Three-Level Longitudinal Mediation With Nested Units: How Does an Upper Level Predictor Influence a Lower Level Outcome via an Upper Level Mediator Over Time?

• In Event: 25.018 - Advanced Analysis and Inference Issues

Abstract

Extending from the cross-lagged panel models and the 2-2-1 cross-sectional multilevel mediation model, we proposed a three-level longitudinal mediation model for modeling the causal process among variables across levels over time. Given the complexity of the proposed model, Bayesian estimation was used. A simulation study was conducted to examine the estimation accuracy of Bayesian estimation for the proposed model. Factors considered included average sample sizes for the lower-level units within each upper-level unit, numbers of upper-level units (J), numbers of time points (T), indirect/direct effect sizes, and variances and covariances of upper-level random effects. Preliminary results showed that a larger number of J could produce slightly more accurate average mediation effect estimates while fixing the overall sample size.

Author

• Qian Zhang, Florida State University