The Wilcoxon-Mann-Whitney test and other nonparametric tests based on rank randomization are distribution-free with respect to the probability of Type I errors; however, they are not distribution-free with respect to the probability of Type II errors and power to detect differences. Various shapes of nonnormal distributions substantially reduce the power of nonparametric tests. However, these densities influence parametric counterparts such as t and F to an even greater extent so nonparametric methods acquire an advantage by comparison.
Get full access to this article
View all access options for this article.
References
1.
BoxG. E. P.MullerM. (1958) A note on the generation of random normal deviates. Annals of Mathematical Statistics, 29, 610–611.
2.
HodgesJ.LehmannE. (1956) The efficiency of some nonparametric competitors of the t-test. Annals of Mathematical Statistics, 27, 324–335.
3.
RandlesR. H.WolfeD. A. (1979) Introduction to the theory of nonparametric statistics. New York: Wiley.
4.
ZimmermanD. W.ZumboB. D. (1993) The relative power of parametric and nonparametric statistical methods. In KerenG.LewisC. (Eds.), A handbook for data analysis in the behavioral sciences: methodological issues. Hillsdale, NJ: Erlbaum. Pp. 481–517.
