Statistics & Probability

Instant, private, and free

Outlier Calculator.

Check a candidate against Tukey's 1.5×IQR outlier fences.

On-device calculationNo signup

Set your values

Results update as you type.

02 / ANSWER

Reference data

Candidate value

2.500000

6.500000

iqr

4.000000

lowerFence

-3.500000

upperFence

12.500000

isOutlier

true

Results update automatically as you type.

Use Cases

Quality Control in Manufacturing

Use the calculator to quickly check if a new measurement from a production batch falls outside the expected range based on historical data, helping identify defects or process shifts.

Example: Reference data: lengths of 50 parts; candidate: length of a new part.

Data Cleaning for Analysis

Before running statistical analyses, use the calculator to flag potential outliers in your dataset that might skew results, allowing you to decide whether to exclude or investigate them.

Example: Reference data: survey responses; candidate: an unusually high response.

Frequently Asked Questions

How does the Outlier Calculator determine if a value is an outlier?: The calculator uses the Tukey IQR method. It computes the first quartile (Q1) and third quartile (Q3) of your reference data, then calculates the interquartile range (IQR = Q3 - Q1). Fences are set at Q1 - 1.5*IQR and Q3 + 1.5*IQR. If your candidate value falls outside these fences, it is flagged as an outlier.
What is the significance of the 1.5 multiplier in the IQR method?: The 1.5 multiplier is a standard convention in Tukey's boxplot rule. It defines the 'inner fences' that identify potential outliers. Values beyond these fences are considered unusual relative to the bulk of the data. This multiplier balances sensitivity and specificity, but you can adjust it if you have domain-specific needs.
Can I use this calculator for small sample sizes?: Yes, but interpret results with caution. With small reference data sets, quartiles may be less stable, and the IQR method might flag values that are not truly anomalous. It's best used with at least 5-10 data points for more reliable outlier detection.

Tips & Common Mistakes

Tips

Ensure your reference data is representative of the normal range you expect. Including outliers in the reference data can skew the quartiles and affect the fences.
The calculator uses the Tukey IQR method with a 1.5 multiplier. For more conservative outlier detection, you can use a higher multiplier like 3, but this calculator uses the standard 1.5.
If your candidate value is exactly on the fence, it is not considered an outlier by this method. Only values strictly outside the fences are flagged.
For small reference datasets, consider using other outlier detection methods or visual inspection, as quartiles may be less reliable.

Common Mistakes to Avoid

Including the candidate value in the reference data. The reference data should only contain the baseline values you want to compare against, not the candidate you are testing.
Misinterpreting the result: a value flagged as an outlier is not necessarily 'bad' or erroneous; it may be a legitimate extreme value that warrants investigation.
Using the calculator without checking the distribution of your reference data. The IQR method assumes a roughly symmetric distribution; for highly skewed data, other methods may be more appropriate.

Last updated: August 13, 2026

Use Cases

Quality Control in Manufacturing

Use the calculator to quickly check if a new measurement from a production batch falls outside the expected range based on historical data, helping identify defects or process shifts.

Example: Reference data: lengths of 50 parts; candidate: length of a new part.

Data Cleaning for Analysis

Before running statistical analyses, use the calculator to flag potential outliers in your dataset that might skew results, allowing you to decide whether to exclude or investigate them.

Example: Reference data: survey responses; candidate: an unusually high response.

Frequently Asked Questions

How does the Outlier Calculator determine if a value is an outlier?

The calculator uses the Tukey IQR method. It computes the first quartile (Q1) and third quartile (Q3) of your reference data, then calculates the interquartile range (IQR = Q3 - Q1). Fences are set at Q1 - 1.5*IQR and Q3 + 1.5*IQR. If your candidate value falls outside these fences, it is flagged as an outlier.

What is the significance of the 1.5 multiplier in the IQR method?

The 1.5 multiplier is a standard convention in Tukey's boxplot rule. It defines the 'inner fences' that identify potential outliers. Values beyond these fences are considered unusual relative to the bulk of the data. This multiplier balances sensitivity and specificity, but you can adjust it if you have domain-specific needs.

Can I use this calculator for small sample sizes?

Yes, but interpret results with caution. With small reference data sets, quartiles may be less stable, and the IQR method might flag values that are not truly anomalous. It's best used with at least 5-10 data points for more reliable outlier detection.

Tips & Common Mistakes

Tips

Ensure your reference data is representative of the normal range you expect. Including outliers in the reference data can skew the quartiles and affect the fences.
The calculator uses the Tukey IQR method with a 1.5 multiplier. For more conservative outlier detection, you can use a higher multiplier like 3, but this calculator uses the standard 1.5.
If your candidate value is exactly on the fence, it is not considered an outlier by this method. Only values strictly outside the fences are flagged.
For small reference datasets, consider using other outlier detection methods or visual inspection, as quartiles may be less reliable.

Common Mistakes to Avoid

Including the candidate value in the reference data. The reference data should only contain the baseline values you want to compare against, not the candidate you are testing.
Misinterpreting the result: a value flagged as an outlier is not necessarily 'bad' or erroneous; it may be a legitimate extreme value that warrants investigation.
Using the calculator without checking the distribution of your reference data. The IQR method assumes a roughly symmetric distribution; for highly skewed data, other methods may be more appropriate.

Use Cases

Quality Control in Manufacturing

Use the calculator to quickly check if a new measurement from a production batch falls outside the expected range based on historical data, helping identify defects or process shifts.

Example: Reference data: lengths of 50 parts; candidate: length of a new part.

Data Cleaning for Analysis

Before running statistical analyses, use the calculator to flag potential outliers in your dataset that might skew results, allowing you to decide whether to exclude or investigate them.

Example: Reference data: survey responses; candidate: an unusually high response.

Frequently Asked Questions

How does the Outlier Calculator determine if a value is an outlier?

What is the significance of the 1.5 multiplier in the IQR method?

Can I use this calculator for small sample sizes?

Tips & Common Mistakes

Tips

Ensure your reference data is representative of the normal range you expect. Including outliers in the reference data can skew the quartiles and affect the fences.
The calculator uses the Tukey IQR method with a 1.5 multiplier. For more conservative outlier detection, you can use a higher multiplier like 3, but this calculator uses the standard 1.5.
If your candidate value is exactly on the fence, it is not considered an outlier by this method. Only values strictly outside the fences are flagged.
For small reference datasets, consider using other outlier detection methods or visual inspection, as quartiles may be less reliable.

Common Mistakes to Avoid

Including the candidate value in the reference data. The reference data should only contain the baseline values you want to compare against, not the candidate you are testing.
Misinterpreting the result: a value flagged as an outlier is not necessarily 'bad' or erroneous; it may be a legitimate extreme value that warrants investigation.
Using the calculator without checking the distribution of your reference data. The IQR method assumes a roughly symmetric distribution; for highly skewed data, other methods may be more appropriate.