Psychometric properties and measurement invariance of an academic self-efficacy scale in college students from five Latin American countries

Our objective was to analyze the measurement invariance of the Academic Situations Specific Perceived Self-Efficacy Scale (ASSPSES) among Peruvian, Mexican, Colombian, Argentinean, and Brazilian college students, and its relationship with academic emotional exhaustion (AEE). In addition to the ASSPSES, the Emotional Exhaustion Scale (EES) was used. Two thousand one hundred forty-two college students (70.26% women) between the ages of 16 and 35 (M = 21.79 years) participated. The one-dimensional structure of ASSPSES was corroborated in each sample with a confirmatory factor analysis and the measurement invariance analysis was carried out with a multi-group factor analysis (MGFA) and with the differential item functioning. The relationship with the AEE was tested with the Pearson correlation coefficient. Regard the results, in all of the samples, the one-dimensional model presented adequate psychometric indicators with respect to both dimensionality and reliability. Similarly, regarding the analysis of measurement invariance, a strong variance was attained, and DIF is weak, which together with the MGFA results provides favorable evidence of measurement invariance. We conclude that ASSPSES is an invariant measure among the analyzed groups although replicating the study is necessary for the consolidation of the findings. These findings contribute to the understanding of the internal structure of the ASSPSES, something that had been awaiting evaluation, given how the scale is used in various contexts.