Visual Methods: Box Plots and Scatter Plots
You'll learn how box plots and scatter plots help you spot outliers and understand their patterns at a glance.
Why Visualize Outliers?
Numbers alone don't always tell the full story. Visual methods transform abstract statistics into pictures your brain can process instantly. Two powerful tools—box plots and scatter plots—make outliers jump off the page.
Box Plots: The Outlier Spotlight
Remember the five-number summary and IQR you learned earlier? A box plot displays this visually. The "box" spans from the first quartile (Q1) to the third quartile (Q3), with a line at the median. "Whiskers" extend to show typical data range.
Here's the key: points plotted beyond the whiskers are potential outliers. The standard rule marks any value below Q1 - 1.5×IQR or above Q3 + 1.5×IQR as suspicious. These appear as individual dots, instantly visible and separate from the pack.
Analogy: Think of a box plot like a crowd photo where most people cluster together (the box), a few stand slightly apart (the whiskers), and outliers are way off in the corner—impossible to miss.
Scatter Plots: Outliers in Context
A scatter plot shows two variables as points on an X-Y grid. Outliers reveal themselves as dots far from the main cluster. Unlike box plots (which examine one variable at a time), scatter plots show relationships—helping you understand whether an outlier is unusual in one dimension or both.
For example, plotting height vs. weight might show most points in a diagonal band, but one point far above could indicate a data entry error or a genuinely exceptional case.
When to Use Each
- Box plot: Compare distributions across categories, quickly spot extreme values in single variables
- Scatter plot: Explore relationships between two variables, identify multivariate outliers
Key Takeaway: Box plots highlight outliers using quartile boundaries, while scatter plots reveal outliers through their position relative to patterns between two variables—both make the invisible visible.