bmtest: A Jamovi Module for Brunner–Munzel’s Test—A Robust Alternative to Wilcoxon–Mann–Whitney’s Test

Abstract

In psychological research, comparisons between two groups are frequently made to demonstrate that one group exhibits higher values. Although Welch’s unequal variances t-test has become the preferred parametric test for this purpose, surpassing Student’s equal variances t-test, the Wilcoxon–Mann–Whitney test remains the predominant nonparametric approach despite sharing similar limitations with Student’s t-test. Specifically, the Wilcoxon–Mann–Whitney test is associated with strong, unrealistic assumptions and lacks robustness when these assumptions are violated. The Brunner–Munzel test overcomes these limitations, featuring fewer assumptions, akin to Welch’s t-test in the parametric domain, and has thus been recommended over the Wilcoxon–Mann–Whitney test. However, the Brunner–Munzel test is currently unavailable in user-friendly statistical software, such as SPSS, making it inaccessible to many researchers. In this paper, I introduce the bmtest module for jamovi, a freely available user-friendly software. By making the Brunner–Munzel test accessible to a wide range of researchers, the bmtest module has the potential to improve nonparametric statistical analysis in psychology and other disciplines.

Keywords

• Brunner–Munzel test
• jamovi
• Wilcoxon–Mann–Whitney test
• nonparametric
• robust