SAVING FACTOR SCORES FOR MULTIPLE GROUPS PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Archana Jajodia posted on Thursday, March 30, 2006 - 1:51 pm
Hi,
I have 6 groups in my analysis and 4 latent variables and I am trying to save the Factor Scores for all of these. When I do so, I find that the factor scores for the first group appear to be fine, but beyond that there is no correlation between the indicator variables and the factor scores saved. (My model appears to be fine with factor structure invariant across the groups, and decent fit CFI= .95)
 Linda K. Muthen posted on Thursday, March 30, 2006 - 1:56 pm
You would need to send your input, data, output, and license number to support@statmodel.com for us to answer this. I suspect you may be using an older version of Mplus.
 yang posted on Thursday, June 07, 2007 - 2:08 pm
Drs. Muthen,

I am fitting a unidimensional model for 20 dichotomous indicators, and want to save the factor score for each of the subjects in the sample. Beside the 20 indicators, there is a dummy variable for groups, which I also want to save in the output data file. I am using the following codes:

SAVEDATA:
FILE = Data_with_Score.txt;
SAVE = FSCORES;

The problem is that, the output datafile only contains the 20 indicators and the estimated factor score, without the dummy variable for groups. Would you mind kindly telling how to solve this problem?

Thank you very much.
 Linda K. Muthen posted on Thursday, June 07, 2007 - 2:33 pm
See the AUXILIARY option of the VARIABLE command.
 yang posted on Thursday, June 07, 2007 - 2:51 pm
I got it. Thank you very much.
 J.D. Haltigan posted on Tuesday, June 21, 2011 - 9:44 pm
Quick question relevant to the Auxiliary option. In running a single group CFA, I want to use the Auxiliary command for a 4-level classification variable so that I can see the scatter of the two specified factors as a function of the class group in the plot command. Is this possible? Or is it only possible to see the scatter as a function of a groping variable in a multi-group FA (which is not relevant to my research question--the FA is on the entire group on a set of scales which are used to inform a classification but which is not part of the formal analyses).
 Linda K. Muthen posted on Wednesday, June 22, 2011 - 9:32 am
I don't believe this would work.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: