Relationships among anxiety, psychological resilience, and physical activity in university students: variable-centred and person-centred perspectives

Background Anxiety is significantly correlated with levels of physical activity in university students. This research assessed the effects of anxiety on engagement in physical activity and explored the potential mediating function of psychological resilience. Additionally, latent profile analysis (LPA) was employed to identify distinct subtypes based on anxiety and resilience levels, and to explore their associations with physical activity. Methods Utilizing a non-probability convenience sampling approach, this cross-sectional study recruited a total of 1,436 collegiate participants from multiple universities. Data collection was carried out with the Generalized Anxiety Disorder Scale (GAD-7), the abbreviated Connor-Davidson Resilience Scale (CD-RISC-10), and the Physical Activity Rating Scale (PARS-3). Data analysis included mediation effect analysis via Bootstrap methods (Model 4) and latent profile analysis (LPA). Results Anxiety demonstrated a significant negative association with physical activity ( β = −0.307, p < 0.001) and was identified as a substantial negative predictor of psychological resilience ( β = −0.413, p < 0.001). A significant positive association was observed between psychological resilience and physical activity ( β = 0.178, p < 0.001). The analysis confirmed the role of psychological resilience as a significant mediating variable in the pathway linking anxiety to physical activity, accounting for 24% of the total effect. Furthermore, latent profile analysis identified three distinct typologies: High Anxiety-Low Psychological Resilience (HA-LPR; 10.38%), Moderate Anxiety-Moderate Psychological Resilience (MA-MPR; 62.74%), and Low Anxiety-High Psychological Resilience (LA-HPR; 26.88%). These latent classes demonstrated statistically significant differences in physical activity levels ( F = 209.655, p < 0.001). Conclusion Results demonstrated that anxiety affects physical activity both directly and indirectly, with the latter effect occurring through the channel of psychological resilience. Latent profile analysis identified three distinct profiles among college students based on anxiety and psychological resilience: High Anxiety-Low Psychological Resilience, Moderate Anxiety-Moderate Psychological Resilience, and Low Anxiety-High Psychological Resilience. Marked variations in physical activity levels were observed among these subgroups. The results underscore the complex relationships among mental health indicators and health behaviors within the collegiate population. The delineation of distinct profiles offers practical implications for designing tailored intervention strategies.