Introduction

This chapter discusses latent growth curve models (LGCM) in detail and provides an introduction to LGCMs, including the estimation of univariate growth curves, the interpretation of growth parameters, and an examination of covariates of growth curves. There are a number of traditional analytical approaches that allow researchers to document change across time, including regression methods, mean comparisons, and repeated measures analysis of variance (RMANOVA). The advent of structural equation modeling (SEM) and multi-level modeling, methods of estimation that are sensitive to within-individual change and inter-individual differences in within-individual change are gaining popularity. Latent growth curve modeling addresses important methodological concerns for analyzing panel data. A latent confirmatory factor model (LCFM) extended to a curve-of-factors growth curve model (CFM) and a parallel process latent growth curve model (PPM) extended to a factor-of-curves growth curve model. The measurement errors can be accounted for in first-order growth curve models (PPM), which serve as a measurement model for a factor-of-curves growth curve model (FCM).