Abstract
BACKGROUND
Physical exercise promotes mental health among college students, but existing studies often assess general mental health or limited psychological indicators, making it difficult to capture comprehensive healthy psychological effects after exercise. This study aimed to develop and validate the Healthy Psychological Effects of Physical Exercise Scale for College Students .
METHODS
An initial scale was developed through theoretical analysis, literature review, semi-structured interviews, and open-ended questionnaires. In an initial sample of 612 college students, item analysis and exploratory factor analysis with varimax rotation were conducted to screen items and identify factor structures. A formal sample of 945 participants was used to examine reliability and validity through Cronbach's α, composite reliability (CR), average variance extracted (AVE), confirmatory factor analysis (CFA), and criterion-related validity testing. Test-retest reliability was assessed in 597 participants after 1 month.
RESULTS
The final scale contained 55 items across five subscales: emotional feelings, self-cognition, frustration coping, self-control and social adaptation efficacy, and interpersonal interaction, comprising 15 subdimensions. Exploratory factor analyses supported clear three-factor structures for each subscale, with cumulative variance explained ranging from 62.546 to 73.711%. CFA indicated good fit for the five subscales (χ 2 / df = 2.978-4.391; RMSEA = 0.046-0.060; SRMR = 0.016-0.021; CFI = 0.978-0.988; TLI = 0.971-0.984), and acceptable fit for the overall model ( χ 2 / df = 3.037; RMSEA = 0.046; SRMR = 0.041; CFI = 0.925; TLI = 0.921). Cronbach's α and CR for the 15 subdimensions ranged from 0.873 to 0.931, and AVE ranged from 0.680 to 0.736. Cronbach's α for the five subscale total scores ranged from 0.861 to 0.894. Test-retest reliability ranged from 0.706 to 0.781 for subdimensions and from 0.686 to 0.771 for subscale total scores. Criterion-related validity was supported by significant positive correlations with positive psychological indicators ( r = 0.229-0.383) and negative correlations with negative psychological indicators ( r = -0.398 to -0.224; all p < 0.001).
CONCLUSION
The scale demonstrates sound reliability and validity and can be used to assess exercise-related healthy psychological effects among college students.