An empirical study on exercise addiction and grit among college students: based on 35 exercise sessions

This prospective observational study used the number of mandatory running sessions as a core between-subject variable (three groups: < 35 sessions, = 35 sessions, > 35 sessions) and measurement time as a within-subject variable (pre-test, post-test). It mainly investigated two issues: first, between-group differences in exercise addiction tendency and grit among college students with different frequencies of mandatory running; second, the correlation characteristics between exercise addiction tendency and grit, so as to provide empirical evidence for college physical education practice.

A total of 1,611 college students (males: n = 930, females: n = 681, aged 18-22 years) were enrolled and divided into three groups by the number of mandatory running sessions: < 35 sessions (n = 380), = 35 sessions (n = 1,132), and > 35 sessions (n = 99). All participants joined regular campus mandatory running activities, with 35 sessions set as the compliance threshold, and no extra experimental interventions were applied. The Shapiro-Wilk test was used to examine data normality, and the Levene test to verify homogeneity of variance. Standard ANOVA was used for homogeneous data, and Welch ANOVA for heterogeneous data. Mixed-design ANOVA was conducted to test the main effects of group, main effects of time, and time × group interaction, accompanied by appropriate post-hoc tests. Correlation correction, sensitivity analysis, and Cronbach's α coefficient were adopted to ensure methodological rigor.

Baseline scores of exercise addiction tendency and grit in the > 35 sessions group were significantly higher than those in the other two groups, with no significant differences between the latter two. No significant pre-to-post changes were found in the Exercise Addiction Inventory (EAI) in any group. For the Original Grit Scale (Grit-O), marginally significant pre-to-post differences were detected in the > 35 sessions and < 35 sessions groups, while no significant difference was observed in the = 35 sessions group. The time × group interaction was not significant for EAI but significant for Grit-O. There was no substantial intrinsic correlation between exercise addiction tendency and grit; their weak correlation resulted from between-group differences. The number of running sessions showed a weak linear positive correlation with both indicators. Both scales presented good internal consistency (α > 0.7).

There are significant between-group differences in exercise addiction tendency and grit among college students with different numbers of mandatory running sessions. The baseline psychological traits of the high-frequency running group (> 35 sessions) are significantly higher than those of the medium- and low-frequency groups. Only the high-frequency group shows an improving trend in grit, and the number of running sessions has a significant moderating effect on grit enhancement, whereas the change pattern of addiction tendency is not affected by running frequency. Exercise addiction tendency and grit have no substantial intrinsic connection; their apparent correlation arises from superimposed between-group differences. The number of running sessions shows a linear dose-response relationship with both indicators. This study provides empirical references for optimizing campus running intervention programs and cultivating psychological qualities among college students.