st: Re:Related question -- CI for proportions

[Susan Clemens <susan.clemens@mcri.edu.au>]

I have a related question.
I have a variable with 4 categorical outcomes.  I need to calculate CIs on 
the proportions and some of the proportions are very low.  Due to the low 
proportion of some responses, I would prefer to use a binomial exact CI but 
pweights are not permitted for this command.  This is a cross-sectional 
survey so I am currently using the svymean command but only specifying the 
weight (psu, strata, fpc are not indicated in my data).  This lets me 
recode the categorical variable into four binomial 
variables.  Unfortunately, I cannot find a binomial exact CI option for the 
svy commands, nor a method that would allow the correct calculation using 
the original multinomial variable (including pweights).  Any help in 
resolving this problem is appreciated.
Cheers,
Susan

At 05:17 PM 18/03/2004 +0000, you wrote:

> At 16:37 18/03/04 +1100, Garry Anderson wrote:
> > Hi,
> >
> > I am enquiring if a more appropriate method could be used please to 
> > calculate the 95% CI of the difference between two proportions in the 
> > -csi- command?
> >
> > At the moment it is possible for the upper bound of the confidence 
> > interval of the difference between two proportions to be greater than 
> > 1.0. I realize that the approximation that is used is not appropriate for 
> > small sample sizes, however I think that reporting of results that are 
> > impossible should be avoided.
*
*   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/