Three or More Related Populations

One-Way Analysis of Variance

Friedman Test

The parametric F test of equality of several population means cannot be used when data are measured on an ordinal scale because the F test is derived from the normal distribution.1 The Friedman test performs an analysis of variance (ANOVA) on ordinal data without relying on normality. The parametric counterpart of the Friedman test is the repeated measures analysis of variance in a randomized block design. Therefore, the Friedman test is also known as the Friedman two-way analysis of variance by ranks. As in the randomized block design, the expectation is that members of a group in a block are more homogenous than members from other blocks.

Additional advantages ...

Get A Primer on Nonparametric Analysis, Volume II now with the O’Reilly learning platform.

O’Reilly members experience live online training, plus books, videos, and digital content from nearly 200 publishers.