Including Time-Invariant Covariates in the Latent Growth Curve Model

Reinoud D. Stoel, Godfried van den Wittenboer, Joop J. Hox

Structural Equation Modeling A Multidisciplinary Journal · 2004 · 30 citations · 14 references

DOIFull text

Open access

Concepts

Abstract

Within the latent growth curve model (McArdle, 1986; Meredith & Tisak, 1990), timeinvariant covariates are generally modeled on the subject level, thereby estimating the effect of the covariate on the latent growth parameters (Willett & Sayer, 1994). Incorporating the time-invariant covariate in this manner may have some advantages regarding the interpretation of the effect, but may also be incorrect in certain instances. In this paper we will discuss a more general approach for modeling time-invariant covariates in latent growth curve models, in which the covariate is directly regressed on the observed indicators. The approach can be used on its own, to get estimates of the growth curves corrected for the influence of a third variable; or it can be used to test the appropriateness of the standard way of modeling the time-invariant covariates. It thus provides a test of the assumption that the relation between the covariate and the observed indicators is fully mediated by the growth parameters.

References

14