In SPSS, you first have run binary logistic regression
with the predictor variables/covariates of interest
and save the predicted probabilities. The predicted
probabilities are then used in the ROC curve analysis
in place of the single variable.
I don't know if you can conduct an inferential test on
the difference in AUCs in SPSS. However, this can
easily be done with Stata.
Scott Millis
--- William Paulson <
[hidden email]> wrote:
> Hello,
>
> I am considering purchasing SPSS, but have been
> unable to find an answer to this question on the web
> site:
> Can SPSS compute a ROC curve based upon more than
> one predictive (independent) variable, as opposed to
> the usual single variable ROC curve?
> Can the AUCs of such curves be compared?
>
> Thanks.
>
> Bill Paulson
> Medical College of Georgia
Scott R Millis, PhD, MEd, ABPP (CN,CL,RP), CStat
Professor & Director of Research
Dept of Physical Medicine & Rehabilitation
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email:
[hidden email]
Tel: 313-993-8085
Fax: 313-966-7682
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
