Indirect effects in MLR PreviousNext
Mplus Discussion > Categorical Data Modeling >
 Douglas Bruce posted on Tuesday, December 17, 2013 - 9:37 am
I'm testing a mediated path model that includes: (1) exogenous continuous variables, (2) dichotomous endogenous mediators, (3) dichotomous endogenous outcome variables

I've included the following commands:
Analysis: Estimator = MLR;
Output: Sampstat standardized;

When estimating indirect effects in my model using MLR, I receive the following error message:

MODEL INDIRECT is not available for analysis with ALGORITHM=INTEGRATED

Any suggestions how to correct this?

 Bengt O. Muthen posted on Tuesday, December 17, 2013 - 3:59 pm
I would not use ML (or MLR) for indirect effects when the mediators are binary. You should use either WLSMV or Bayes estimation in which case you work with continuous latent response variables for the binary variables and this makes regular definitions of indirect effects reasonable. This is explained more fully in my paper on our website:

Muthén, B. (2011). Applications of causally defined direct and indirect effects in mediation analysis using SEM in Mplus.
 Douglas Bruce posted on Wednesday, December 18, 2013 - 8:07 am
Thank you Bengt!
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message