Re: [S] strange p-value for Fisher exact's test

Wim Kimmerer (
Mon, 23 Mar 1998 16:30:42 -0800

The Fisher exact test problem also comes out wrong in S+ 3.3 for Windows,
so EVEN I have to admit (although I hated 4.0 so much I reverted) that it
is not a Version 4 problem.

fisher.test(rbind(c(1, 1), c(91, 98)))

Fisher's exact test

data: rbind(c(1, 1), c(91, 98))
p-value = 1.0001
alternative hypothesis: two.sided

Dr. Wim Kimmerer
Romberg Tiburon Center
San Francisco State University
PO Box 855
Tiburon CA 94920
Ph. (415) 338-3515
Fax (415) 435-7120

This message was distributed by To unsubscribe
send e-mail to with the BODY of the
message: unsubscribe s-news