Course contentsShow
← Data Science
Lesson 55 of 2,1452. Descriptive Statistics FoundationsFree lesson

Visualizing Spread

Interpret boxplots, error bars, and violin plots to see variability graphically.

Visualizing Spread

What you'll learn: How to see variability in data using boxplots, error bars, and violin plots.

Numbers like standard deviation and IQR are useful, but seeing spread makes patterns jump out instantly. Three visual tools help you grasp variability at a glance.

Boxplots: The Five-Number Summary

A boxplot draws a box from the first quartile (Q1) to the third quartile (Q3), with a line at the median. "Whiskers" extend to show the range (or a modified range), and dots mark outliers. Wide boxes mean high variability; narrow boxes mean data cluster tightly. You can immediately compare spread across multiple groups side-by-side.

Think of it like: A visual IQR calculator that also flags unusual values.

Error Bars: Uncertainty Around a Point

Error bars attach vertical or horizontal lines to a point (often a mean) showing ±1 standard deviation, ±2 SE, or confidence intervals. Longer bars = more spread or uncertainty. They're common in scientific papers to show measurement precision. If error bars overlap heavily between groups, their means might not be meaningfully different.

Think of it like: A "margin of doubt" drawn right on your chart.

Violin Plots: The Full Distribution Shape

A violin plot combines a boxplot with a smoothed density curve mirrored on both sides. The width at any height shows how many data points fall there. You see not just spread, but whether data is symmetric, bimodal, or skewed. It's richer than a boxplot but takes more practice to read.

Think of it like: An x-ray that reveals the internal structure of your data's distribution.


Key Takeaway: Boxplots show quartiles and outliers, error bars display uncertainty around estimates, and violin plots reveal full distribution shapes—each turning abstract measures of spread into intuitive visual insights.