Growth Curve Episode 9: Choosing Between Multilevel & Structural Equation Approaches
In this final installment of our growth curve modeling series, Patrick explores the various advantages and disadvantages to estimating growth models within a multilevel model (MLM) and structural equation model (SEM) framework...
Patrick begins by briefly reviewing how the MLM and SEM each conceptualize growth from different traditions. He then describes the broad conditions under which the MLM and SEM are numerically identical and will provide equivalent results when applied to the same data. He then focuses on the specific conditions under which the standard MLM and SEM approaches differ from one another and he describes potential advantages of each. He concludes with recommendations for the choice of ideal approach within a given application. A few suggested readings include:
Bauer, D. J. (2003). Estimating multilevel linear models as structural equation models. Journal of Educational and Behavioral Statistics, 28, 135-167.
Chou, C. P., Bentler, P. M., & Pentz, M. A. (1998). Comparisons of two statistical approaches to study growth curves: The multilevel model and the latent curve analysis. Structural Equation Modeling: A Multidisciplinary Journal, 5, 247-266.
Curran, P.J. (2003). Have multilevel models been structural equation models all along? Multivariate Behavioral Research, 38, 529-569.
Willett, J. B., & Sayer, A. G. (1994). Using covariance structure analysis to detect correlates and predictors of individual change over time. Psychological bulletin, 116, 363-381.