Evaluating the Associations Between Exercise Adherence and Related Factors among Students Enrolled in University General Physical Education Courses : A Meta-Analysis

Purpose: This study aims to evaluate the associations between exercise adherence, operationalized as adherence behavior and adherence intention, and related variables among students enrolled in university general physical education courses using a meta-analytic approach. Method: Relevant studies were systematically searched in Korean academic databases and Google Scholar. A total of 21 studies, yielding 42 effect sizes, were included based on PRISMA guidelines and predefined eligibility criteria. To account for the non-independence of multiple effect sizes extracted from the same study, a three-level random-effects model was applied using the rma.mv function in the metafor package in R. Effect sizes were computed by transforming correlation coefficients (r) into Fisher’s Z values. Subgroup analyses were conducted based on the operationalization of exercise adherence (behavior vs. intention) and the categories of correlates (intrapersonal psychological factors vs. learning-experience/environment factors). Publication bias was rigorously assessed using funnel plots, Egger’s regression test, the limit-estimate method, and trim-and-fill analysis. Results: First, the overall pooled correlational effect between exercise adherence and related variables was moderate (ES(r)=.405). Second, effect sizes for exercise adherence behavior (ES(r)=.415) and exercise adherence intention (ES(r)=.388) were both moderate, and the difference between these outcome levels was not statistically significant. Third, the learning-experience/environment factor category showed a significantly larger effect size (ES(r)=.465) than the intrapersonal psychological factor category (ES(r)=.278;t=2.810, p=.014). Fourth, among specific constructs, exercise flow showed the strongest association with exercise adherence behavior (ES(r)=.646), whereas class satisfaction showed the strongest association with exercise adherence intention (ES(r)=.543). Worry showed a negative association with exercise adherence intention (ES(r)=−.233). Effect sizes based on a single study were reported as descriptive statistics only and should be interpreted with caution. Conclusion: These findings suggest that both intrapersonal and environmental/learning factors are critical to understanding exercise adherence in university physical education. As this study is based on a correlational meta-analysis, causal inferences are precluded; however, the identified correlates provide a robust foundation for future experimental and longitudinal research aimed at promoting sustained physical activity among students.