Marginal models & level 2 variable as... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 mlg posted on Tuesday, March 20, 2012 - 6:49 pm
Hello,

1) Can population average (marginal) models be specified in Mplus rather than conditional RE models?

I'm collecting cross sectional data for individuals (i=100) clustered within neighborhoods (j=40). I'm interested in group effects on individual outcomes and since I won't have variability within a neighborhood at the group level variable, I was looking at using marginal models.

2) Can a variable for the group (ambient air pollution concentrations) be used as a mediator of two individual level variables in the model? Individual measurements are not possible.

Thanks!
 Bengt O. Muthen posted on Wednesday, March 21, 2012 - 8:25 pm
Why not do a 2-level model with cluster=neighborhood? You specify group level variables on the Between= list.

Mediational modeling can be done in 2-level model where cluster-level variables are mediators for the between-level component of individual-level variables.
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: