Course contentsShow
← Data Science
Lesson 1222 of 2,14531. Data Visualization PrinciplesFree lesson

Scatter Plots for Relationships

Use scatter plots to visualize relationships between two continuous variables and identify correlations or patterns.

Scatter Plots for Relationships

What you'll learn: How to use scatter plots to reveal and explore relationships between two continuous variables.

The Power of Dots on a Grid

A scatter plot places individual data points on a two-dimensional plane, with one variable on the x-axis and another on the y-axis. Each dot represents a single observation, and the pattern these dots form tells you whether—and how—the two variables are related.

Think of it like plotting houses on a map: one axis shows square footage, the other shows price. The resulting cloud of points reveals whether bigger homes tend to cost more, and how tight or loose that relationship is.

What Scatter Plots Reveal

When you create a scatter plot, you're looking for patterns:

  • Positive correlation: Points trend upward from left to right (as one variable increases, so does the other)
  • Negative correlation: Points trend downward (as one increases, the other decreases)
  • No correlation: Points scatter randomly with no clear direction
  • Nonlinear relationships: Curves, U-shapes, or other patterns that aren't straight lines
  • Clusters: Groups of points separated from others, suggesting subpopulations
  • Outliers: Individual points far from the main pattern, warranting investigation

Why Scatter Plots Matter

Scatter plots are your first line of defense against Anscombe's quartet—the famous cautionary tale where four datasets have identical summary statistics but wildly different relationships that only visualization reveals. A correlation coefficient might say "0.8," but the scatter plot shows whether that's a clean linear trend, a curved relationship, or driven by a single outlier.

They bridge exploratory and explanatory visualization: you use them to discover relationships during EDA, then refine them to communicate findings.

Key Takeaway: Scatter plots transform pairs of numbers into visual patterns, making relationships, correlations, trends, and anomalies immediately apparent in ways summary statistics alone cannot capture.