Re: [R] Non-parametric test for location with two unpaired setsof data measured on ordinal scale.

2009-09-25 Thread Greg Snow
Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare greg.s...@imail.org 801.408.8111 > -Original Message- > From: John Sorkin [mailto:jsor...@grecc.umaryland.edu] > Sent: Thursday, September 24, 2009 2:52 PM > To: Greg Snow; r-help@r-project.org >

Re: [R] Non-parametric test for location with two unpaired setsof data measured on ordinal scale.

2009-09-24 Thread John Sorkin
Greg, I used the term location because I did not want to use the terms mean or median for the exact reason that you gave; these to values can be different in a given distribution. I want to test the null hypothesis that the data come from a single distribution. This is often done by comparing a