SAT scatterplots: read the line of best fit like a pro
A scatterplot’s line of best fit turns a cloud of points into a prediction. Learn to interpret its slope and intercept in context, make predictions, and avoid the classic causation trap.
The short version
What you'll take away
- The slope of a line of best fit is the predicted change in y for each one-unit increase in x.
- Points above the line have actual values greater than predicted; points below have smaller ones.
- A strong association does not show cause, and predictions far outside the data are unreliable.
In this article
From a cloud of points to a line
The study lab
Practice hours vs. quiz score
Nine students (illustrative data) and the line of best fit y = 1.1x + 5.
- Line of best fit: y = 1.1x + 5
Interpret slope and intercept in context
Predictions and residuals
Your turn
Quick practice
Using the line of best fit y = 1.1x + 5, what is the predicted quiz score for a student who practices 6 hours?
Hint: The question asks for the value from the line, not the plotted point.
Direction, strength, and the causation trap
See the skill
Describe what you see
Points rise left to right
Positive
As x increases, y tends to increase.
Points fall left to right
Negative
As x increases, y tends to decrease.
Tight around the line
Strong
Predictions from the line are fairly reliable.
Association only
Not causation
An observational study shows a link, not a cause. Only a randomized experiment can support a causal claim.
Practice: try these before checking the answers
- 1. Interpret the slope −0.5 in context.
- 2. What remaining battery percent does the model predict after 20 hours?
- 3. A phone has 37% left after 10 hours. Is it above or below the line, and by how much?
- 4. Should you use the model to predict battery after 100 hours? Why or why not?
Sources & further reading
One step from here
Put the idea to work.
SAT Practice helps you turn questions into useful feedback, so each session has a next step.
Try a practice set