OBJECTIVE: This study aims to explore the mechanisms underlying exercise commitment, exercise atmosphere, exercise self-efficacy, and exercise adherence among college students, as well as the impact of gender differences on these mechanisms.
METHODS: Using a stratified cluster sampling method, 984 college students (aged 19.74 �� 1.25 years) from six universities in Anhui, Shandong, and Shaanxi provinces were randomly selected, with 403 male students and 581 female students. They had completed the Exercise Commitment Scale (ECC), Physical Exercise Atmosphere Scale (PEAS), Exercise Self-Efficacy Scale (ESES), and Exercise Adherence Scale (EPS). The study variables were analyzed in sequence for reliability and validity, correlation analysis, regression analysis, structural equation model testing, and bias-corrected percentile Bootstrap testing using SPSS 23.0 software and Hayes' (2013) Process plug-in.
RESULTS: Exercise commitment significantly positively predicted exercise adherence (�� = 0.796, p < 0.01), exercise commitment significantly positively predicted exercise environment and exercise self-efficacy (�� = 0.645, p < 0.01, �� = 0.356, p < 0.01), exercise environment significantly positively predicted exercise self-efficacy and exercise adherence (�� = 0.344, p < 0.01, �� = 0.144, p < 0.01), and exercise self-efficacy significantly positively predicted exercise adherence (�� = 0.934, p < 0.01). The mediating role of exercise environment in the relationship between exercise commitment and exercise adherence does not exist in male college students, while it exists in female college students.
CONCLUSIONS: Exercise atmosphere and exercise self-efficacy play a mediating role between exercise commitment and exercise adherence, with a total mediating effect value of 0.796. This study indicates that exercise commitment not only directly predicts exercise adherence, but also indirectly predicts exercise adherence through the chained mediating effects of exercise atmosphere and exercise self-efficacy. The above-mentioned mediating effect exists in female college students, while some mediating effects do not exist in male college students, thus gender has a significant impact on this mediating effect.
