Date:         Thu, 9 Feb 2006 21:14:00 -0700
Reply-To:     Aric Zion <Aric.Zion@asu.edu>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Aric Zion <Aric.Zion@asu.edu>
Subject:      Re: Odds ratio question
Content-type: text/plain; charset=UTF-8

No, it means a 77% decrease in the odds (i.e. 1 - .23 = .77).  If you think of it this way... that .866 means that a 1 unit change in x means that the odds gets multiplied by .866 (i.e. a 13.4% decrease), then compounding .866 ten times (i.e. .866^10) gives you .237 as an odds multiplier and 1 - .237 = .763.

Aric

        -----Original Message-----
        From: SPSSX(r) Discussion on behalf of S Crawford
        Sent: Thu 2/9/2006 8:01 PM
        To: SPSSX-L@LISTSERV.UGA.EDU
        Cc:
        Subject: Odds ratio question



        Hi, everyone.

        I'm having problems with an odds ratio from a logistic regression.  We are
        trying to predict a binary outcome (yes or no) based on one continuous
        variable (variable X).  We get a significant overall regression.  The B
        coefficient for this variable was -.144, and Exp(B) from the SPSS printout
        for variable X was .866.

        If we want to find out the change in odds for every 10 point increase in
        variable X, do we just take:
        .866 and raise it to the 10th power?

        That would be .23 rounding off - so does that mean a 23% decrease in odds of
        an abnormal outcome for every 10 point increase in variable X?

        Many thanks for the help.

        Susan

        _________________________________________________________________
        Scan and help eliminate destructive viruses from your inbound and outbound
        e-mail and attachments.
        http://join.msn.com/?pgmarket=en-ca&page=byoa/prem&xAPID=1994&DI=1034&SU=http://hotmail.com/enca&HL=Market_MSNIS_Taglines
          Start enjoying all the benefits of MSN® Premium right now and get the
        first two months FREE*.