RE: st: _tabi, exact_ gives obviously wrong P-value

[Richard Williams <Richard.A.Williams.5@nd.edu>]

At 12:58 PM 12/22/2004 -0500, Steichen, Thomas J. wrote:


> What version of tabi and Stata are you running, Marcello?
>
> I get:
>
> .. tabi 27879 35177 29377 17201 \ 118 187 183 65, e chi2
>
>            |                     col
>        row |         1          2          3          4 |     Total
> -----------+--------------------------------------------+----------
>          1 |    27,879     35,177     29,377     17,201 |   109,634
>          2 |       118        187        183         65 |       553
> -----------+--------------------------------------------+----------
>      Total |    27,997     35,364     29,560     17,266 |   110,187
>
>           Pearson chi2(3) =  17.7073   Pr = 0.001
>            Fisher's exact =                 1.000
I'm with Thomas on this one.  I ran it this morning and got the above 
results.  Then I did an update all (there is a Dec. 20 update for those who 
haven't noticed) and again got the above results.

HOWEVER -- I ran the problem in Stata 7 -- and got

          Pearson chi2(3) =  17.7073   Pr = 0.001
           Fisher's exact =                 0.001

There has been some fiddling around with the algorithm for Fisher's exact 
test to make it faster -- has accuracy perhaps been compromised in the process?


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   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/