CFA ascii file dataset question PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Amanda M. White posted on Wednesday, October 17, 2007 - 9:51 am
I have a question regarding what to include in the ascii file from SPSS to be used in Mplus. I've run both an EFA and a CFA to explore and confirm latent constructs, however now that I want to run a path analysis, I'm having some problems. In the EFA and CFA, I included only the variables that were being exmained. I did not include the dependent variables or any control variables in the original EFA and CFA since I didn't want them to be part of the latent consructs. My main question is should I keep these variables in the ascii file even though I don't want them to be used for the EFA and CFA. But, I do want to use them in the path analysis. Any thoughts?
 Linda K. Muthen posted on Wednesday, October 17, 2007 - 10:09 am
I would have all of the variables in the ASCII file. The USEVARIABLES option of the data set is used to select a subset of variables from a data set to be used in the analysis.
 Amanda M. White posted on Wednesday, October 17, 2007 - 12:02 pm
Great! Thanks for your help!
 Amanda M. White posted on Wednesday, October 17, 2007 - 12:17 pm
One more question regarding SEM - I can only find an example for SEM using continuous variables. Is the syntax not different for SEM using continous and categorical?
 Linda K. Muthen posted on Wednesday, October 17, 2007 - 2:13 pm
Observed factor indicators in an SEM model can be continuous, censored, binary, ordered categorical (ordinal), counts, or combinations of these variable types. You can use the CENSORED, CATEGORICAL, and COUNT options of the VARIABLE command to specify the scale of dependent variables that are not continuous.
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: