Course contentsShow
← Data Science
Lesson 1176 of 2,14530. Exploratory Data Analysis (EDA) ProcessFree lesson

Box Plots for Spread and Outliers

Interpreting box plots to assess quartiles, IQR, symmetry, and potential outlier presence.

Box Plots for Spread and Outliers

What you'll learn: How to read box plots to understand data spread, quartile positions, symmetry, and identify potential outliers at a glance.

The Core Idea

A box plot (or box-and-whisker plot) is a visual summary that packs five key statistics into one compact diagram. It shows you where the middle 50% of your data lives, how symmetric your distribution is, and which values might be unusual.

Here's what each part represents:

  • The box itself spans from Q1 (25th percentile) to Q3 (75th percentile)—this is the interquartile range (IQR), containing the middle half of your data
  • The line inside the box marks the median (Q2, 50th percentile)
  • The whiskers extend to the smallest and largest values within 1.5 × IQR from the box edges
  • Individual points beyond the whiskers are flagged as potential outliers

Interpreting What You See

Symmetry: If the median line sits near the box's center and both whiskers are roughly equal length, your data is fairly symmetric. A median closer to Q1 or Q3 suggests skewness.

Spread: A tall box means high variability in your central data; a short box means values cluster tightly. Long whiskers indicate data stretches far from the core.

Outliers: Those lonely dots beyond the whiskers deserve investigation—they might be data errors, rare events, or the most interesting observations in your dataset!

Think of a box plot like a building's elevation view: the box is the main structure housing most occupants, whiskers are reasonable extensions, and outliers are the penthouse suites or basement levels that stand apart.

Key Takeaway: Box plots efficiently reveal quartile positions, IQR spread, distribution symmetry, and highlight potential outliers—making them essential for quickly assessing numerical variable behavior during univariate analysis.