Path Analysis of Psychological Resilience as a Mediator in the Relationship Between Physical Activity and Depressive Symptoms in University Students

Background This study aims to examine the relationship between physical activity, psychological resilience, and depressive symptoms, specifically investigating the effects of exercise intensity, duration, and frequency on depressive symptoms and the individual subcomponents of psychological resilience. Methods A total of 1,156 university students were recruited from a university in Shanghai through convenience sampling. Data were gathered using the Physical Activity Rating Scale (PARS-3), Connor-Davidson Resilience Scale (CD-RISC), and Beck Depression Inventory-Ⅱ (BDI-Ⅱ). Statistical analyses, including one-way ANOVA, correlation analysis, regression analysis, and mediation analysis, were conducted to analyze the data. Results Significant correlations were found between PARS-3, CD-RISC, and BDI-Ⅱ. Depressive symptoms in university students showed significant negative correlations with CD-RISC (r=-0.357, P < 0.001), resilience (r=-0.346, P < 0.001), strength (r=-0.349, P < 0.001), and optimism (r=-0.280, P < 0.001). PARS-3 was significantly negatively correlated with depressive symptoms (r=-0.121, P < 0.001), and positively correlated with CD-RISC (r = 0.191, P < 0.001), resilience (r = 0.183, P < 0.001), strength (r = 0.180, P < 0.001), and optimism (r = 0.169, P < 0.001). Exercise intensity was positively correlated with CD-RISC (r = 0.148, P < 0.001), resilience (r = 0.150, P < 0.001), strength (r = 0.150, P < 0.001), and optimism (r = 0.115, P < 0.001). Exercise duration was positively correlated with CD-RISC (r = 0.169, P < 0.001), resilience (r = 0.159, P < 0.001), strength (r = 0.167, P < 0.001), and optimism (r = 0.149, P < 0.001). Exercise frequency showed significant positive correlations with CD-RISC (r = 0.206, P < 0.001), resilience (r = 0.180, P < 0.001), strength (r = 0.219, P < 0.001), and optimism (r = 0.192, P < 0.001). Exercise intensity was negatively correlated with BDI-Ⅱ (r=-0.086, P < 0.001), exercise duration was negatively correlated with BDI-Ⅱ (r=-0.095, P < 0.001), and PARS-3 was negatively correlated with BDI-Ⅱ (r=-0.121, P < 0.001). Additionally, the subdimensions of CD-RISC partially mediated the relationship between PARS-3 and depressive symptoms, with the direct effect reaching 48.89%. The effect of mediation 1 was 26.67%, and mediation 2 accounted for 22.22%. Conclusion CD-RISC partially mediates the relationship between PARS-3 and depressive symptoms among university students. Lower PARS-3 levels are associated with higher depressive symptom scores and lower CD-RISC scores. This effect is closely linked to exercise intensity, duration, and frequency, demonstrating a significant "dose-response" relationship. It is recommended to implement appropriate exercise interventions based on the severity of symptoms. Universities should encourage students to engage in more high-intensity, long-duration, and frequent PARS-3 exercises to prevent and alleviate depressive symptoms in university students.