s-news
[Top] [All Lists]

Mantel-Haenszel analysis

To: <s-news@lists.biostat.wustl.edu>
Subject: Mantel-Haenszel analysis
From: "Neal Alexander" <Neal.Alexander@lshtm.ac.uk>
Date: Fri, 15 Nov 2002 10:50:53 +0000
S+ has the function 'mantelhaen.test' which, in epidemiological terms, tests 
for an association between an exposure and an outcome, stratifying on a 
possible confounder.  This function produces the chi-squared and p value.  It 
would be useful to have, in addition, the crude (unadjusted) odds ratio, the 
stratum-specific odds ratios, the M-H summary odds ratio (a weighted average of 
the stratum ORs), the CI for the M-H OR, and a test for interaction.  Is anyone 
aware of such a function (whether or not based on the built-in one), either 
public domain or which you would be willing to share?  Thanks, Neal

Neal Alexander, PhD, Senior Lecturer in Medical Statistics & Epidemiology
   Infectious Disease Epidemiology Unit, Dept of Infectious & Tropical Diseases
   London School of Hygiene&Tropical Medicine,Keppel St,London WC1E 7HT,UK
   neal.alexander@lshtm.ac.uk; tel (+44) 20 7927 2826; fax 20 7636 8739



<Prev in Thread] Current Thread [Next in Thread>
  • Mantel-Haenszel analysis, Neal Alexander <=