Robustness without rank order statistics

Abstract

An alternative to conventional rank tests based on a Euclidean distance analysis space is described. Comparisons based on exact probability values among classical two-sample t-tests and the Wilcoxon–Mann–Whitney test illustrate the advantages of the Euclidean distance analysis space alternative.

Keywords:

• analysis space
• Euclidean distance
• rank-order statistics
• robustness

Notes

Bergmann, Ludbrook, and Spooren noted deficiencies in the computation of probability values for the Wilcoxon–Mann–Whitney test by statistical software packages, concluding that “the only accurate form of the Wilcoxon–Mann–Whitney procedure is one in which the exact permutation null distribution is compiled for the actual data” [2, p. 72]