How to Read a Box Plot: Quartiles, Whiskers & Outliers
A step-by-step walkthrough of median, Q1/Q3, Tukey fences, and how to compare multiple experimental groups accurately.
Five-number summary, Tukey 1.5×IQR outliers, jitter points, and violin overlay.
The five-number anatomy
Every box plot condenses a numeric sample into five robust landmarks: minimum, first quartile (Q1, 25th percentile), median (50th percentile), third quartile (Q3, 75th percentile), and maximum. Because quartiles rely on rank order rather than squared deviations, they remain stable even when a dataset contains extreme measurements.
Understanding Tukey 1.5 × IQR whiskers
The interquartile range (IQR = Q3 − Q1) sets the height of the central box. Under John Tukey's standard convention, the upper whisker reaches the largest observation no greater than Q3 + 1.5 × IQR, and the lower whisker reaches the smallest observation no less than Q1 − 1.5 × IQR. Points beyond those fences are plotted individually as candidate outliers—not automatically errors, but observations worth investigating.
Reading skewness and comparing groups
If the median line sits closer to Q1 and the upper whisker stretches longer, the distribution is right-skewed. When comparing treatment arms side by side, check both median shift and box overlap—and consider enabling a jitter or violin overlay to confirm whether groups are unimodal.