Input and settings
Configure the calculation
Formula used
Interquartile range and outlier rules
IQR = Q3 − Q1
The IQR measures the middle half of observations.
Lower = Q1 − k(IQR)
Values below this fence are potential outliers.
Upper = Q3 + k(IQR)
Values above this fence are potential outliers.
How to use
Calculate and interpret IQR
- Select the correct input mode.
- Paste data or import a CSV file.
- Choose a quartile method and fence multiplier.
- Configure missing values, filters, and duplicate handling.
- Calculate, review charts, then export results.
| Example data | Q1 | Median | Q3 | IQR | Likely outlier |
|---|---|---|---|---|---|
| 4, 5, 7, 8, 9, 10, 12, 13, 15, 18, 22, 45 | 7.5 | 9.5 | 16.5 | 9 | 45 |
Frequently asked questions
Interquartile range questions
What does the IQR measure?
It measures the spread of the middle fifty percent.
Why is IQR useful in machine learning?
It supports robust outlier detection and feature scaling.
Which quartile method should I choose?
Use the method matching your software or reporting standard.
What does an IQR of zero mean?
The middle half contains no measurable spread.
Are all flagged values incorrect?
No. Outliers may be valid, rare, or domain-important observations.
Should I remove detected outliers?
Review causes first. Removal can distort useful model information.
How does robust scaling use IQR?
It centers values by median and divides by IQR.
Can I compare several features?
Yes. Use feature-table mode with CSV column headers.
What multiplier should detect extreme outliers?
A three-IQR multiplier commonly flags extreme observations.