Mood’s Median

Mood’s Median test is one of the most straightforward non‑parametric tools for comparing medians across multiple groups. Unlike Kruskal‑Wallis, which compares distributions, Mood’s Median focuses specifically on medians, making it particularly useful when the central tendency—not the overall distribution—is your primary interest. 

The test is robust to outliers and works well when data is heavily skewed or when distributions differ in shape. This makes it ideal for processes where extreme values are common, such as service times, repair durations, or financial transaction amounts. Because the test is based on counts rather than ranks, it is less sensitive to distributional differences and easier to interpret. 

The process begins by calculating the overall median across all groups. Each observation is then classified as above or below the median. A contingency table is created, and a chi‑square statistic is used to evaluate whether the proportion of values above and below the median differs across groups. If the proportions differ significantly, it suggests that at least one group has a different median. 

Mood’s Median is not as powerful as Kruskal‑Wallis when distributions are similar, but it is more robust when distributions differ in shape or when extreme values distort rank‑based methods. This trade‑off makes it a valuable tool in situations where data quality is imperfect or where distributional assumptions are clearly violated. 

In the Analyze phase, Mood’s Median helps you identify meaningful differences in central tendency across multiple groups without being misled by outliers or skewness. It provides a clear, intuitive way to compare medians and supports confident decision‑making in complex, real-world environments.