Physical activity and short video addiction among college students: the serial mediating roles of self-efficacy and loneliness

Objective This study examined the relationship between physical activity and short video addiction among college students, with a particular focus on the mediating roles of self-efficacy and loneliness. Methods A cross-sectional survey was conducted among 581 students from three universities. Measures included validated scales assessing physical activity, self-efficacy, loneliness, and short video addiction, with demographic covariates (age, grade, major type [STEM vs. non-STEM], and urban–rural background). Data were analyzed using SPSS 27.0 and Hayes’ PROCESS macro (Model 6). Bootstrap analysis (5,000 resamples) was employed to test mediation effects with 95% confidence intervals. Results Physical activity was positively associated with self-efficacy and negatively associated with loneliness and short video addiction. Self-efficacy was negatively associated with both loneliness and short video addiction, whereas loneliness showed a positive association with short video addiction. Mediation analyses showed that physical activity reduced short video addiction both directly ( β = −0.074, p < 0.001) and indirectly through self-efficacy and loneliness, including a serial mediation pathway. The total indirect effect was −0.0330 (95% CI [−0.0502, −0.0179]), accounting for 30.8% of the total effect. Conclusion Physical activity reduces short video addiction directly and indirectly by enhancing self-efficacy and reducing loneliness, with a significant sequential mediation pathway. These findings highlight potential psychological mechanisms and provide practical implications for interventions targeting short video addiction among college students.