Dichotomous Mediator PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Philipp B. posted on Thursday, September 06, 2012 - 9:34 am
Hello B., hello L.,

I would like to test following mediator relationship/path model in Mplus:

Continuous/Dichotomous var --> Dichotomous var --> Dichotomous var

I treat the Dichotomous mediator (Yes/No) as a continuous var and use a logistic regression (ML/Categorical are) for the equation of the endogenous Variable.

Is this okay?

Kind regards,
P.
 Bengt O. Muthen posted on Thursday, September 06, 2012 - 9:55 am
Short answer: No. Long answer: You should read the paper on our web site under Mediational Modeling:

Muthén, B. (2011). Applications of causally defined direct and indirect effects in mediation analysis using SEM in Mplus.

The paper shows that there is a paradigm shift going on in mediational modeling with categorical outcomes and many SEMers are not sufficiently aware of this. The good news is that it can be done in Mplus as the paper's scripts show.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: