Mediation analysis in latent growth c... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Caiyan Zhang posted on Wednesday, September 03, 2014 - 6:31 am
Dear Dr. Muthén,

I am running a mediation analysis in LGCM framework. I wonder whether bootstrap with CI for the mediation effect is available for categorical outcomes using ML(R) estimator?

Many thanks in advance,

 Bengt O. Muthen posted on Wednesday, September 03, 2014 - 3:20 pm
Mplus Version 7.2 added bootstrap with numerical integration - which is what ML(R) calls for with categorical outcomes.
 Caiyan Zhang posted on Wednesday, September 03, 2014 - 5:18 pm
Thank you for your answer!
 Caiyan Zhang posted on Thursday, September 18, 2014 - 11:36 am
There is a follow-up question that I want to ask. How to use model constraint command to specify and test the indirect effect when there are multiple pathways from X to the mediator? More specifically, in multiple growth process where growth factors in latent growth curve models are used as x(i1,s1) and mediator(i2,s2).

MODEL: ...snip..
i2 on i1;
s2 on s1;

outcome on i2;
 Bengt O. Muthen posted on Friday, September 19, 2014 - 6:04 am
You simply have several a*b parts, one for each X and M. They can be added up for a total indirect effect.
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