AP Statistics · Topic 2.8

Least Squares Regression Practice

Part of Exploring Two-Variable Data.(DAT-1.E)

Practice questions

14

Want a predicted score for the whole AP STAT exam? Take the 20-question diagnostic and Lumi will plan the rest.

Sample questions

5 of 14 — sign in to practice the rest with adaptive difficulty and mastery tracking.

  1. Sample 1difficulty 1/5

    Consider r-squared from a least-squares regression.

    0 1 Range of R-squared

    Which set best describes the possible values of r-squared?

    • A

      0 to 1 (or 0% to 100%)

      check_circle
    • B

      any real number

    • C

      0 to 100

    • D

      -1 to 1

    Why

    r-squared is a proportion between 0 and 1 (commonly written as a percent 0%-100%).

  2. Sample 2difficulty 2/5

    The means of x and y in a sample are 50 and 100. The slope of the least-squares regression line is 1.5.

    Sample summary x-bar = 50 (units) y-bar = 100 (units) slope b = 1.5 intercept a = ?

    What is the y-intercept?

    • A

      75

    • B

      50

    • C

      150

    • D

      25

      check_circle

    Why

    The line passes through (x-bar, y-bar): a = y-bar - b * x-bar = 100 - 1.5(50) = 25.

  3. Sample 3difficulty 2/5

    Two models compete: Model A has r-squared = 0.40 and Model B has r-squared = 0.85.

    Comparison of two models Model A: R-Sq = 0.40 Model B: R-Sq = 0.85 Which explains more variation?

    Which statement is best supported?

    • A

      Cannot be compared without more info

    • B

      Model A is more accurate

    • C

      The models are equally good

    • D

      Model B explains more of the variability in y

      check_circle

    Why

    Higher r-squared indicates a greater proportion of variability in y explained by the linear model.

  4. Sample 4difficulty 2/5

    Output is shown for a regression of sales on ad spend.

    Regression of Sales on Ad Spend y-hat = 200 + 12 x r = 0.85 R-Sq = 72.25%

    Which is the best interpretation of r-squared?

    • A

      About 72% of advertisements lead to sales

    • B

      Sales increase by 72% per ad dollar

    • C

      The model predicts sales correctly 72% of the time

    • D

      About 72% of the variation in sales is explained by the linear regression on ad spend

      check_circle

    Why

    r-squared is the proportion of variability in the response (sales) accounted for by the linear model with the predictor (ad spend).

  5. Sample 5difficulty 2/5

    For a linear regression, sx = 4, sy = 10, and r = 0.6.

    Summary Statistics sx = 4 (units of x) sy = 10 (units of y) r = 0.6 slope b = ?

    What is the slope of the least-squares regression line predicting y from x?

    • A

      1.5

      check_circle
    • B

      2.4

    • C

      0.6

    • D

      0.24

    Why

    Slope b = r * (sy/sx) = 0.6 * (10/4) = 1.5.