Hello, I'm doing a simulation experiment, and I've a population with 80 groups. What I would like to do is performing a factor model to appreciate the hierarchical structure (rather than performing many models for each group). But what I expect to see at the end are the estimates of the parameters for each group. I don't have so much experience with these models, and so, does anybody know some references about how to do that via Mplus? If you know some good references will be really appreciated. Thanks. Angelo
You say that you have 80 groups and that you want to use:
"a factor model to appreciate the hierarchical structure"
That sounds like you want to view group as a random mode of variation as opposed to the fixed mode approach of multiple-group analysis. If I am right, you should take a look at the paper on our website:
Muthén, B. & Asparouhov, T. (2013). New methods for the study of measurement invariance with many groups. Mplus scripts are available here.
Thank you for your help! Yes, rather then doing separate factor models on the clusters, as I have a hierarchical structure in the data, I would like to include that in the model. And, given my purpose, In the end I will be only interested in the factors, and not in the model as a whole. The paper is very useful and relevant for what I am doing.