Two-variable data on the SAT
A best-fit line turns a scatterplot into a prediction. Its slope is the predicted change in y for each one-unit increase in x.
How to solve two-variable data questions
- Read the slope with its units.
- Read the intercept as the predicted value at x = 0.
- Plug in x to predict y. Actual minus predicted is the residual.
Quick tip: The slope of a best-fit line is the predicted change in y per one unit of x.
Worked example
The line of best fit for a scatterplot of hours studied, h, and test score, s, is s = 4.5h + 62. Which is the best interpretation of 4.5?
- AStudying causes scores to rise by 4.5 points per hour.
- BEvery student's score increases by exactly 4.5 points per hour.
- CA student who studies 0 hours is predicted to score 4.5 points.
- DFor each additional hour studied, the predicted score increases by about 4.5 points.
Show the answer
Answer: D · For each additional hour studied, the predicted score increases by about 4.5 points.
- Name x and y with units. h is hours, and s is points.
- Read the slope. 4.5 is the change in s for each extra hour: 4.5 points per hour.
- Use predicted language. It's a trend line, so it predicts. It isn't a rule for every student.
- A. A scatterplot shows an association, not cause.
- B. A best-fit line predicts. It isn't exact for every student.
- C. The prediction at 0 hours is 62, the intercept.
The common trap
Answers should say predicted or estimated, since a best-fit line is not exact. An association on its own does not prove one thing causes another.
A line of best fit predicts a score of 80 for a student who actually scored 74. What is the residual?
- A−6
- B74
- C6
- D80
Show the answer
Answer: A · −6
Residual = actual − predicted. A point below the line has a negative residual.
- B. 74 is the actual score.
- C. The residual is actual minus predicted, not the other way around.
- D. 80 is the predicted score.
Remember
- The slope is the predicted change in y per one unit of x, with units.
- The intercept is the prediction at x = 0.
- Say predicted. And association isn't cause.
Practice questions
Try each one, then open the answer to check your work.
The equation y = 73 + 7x models the number of visitors y to a museum x days after an exhibit opened. According to the model, how many visitors were there on the day the exhibit opened?
- A7
- B66
- C73
- D80
Show the answer
Answer: C · 73
Opening day is x = 0, and y = 73 + 7(0) = 73.
- A. 7 is the number of visitors added each day. Opening day is x = 0.
- B. Opening day is x = 0, so nothing is subtracted: y = 73.
- D. That's day 1. The exhibit opened on day 0.
The equation y = 9x + 42 models a plant's height y, in centimeters, x weeks after planting. What is the best interpretation of 9?
- AThe predicted height after 42 weeks is 9 centimeters.
- BThe predicted height increase each week is 9 centimeters.
- CThe predicted height at planting is 9 centimeters.
- DThe plant is predicted to grow for 9 weeks.
Show the answer
Answer: B · The predicted height increase each week is 9 centimeters.
The slope is the predicted change in y per one unit of x. The starting height is 42.
- A. At x = 42, the model gives 420, not 9. 9 is the change per week.
- C. The height at planting is the y-intercept, 42. 9 is the slope.
- D. Nothing in the model says when growth stops. 9 is the change in height per week.
A line of best fit is y = 3x + 40. At x = 5, the actual value is 65. What is the actual value minus the predicted value?
Type your answer. It can be a whole number, a decimal or a fraction.
Show the answer
Answer: 10
Predicted: 3(5) + 40 = 55. Actual minus predicted: 65 − 55 = 10.