Friedman Twoway Analysis of Variance (ANOVA) by Ranks

Abstract

The Friedman Twoway Analysis of Variance (ANOVA) by Ranks Test is often viewed as the nonparametric equivalent of the parametric Twoway Analysis of Variance (ANOVA). Both the nonparametric Friedman Test and parametric Twoway ANOVA are used to determine if there are statistically significant differences for comparisons of multiple groups, with different factors for each group. However, it may be too convenient to view these tests as being mere complements of each other. The Friedman Test, as a nonparametric test, is used with ranked data, particularly for when: (1) the data do not meet the rigor of interval data, (2) there are serious concerns about extreme deviation from normal distribution, and (3) there is considerable difference in the number of subjects for each breakout group. The use of a block-type research design, a factorial design typically associated with ANOVA, is introduced in this lesson. This lesson also reinforces the many quality assurance measures that should be attempted before actual implementation of this type of inferential analysis.