LCA trend in LMR(p) PreviousNext
Mplus Discussion > Latent Variable Mixture Modeling >
 Paul Sacamano posted on Thursday, December 21, 2017 - 4:53 am
I have run 2, 3, 4, and 5 class two level mixture LCA with 7 ordinal indicators (3-levels).

The -2LL, AIC, and BIC all decline from 2 to 5 classes. Entropy also declines.

My question is that I'm seeing an inconsistent trend in the LMR p value with increasing number of classes.
2 classes <0.0001
3 classes <0.0001
4 classes 0.1241
5 classes 0.0457

Is there a reason/problem that the LMR p value goes from significant to not significant and then significant again with increasing number of classes?
 Bengt O. Muthen posted on Thursday, December 21, 2017 - 3:26 pm
Stop using the LMR test after you get your first non-significant value.
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