Mann-Whitney test

Also found in: Dictionary, Encyclopedia, Wikipedia.

Mann-Whitney test, Mann-Whitney U test

Mann-Whitney test

A statistical test of the probability that two independent sets of observations come from the same population. The Mann-Whitney test is independent of distribution and can be used when the t test is inappropriate.


Henry Berthold, U.S. mathematician, 1905–.
Mann-Whitney test - rank sum test.


Donald Ransom, U.S. statistician, 1915–.
Mann-Whitney test - see under Mann, Henry Berthold
References in periodicals archive ?
Neither the i-test nor the Mann-Whitney test gave any indication of or took into account the bimodality of the data distribution in the group with subsequent seizures.
The Mann-Whitney test considers median responses, rather than mean responses, and thereby allows comparisons to be made even when data are non-interval and/or non-normal.
Mann-Whitney test showed that the difference between percentage of inhabitants with middle level of education that use Internet in European developed and European post-communist countries is not statistically significant for age group 16-24, and is statistically significant for the age groups 25-54 and 55-74 in both 2006 and 2010 year.
Table 7 presents the results of the non-parametric Mann-Whitney test for significant differences between our three fund categories for each of our return variables.
The Mann-Whitney test is a powerful nonparametric test for comparing two independent populations.
Vegetation data from within quadrats and between quadrats and transects were compared between sites utilizing the Wilcoxon Mann-Whitney test of SAS 8.
Significance of differences and frequency distribution of values between the groups were assessed by the Mann-Whitney test and the [chi square] test (for categoric variables).
We calculated that we would need 34 patients per group to have an 80% chance of detecting a 35% reduction in 24-hour morphine consumption at a 5% significance level, using a Mann-Whitney test with a 0.
The twotail Mann-Whitney test supports the conclusion that these samples are significantly different at the 1% level.
This is statistically confirmed by a Mann-Whitney test (p = 0.
P value generated by non-parametric tests such as chi-square or Mann-Whitney test.
The Mann-Whitney test was used to compare the SF-36 and BI responses with seizure status and duration of epilepsy, while the Kruskal-Wallis test was chosen to compare the SF-36 and BI responses with seizure frequency in both syndromes.