Does anyone know a reference for estimating the power of a Mantel-Haenszel test for stratified 2xk contingency tables? I have one for 2x2 but it's not clear to me how to generalize it. Thanks, Jeph -- Jeph Herrin Statens institutt for folkehelse/National Institute of Public Health Postboks 4404 Torshov N-0403 Oslo http://www.folkehelsa.no/ Tlf. (47) 22042346 Fax. (47) 22042595 E-mail:[log in to unmask]