Date:         Fri, 30 Jan 1998 10:54:56 +0100
Reply-To:     Paul Dickman <Paul.Dickman@ONKPAT.KI.SE>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Paul Dickman <Paul.Dickman@ONKPAT.KI.SE>
Subject:      Re: Simple program for crosstabs? Excel Macro available.
Content-Type: text/plain; charset="us-ascii"

> I am an attorney that works with statistical evidence, and I was
> wondering whether there is a simple statistical program that I
> could use to get p values for two-by-two crosstabs.  Ideally, I'd
> like something that would extract the values from a data base,
> but I can imagine a simpler program that would give me statistics
> if I just enter the four values.
 
I wrote an Excel macro, which requires you to enter the four cell
counts and then calculates the chi-square test statistic, chi-square
with continuity correction, Fisher's exact test, RR, OR, and expected
cell counts under the null hypothesis.
 
The worksheet is 30Kb in size and I'm happy to send a copy to
anyone who asks.
 
I also have another Excel worksheet for estimating relative risks
from stratified 2x2 tables.  It calculates the RRs for each table,
a test of homogeneity of the stratum-specific RRs, and the
Mantel-Haenszel estimate of the stratified RR.
 
This worksheet is also 30Kb in size and I'm happy to send a
copy to anyone who asks.
 
I'll send the documents as MIME attachments, let me know if
you'd prefer BinHex or UUENCODE instead.
 
Paul Dickman
---
Paul Dickman,  Paul.Dickman@onkpat.ki.se
Department of Cancer Epidemiology, Radiumhemmet,
Karolinska Hospital, 171 76 Stockholm, Sweden
Ph: +46 8 5177 5375   Fax: +46 8 +46 8 326 113