Date:         Tue, 14 Dec 1999 22:17:51 GMT
Reply-To:     olmsted_a@MY-DEJA.COM
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         olmsted_a@MY-DEJA.COM
Organization: Deja.com - Before you buy.
Subject:      exact CMH (Cochran-Mantel-Haenszel) test for 2x2xk tables

As far as I know, proc freq does not
compute exact CMH p-values for a set of
2x2 tables. There was a Fortran program
written by Donald Thomas of the
National Cancer Institute that did
(called CMB2x2CL, I think). I was not
able to find it by visiting the NCI
web site, though. Does anyone know if
it can be downloaded or requested by mail,
or of any (free) alternative?

On another front, does anyone know if the
BMDP example datasets are downloadable from
anywhere? I visited the SPSS web site
(www.spss.com), but the only example datasets
there are SPSS and Systat ones, I think.

-- Ann Olmsted


Sent via Deja.com http://www.deja.com/
Before you buy.