ABSTRACT
Health and medical researchers are increasingly able to take advantage of data on large numbers of patients collected over longer periods of time. Such data may enable researchers to overcome limitations inherent in cross-sectional studies and make stronger inferences about hypothesized causality. Longitudinal analyses also necessitate complex modeling decisions regarding the relationships between time-varying variables. This chapter provides an overview and some examples of how structural equation models, including autoregressive and latent growth models, can be applied to answer some common types of research questions for data.
