Multilevel multiple group of unequal CFA PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Skylar Son posted on Wednesday, September 12, 2018 - 12:04 am
I took a look-see multiple group multilevel analysis (webnote 16).
And then I have a two question about "within level group variable analysis" in case of unequal factor structure CFA (e.g., one factor within level and two factor between level OR two factor within level and one factor between level).

1. how can I estimate the latent mean between two groups using the suggested models in this paper? Because the latent mean is estimated at the between model, the estimates are expected to decompose or composite at the between level.

2. Is it possible to test metric (weak) invariance and intercept (strong) invariance? Metric invariance has to constraint cross level factor loading in multilevel data, but how can I do in case of unequal factor structure CFA?
 Bengt O. Muthen posted on Wednesday, September 12, 2018 - 5:01 pm
Are you saying that you have 2 groups and they have different number of factors on both within and between? If so, they can't be compared.
 Skylar Son posted on Thursday, September 13, 2018 - 6:57 am
Thank you for the reply.
 Skylar Son posted on Monday, September 17, 2018 - 9:22 pm
I have an additional question about multiple group multilevel analysis (webnote 16).

In within level group variable analysis, the estimator used ML, not MLR.
In MSEM, I was aware of that the default estimator is MLR, so I wonder why ML is used instead of MLR.
 Bengt O. Muthen posted on Tuesday, September 18, 2018 - 5:42 pm
Either can be used.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message