Outlier calculator

Paste your numbers separated by commas, spaces or new lines and choose a rule. The tool lists the values outside the limits, shows the limits it used and gives the data set without them.

At least 4 numbers, separated by commas, spaces or semicolons.
Used by the IQR rule. 1.5 is the usual choice; 3 marks only extreme outliers.
Used by the z-score rule. Values more than this many standard deviations from the mean are flagged. 3 is common.
Outliers48
First quartile Q1
15
Third quartile Q3
19
IQR (Q3 − Q1)
4
Number of outliers
1
Lower limit
9
Upper limit
25
Data without outliers
12, 14, 15, 15, 16, 17, 18, 19, 21

Runs in your browser: nothing you enter is sent anywhere.

How to use it

  1. Paste the numbers.
  2. Choose the IQR rule or the z-score rule and, if you wish, change the multiplier or cut-off.
  3. Read the outliers, the limits and the cleaned list.

How it works

IQR rule: the tool sorts the data, finds Q1 and Q3 by the chosen method and flags values outside Q1 − k × IQR and Q3 + k × IQR. Z-score rule: it flags values whose distance from the mean is more than the cut-off times the sample standard deviation. The limits shown are mean ± cut-off × s.

Worked example

12, 14, 15, 15, 16, 17, 18, 19, 21, 48. By the median-of-halves method Q1 = 15 and Q3 = 19, so IQR = 4 and the fences are 15 − 6 = 9 and 19 + 6 = 25. Only 48 lies outside.

Tips

  • Use the z-score rule on roughly bell-shaped data with many values; the IQR rule copes better with skewed data and small samples.
  • For the full five-number summary and a picture, use the quartile calculator or box plot maker.

Frequently asked questions

What is the 1.5 × IQR rule?

Tukey's fences. The interquartile range is IQR = Q3 − Q1. A value below Q1 − 1.5 × IQR or above Q3 + 1.5 × IQR is flagged as an outlier. Box plots draw their whiskers and dots from the same rule.

Which quartile method should I use?

Quartiles are defined several ways and the fences move slightly with the method. The default, the median of each half with the middle excluded, is the textbook one; the interpolated method matches Excel's QUARTILE.INC.

How is the z-score computed?

z = (x − mean) / s, using the sample standard deviation s with n − 1 in the divisor. The extreme values also pull the mean and s, so in small samples a z-score rule can miss a clear outlier.

Should I delete outliers?

Not automatically. An outlier can be a typing error or a real, informative case. Check the source first and report what you removed.

More in Math

All tools in this category: Math ·Updated November 3, 2026

Privacy choices

Google Analytics shows us which tools are used and whether they work. It may set cookies. Your choice is kept in this browser.