Interquartile Range (IQR)
IQR = Q₃ - Q₁Measures statistical dispersion (middle 50% of data)
Larger IQR = more spread out data
Generate box plots (box-and-whisker plots) with five-number summary: min, Q1, median, Q3, max.
https://www.solvezi.com/statistics/box-plot-generator
Visualize data distribution and identify outliers
| # | Value | Action |
|---|---|---|
| 1 | ||
| 2 | ||
| 3 | ||
| 4 | ||
| 5 | ||
| 6 | ||
| 7 | ||
| 8 | ||
| 9 | ||
| 10 |
10 values
Minimum
0.0000
Q₁ (25th)
0.0000
Median (Q₂)
0.0000
Q₃ (75th)
0.0000
Maximum
0.0000
Mean (x̄)
0.0000
Box Plot Details
Understanding the Box Plot
• Box: Shows the interquartile range (IQR) from Q₁ to Q₃
• Line inside box: Median (Q₂) - the middle value
• Green dot: Mean (average) of the data
• Whiskers: Extend to min/max within 1.5×IQR of Q₁/Q₃
• Red dots: Outliers - values outside the whiskers
Box Plot Generator • Visualize data distribution and identify outliers
Master box plots (box-and-whisker plots) — learn to create, interpret, and compare data distributions. Understand quartiles, outliers, skewness, and when to use this powerful visualization tool.
A box plot (or box-and-whisker plot) is a standardized way of displaying the distribution of data based on the five-number summary: minimum, first quartile (Q₁), median (Q₂), third quartile (Q₃), and maximum.
Smallest value in the dataset
The lowest data point
25th percentile - lower quartile
Median of lower half of data
50th percentile - middle value
The central data point
75th percentile - upper quartile
Median of upper half of data
Largest value (within whiskers)
Highest data point within fences
IQR = Q₃ - Q₁Measures statistical dispersion (middle 50% of data)
Larger IQR = more spread out data
Lower Fence = Q₁ - 1.5 × IQRThreshold for lower outliers
Values below = outliers
Upper Fence = Q₃ + 1.5 × IQRThreshold for upper outliers
Values above = outliers
Range within 1.5×IQR of Q₁/Q₃Extend to min/max within fences
Show data range excluding outliers
Values < L.Fence or > U.FenceExtreme values beyond whiskers
Points to investigate further
x̄ = (Σxᵢ) / nArithmetic average
May differ from median in skewed data
Median centered in box, whiskers roughly equal length
Data evenly distributed around center
Median closer to bottom, longer upper whisker
Tail extends to larger values
Median closer to top, longer lower whisker
Tail extends to smaller values
Large IQR, spread out box
High variability in middle 50%
Small IQR, compact box
Low variability, consistent data
Whiskers much longer than box
Data has extreme values
Identify production issues and maintain quality standards
Compare student performance and identify struggling students
Understand return distributions and risk factors
Compare treatment outcomes and identify patterns
Evaluate athlete consistency and team performance
Quickly understand data distributions and outliers
Arrange all values in ascending order from smallest to largest
Locate the middle value. For even n, average the two middle values
Q₁ = median of lower half. Q₃ = median of upper half
IQR = Q₃ - Q₁ (the 'box' height)
L.Fence = Q₁ - 1.5×IQR, U.Fence = Q₃ + 1.5×IQR
Extend to min and max values within fences
Mark points beyond fences as separate dots
Mark the arithmetic mean for comparison
Best for:
Distribution shape, modality, bin details
Limitation:
Requires bin selection, less compact
Best for:
Summary statistics, outliers, multiple comparisons
Limitation:
Hides multimodality, distribution detail
Best for:
Distribution shape + summary stats combined
Limitation:
More complex, harder for non-statisticians
Best for:
Individual data points for small datasets
Limitation:
Overcrowding with large n
Confusing whiskers with min/max including outliers
Whiskers stop at the smallest/largest values WITHIN the fences, not including outliers
Outliers are plotted as individual points beyond whiskers
Using mean instead of median for box center
The line in the box always represents the median, not the mean
Mean can be added as a separate marker (like a dot or plus sign)
Incorrect quartile calculation for small datasets
For n < 5, quartiles may not be well-defined. Consider alternative visualizations
Box plots work best with at least 5-10 data points
Ignoring outliers in analysis
Outliers deserve investigation — they may be errors or important discoveries
Always examine outliers before removing them
The 1.5×IQR rule was proposed by John Tukey (the inventor of box plots) as a reasonable threshold for identifying potential outliers. Values beyond 1.5×IQR from the quartiles are considered 'outside' values worth investigating. The multiplier 1.5 was chosen empirically and works well for normally distributed data.
Box plots can hide multimodality because they only show summary statistics. For bimodal or multimodal distributions, the box plot may look like a single box with wide IQR. Always combine box plots with histograms or density plots when exploring distribution shape.
Tukey's original box plot uses the 1.5×IQR rule for whiskers. Modified box plots can use different multipliers (e.g., 3×IQR for extreme outliers) or percentiles (e.g., 2nd and 98th percentiles). Our calculator uses the standard Tukey method with optional outlier display.
Yes! Side-by-side box plots are excellent for comparing multiple groups, treatments, or categories. They reveal differences in central tendency, spread, and outliers across groups. Just ensure consistent scales for meaningful comparison.
While box plots can be created with as few as 5 data points, they become more reliable with larger samples (n ≥ 20-30). With very small samples, quartiles may not be stable, and patterns may not represent the true population.
Skewness appears as asymmetry: Right-skewed data has median closer to Q₁ with longer upper whisker. Left-skewed data has median closer to Q₃ with longer lower whisker. Symmetric data has median centered and whiskers roughly equal.
Alternatives include: violin plots (show density shape), beeswarm plots (show individual points), notched box plots (show confidence intervals), and bean plots (combine density and box plot elements). Each has strengths for different analysis needs.
For large datasets (n > 1000), consider: 1) Sampling a subset, 2) Using a quantile box plot (fewer displayed quantiles), 3) Using a violin plot, or 4) Creating multiple box plots by grouping (e.g., by time period). Our calculator handles moderate datasets efficiently.
1915-2000
Invented the box plot in 1970
Also coined terms 'bit', 'software', and developed FFT algorithm
1897-1986
Earlier 'range bar' visualization (1952)
Pioneered statistical graphics for business applications
Active 1980s-present
Extended box plots to 'letter-value plots'
Developed many modern statistical visualization techniques
Did You Know? John Tukey, who invented the box plot in 1970, also coined the terms "bit" (binary digit), "software," and "Fourier transform" in computing contexts. He was a pioneer in exploratory data analysis and believed in letting data "speak for itself" before formal statistical testing.
Box plot generator and complete guide. Create box-and-whisker plots to visualize data distributions, identify outliers, and compare multiple datasets. Learn how to calculate quartiles, IQR, and interpret box plot patterns including skewness and spread. Perfect for statistics students, data analysts, quality control professionals, and researchers.