Is it possible to statistically compare the resulting CRMs for each latent profile to identify which profiles significantly differ from one another in Mplus? If not, is there a way to do this either manually or in another program using the standard errors that are provided in Mplus?
I don't know what CRM stands for. If it is the conditional means of outcomes in the different classes, you could label the mean parameters and test differences for all means by a Wald test in MODEL TEST.
Thank you Dr. Muthen. Yes, I should have been more clear initially. By CRM, I was referring to Conditional Response Means. I am currently still using version 3, is the MODEL TEST option available in that version? Thank you very much for your help with this, I greatly appreciate it.