Have you tried it? Here's an example. Modify the path on the GET FILE line as needed.
new file.
dataset close all.
GET FILE='C:\SPSSdata\1991 U.S. General Social Survey.sav'
/keep = sex race region age.
dataset name raw.
compute male = sex EQ 1.
formats male(f1).
crosstabs sex by male.
LOGISTIC REGRESSION VARIABLES male
/METHOD=ENTER race region age
/CONTRAST (race)=Indicator
/CONTRAST (region)=Indicator
/PRINT=CI(95)
/CRITERIA=PIN(0.05) POUT(0.10) ITERATE(20) CUT(0.5).
* Now aggregate the data, getting the count of cases for
* each combination of the variables (explanatory AND outcome).
DATASET DECLARE Agg.
AGGREGATE
/OUTFILE='Agg'
/BREAK=race region age male
/Kount=N.
dataset activate Agg.
WEIGHT by Kount.LOGISTIC REGRESSION VARIABLES male
/METHOD=ENTER race region age
/CONTRAST (race)=Indicator
/CONTRAST (region)=Indicator
/PRINT=CI(95)
/CRITERIA=PIN(0.05) POUT(0.10) ITERATE(20) CUT(0.5).
* Compare the -2LL and coefficients etc from the two models.
Kurt Taube wrote
Is there any way to weight cases in SPSS logistic regression without manually repeating lines of data, or to allow the dependent variable to take a probabilistic value between 0 and 1 rather than just a binary 0 or 1? (I know I can input successes and trials in probit but I haven’t been able to get the analysis to run.)
Thanks in advance.
Kurt Taube
Columbus, Ohio, USA
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/"When all else fails, RTFM."
PLEASE NOTE THE FOLLOWING:
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (
https://listserv.uga.edu/).
Locked
