Wilcoxon-Mann-Whitney or t-test? On assumptions for hypothesis tests and multiple interpretations of decision rules

• DOI: 10.1214/09-ss051
• Publication Date: 2010-02-17T15:19:30Z
• Authors: Michael P. Fay; Michael A. Proschan
• Keywords: Decision rule; Neutrality

In a mathematical approach to hypothesis tests, we start with clearly defined set of hypotheses and choose the test best properties for those hypotheses. practice, often less precise For example, researcher wants know which two groups generally has larger responses, either t-test or Wilcoxon-Mann-Whitney (WMW) could be acceptable. Although both t-tests WMW tests are usually associated quite different hypotheses, decision rule p-value from many sets assumptions, call perspectives. It is useful have perspectives may applied collected in one place, since each perspective allows interpretation p-value. Here collect such two-sample t-test, other related tests. We discuss validity consistency under recommendations between light these Finally, briefly testing genetic neutrality where knowledge vital proper rule.