When the normality assumption is not met in a randomized block design, which of the following tests should be used?
A) Wilcoxon rank sum test
B) Wilcoxon signed ranks test
C) Kruskal-Wallis test
D) Friedman rank test

Question

Quizplus · Accepted Answer

When the normality assumption is not met in a randomized block design, the appropriate test is the Friedman rank test, which is a non-parametric test. Option D is the correct answer as it is the only non-parametric test listed among the options. The Wilcoxon rank sum test and the Wilcoxon signed ranks test are both used for independent or paired samples with continuous data, but they rely on the assumption of normality. The Kruskal-Wallis test is used for comparing three or more independent groups, but it also assumes normality.

When the Normality Assumption Is Not Met in a Randomized