Unidimensional, bidimensional, or bifactorial? Structural evidence for the Rosenberg Self-Esteem Scale in Mexican adults
Description
The Rosenberg Self-Esteem Scale (RSES) is one of the most widely used measures worldwide; however, its factorial structure remains under debate. The aim of this study was to examine the internal structure of the RSES, compare alternative factorial models proposed in the literature, and estimate the reliability of its scores in Mexican adults aged 18 to 60 years. A total of 470 participants completed an online questionnaire. Confirmatory factor analyses were conducted using Weighted Least Squares with Mean and Variance adjusted estimation to compare four models: unidimensional, bidimensional, unidimensional with method effects, and bifactor. In addition, bifactor indices (Explained Common Variance, Percentage of Uncontaminated Correlations, and hierarchical omegas) were calculated. The bifactor model showed the best fit to the data. However, although bifactor indices indicated a predominance of the general factor, its hierarchical reliability was very low. These findings provide evidence on the functioning of the RSES in Mexican adults and suggest that both total and subscale scores should be interpreted with caution. Their use should not be arbitrary, but rather based on their empirical performance within the target population.
Files
Institutions
- Universidad Nacional Autónoma de MéxicoMexico City, Mexico City