st: Re: Testing for differences in ratios

[Roger Newson <roger.newson@kcl.ac.uk>]

At 13:16 22/12/03 -0600, Craig Goodman wrote:
> I was hoping someone could provide some guidance on whether there is a 
> test in Stata for determining whether there is statistically significant 
> difference in ratios.  In my case, I am examining committee seats and the 
> ratio of majority members to minority members and comparing that to the 
> chamber as a whole.  Any thoughts?
Usually, when comparing 2 ratios, we measure ratios between ratios, not 
differences between ratios. In your case, you seem to be looking for 
logistic regression, which in Stata is handled by the -logit- procedure. 
The odds in favour of being a minority member is defined as the probability 
of being a minority member divided by the probability of being a majority 
member. Logistic regression is used to derive confidence intervals for odds 
and their ratios. In your case, you might have a binary variable 
-minority-, which is 1 for minority members and 0 for majority members, and 
a variable -committee-, which is 1 for committee members and 0 for 
non-committee legislators. If you type

logit minority committee, or

then Stata will give you a confidence interval for the ratio between the 
odds of being minority for committee members and the odds of being minority 
for non-committee legislators. And if you typed

gene byte baseline=1
logit minority committee baseline, or noconst

then Stata will give you confidence intervals for the baseline odds of 
being a minority (assuming that you are a  non-committee legislator) and 
also for the odds ratio as before.

I hope this helps.

Roger


--
Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
King's College London
5th Floor, Capital House
42 Weston Street
London SE1 3QD
United Kingdom

Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
  or 020 7848 6605 International +44 20 7848 6605
Email: roger.newson@kcl.ac.uk
Website: http://www.kcl-phs.org.uk/rogernewson

Opinions expressed are those of the author, not the institution.

*
*   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/