. Serial and moderated serial indirect associations were estimated using PROCESS v4.1 (Model 6; a custom moderated specification equivalent to PROCESS Model 89 for the joint outcome equation; 5000 bootstrap resamples), adjusting for gender, grade, and hometown, with three interactions entered simultaneously in the same outcome equation. Results: Physical activity was negatively associated with depressive symptoms (total association: B = −0.175, 95% CI [−0.215, −0.138]; direct association: B = −0.079, 95% CI [−0.120, −0.039]), and the total indirect association was B = −0.097 (95% CI [−0.126, −0.070]), with specific indirect associations through self-esteem (B = −0.031), positive psychological capital (B = −0.048), and the serial pathway (B = −0.017); the positive psychological capital pathway was significantly larger than the serial pathway (difference = −0.031, 95% CI [−0.048, −0.016]). Perceived social support moderated the associations of physical activity (B = 0.0043, p = 0.009), self-esteem (B = 0.0256, p < 0.001), and positive psychological capital (B = 0.0062, p < 0.001) with depressive symptoms in the simultaneous custom moderated outcome equation (R2 = 0.449); all three indices of moderated mediation were significant, and conditional indirect associations were stronger at lower support, with the self-esteem pathway no longer significant at high support. Conclusions: Physical activity was cross-sectionally associated with fewer depressive symptoms via independent and serial indirect associations involving self-esteem and positive psychological capital, with perceived social support showing a compensatory moderating pattern; temporal or causal ordering was not established.