Scatter Plots for Two Numeric Variables
You'll learn how to visualize relationships between two continuous variables to spot patterns, trends, and unusual observations.
What Is a Scatter Plot?
A scatter plot displays data points where each point represents one observation with values for two numeric variables—one on the x-axis, one on the y-axis. Think of it like plotting coordinates on a map: each person's height (x) and weight (y) becomes a single dot. When you plot many observations, patterns emerge that reveal whether and how the variables relate.
What to Look For
Trends: Do points generally move upward (positive relationship), downward (negative), or stay flat (no relationship)? For example, house size and price typically trend upward together.
Strength: Are points tightly clustered along a line/curve (strong relationship) or widely scattered (weak relationship)?
Shape: Is the pattern linear (straight-line trend) or curved (nonlinear)?
Outliers: Do any points sit far from the main cluster? These unusual cases deserve investigation—they might be data errors or genuinely exceptional observations.
Why Scatter Plots Matter in Bivariate Analysis
After examining variables individually in univariate analysis, scatter plots are your first tool for exploring relationships. They're visual hypothesis generators: "Does income correlate with spending?" "Is there a threshold effect?" The plot answers these questions before you run formal statistical tests.
Unlike correlation coefficients (which reduce everything to a single number), scatter plots show the full story—including nonlinear patterns, clusters, and outliers that summary statistics miss.
Key Takeaway: Scatter plots reveal the nature, direction, and strength of relationships between two numeric variables, making them essential for spotting patterns and outliers during exploratory bivariate analysis.