Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: interpretation of OR in logistic regression


From   Blau Blau <BlauBlau@seznam.cz>
To   statalist@hsphsun2.harvard.edu
Subject   st: interpretation of OR in logistic regression
Date   Wed, 25 Dec 2002 12:09:50 +0100 (CET)

Dears,
if I have in logistic regression these results

                       coefficients      or
fair (versus poor)       0,862          2,37 
good (versus poor)       1,047          2,85 
excellent (versus poor)  1,138          3,12

can I say that odds excellent (versus poor) are 3,12 times higher,
controlling for other variables in model (other things being equal),
as I say in many textbooks or should not I say that odds excellent
(versus poor) are just 2,12 higher (3,12 - 1), because 1 is the start
point in interpretation of OR in logistic regression? I mean that 1
means no influence of independence variable to dependence variable,
if we talk in terms of OR, that is why it seems to me that I have to
use this formula 3,12 - 1 for interpretation of OR. Am I right or
completely wrong? 
 
Thanks Tomas

Tomas Katrnak, PhD
Masaryk University
School of Social Studies
Department of Sociology
Gorkeho 7
Brno 602 00
Czech Republic

______________________________________________________________________
Reklama:
E-mail zdarma na cely zivot: http://email.seznam.cz
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index