Overview
Find possible outliers in a small data set using Tukey's fences, the same rule box plots use.
Enter your values in any order.
Worked examples
| Inputs | Result |
|---|---|
| Your numbers12, 16, 14, 21, 19 | Outlier fences: [6.5, 26.5]Outliers: none |
Show 2 more examples
| Your numbers5, 7, 8, 12, 30 | Outlier fences: [-0.5, 19.5]Outliers: 30 |
| Your numbers-40, 10, 11, 12, 13 | Outlier fences: [7, 15]Outliers: -40 |
How it works
The fences are Q1 − 1.5 × IQR and Q3 + 1.5 × IQR. Values outside them are flagged. Quartiles use linear interpolation, the method in Excel's QUARTILE.INC, so other tools may give slightly different values for small data sets.
The formula
Fences = Q1 − 1.5 IQR, Q3 + 1.5 IQR
Practical tips
- For 12, 14, 16, 19 and 21, the fences are 6.5 and 26.5, so there are no outliers.
- For 5, 7, 8, 12 and 30, the upper fence is 19.5, so 30 is flagged.
- Investigate outliers before removing them, since they may be real.
Frequently asked questions
Should I delete outliers?
Only if they're errors. Real extreme values tell you something.
Why 1.5 × IQR?
It's Tukey's convention, and it flags few values in normal data.
Is there another method?
Z-scores above 3 are common for larger, roughly normal data sets.