Non-parametric or distribution free test is a statistical procedure whereby the data does not match a normal distribution. The data used in non-parametric test is frequently of ordinal
data type, thus implying it does not depend on arithmetic properties. Consequently, all tests
involving the ranking of data are non-parametric and also no statement about the distribution
of data is made.
The Kruskal-Wallis H-test is a non-parametric statistical procedure for comparing more than
two samples that are independent. The parametric equivalent to this test is the one-way
analysis of variance (ANOVA).
Yet ANOVA is used for normally distributed data, but Kruskal Wallis can be perform
without the data being normally distributed
The H-test is a generalization of the Mann-Whitney test, a test for knowing whether the two
samples chosen are taken from the same population. The p value for both the Kruskal Wallis
and the Mann- Whitney test are equal.
However Kruskal Wallis unlike Mann- Whitney test is used in samples to evaluate their
degree of association.

свернуть