Outlier Calculator
Enter a list of numbers to calculate outlier calculator with the existing statistics engine.
Calculator
Results
Enter your numbers, then press Calculate.
Your mean, median and mode appear here, together with the working, the frequency table and the advanced statistics.
Frequency table
Advanced statistics
Step by step
Saved calculations
History is stored only in this browser. Nothing is sent to our server, and Clear history removes it.
Results are for educational purposes. Please verify important calculations. Read the disclaimer, the privacy policy and the guide to how to use the calculator.
Learn about outliers using the 1.5 × IQR rule
What does this calculator measure?
An outlier is a value unusually far from most of the data. The interquartile range, or IQR, is Q3 minus Q1.
How is it calculated?
The lower fence is Q1 minus 1.5 times IQR. The upper fence is Q3 plus 1.5 times IQR. Values outside these fences are flagged for review, not automatically removed.
Can you see a worked example?
For 1, 2, 3, 4, 5, 20, the value 20 may be flagged as an outlier, depending on the quartile convention. Always inspect the actual quartiles and fences in the result.
How do you use this calculator?
Enter your numbers separated by spaces or commas. Check the selected result at the top, press Calculate, and read the working. You can change your numbers and calculate again. Other statistics are optional, so you can focus on the result you need.
When is this useful?
This tool is useful for school exercises, practice questions, checking homework and understanding patterns in a small data set. Start with a few values that you can verify by hand. When you use a large data set, check for typing errors before interpreting the answer.
What mistakes should you avoid?
Check negative signs, repeated numbers, decimal points and the number of observations. Do not mix sample and population formulas. Different quartile and percentile conventions can give different answers. Keep your original data and explain the method you used.
How should you explain your result?
Write down what the numbers represent, the units if any, and the statistic you selected. Explain what the result says about the data rather than simply copying a number. For comparisons, make sure both data sets were measured in the same way.
What should you do before sharing a result?
Read the result in the context of your question. A statistic is a summary, not a complete description of every observation. Include the data source and sample size when they are known. If you are making a decision, check whether the data is representative and whether a different summary would be more useful.
How can you check the answer?
Try a short example and calculate the result on paper. Look at the step-by-step working and compare it with your own steps. If your answer differs, check the input and the formula. Read the guide for more background before changing the numbers.
Frequently asked questions
Can I enter negative numbers?
Yes. Enter negative values with a minus sign.
Can I use decimal numbers?
Yes. Use a decimal point for each decimal value.
Do repeated values count?
Yes. Repeated values remain part of your data.
Does the tool change my original numbers?
No. You can edit the input and calculate again.
Can methods differ between tools?
Yes. Check the stated statistical method when comparing results.
Continue learning with the calculation methods guide, statistics practice questions and main calculator.
How does this calculator decide that a value is an outlier?
It uses the 1.5 × IQR rule. It finds Q1 and Q3 by the median-of-halves method (with an odd count, the median is included in both halves), then works out the interquartile range IQR = Q3 − Q1. The lower fence is Q1 − 1.5 × IQR and the upper fence is Q3 + 1.5 × IQR. A value is flagged only if it is below the lower fence or above the upper fence.
The quartile method is explained with examples in the five-number summary guide. A different quartile method can move the fences slightly.
Worked example with an outlier: 1, 2, 3, 4, 5, 20
The median is 3.5. The lower half is 1, 2, 3, so Q1 = 2, and the upper half is 4, 5, 20, so Q3 = 5. The IQR is 5 − 2 = 3, and 1.5 × 3 = 4.5.
| Step | Working | Result |
|---|---|---|
| Lower fence | 2 − 4.5 | −2.5 |
| Upper fence | 5 + 4.5 | 9.5 |
| Check | 20 is above 9.5 | 20 is flagged |
| Check | 1 to 5 are inside the fences | not flagged |
With the 20 included, the mean is about 5.83. Without it, the mean is 3. The median moves only from 3.5 to 3, which is why the median is called resistant to outliers.
Worked example with no outliers: 12, 15, 15, 18, 20
Here n = 5, the median is 15, and it is included in both halves. Q1 = 15 and Q3 = 18, so IQR = 3 and 1.5 × 3 = 4.5. The fences are 15 − 4.5 = 10.5 and 18 + 4.5 = 22.5. Every value sits between them, so nothing is flagged.
What should I do with a flagged value?
Check it first. It may be a typing slip, a different unit or a real but unusual result. Do not delete it just because it is flagged. A good habit is to report the mean and the median with and without the value, and say which one you used.
To see how outliers change an average, read the mean vs median vs mode guide. For the quartile method, read the five-number summary guide, or find the spread with the range calculator.
Last updated:
Other calculators
How are these results calculated?
The calculator uses the median-of-halves method for quartiles. For an odd number of values, the middle value is included in both halves. Percentiles use linear interpolation with position (p ÷ 100) × (n − 1). Other tools may use different rules.
Method and references
For an independent explanation of statistical methods, see the NIST/SEMATECH e-Handbook of Statistical Methods. This is a reference for general methods, not a claim of endorsement or external review. Check the calculator’s displayed steps for the exact operation used.