Choosing variables for LCA analysis PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Rongjin Guan posted on Sunday, April 17, 2016 - 2:11 pm
Last time when i did LCA, we had a lot of variables, and used factor analysis to narrow down the variables to be used in LCA.

Now we have a new data set with 5 variables. Is there a general rule how to choose which ones to use for LCA?

If all of them can separate the classes well, it will be fine. But I am not sure if any of them cannot separate the classes well.
 Bengt O. Muthen posted on Sunday, April 17, 2016 - 3:59 pm
You should try this question on SEMNET.
 samah Zakaria Ahmed posted on Sunday, January 29, 2017 - 5:15 pm
The ON statement specifies the multinomial logistic regression of the categorical latent variable c on the continuous latent variable f (c on f)
my question is:
how to specify the multinomial logistic regression of the categorical latent variable c1 on another categorical latent variable c2 ? (the ON statement doesn't run)
 Bengt O. Muthen posted on Monday, January 30, 2017 - 3:16 pm
c1 on c2;

But in the classes= statement, c2 then has to precede c1.
 samah Zakaria Ahmed posted on Tuesday, January 31, 2017 - 6:29 am
i am sorry, but i can't understand your comment.
please clarify
 samah Zakaria Ahmed posted on Wednesday, February 01, 2017 - 1:34 am
ooh, many thanks for your help.
i finally understood what did you mean and the command is run.
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