Modeling latent growth with multiple indicators: a comparison of three approaches

Abstract

Latent growth curve models (LGCMs) are widely used methods for analyzing change in psychology and the social sciences. To date, most applications use first-order (single-indicator) LGCMs. These models have several limitations that can be overcome with multiple-indicator LGCMs. Currently, almost all multiple-indicator applications use the so-called second-order growth model (SGM; McArdle, 1988). In this article, we review the SGM and discuss 2 alternative, but less well-known, multiple-indicator LGCMs that overcome some of the limitations of the SGM: the generalized second-order growth model (GSGM) and the indicator-specific growth model (ISGM). In contrast to the SGM, the GSGM does not involve a proportionality constraint on the ratio of general to specific variance. The ISGM allows researchers to model indicator-specific growth. Both of these alternative models allow testing measurement invariance across time for state-variability components. We also present an empirical application regarding changes in self-reported levels of anxiety and discuss implications of the differences between the 3 models for applied research.