Latent Change Scores

Latent change scores represent differences across two or more sequential measurements within a structural equation model framework. Modeling change is important because it evaluates whether an effect is dependent on the amount of time elapsed and can examine the temporal order of effects. The general process for use of latent change scores is (1) behavioral, cognitive, or psychological data are collected at two or more consecutive measurements; (2) the lag between measurements is optimized to evaluate change; (3) data are fit to the structural components of the latent change score; (4) the latent change score is formed; and (5) evaluation of whether latent change scores are predictive of, and/or predicted by, variables of interest. The goal of this approach is often to evaluate changes within ...