Problematic social media use and academic burnout among university students: the mediating role of anxiety and the moderating role of physical activity

Background Problematic social media use (PSMU) has increasingly been identified as a behavioral risk factor for academic burnout among university students, yet the psychological mechanism that may account for this association and the conditions under which it is weaker remain unclear. This study examined whether anxiety carries part of the association between PSMU and academic burnout, and whether physical activity moderates the strength of this association. Methods A cross-sectional online survey was administered to 2,572 students at one university in Changsha, China, using the Bergen Social Media Addiction Scale (BSMAS), the Generalized Anxiety Disorder Scale (GAD-7), the Physical Activity Rating Scale-3 (PARS-3), and the Learning Burnout Scale for College Students. After adjustment for age, gender, and socioeconomic status (SES), a moderated mediation model was estimated using Model 5 of the PROCESS macro, with PSMU as the focal predictor, anxiety as the mediator, academic burnout as the outcome, and physical activity as the moderator of the residual direct path. Results PSMU was positively associated with academic burnout; at the mean level of physical activity, the conditional association was B = 0.260 ( p < 0.001). The indirect association through anxiety was 0.087, 95% CI [0.070, 0.105], corresponding to 25.1% of the conditional total association at the mean of the moderator; this proportion is reported as a descriptive quantity and is not given a causal interpretation. Physical activity moderated the residual association (interaction term B = −0.120, p < 0.001), such that the association was strongest among students with the lowest observed physical activity ( B = 0.351) and weakest, though still statistically significant, among those one SD above the mean ( B = 0.140). Conclusion Higher physical activity was associated with a weaker positive association between PSMU and academic burnout, but it did not eliminate this association; physical activity therefore cannot substitute for efforts that target problematic use itself. Because all variables were measured at a single time point, these findings describe patterns of covariation rather than causal effects. They suggest that reducing problematic use and promoting physical activity may be complementary goals, although no intervention was tested.