Wilcoxon-Mann-Whitney Test

The non-parametric Wilcoxon-Mann-Whitney test (a.k.a. Mann–Whitney U test, Wilcoxon rank-sum test) tests if two independent samples, with ordinal attributes, originate from populations with the same distribution.

More information can be found at Wilcoxon-Mann-Whitney test on Wikipedia.

Hypotheses:
H0: the distributions of the populations are the same.
HA: the distributions of the populations are different.

Options

Test Column
Column containing independent, ordinal samples.
Group Column
Column containing group labels.
Group One
Label of first group.
Group Two
Label of second group.
Missing Value Strategy
Strategy on how missing and Double.NaN values are treated.
  • Fixed: Missing values will be replaced with Double.NaN.
  • Failed: An exception is thrown and the node fails to execute.
  • Maximal: Missing values are considered to be maximal in the ranking.
  • Minimal: Missing values are considered to be minimal in the ranking.

Input Ports

Icon
Input table with at least one ordinal column and one nominal column (groups).

Output Ports

Icon
Output table with Wilcoxon-Mann-Whitney test statistics, p-Value, and descriptive statistics of group ranks.

Views

This node has no views

Workflows

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.