Mood's Median Test

Mood's median test is a nonparametric test of whether two or more groups share the same median, based on counts of observations above and below the pooled median.

Mood 중앙값 검정은 전체 중앙값보다 큰 값과 그렇지 않은 값의 개수를 집단별로 세어 둘 이상의 집단 중앙값이 같은지 검정하는 비모수 방법이다.

Can I compare medians across groups whose distributions have different shapes?

집단마다 분포 모양이 다를 때도 중앙값을 비교할 수 있나요?

Mood's median test classifies each observation as above or not above the grand median and applies a chi-square test of independence to the resulting group-by-position table. It is robust to outliers and shape differences but has lower power than the Kruskal-Wallis test when distributions are similar in shape.

When to use Mood's Median Test

• You want to compare medians across two or more independent groups.
• The data contain extreme outliers or heavily skewed values.
• Group distributions may differ in shape.

When to use something else

• Distributions have similar shapes and you want more power; use the Kruskal-Wallis test instead.
• The data are normal with equal variances; use one-way ANOVA instead.

Assumptions and requirements

• Observations are independent.
• Expected counts in the 2 × k table are large enough for the chi-square approximation.

What the analysis reports

• Grand median
• Counts above and at-or-below the median by group
• Chi-square statistic, df and p-value
• Fisher exact p-value when there are two groups

How to report it

Mood's median test showed that the proportion of scores above the grand median (Mdn = [value]) [differed/did not differ] across groups, χ²([df], N = [value]) = [value], p = [value].

Related methods

• Kruskal-Wallis Test
• Mann-Whitney U Test
• Chi Square Test

Run Mood's Median Test on Quorant