Alignment and cross loadings PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 Olev Must posted on Wednesday, September 27, 2017 - 12:19 am
I need to compare the test scores over 2 time points.I have binary items and models use cross loadings of items (WLSMV method). In some subtests I obtained the estimation of the measurement invariance via classical MGCFA. In some subtests I had problems with model convergence (scalar invariance) although the configural invariance was good. I tried to use alignment opimization, but I got the following feedback:
Cross loadings are not allowed for invariance analysis with ALIGNMENT=FREE or ALIGNMENT=FIXED.
Do you have suggestions how I should continue my modelling?
Thank you!
 Bengt O. Muthen posted on Wednesday, September 27, 2017 - 3:27 pm
Note that if you have 2 time points, the subjects need to be different at the 2 time points in order to use MGCFA and Alignment.

For analysis strategies, try SEMNET.
 Olev Must posted on Wednesday, September 27, 2017 - 11:42 pm
Thank you for suggestion!
Yes, the subjects are different: different age cohorts.
I shall try SEMNET possibilities.
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