Describes an exact nonparametric test suitable for testing whether 1 sample is stochastically larger than another. The advantage of the test over currently available forms of the Mann-Whitney U test is that it can handle small samples with numerous ties. Comparisons with the Mann-Whitney and normal approximations to it are presented. The use of the statistic is illustrated with data drawn from current research literature. (PsycINFO Database Record (c) 2006 APA, all rights reserved).
- nonparametric comparison of 2 small samples with many shared variables
ASJC Scopus subject areas