17.5.4 Mann-Whitney Test


The Wilcoxon/Mann-Whitney test is a common non-parametric test for comparing two samples with non-normal distributions and, thus, is a powerful substitute for the two-sample t-test. Mann-Whitney tests whether the samples are taken from populations with identical distributions. This test is performed on ranked data and therefore can be used for ordinal data.

Handling Missing Values

The missing values in the data range will be excluded in the analysis

From Origin 2015, missing values in the grouping range and the corresponding data values will be excluded in analysis. In the previous version, missing values in the grouping range will be considered as a group.

Performing Mann-Whitney Test

To perform a Mann-Whitney test:

  1. Select Statistics: Nonparametric Tests: Mann-Whitney Test. This opens the mwtest dialog box.
  2. Specify the Data Range and the desired Alternate Hypothesis. Then check Exact P Value box for exact p-value.
  3. Upon clicking OK, an analysis report sheet is generated that lists the ranks table, degrees of freedom, the U and Z statistic, the associated p-value, and the test conclusion.
Note: Origin provides both asymptotic and exact p-value as result for Mann-Whitney test.

Topics covered in this section: