rangsomtest

Rangsomtest is a term that appears in some non-English statistical texts to denote the rank-sum test, a nonparametric method for comparing two independent samples. In standard English-language usage, the equivalent concept is known as the Mann-Whitney U test or the Wilcoxon rank-sum test. The term rangsomtest is not widely used in major reference works, and it is often considered a translation variation or a misspelling of rangsumtest.

The rank-sum test assesses whether values from one population tend to be larger than those from another without assuming normality. The typical approach is to combine the two samples, assign ranks to all observations, and then compare the sums of ranks between the two groups. The test statistic can be taken as the U statistic or the sum of ranks for one group, from which a p-value is derived to evaluate the null hypothesis that the distributions are identical.

Assumptions for rangsomtest (rank-sum test) are minimal: observations should be independent, and the data should be at least ordinal. The test does not require normality and is therefore preferred for small samples or when distributions are skewed. It is sensitive to differences in central tendency, particularly medians, but less so to other distributional differences.

Variations and related tests include the Wilcoxon rank-sum test and the Mann-Whitney U test, which are mathematically equivalent under certain definitions. Common extensions address tied values and exact p-value calculations for small samples. See also nonparametric statistics, rank tests, and related two-sample procedures.