The Impact of Time and Age Heterogeneity on Latent Growth Model Estimates

• In Event: Roundtable Session 15
  In Roundtable Session: Nuanced Methods for Latent Growth Models

Abstract

Latent growth modeling (LGM) within the structural equation modeling (SEM) arena has become a popular method with which longitudinal data may be analyzed. Various time-coding strategies may be used with latent growth models that render different interpretations of the parameters of interest. Studies, though limited, have shown that increased interindividual heterogeneity can lead to inaccurate estimates in LGM. This paper assessed the impact of both ignoring and modeling time and age heterogeneity in latent growth models. Preliminary results indicate that parameter bias is substantial when interindividual heterogeneity is not appropriately modeled. It is hoped that this study will provide applied researchers with more information about the impact of time and age heterogeneity in latent growth models under various conditions.

Authors

• Tiffany Ann Whittaker, The University of Texas - Austin

• Wanchen Chang, The University of Texas - Austin

• Rommel Lazo Bunuan, The University of Texas - Austin