Psychometric Evaluation of the Revised Exercise Addiction Inventory (EAI-R) Among Chinese College Students.

The present study tested the psychometric properties of the Chinese Revised Exercise Addiction Inventory (EAI-R). In Study 1, 800 non-sport and 357 sports science students completed the Chinese EAI-R, the Exercise Dependence Scale (EDS), the Compulsive Exercise Test (CET), and the Eating Disorder Inventory (EDI). They also reported their weekly exercise frequencies and exercise length per workout. An exploratory factor analysis (EFA) was conducted. Three weeks later, 94 students were re-administered the Chinese EAI-R to determine its test–retest reliability. In Study 2, 398 college students completed the Chinese EAI-R. A confirmatory factor analysis (CFA) was conducted and measurement invariance across gender was evaluated. In Study 1, the EFA yielded a one-factor structure after excluding one item (interpersonal conflict) from the Chinese EAI-R. The correlation between the Chinese EAI-R and EDS (r = 0.60) reflected concurrent validity. The risk of exercise addiction was higher among males than females in both samples and among sports science students compared to non-sports science students. The Chinese EAI-R also correlated with the CET score (r =0.26) but not the EDI score (r = 0.01). Therefore, the convergent and divergent validity of the Chinese EAI-R was supported. McDonald's omega was 0.79 for the Chinese EAI-R items among non-sports science students and 0.84 among sports science students. The scale's test–retest reliability was 0.75 over 3 weeks. Finally, Study 2 confirmed the unidimensional structure of the scale while also supporting its measurement invariance between males and females. The Chinese EAI-R appears to be a valid and reliable tool for assessing the risk of exercise addiction among Chinese college students. [ABSTRACT FROM AUTHOR]