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

st: Confidence interval of difference between two proportions and -csi-


From   Garry Anderson <g.anderson@unimelb.edu.au>
To   statalist@hsphsun2.harvard.edu
Subject   st: Confidence interval of difference between two proportions and -csi-
Date   Thu, 18 Mar 2004 16:37:54 +1100

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 example
-csi 5 1 0 4,e-


                 |   Exposed   Unexposed  |     Total
-----------------+------------------------+----------
           Cases |         5           1  |         6
        Noncases |         0           4  |         4
-----------------+------------------------+----------
           Total |         5           5  |        10
                 |                        |
            Risk |         1          .2  |        .6
                 |                        |
                 |      Point estimate    |  [95% Conf. Interval]
                 |------------------------+----------------------
 Risk difference |               .8       |   .449391    1.150609 
      Risk ratio |                5       |   .866228    28.86076 
 Attr. frac. ex. |               .8       | -.1544305    .9653509 
 Attr. frac. pop |         .6666667       |
                 +-----------------------------------------------
                                1-sided Fisher's exact P = 0.0238
                                2-sided Fisher's exact P = 0.0476



The value of 1.15 is not desirable.

One potential solution to this is to use a method that is described in
Newcombe RG (1998) Interval estimation for the difference between independent proportions: comparison of eleven methods. Statistics in Medicine 17: 873-890

The -ci- command gives an exact 95% CI for a proportion, therefore I think that it is desirable that the difference between two proportions have sensible 95% confidence intervals.

Kind regards, Garry



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