Log In
Home Annual Meeting Journal of Behavioral Data Science Log in

Trace: isdsa2019c1

isdsapress:books:isdsa2019:isdsa2019c1

Latent Growth Curve Models with VAR Residuals for Longitudinal Mediation Analysis

Xiao Liu
University of Notre Dame, Notre Dame IN 46556, USA
xliu19@nd.edu

Abstract. Mediation analysis using longitudinal data has become increasingly popular. To perform longitudinal mediation analysis, different models have been proposed, such as the latent growth curve mediation model (LGCM) and the cross-lagged panel mediation model (CLPM). In the current study, we proposed an alternative longitudinal mediation model (referred to as LGCM-CLRM), where a system of latent growth curve models are used to describe the deterministic inherent trajectories of each individual and a vector autoregressive model is used to describe the within-individual stochastic deviations from the latent trajectory. Compared to the existing longitudinal mediation models, the proposed model allows the mediation effects in both level-1 and level-2 models, and thus could disentangle different types of mediation effects. The proposed model can be estimated in the multilevel structural equation modeling framework. Simulation studies were performed to evaluate the estimation quality. We also provided a real data example for illustration.

Keywords: Longitudinal mediation analysis • Growth curve modeling • Vector autoregressive residuals.

DOI: https://doi.org/10.35566/isdsa2019c1


ISDSA About Membership Jobs at ISDSA Privacy ISDSA Press About Journal of Behavioral Data Science Books Annual Meeting Current Meeting Donate ISDSA is an exempt organization under section 501(c)(3) of the Internal Revenue Code. To make tax deductible contribution for the growth of ISDSA, click here.