Assessing Measurement Equivalence in Ordered-Categorical Data

Résumé 0

"Assessing measurement equivalence in the framework of the common factorlinear models (CFL) is known as factorial invariance. This methodology isused to evaluate the equivalence among the parameters of a measurement model among different groups. However, when dichotomous, Likert, or ordered responses are used, one of the assumptions of the CFL is violated: the continuous nature of the observed variables. The common factor analysis of ordered-categorical data (CFO) has been described in several works, ut none evaluate its power and Type I error rate in the evaluation ofmeasurement equivalence (ME). In this simulation study, we evaluated MEunder four different conditions: size of group (300, 500 and 1000), type ofDIF (thresholds, loadings), amount of DIF (0.25, 0.40), and equality/impactof the distributions. The parameters used for the data generation came fromone scale with nine items with three ordered categories. The results wereevaluated according to three decision rules: a) the significance of thedifference in chi-square values obtained in two nested models, b) thesignificance of the difference in chi-square values between two nestedmodels with Bonferroni corrections, and c) the difference between thevalues of the Comparative Fix Index (CFI) obtained in two nested models.The results showed good power as well as good control of the false positivesfor both the chi-square Bonferroni correction and CFI difference index."