ABSTRACT
The structural equation modeling (SEM) methodology offers a highly useful and readily applicable alternative framework allowing one to address comprehensively issues pertaining to studying change over time. This chapter refers this framework as latent change analysis (LCA), since its focus is the study of change over time in latent dimensions. LCA has been primarily formalized and popularized in the social and behavioral sciences over the past couple of decades as a general means for studying growth or decline at the unobserved variable level as well as their correlates and predictors. To permit the study of correlates and predictors of change, the chapter extends the level-and-shape (LS) model to include putative covariates and relate them to the Level and Shape factors. Specifically, by focusing on the correlations of the Shape factor with presumed predictors, one obtains estimates of the degrees of interrelationship between the latter and overall latent change.
