Interaction between Cg and C on dista... PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
Message/Author
 eli puterman posted on Thursday, April 14, 2011 - 1:26 am
Dear Drs. Muthen,

We are examining the effects of eating over 10 years across adolescence on depression at year 10 in black/white girls, and want to examine the (1) independent effects of race and latent class eating trajectories on depression, and (2) to see if there are different (a) patterns for each race; and (b)whether patterns have different associations with depression as a function of race. Below is syntax of the first part - 6 different patterns, 3 per race- but need to get (b). In other words, how do we test an interaction on depression? We added after the %overall% rules:
%c1#1%
c2 on y10_cesd;
%c1#2%
c2 on y10_cesd;

We get: *** ERROR in MODEL command Unknown class label in MODEL :
%C1#1%

Here is the model we are working with:

USEVariables=EEA1 EEA2 EEA3 EEA4 EEA5 EEA6 EEA7 EEA9 y10_CESD;
classes =c1(2) c2(3);
knownclass = c1 (y0_race=1 y0_race = 2);
missing = all (-999);
analysis: type = mixture ;
starts=100 10;
Model:
%overall%
i s | EEA1@0 EEA2@1 EEA3@2 EEA4@3 EEA5@4 EEA6@5 EEA7@6 EEA9@8 ;
c2 on c1 y10_CESD ;
 Bengt O. Muthen posted on Thursday, April 14, 2011 - 2:09 pm
With more than one latent class variable, you should say

Model c1:

%c1#1%
c2 on y10_cesd;
%c1#2%
c2 on y10_cesd;
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: