|
Activity Number:
|
149
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #309986 |
|
Title:
|
When Does the Wilcoxon-Mann-Whitney Fail?
|
|
Author(s):
|
Abu Minhajuddin*+ and Paul Nakonezny
|
|
Companies:
|
The University of Texas Southwestern Medical Center at Dallas and The University of Texas Southwestern Medical Center at Dallas
|
|
Address:
|
5323 Harry Hines Blvd, Dallas, TX, 75390,
|
|
Keywords:
|
nonparametric methods ; Wilcoxon-Mann-Whitney ; Student's t test ; two-sample tests ; robustness
|
|
Abstract:
|
Wilcoxon-Mann-Whitney (WMW) is perhaps one of the most widely used tests in applied statistics. It is well-accepted as the nonparametric alternative to the Student's two-sample t-test. However, under certain conditions, the WMW may not be the most appropriate test to use. Thus, we explore the general question: "when does the WMW test fail?" Specifically, we address: (1) what are some of the conditions under which the WMW may not be the optimum test of choice? and (2) What is the alternative test in such situations?
|
