A Structural Equation Model for the Study of Factors Associated with School Performance

Abstract

In this paper is made a factorial analysis for the database of Spanish third-grade taken from TERCE 2013 Honduras. This analysis is made by the factor analysis (FA) and structural equation modeling (SEM). It is verified that the sample is valid to carry out FA and meets the assumption of multivariate normality. Exploratory factor analysis is performed to determine the number of factors to use. Additionally, confirmatory factor analysis is used to verify the most relevant fit indices such as CFI, TLI, RMSEA, SRMR which allow for statistical coherence observation. Finally, the structural equation model is carried out, estimation being performed by the maximum likelihood method (MLE) and to validate the model, it was checked for common method variance (CMV) using Harman's single-factor test and the Heterotrait-Monotrait Ratio of Correlations (HTMT) criterion. Regarding the associated factor results, it was observed that not only the characteristics of the institutions influence student performance but also those of the community.