Comparing within-person effects from multivariate longitudinal models

Abstract

Several multivariate models are motivated to answer similar developmental questions regarding withinperson (intraindividual) effects between 2 or more constructs over time, yet the within-person effects tested by each model are distinct. In this article, the authors clarify the types of within-person inferences that can be made from each model. Whereas previous research has focused on detecting whether within-person effects exist over development, the present work can be used to understand the nature of these relationships. The authors compare each modeling approach using an example investigating the concurrent development of mother- child closeness and mother- child conflict. The findings demonstrate that fundamentally different conclusions about developmental processes may be reached depending on which model is used, and a framework is demonstrated for making sense of seemingly contradictory findings.