Dimensionality and Validity of the Rosenberg Self-Esteem Scale for Use With Latino Adolescents

Abstract

This study evaluated the construct validity and factor structure of the Rosenberg Self-Esteem Scale (RSES) using a large sample ( n = 814) of adolescents in Mexican, Salvadoran, and Guatemalan immigrant families who completed surveys in English. Results suggested that the RSES may be best specified as assessing two somewhat distinct yet related constructs. There was a high degree of measurement equivalence across the three nationality groups and also a high degree of equivalence across samples separated by generational status (adolescents born in the United States vs. not). There was relatively strong evidence to suggest that bidimensionality of the RSES for use with Salvadoran and Guatemalan adolescents is related to method effects associated with negatively worded items. Findings also suggested that two RSES factors (positive self-esteem and self-deprecation) assess substantively distinct elements of self-esteem as psychological control by mothers, age, and generational status were more strongly related to a self-deprecation (negatively worded items) factor than to a positive self-esteem factor. Implications for using the RSES in research on Latino adolescents are discussed.