Evaluating the Effect of Planned Missing Designs in Structural Equation Model Fit Measures

In a planned missing design, the nonresponses occur according to researcher’s will, with goal of increasing data quality and avoiding overly extensive questionnaires. When adjusting structural equation model data, there are different criteria evaluate how theoretical fits observed root mean square error approximation (RMSEA), standardized residual (SRMR), comparative fit index (CFI) Tucker–Lewis (TLI) being most common. Here, I explore effect due specific design—the three-form design—on mentioned indices when model. A simulation study was conducted correctly specified one misspecified correlation between factors. The CFI, TLI SRMR affected by nonresponses, particularly small samples, low factor loadings numerous variables. existence considering models causes unacceptable values for all four indexes under analysis, namely strong factors is considered. results shown here were performed simsem package in R full information maximum-likelihood method used handling during fitting.