Generating Non-normal Data for Simulation of StructuralEquation Models Using Mattson’s Method

SEM researchers use Monte-Carlo simulations to ascertain the robustness of statisticalestimators and the performance of various fit indices under varying conditions of nonnormality.The efficacy of these Monte-Carlo simulations is closely related to thegeneration of non-normal data. Traditionally, SEM researchers have used approachesproposed by Fleishman (1978) and Vale and Maurelli (1983) to generate multivariate nonnormalrandom numbers. However, both approaches do not provide a method to determineunivariate skewness and kurtosis of the observed variables when a non-normal distributionis specified for some or all of the latent variables...