Hello- Ive been looking for a way to integrate CFA with LPA. I think Ive figured out how I can use factor scores derived from the CFA in an LPA. 1)Ill need to iron out the details, but in general is this an acceptable way to integrate these two analyses? Any potential major problems that may make this strategy untenable? 2)Is there a way to do this all in one model rather than two separate analyses?
Yes, I would use factor mixture modeling instead of first estimating factor scores. Regarding FMM, you may want to study
Muthén, B. (2008). Latent variable hybrids: Overview of old and new models. In Hancock, G. R., & Samuelsen, K. M. (Eds.), Advances in latent variable mixture models, pp. 1-24. Charlotte, NC: Information Age Publishing, Inc.
which you find under Papers, General Mixture Modeling on our web site.
I am researching LPA (or LCA, as in Example 7.9) and mixture CFA modeling (e.g., Example 7.17) and have not been able to find information on the advantages/disadvantages of using each approach. It seems that either approach would be theoretically plausible and appropriate (assuming that a mixture CFA model has only one factor). Are the benefits to using one approach over?