1 categorical variable vs 2 categoric... PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Vlad posted on Thursday, October 29, 2009 - 7:32 am
Hello,
I have conducted FA and results shows that 2 factors are enough to explain all variables. Then I would like to estimate LCA. But I have questions:
1)Should I specify 1 or 2 categorical variables?
2)If 2 cat.variables then what statistics, except criteria, should I use to decide the proper number of classes for each cat.variable(LMR and Bootstrapped LR tests are not availble any more)?
3)Can be you recommend me paper for LCA with 2 cat.variables?

Regards,
V
 Linda K. Muthen posted on Thursday, October 29, 2009 - 9:26 am
I would use one categorical latent variable and look at different numbers of classes for that variable. You should look at BIC, TECH14, TECH11, loglikelihood etc. and theory.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: