Psychometric Properties of the General Self-Esteem Test (GSET) in Peruvian Teenagers’ students from Lima

Grajeda Montalvo, Alex Teófilo; Corvetto Castro Giovanni Jeffrey; Danielli Roca, Juan José; Quintana Peña, Alberto Loharte; Roggero Rebaza, Susana Consuelo; Solano Guillen, Ynes Eliana; Sotomayor Prado, Juan Carlos; Ore Sandoval, Shirley Rossmery

doi:10.61707/ezdhsg71

Authors

Grajeda Montalvo, Alex Teófilo Docente de la Facultad de Psicología de la Universidad Nacional Mayor de San Marcos. Autor para correspondencia
Corvetto Castro Giovanni Jeffrey Docente de la Facultad de Educación de la Universidad Nacional Mayor de San Marcos
Danielli Roca, Juan José Docente de la Facultad de Psicología de la Universidad Nacional Mayor de San Marcos
Quintana Peña, Alberto Loharte Docente de la Facultad de Psicología de la Universidad Nacional Mayor de San Marcos
Roggero Rebaza, Susana Consuelo Docente de la Facultad de Psicología de la Universidad Nacional Mayor de San Marcos
Solano Guillen, Ynes Eliana Docente de la Facultad de Psicología de la Universidad Nacional Mayor de San Marcos
Sotomayor Prado, Juan Carlos Docente de la Facultad de Psicología de la Universidad Nacional Mayor de San Marcos
Ore Sandoval, Shirley Rossmery Docente de la Escuela de Psicología de la Universidad César Vallejo

DOI:

https://doi.org/10.61707/ezdhsg71

Keywords:

Saving Self-esteem, Psychometric properties, Validity, Reliability, Scale

Abstract

The psychometric properties of the General Self-Esteem Test (GSET) have been investigated in Peruvian teenagers’ students from Lima. For this, 1,392 participants of both sexes, with ages ranging from 12 to 18 years, took the test. The results have shown that statistically the items meet the necessary values to be accepted. Likewise, the validity of the structure assigned to the original four-domain model has been optimally accepted since the CFA determined the following: CFI=.99; TLI=.99; GFI=1.00, PFGI=.70; RMSEA=.025; SRMR=.030. On the other hand, with respect to reliability, two formulas were worked on (split half and KR20), obtaining in both cases a coefficient of .89, which is considered very high reliability. Finally, a single scale has been developed, without differentiating by sex, with the 25th, 50th and 75th percentiles, because the study of significant differences shows a small size in the magnitude of the effect.