ABSTRACT
Latent growth modeling (LGM) methodology provides a number of advantages to researchers studying change and development over time. The LGM describes a single individual's developmental trajectory and captures individual differences in these trajectories over time. Within the two-factor LGM specification, it is possible to test the adequacy of the hypothesized growth form as linear, using a specified growth function, or as nonlinear, using freely estimated parameters that describe the growth form with maximal fit to the data. In addition, with the use of multivariate LGMs, it is possible to determine whether development in one behavior covaries with other behaviors when assessed repeatedly. The flexibility of the multilevel covariance structure model makes it a particularly attractive analytic tool for LGM, and for investigations of growth and development among variables of interest with clustered and multilevel data. Despite numerous attractions, LGM is not always the appropriate analytical choice.
