Mediation Analysis with an Event-History Mediator: An Empirical Data Illustration and Simulation Study

This study introduces an event-history mediation model, demonstrating how event-history indicators can function as mediators within a Bayesian framework. To illustrate its practical application, we apply the method to data from the National Longitudinal Study of Youth, examining indirect effects in an event-history mediation context. We detail the process of model comparison, testing the mediation effects, and interpreting results within the Bayesian framework. Additionally, we conducted a small Monte Carlo simulation study to the models’ performance under various conditions, confirming its robustness in detecting mediated effects. The study concludes with implications for future research, highlighting the potential for expanding the application of mediation models to include event-history indicators as mediators, thereby broadening the scope and utility of mediation analysis in longitudinal research contexts.