Performing a Friedman test of the relative values of three or more numeric columns

A Friedman test indicates whether values in some columns tend to be larger or smaller than values in other columns within the same row. It is similar to a repeated-measures ANOVA test in that it can be used to analyze related observations (for example, test scores from the same students across time). Unlike repeated-measures ANOVA, the Friedman test makes no assumptions about the underlying distribution of values.

To perform a Friedman test:

1. In the Summary view, use the confidence slider to choose a desired confidence level
2. Command-click three or more numeric columns of interest on the left
3. Check the box labeled Repeated measures below the summary picture
4. Select the button below the Covariate view labeled Boxes
5. The result of the Friedman test will appear in the Bottom Line

See also:

• Performing a repeated measures ANOVA test of three or more numeric columns’ means
• Performing a Wilcoxon signed-rank test of the median difference between two numeric columns
• Adjusting a p-value to account for multiple comparisons
• Performing a Kruskal-Wallis test of three or more numeric columns’ medians
• Formatting descriptive and inferential statistics
• About the Summary view

---

Back to Wizard Help