Structural Equation Modeling and selection tests - Case of the selection process of the State University of Ceará

Abstract

The objective of this study is to make use of Structural Equation Modeling (SEM) in evaluating the model employed by the State University of Ceará (Uece) in the selection process of prospective students, which is based on the Classical Test Theory (CCT), as well as to propose adjustments to this model through the use of Second Order Factor Analysis and Regression Analysis, via MME. The data set used is comprised of results of 11,060 admission exams from the year 2018.1. AMOS/SPSS-v.22 software was used to process the data, obtaining the following quality indicators: CFI (Comparative Fit Index) = 0.925; GFI (Goodness-of-fit Index) = 0.965; TLI (Tucker Lewis Index) = 0.922, and RMSEA (Root Mean Square Error of Approximation) = 0.019. Together, these indicators demonstrate that the model is robust and highly consistent, presenting a R² (Pearson Correlation Coefficient) = 0.965, indicating that the proportion of covariates, observed among the manifest variables and explained by the adjusted model, is very significant. All the variables of the adjusted model present elevated regression coefficients with values between 0.874 e 0.996, allowing for good discrimination between applicant scores, especially for those with the same number of correct question responses.

Educational Evaluation; Structural Equation Modeling (SEM; Confirmatory factor analysis (CFA; University admission examination