Description | Kruskal-Wallis test is a non-parametric version of one way ANOVA. It determines whether the medians of two or more groups are different. |
Why to use | To identify whether there is a significant difference between the medians in the groups. |
When to use | - When all the independent variables are numerical.
- When there are more than two variables in single or multiple datasets.
- When there is no relationship between the members of all groups.
| When not to use | - In the case of non-continuous, categorical, or textual variables.
- If a group contains any dependent variable and is not on the Ordinal, Ratio, or Interval scale.
- If a group contains constants, discrete, and empty/missing values.
|
Prerequisites | - The independent variable should be a numerical value. It should not possess any infinite or missing value.
- All the groups should have the same size of distributions.
|
Input | Numeric dataset | Output | |
Statistical Methods Used | - H-statistic
- p Value
- Alpha (α)
| Limitations | this method cannot identify which dataset/column has a different median.
|