ESEM - Multiple groups & factors PreviousNext
Mplus Discussion > Exploratory Factor Analysis >
 Elizabeth Salib posted on Thursday, October 03, 2013 - 8:15 am

I'm working on the measurement model for an upcoming SEM analysis. The original scale has 8 factors and 30 items. Our data reflects 6 groups (countries) with approx 250 participants per group.

Based on our data we do not expect the original scale structure to be upheld - and a quick MGCFA (using model = configural metric scalar) did not establish configural invariance.

Can you suggest an approach to establishing the factor structure of this scale? The number of groups and items is making an ESEM analysis unwieldy.
 Elizabeth Salib posted on Thursday, October 03, 2013 - 9:30 am
We have been considering using the alignment method to the items that are non-invariant within the established structural model. Does this seem like the right approach?
 Bengt O. Muthen posted on Thursday, October 03, 2013 - 10:27 am
The alignment method requires that you have configural invariance, that is, the zero loadings in the same places.

It sounds like you may have many cross-loadings and need to go back to EFA in each group, perhaps followed by multiple-group ESEM.
 Elizabeth Salib posted on Tuesday, October 08, 2013 - 7:28 am
Thank you!
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