Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: Logistic regression interpretation

From   Maarten buis <>
Subject   Re: st: RE: Logistic regression interpretation
Date   Wed, 22 Sep 2010 09:23:07 +0000 (GMT)

--- On Wed, 22/9/10, Kieran McCaul wrote:
> low risk, control group: OR=1.00 (ref)
> high risk, control group: OR=2.11
> low risk, treatment group: OR=4.79
> high risk, treatment group: OR=2.11*4.79=10.11

--- I wrote:
> There are no interactions so the odds increases
> with a factor 2.1 when one receives the treatment
> regardless of whether one is in the high or
> low risk group 

These two statements appear to be a contradiction,
but they are not. The difference is in what we define 
as our reference group.

I compared control versus treatment both in high risk 
group and control versus treatment both in the low 
risk group. This is how I think of "controlling for
a variable".

Kieran compared all groups with the low risk control

As always, an effect is just a comparison of groups,
so think about what kind of comparison would answer 
your research question, and choose your definition
of reference group accordingly.

Hope this helps,

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen


*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index