Interpreting Scatterplots
A scatterplot shows the relationship between two variables. You read the overall trend, the strength and direction of the correlation, and any outliers.
Reading the concept isn't enough. The score comes from practicing the real Bluebook-style format and finding the exact mistakes you keep making.
What the SAT tests here
A rising cloud is a positive correlation; a falling one is negative. Correlation isn't causation, and a clear downward trend is still a correlation, not "no relationship."
How to solve one, step by step
Example: points drift downward as you move right.
- As $x$ increases, $y$ tends to decrease.
- That's a negative correlation.
The mistakes that cost points
- Reading correlation as causation. A relationship in the data doesn't prove one variable causes the other.
- Calling a negative trend "no correlation." A downward trend is a (negative) correlation.
Practice questions
Try these the way you would on test day, then open the solution to check your method.
Easy
A scatterplot shows a strong positive correlation between the number of hours students spend studying and their test scores. A researcher concludes that studying more causes higher scores. Which of the following best evaluates this conclusion?
- AThe conclusion is not supported — correlation does not establish causation. Other variables may explain the relationship.
- BThe conclusion is supported because the trend is linear.
- CThe conclusion cannot be evaluated because no outliers are present.
- DThe conclusion is supported because the correlation is strong and positive.
Show solution
Answer: A, The conclusion is not supported — correlation does not establish causation. Other variables may explain the relationship.. A scatterplot shows association, not causation. Even a strong positive correlation cannot prove that one variable causes the other; there may be confounding variables such as student motivation or prior knowledge.
Medium
A scatterplot shows the relationship between a city's average temperature and its monthly ice cream sales. The correlation is $r = 0.92$. A student claims that higher temperatures cause people to buy more ice cream. Which of the following is the most accurate interpretation?
- AThe outliers in the data disprove the relationship entirely.
- BBecause $r = 0.92$ is close to 1, temperature must cause higher ice cream sales.
- CThe relationship is too weak to draw any conclusion.
- DThe data show a strong positive association, but the scatterplot alone cannot confirm that temperature causes higher sales.
Show solution
Answer: D, The data show a strong positive association, but the scatterplot alone cannot confirm that temperature causes higher sales.. Even a very high correlation ($r = 0.92$) does not imply causation. Other factors — such as more outdoor events in warm months — might explain the pattern. Scatterplots reveal association only.
Hard
A scatterplot shows 30 data points with a strong negative trend. However, two points in the upper-right cluster deviate significantly from the rest. A student removes these two outliers and recalculates the correlation. The correlation becomes even more negative. What does this tell us about the outliers' effect on the original correlation?
- AThe outliers indicate that the true relationship is actually positive.
- BThe two outliers were pulling the correlation toward zero (weakening it), because they appeared in the upper right, opposite the negative trend.
- CThe outliers had no effect because correlation is not influenced by individual points.
- DThe outliers were strengthening the negative correlation, so removing them weakened it.
Show solution
Answer: B, The two outliers were pulling the correlation toward zero (weakening it), because they appeared in the upper right, opposite the negative trend.. Points that go against the overall trend (here, high-x and high-y values in a negative-trending dataset) reduce the magnitude of the negative correlation. Removing them makes the correlation stronger (more negative), confirming the outliers were dampening the relationship.
Find your exact gaps
Take a free, full-length practice test and see precisely which question types trip you up.
Practice this for freeNo credit card. Register once, take as many tests as you like.