Take a look at SPLIT FILE for the first part.
For saving the coefficients, you have a general mechanism that works for any output table via OMS. For REGRESSION in particular, you can use the OUTFILE subcommand to create a sav file or dataset with this information. Here is an example using OUTFILE.
DATASET DECLARE coef.
sort cases by jobcat.
split file by jobcat.
REGRESSION
/DEPENDENT salary
/METHOD=ENTER salbegin prevexp jobtime
/OUTFILE=COVB(coef) .
HTH,
Jon Peck
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of Kathy McKnight
Sent: Donnerstag, 7. Juni 2007 10:09
To:
[hidden email]
Subject: [SPSSX-L] saving individual parameters
Hi all,
I want to demonstrate the "slopes-as-outcomes" approach, a.k.a. "individual
regressions" to a group to show how individual growth curves have been
addressed prior to multilevel modeling. To do so, I need to be able to carry
out regressions per individual and save the individual intercepts & slopes
for a second step analysis in which these will become DVs to be explained.
I've done these in SAS by simply stating that the regression procedure is
done BY individual. I don't see where I'd do that in SPSS. I can also save
the intercepts & slopes in SAS in a separate, SINGLE data set and I'm
looking for the same option in SPSS.
Is anyone aware of how this can be done in SPSS?
Thanks very much!
Katherine McKnight
Locked
