Daniel posted on Tuesday, December 07, 2004 - 8:23 am
Do you have any experience with more than two latent categorical variables, with growth mixture modeling? My super is interested in running a model that has three processes, each of which has 2 latent classes. Am I to expect trouble with this extension?
bmuthen posted on Tuesday, December 07, 2004 - 11:26 am
I have worked with 3-4 latent class variables in latent transition modeling, but not with growth mixture modeling as far as I remember. Given that you have only 2 classes for each, working with 8 total classes should be fine - I have worked with 40 or so.