IQR Outlier Detector Identify Anomalies with Interquartile Range Method

Detect outliers using the robust Interquartile Range (IQR) method. Supports both Tukey fences and MAD detection with 100% client-side processing for complete data privacy.

πŸš€ Quick Start

1

Enter Your Data

2

Choose Detection Method

3

View Detected Outliers

Data Input

Waiting for data input...
πŸ”’ 100% local computing, data never leaves browser

Supported Data Formats:

  • Comma-separated: 1.5, 2.8, 9.1, 16.2
  • Space-separated: 1.5 2.8 9.1 16.2
  • Newline-separated: one number per line
  • Scientific notation: 1.23e-4, 5.67E+8
  • Series mode: use "Group Name: value1, value2, value3" per line to compare multiple groups
  • Automatically ignores text and special characters

Outlier Detection Results

Outliers Detected: -
IQR: -
Lower Fence: -
Upper Fence: -

Interpretation: Enter data to see interpretation

Data Distribution & Outliers

Enter data to generate box plot

Please enter at least 4 numbers above to start calculating

Export Results

Detection Method

Outlier detection identifies data points that fall significantly outside the expected range. We support two robust methods:

Tukey Method (1.5Γ—IQR)

Uses the Interquartile Range (IQR) to define fences. Any point beyond Q1 - 1.5Γ—IQR or Q3 + 1.5Γ—IQR is flagged as an outlier. This is the standard method used in box plots.

MAD Method

Uses Median Absolute Deviation for robust outlier detection. Better suited for skewed distributions or when the data contains many outliers.

Frequently Asked Questions

What is the IQR method for outlier detection?

The IQR (Interquartile Range) method uses quartiles to identify outliers. Any data point beyond 1.5 times the IQR from Q1 or Q3 is considered an outlier. This is the standard method used in box plots.

When should I use Tukey vs MAD method?

Use Tukey method for general-purpose outlier detection with normally distributed or moderately skewed data. Use MAD method when your data is heavily skewed or contains many outliers, as MAD is more robust to extreme values.

What does it mean if no outliers are detected?

If no outliers are detected, all your data points fall within the expected range (within the fences). This suggests your data is relatively consistent, though it doesn't guarantee there are no unusual valuesβ€”just that they don't exceed the statistical thresholds.

Should I remove outliers from my dataset?

Not necessarily. Outliers may represent valid extreme values (e.g., top performers, rare events) or data errors. Always investigate outliers before removing them. Consider the context and whether they provide meaningful insights or indicate data quality issues.

Is my data safe with this detector?

Yes! All calculations happen in your browser. Your data never leaves your device and is not sent to any servers. The tool works completely offline.

Recent Updates

Added MAD outlier detection

Now supports both Tukey and MAD methods for robust outlier detection

Enhanced visualization

Outliers are now clearly highlighted in the box plot visualization