CFA in two samples PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 J Lydecker posted on Tuesday, January 19, 2010 - 12:17 pm
I am trying to test whether factor loadings are equal in two different samples (of 400 and 298 subjects). I have a grouping variable in the syntax, but am having trouble interpreting the output. It looks like I get overall model fit statistics, and then factor loadings (and other results) for each sample separately, but without information as to whether differences are significant.

Where can I find information about whether there are significant differences between the factor loadings in different samples?
 Linda K. Muthen posted on Tuesday, January 19, 2010 - 12:44 pm
See the Topic 1 course handout under multiple group analysis to see the models to use for testing differences in factor loadings across groups. If the handout is not sufficient, you can listen to the Topic 1 video.
 J Lydecker posted on Tuesday, January 26, 2010 - 8:52 am
When I follow the procedure outlined in the video--exactly what I was looking for--the multiple group CFA with no measurement invariance model was not identified. I ran this model in order to compare it to models with measurement invariance of factor loadings, and invariance of factor loadings and intercepts. I looked up the parameter that was problematic in Tech1. Is there anything I can do to make the model identifiable so that I can compare it to models with measurement invariance?

 Linda K. Muthen posted on Tuesday, January 26, 2010 - 10:15 am
You are probably freeing the first factor indicator in the group-specific MODEL command. The factor indicator that sets the metric of the factor should not be mentioned in the group-specific MODEL commands.
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