Date:         Fri, 18 Nov 2005 11:35:59 +0100
Reply-To:     Marta García-Granero
              <biostatistics@terra.es>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Marta García-Granero
              <biostatistics@terra.es>
Organization: Asesoría Bioestadística
Subject:      Re: Logistic regression's question
In-Reply-To:  <20051118093054.28960.qmail@web25411.mail.ukl.yahoo.com>
Content-Type: text/plain; charset=ISO-8859-1

Bon jour Jan

kj> In a multiple logistic regression model, the parameter
kj> estimate of a continuous variable is significantly different from
kj> zero (p=0.03), but the odds ratio is 1.00 and the confidence
kj> interval of OR is [1.00;1.0010].  can I nevertheless say that this
kj> variable has an effect ?

You need to rescale the continuous DV. Try multiplying it by 10, or
100. The OR will then reflect the change in risk associated to a 10 or
100 units increase of the DV.

--
Regards,
 Marta                            mailto:biostatistics@terra.es