Model identification in twolevel CFA/SEM PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Student 09 posted on Friday, June 19, 2009 - 7:56 am

I wonder how twolevel SEM's are identified in Mplus.

a) Is it correct that in Mplus, the number of elements on the between-level equals (p(p+1)/2) + intercepts, where p are the observed indicators, just as in 'regular' covariance and mean structure analysis?

b) Is it correct that on the within level, identification follows (p(p+1)/2), that is without the mean (intercept) information, because the intercepts are modelled only on the between-level?

Many thanks!
 Linda K. Muthen posted on Friday, June 19, 2009 - 3:31 pm
a) Yes.
b) Yes.
 Student 09 posted on Friday, June 19, 2009 - 4:58 pm
so this implies that the covariance structure of a twolevel model can be identified, while the mean structure on the between level is not identified?

Thanks for your response
 Bengt O. Muthen posted on Friday, June 19, 2009 - 5:13 pm
No, it doesn't imply that. The sample means identify the p means in the model (which appear on Between).
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