Problem with categorical variables PreviousNext
Mplus Discussion > Structural Equation Modeling >
 Matteo Cristofaro posted on Friday, January 29, 2016 - 9:29 am
Hi everyone, I'm in trouble!
I'm carrying out a SEM in which I have a categorical variable (no Likert) measuring a latent construct. Do I have to code different dummy variables (0-1) in order to find which category affects the latent construct? Or do I need to have only one predictor variable composed of different categories? In this last case, how can I interpret which one affects the latent construct?
Thanks in advance
 Bengt O. Muthen posted on Friday, January 29, 2016 - 5:59 pm
Sounds like you have a nominal variable. I would not specify a factor as measured by this nominal variable if that is the only factor indicator.

You may want to discuss this on SEMNET.
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