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

RE: st: odds ratio vs. RRR in multinomial logistic regression

From   "Nick Cox" <>
To   <>
Subject   RE: st: odds ratio vs. RRR in multinomial logistic regression
Date   Wed, 10 Aug 2005 21:04:30 +0100

I don't think this could be much plainer. It 
could be tinkered with as a matter of style, 
but that seems a different issue. Some readers
prefer percents, so you could consider variations

given X, the relative risk of outcome 3 is 85% that of 
outcome 1 


> Some of the statalist members were kind enough to
> respond to my inquiry a few months ago about
> interpreting multinomial logistic regression and the
> resulting RRR. 
> I received this response about reporting the results: 
> "For the RRR it would be, for example, 'the relative
> risk of outcome 3 associated with X is 0.85 times
> the relative risk of outcome 1 associated with X'."
> This was helpful, but when I wrote it up in my
> dissertation this way, my advisor has told me that I
> need to put it in "plain English." Can anyone help me
> put this sentence into more interpretable language for
> someone who is not very familiar with RRR? 
> I had already written up a detailed description of
> where the RRR comes from and then followed it with the
> results reported as above, but my advisor is still
> requiring a more intuitive or interpretable sentence. 

*   For searches and help try:

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