I have a data set in a SAV file that has not changed but I get 2 different results with the same program when using v16 of SPSS.
How the source file is created:
- The original source file has pretest and posttest records (not every case has both).
- I use CASESTOVARS to create a working file with one record per case.
- CASESTOVARS adds a 0 or 1 to variable names to distinguish the Pre and Post versions of the same variable.
- I compute a new variable (PrePost) to tell me how many matched Pre - Post records i have.
If i start with only one data file open I always seem to get the same correct result (number with both Pre and Post is 96 out of 187).
If i have a second file already open when i run the CASESTOVARS, I often get a noticeably smaller result (88 out of 187).
My conclusion is that CASESTOVARS can return the wrong results if there was a second file open (though there appears to be an additional factor i could not identify as it does not seem to do this 100% of time).
Another related bug is that the working file created by CASESTOVARS is sometimes dropped by SPSS when I stop to check
intermediate output before finishing additional steps to the analysis. Again, this only appears to happen if a second file
was already open when I run CASESTOVARS.
Dennis Deck, PhD
RMC Research Corporation
111 SW Columbia Street, Suite 1200
Portland, Oregon 97201-5843
voice: 503-223-8248 x715
voice: 800-788-1887 x715
fax: 503-223-8248
[hidden email]
Locked
