Date:         Mon, 14 Jul 1997 18:06:45 GMT
Reply-To:     "LINDA J. FELCH" <lfelch@WELCHLINK.WELCH.JHU.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         "LINDA J. FELCH" <lfelch@WELCHLINK.WELCH.JHU.EDU>
Organization: HCF - Johns Hopkins University, Baltimore, Maryland, USA
Subject:      dichotomous data, repeated measures, grouping factor
Comments: To: STAT-L@VM1.MCGILL.CA

I am trying to use the Cochran-Mantel-Haenzel statistic in SAS to
analyze a dichotomous variable collected across time for 3 groups.
 
As a first step, I tried to run the SAS code in Agresti's _Categorical
Data Analysis_ (1990) for the example he gives on page 233 (SAS code on
page 495). I don't get the same values for the statistics he reports in
the text. Is this an error in the text? Has SAS changed it's
implementation of the CMH test? Or is it just me?
 
Is there a better statistic to use for this dataset?
 
Thanks for your help!
--
Linda Felch
lfelch@welchlink.welch.jhu.edu
Baltimore, MD