Two-sample tests using Wilson confidence intervals

[Nestor Mojica]

Is there a way that I can use the two-sample tests of proportion (prtest) using wilson confidence intervals? Is there another command for this?

Thanks!

[Oded Mcdossi]

I just typed

search wilson confidence intervals

and this is the result I found.

. search wilson confidence intervals

Keyword search

        Keywords:  wilson confidence intervals
          Search:  (1) Official help files, FAQs, Examples, SJs, and STBs

Search of official help files, FAQs, Examples, SJs, and STBs


[R]     ci  . . . . .  Confidence intervals for means, proportions, and counts
        (help ci)

STB-52  sg119 . . . . . Improved confidence intervals for binomial proportions
        (help propci, propcii if installed) . . . . . . . . . .  J. R. Gleason
        11/99   pp.16--18; STB Reprints Vol 9, pp.208--211
        adaptation of ci (and cii) providing three alternative
        endpoint-adjusted methods (enhanced Wald, Wilson, and
        enhanced arcsine) for computing confidence intervals

Just click on the package to see if this what you are looking for.

[Nestor Mojica]

Not quite. I looked at the command and what it gives are confidence intervals. What I am looking for is the p-value that compares two binary proportions, but using wilson confidence intervals instead of the standard CI intervals. This is the original command, but not what I want:

prtest fluoros1 if Salmonella==2, by(dates)

Two-sample test of proportions                     1: Number of obs =      109
                                                   2: Number of obs =      267
------------------------------------------------------------------------------
    Variable |       Mean   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
           1 |   .1100917   .0299804                      .0513313    .1688522
           2 |   .2771536   .0273922                      .2234657    .3308414
-------------+----------------------------------------------------------------
        diff |  -.1670618   .0406098                     -.2466556   -.0874681
             |  under Ho:   .0477403    -3.50   0.000
------------------------------------------------------------------------------
        diff = prop(1) - prop(2)                                  z =  -3.4994
    Ho: diff = 0

    Ha: diff < 0                 Ha: diff != 0                 Ha: diff > 0
 Pr(Z < z) = 0.0002         Pr(|Z| < |z|) = 0.0005          Pr(Z > z) = 0.9998

The bold parts are what I'm interested in. Hope this makes sense!

Thanks, NM