The general self-efficacy scale: New evidence of structural validity, measurement invariance, and predictive properties in relationship to subjective well-being in Serbian samples

Abstract

The General Self-Efficacy Scale (GSE) is one of the most frequently used measures of positive expectations, but its psychometric properties have been rarely examined using a prospective design. The research presented here evaluated the validity (construct, convergent, predictive), reliability (internal consistency, test-retest), and measurement invariance (across gender and time) of the General Self-Efficacy Scale (GSE). A total of 3667 undergraduate students from Serbia participated in three studies. Cross-sectional data were used in Study 1 and Study 2, and longitudinal data were used in Study 3. The results supported the GSE’s modified single-factor structure as well as strict measurement invariance across gender and time. The GSE demonstrated adequate internal consistency, moderate 4-month and 2-year test-retest reliability, and good convergent validity in relation to measures of positive expectations and subjective well-being. After controlling for the initial levels of well-being, the GSE showed limited predictive utility in predicting subjective well-being (positive affect, life satisfaction, emotional distress) over periods of 4 months and 2 years. Our findings suggest that use of the GSE for predictive purposes should be carefully examined in future studies.