Latent growth curve models (LGCM) became in recent years a very popular technique for longitudinal data analysis: they allow individuals to have distinct growth trajectories over time. Although the LGCM specified model structure imposes normality assumptions, the data analyst often faces data deviations from normality, implying mild, moderate or even severe values for skewness and or kurtosis. In the current research, a Monte Carlo simulation study was conducted in order to investigate the effect of observed data deviations from normality on goodness-of-fit indices. A new approach to generate multivariate non-normal distributed data was used: the VITA method. This method is a covariance model simulation method using regular vines. The dependency structure is determined by bivariate copulae and a nested set of trees. One thousand datasets were randomly generated from regular vines using Clayton copula and three marginal distributions (Normal, Student 3 and Gamma). The multivariate normal distribution was also used for data generation. LGCM with unconditional linear growth was considered. Three time points and distinct combi- nations of sample sizes were used. The impacts of such deviations on goodness-of-fit measures are discussed.

--

goodness-of-fit indices; LGCM,non-normality data,VITA method