2.4 Representing the Relationship Between Two Quantitative Variables

A scatterplot shows the relationship between two quantitative variables, with the explanatory variable on the x-axis and the response variable on the y-axis. To describe one, check four things every time: form, direction, strength, and unusual features.

Why This Matters for the AP Statistics Exam

Scatterplots are the starting point for the entire two-variable data unit. Before you calculate a correlation, fit a line, or check residuals, you have to read the plot and describe what you see. On the AP Statistics exam, you can be asked to identify which variable is explanatory, build or read a scatterplot, and describe its features in words.

This skill shows up in both multiple-choice questions and free-response questions. Multiple-choice items often ask you to match a description to a plot or pick the correct direction or form. Free-response questions reward clear, in-context descriptions, so getting comfortable with form, direction, strength, and unusual features now pays off across the rest of Unit 2.

more resources to help you study

practice multiple choiceFRQ practice & scoringcheatsheetsscore calculatorkey terms

Key Takeaways

• A bivariate quantitative data set has two numeric measurements for each individual, plotted as one point on a scatterplot.
• The explanatory variable goes on the x-axis and is used to explain or predict the response variable on the y-axis.
• Describe every scatterplot with four parts: form, direction, strength, and unusual features.
• Direction is positive (both increase together) or negative (one increases as the other decreases); form is linear or non-linear.
• Strength is strong, moderate, or weak based on how closely points follow the pattern.
• Unusual features include clusters of points and points with large discrepancies between actual and predicted response values.

What Is the Relationship Between Two Quantitative Variables?

The relationship between two quantitative variables is the pattern you see when each individual has two numerical measurements. A scatterplot shows that relationship by plotting one variable on the x-axis and the other on the y-axis. From there, you describe the association using form, direction, strength, and unusual features.

Setting Up a Scatterplot

A bivariate quantitative data set has two different quantitative variables measured on each individual in a sample or population. For example, you might record both the age and the blood pressure of each person in a study.

One variable is the explanatory variable (also called independent), and the other is the response variable (also called dependent). The explanatory variable is used to explain or predict the response variable. With age and blood pressure, age would usually be the explanatory variable used to predict blood pressure.

A scatterplot graphs these pairs. Each observation becomes one point with two numeric values: the explanatory variable on the horizontal x-axis and the response variable on the vertical y-axis.

Describing a Scatterplot

When you describe a scatterplot, cover four things every time: form, direction, strength, and unusual features. Missing one is a common way to lose easy points.

Form

The form is the general shape of the points. It can be linear, where the points follow a straight-line pattern, or non-linear, where they follow a curve. Linear and non-linear can each hold to varying degrees, so look at how cleanly the points track a single shape.

For example, a clearly straight pattern suggests a linear relationship, while a bending pattern suggests something non-linear, like a curve.

Direction

The direction is the trend you see moving from left to right. It can be positive or negative:

• Positive association: as one variable increases, the other tends to increase.
• Negative association: as one variable increases, the other tends to decrease.

Use "tend to" language because the pattern describes a general trend, not a guarantee for every point. For age and height in growing children, a positive association means older children tend to be taller.

Strength

The strength describes how closely the points follow the pattern, and you describe it as strong, moderate, or weak. Points hugging a clear line or curve are strong; points loosely scattered around the pattern are weak. You will quantify strength with the correlation coefficient in the next topic, but here you judge it from the plot.

Unusual Features

Finally, point out anything that breaks the pattern:

• Clusters are groups of points sitting close together, which can signal subgroups in the data.
• Points with large discrepancies between the actual response value and a predicted response value stand out from the overall trend.

Always mention unusual features when you see them, because they can change how you interpret the relationship.

Worked Example

Describe a scatterplot of age at first word versus Gesell score in context.

A strong sample answer: "The scatterplot appears to follow a linear pattern. It shows a negative association, since the Gesell score tends to decrease as the age at first word increases. The association looks moderate, since some points follow the pattern closely while others spread away from it. There is one main cluster of points, and one point stands out with a large discrepancy between its actual Gesell score and the score the pattern would predict."

Notice that this response names the actual variables instead of saying "x and y." Writing in context is how you support a stronger score on this kind of free-response question.

Outliers, Influential Points, and High-Leverage Points

These three terms come up together because they all describe points that can pull a relationship around. They are explored more fully in Topic 2.9, but here is the quick version so you can name them when describing a plot.

• An outlier in regression is a point that does not follow the general trend of the rest of the data and has a large residual once a line is fit.
• A high-leverage point has an x-value that is much larger or smaller than the other x-values in the data set.
• An influential point is any point that, if removed, changes the relationship substantially, such as a clearly different slope, intercept, or correlation. Outliers and high-leverage points are often influential.

How to Use This on the AP Statistics Exam

MCQ

• Match a written description to the correct scatterplot, or pick the correct form, direction, and strength for a given plot.
• Identify which variable is explanatory and which is response based on the wording of the scenario.

Free Response

• Describe a scatterplot using all four parts: form, direction, strength, and unusual features.
• Name the actual variables in your description instead of "x" and "y."
• Use "tend to" and "on average" style language so you describe a trend, not a rule that holds for every point.

Common Trap

Saying a relationship is positive or negative without context. Instead of "there is a negative association," write "as age at first word increases, Gesell score tends to decrease." Connecting the description to the real variables is what makes the answer complete.

Common Misconceptions

• Mixing up explanatory and response variables. The explanatory variable predicts, and it goes on the x-axis. Swapping them changes the plot and any later model.
• Treating "line" as proof of linearity. Saying the points "make a line" is not enough. Describe the form as linear and back it up with how closely the points follow a straight pattern.
• Forgetting unusual features. Form, direction, and strength are not the full description. Always scan for clusters and standout points.
• Confusing strength with direction. A negative association can still be strong. Direction is about which way the trend goes; strength is about how tightly points follow it.
• Calling every standout point an outlier. A point can be high-leverage because of an extreme x-value without being a regression outlier, and not every unusual point changes the relationship.
• Saying correlation proves causation. A clear pattern in a scatterplot shows association, not that one variable causes the other.

Related AP Statistics Guides

• Unit 2 Overview: Exploring Two-Variable Data
• 2.1 Introducing Statistics: Are Variables Related?
• 2.6 Linear Regression Models
• 2.7 Residuals
• 2.8 Least Squares Regression
• 2.5 Correlation