Abstract
Exercise Self-Efficacy Scale (ESES) is widely used to assess individuals' exercise self-efficacy through self-reporting. It includes one factor and 18 items that gauge one's confidence in exercising under various conditions, such as when tired, stressed, or in unfavourable weather. Evidence indicates that differences in the original factor structure and psychometric properties were observed across different populations, including school-age students, university students, and adults. This study examined the factor structure, reliability, convergent validity, and measurement invariance of the Chinese version of the ESES among secondary school students. Data from 856 students (age: M = 13.80, SD = 0.94) were analysed using exploratory and confirmatory factor analyses to identify the best-fitting factor structure. The reliability and convergent validity were tested using the collected data. Configural, metric, and scalar invariances, as well as the likelihood ratio test, were tested for measurement invariance. A 14-item, two-factor structure of the ESES, consistently demonstrated the best fit among secondary school students. The two-factor structure showed strong internal consistency reliability (McDonald's Omega of 0.921 and 0.843) and satisfactory convergent validity (average variance extracted values of 0.582 and 0.478, composite reliability values of 0.917 and 0.845). Multi-group confirmatory factor analysis (likelihood ratio test, p > 0.01) revealed scalar measurement invariance across sex, ethnic backgrounds, grades, and school locations. These findings suggest that the 14-item, two-factor Chinese version of the ESES is suitable for use with secondary-school students. Future studies could confirm these findings by examining the 14-item, two-factor ESES in diverse samples, considering ethnicity, socioeconomic status, and age range.