Std coeff. for paths with binary cova... PreviousNext
Mplus Discussion > Categorical Data Modeling >
 Dino Gibertoni posted on Wednesday, June 26, 2013 - 6:28 am
dear drs. Muthén,
I have a doubt about how to calculate the standardized coefficient of an indirect effect involving binary covariates.

Specifically, I have paths from A to B to Y where A and B are binary and Y is continuous.
For direct effects, I'd estimate the STDY with the ratio of STDYX to stddev(X), but here I have an independent and a mediator. What should I use for stddev(X)?

thanks for your attention,
best regards
 Bengt O. Muthen posted on Wednesday, June 26, 2013 - 3:29 pm
It sounds like B is a binary mediator. The indirect effect is from the binary covariate A to the continuous distal outcome Y.

If this is a correct understanding of your case, the indirect effect is an effect from a binary A to a continuous Y and should therefore use STDY if you want to standardize.

Note that with a binary mediator you should (1) use WLSMV or Bayes, and (2) read my paper:

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

which is on our website.
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