SAVEDATA PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Isaiah Baker posted on Sunday, May 09, 2010 - 8:11 pm
I have used obtained factor scores for my CFA model by using SAVEDATA. In the output of factor scores, however, there is no original case number that matches each factor score. In calculating the factor scores, several cases are left out because they are missing data. So, I have a difficulty in matching the factor scores to each case just based on the order they are shown. so, if there is any procedure that can match the case number and the factor scores automatically, that will help me a lot. (Yes, I can try to match the case number and factor scores by canceling out the missing data from the column of case number. But I do not feel good about this simple procedure because it is very unsure.)

Hope I can get some help from you, thanks.
 Linda K. Muthen posted on Sunday, May 09, 2010 - 9:56 pm
See the IDVARIABLE option of the VARIABLE command.
 Isaiah Baker posted on Monday, May 10, 2010 - 2:04 am
I have tried to use the IDVARIABLE option in VARIABLE command. And I have even used the FORMAT option in the SAVEDATA command.
But the output only shows 10 stars for each case under ID label. The ID variable in my original data is numerical and has 7 digits for each case, no missing data. But why can this happen? I have searched through this discussion forum, no clue yet.
Thanks.
 Linda K. Muthen posted on Monday, May 10, 2010 - 12:17 pm
It sounds like your ID variable is too large. The length cannot exceed 16 although it was less than that in older versions of Mplus. If you can't figure this out, please send your full output, data, and license number to support@statmodel.com.
 Isaiah Baker posted on Monday, May 10, 2010 - 3:23 pm
I think you are right. I changed the ID to the natural case number: 1 2 3... Then it works. Thanks for your guide.

But the weird thing is that in the output, each ID number is followed by three 0 after the dot. I do not know why. In the original format of the data (SPSS), there is no any decimals for each ID number.
 Linda K. Muthen posted on Monday, May 10, 2010 - 3:26 pm
It sounds like when you converted from SPSS to an ASCII file, the format of the ID variable was changed.
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: