The Kruskal-Wallis test is a non-parametric method for comparing three or more what?

• A. Means
• B. Medians
• C. Variances
• D. Proportions

Answer: (A)

Kruskal-Wallis is a nonparametric method used to compare three or more independent groups to see if their central tendency differs. It works by ranking all the data across groups and testing whether those ranks come from the same distribution, which, when group shapes are similar, is interpreted as differences in central tendency. This makes it the nonparametric counterpart to ANOVA, which targets means in the parametric setting; using ranks allows the test to avoid normality assumptions.

Among the options, this test is about central tendency across multiple groups, which is why it’s described as comparing means in many contexts. It does not directly focus on variances or proportions, and it isn’t a test for those aspects.