Date: Wed, 13 May 1998 13:41:17 0400
ReplyTo: Jeff Mather <Jmather@HARTHOSP.ORG>
Sender: "SPSSX(r) Discussion" <SPSSXL@UGA.CC.UGA.EDU>
From: Jeff Mather <Jmather@HARTHOSP.ORG>
ContentType: text/plain
I am trying to estimate kappa form a 3x3 table as follows:
norm rev fixed
norm 300 55 22
rev 18 122 5
fixed 0 0 0
This is paired data, drug a vs drug b. The 0's in the botttom row don't
allow for a calculation (division by zero). Is there any procedure to get
around this (i.e. estimates), so that a kappa can be calculated?
