Impact of Misspecified Turning Points on Growth Trait Estimation in Multiple-Group Analyses

• In Event: 52.076 - Longitudinal Data Modeling

Abstract

In many experimental or quasi-experimental studies that aim to assess treatment effects on individual growth process, multiple-group latent growth models are used to compare groups for differences in change. Because growth rate may change after the treatment, piecewise growth curve models are often used. The conventional practice is to specify turning points a priori. However, it may not always be tenable because causal impacts could be delayed rather than immediate. The study examines the impact of turning points misspecification on growth parameter estimation and assesses the performance of a newly proposed procedure, multiple-group piecewise growth modeling with unknown turning points on growth parameter recovery. The study will draw applied researchers’ attention to the importance of correct specification of turning points.

Authors

• Ling Ning, University of California - Davis

• Wen Luo, Texas A&M University - College Station