Date:         Fri, 15 Mar 2002 17:22:11 -0500
Reply-To:     emaguin@acsu.buffalo.edu
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         emaguin@acsu.buffalo.edu
Subject:      Re: odds ratios for continuous predictors in logistic regression
In-Reply-To:  <1D69068B.61841519.0E7B4242@cs.com>
Content-Type: text/plain; charset=us-ascii

I may be misremembering and you should check Hosmer and Lemeshow's book,
but for continuous variables the log odds is for a one standard deviation
change in the IV. And, which i'll bet is also true for categorical
variables unless you have declared them to be 'categorical' on the Logistic
Regression command.

Gene Maguin