Abstract
We investigate a method to estimate the combined effect of multiple continuous/ordinal mediators on a binary outcome: (a) fit a structural equation model with probit link for the outcome and identity/probit link for continuous/ordinal mediators, (b) predict potential outcome probabilities, and (c) compute natural direct and indirect effects. Step 2 involves rescaling the latent continuous variable underlying the outcome to address residual mediator variance and covariance. We evaluate the estimation of risk-difference- and risk-ratio-based effects (RDs, RRs) using the maximum likelihood (ML), mean-and-variance-adjusted weighted least squares (WLSMV) and Bayes estimators in Mplus. Across most variations in path-coefficient and mediator-residual-correlation signs and strengths, and confounding situations investigated, the method performs well with all estimators, but favors ML/WLSMV for RDs with continuous mediators, and Bayes for RRs with ordinal mediators. Bayes outperforms ML/WLSMV regardless of mediator type when estimating RRs with small potential outcome probabilities and in two other special cases. An adolescent alcohol prevention study is used for illustration.
Original language | English (US) |
---|---|
Pages (from-to) | 368-383 |
Number of pages | 16 |
Journal | Structural Equation Modeling |
Volume | 23 |
Issue number | 3 |
DOIs | |
State | Published - May 3 2016 |
Keywords
- binary outcome
- causal inference, continuous mediators
- causal mediation analysis
- multiple mediators
- ordinal mediators
- structural equation modeling
ASJC Scopus subject areas
- General Decision Sciences
- General Economics, Econometrics and Finance
- Sociology and Political Science
- Modeling and Simulation