EFFECT OF SAMPLE SIZE ON THE ADJUSTMENT INDICES OF THE CONFIRMATORY FACTORIAL ANALYSIS ON THE PHQ - 9 QUESTIONNAIRE APPLIED TO THE PERUVIAN POPULATION

Abstract

In the present investigation, the objective was to determine the effect of the sample size on the adjustment indices of the confirmatory factor analysis on the PHQ - 9 questionnaire applied to the Peruvian population. The population was represented by 34,476 records of the ENDES - INEI - 2018 survey, and the sample size was made up of samples chosen through the Bootstrap poll with 100 replicates in the number of 50, 100, 250, 500, 750, 1000, 2500, 5,000 and 10,000 cases. The sample size had a direct effect on the indices of adjustment of the confirmatory factor analysis on the PHQ - 9 questionnaire applied to the Peruvian population; likewise, a direct effect with upward concavity in X², X²/gl, AIC and BIC, a direct effect with downward concavity in CFI, TLI, NNFI, NFI, RNI, GFI, PNFI and AGFI, and in some measures of goodness of fit an indirect effect on the RMSEA, RMR and SRMR. A direct and inverse effect of the sample size on the measures of goodness of fit is contrasted.