1. Home
  2. Medical Statistics Core
  3. Tests
  4. Mann–Whitney U test (with visualisation)

Mann–Whitney U test (with visualisation)

Mann–Whitney U test (with visualisation) icon

Test a null hypothesis involving the difference between randomly selected values from two samples by performing a two-sample Wilcoxon or Mann-Whitney U test.

Details

The Mann-Whitney U test is used to compare whether there is a difference in the dependent variable for two independent groups. Select a numerical test variable (test_var) and a categorical grouping variable (group_var).

Output

The first section of the output is the printed output of a wilcox.test run in R. There are more details on this printed output in the R documentation for wilcox.test.

The output also includes a boxplot comparing the distributions of the selected groups.

Parameters

 

Variable name Required Constraints Description
test_var Yes Column Input.

Integer, Decimal

Numerical variable upon which the test will be computed.
group_var Yes Column Input.

Text, Integer, Boolean, Date, DateTime

Categorical variable which provides the two independent samples.
mu Yes Decimal The null hypothesis is that the distributions of the samples differ by a location shift of mu
alternative Yes Text, three options. Select between a two sided and one-sided test by choosing either two.sided, greater or less.
confidence_level Yes Decimal between 0.5 and 0.99 Set your confidence interval, default to 0.95.

See Also

Updated on October 18, 2018

Was this article helpful?

Related Articles

Not the solution you were looking for?
Click the link below to submit a support ticket
CONTACT SERVICE DESK