Covariates of the latent class PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Judy Black posted on Tuesday, November 13, 2012 - 12:33 am
Dear Drs.Muthen,

I am going to examine a predictor of the latent class. But this predictor is a categorical variable. Is it still possible to use this command "Auxiliary = X (r3step)"?

 Jon Heron posted on Tuesday, November 13, 2012 - 1:55 am
Yes, but you'll need to turn your predictor into a set of binary dummy variables
 Judy Black posted on Tuesday, November 13, 2012 - 7:15 am
Hi Jon,

Thanks for reply. Suppose X has 3 categories. What commands can I use to turn X into binary dummy variables in Mplus?

Kind regards
 Bengt O. Muthen posted on Tuesday, November 13, 2012 - 1:26 pm
You have to create the dummy variables - no option in Mplus for automatically doing this.
 Laura posted on Wednesday, May 28, 2014 - 3:06 am

I have understood that with auxiliary option the covariates need to be
either binary or continuous (e.g. when using "auxiliary (e)"). When I include the same variables in a latent class growth analysis (e.g. "C ON gender" or "C ON age"), do I have to define which of the covariates are binary or continuous? And with binary (dummy) covariates, is zero automatically the reference category?
 Linda K. Muthen posted on Wednesday, May 28, 2014 - 2:10 pm
In regression, all covariates are treated as continuous. You should not specify their scale.

Yes, zero is the reference category for binary covariates.
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