A0583
Title: Evaluating the effect of planned missing designs in the structural equation models fit measures
Authors: Paula C R Vicente - Lusofona University (Portugal) [presenting]
Abstract: Missing or incomplete data represent a persistent problem in several studies. In a planned missing design, the non-responses occur according to the researcher's will, and the purpose of using such a design is to increase the quality of the data, avoiding the effort of inquiry. On the other hand, the estimation of a structural equation model consists in finding estimates for the model parameters that result in a variance-covariance matrix with the best fit to the theoretical model considered. There are different criteria to evaluate how well the theoretical model fits the observed data. The fit indices usually used in this type of modeling are the root mean square error of approximation (RMSEA), root mean square residual (SRMR), comparative fit index (CFI) and Tucker-Lewis Index (TLI). The aim is to explore the effect of the non-responses due to a planned missing design on the mentioned fit indices. A simulation study was conducted, considering different models, sample sizes, number of indicators, factor loadings and correlation between factors. For each simulated condition, 1000 replications were generated using the simsem package in R. Recommendations for best practices are discussed.
