Performing a Kolmogorov-Smirnov test of two or more numeric columns’ distributions

A Kolmogorv-Smirnov test indicates whether the values of two or more numeric columns tend to be distributed differently. Unlike an ANOVA test, it looks for differences in the entire distribution and not just the mean; however, the Kolmogorov-Smirnov test usually requires more observations than other tests in order to detect a difference.

To perform a Kolmogorov-Smirnov test:

  1. In the Summary view, use the confidence slider to choose a desired confidence level
  2. Command-click two or more numeric columns of interest on the left
  3. Click the button labeled Bars (or Dots) below the Covariate view
  4. The result of the Kolmogorov-Smirnov test will appear in the Bottom Line

See also:

Back to Wizard Help