Kruskal–Wallis H: what it means and how to read it

In the report: Kruskal-Wallis H — How far the mean ranks of the groups are apart, with the tie correction.

How to read it

The rank-based counterpart of the ANOVA F: the larger H, the more the groups differ. Read Asymp. Sig.; in a report: H(2) = 1.71, p = .425.

In plain words

A teacher compared three classes in a reading contest and wanted to know whether they read at the same level. A few scores were very high, so she did not trust the averages. She put all pupils in one line from best to worst and looked at the average place of each class. The Kruskal-Wallis H came out 8.4. The bigger H is, the further apart the average places of the classes are. The chance column next to it said the gap was hard to put down to chance alone. Now she can say the three classes did not read at the same level, though H does not tell which class differs.

How it is computed

H = (12 / (N(N + 1)) · Σ Rᵢ²/nᵢ − 3(N + 1)) / (1 − Σ (t³ − t) / (N³ − N))

Example

The cell with Kruskal-Wallis H is selected in each report; point at a label to see its note.

  • — the sample data (50 rows) the examples use
  • — three groups =STATS_KRUSKAL(Data!A1:T51, "x,x1,x2")

Where you see it

Get it in your sheet

One formula in a cell gives the whole report, Kruskal–Wallis H included. ALS Stats is a free add-on for Google Sheets™.

Install ALS Stats from Google Workspace Marketplace (opens in a new tab)