Latent Class Analysis with covariates... PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Katherine Fiori posted on Friday, June 19, 2009 - 10:35 am
My colleagues and I are trying to run a model in which we use our latent classes (5) to predict various continuous outcomes. We would like to know whether class membership predict differences in these outcomes, beyond the effects of several covariates on the outcomes.

When we ran the model as outlined in the MPlus handbook, we discovered that the covariates predicted the outcome WITHIN each class, rather than across the classes. How can we use the covariates to predict the outcome ACROSS the classes?

Thank you!!
 Linda K. Muthen posted on Friday, June 19, 2009 - 3:42 pm
If the intercepts of the continuous outcomes are different across the classes, this implies that the categorical latent variable predicts the outcomes. This happens by default.

You can regress the categorical latent variable on a set of covariates in the overall part of the model command to see if the covariates predict class membership.
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