Or press and hold the picture and choose “Save image”.
Linear Regression & Correlation — Easy Worksheet 2
Grade 12 · Whoobee Name: ____________________ Date: __________
Scan for answers
Solve each problem. Residual = actual - predicted. Give percents for r^2 like 64%. For lettered choices, write the letter.
How to solve
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Example
The least-squares line ŷ = 640 - 8.4x predicts y = number of tickets sold from x = ticket price in dollars. Predict y when x = 29.
Answer: 396.4
Ways to write your answer
Not quite. The answer is
💡 Substitute x = 68: ŷ = −12 + 0.8 × 68 = 42.4.
✓ Correct!
Not quite. The answer is
💡 Substitute x = 68: ŷ = −12 + 0.8 × 68 = 42.4.
💡 How to solve: Linear Regression & Correlation
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Try a similar one
The least-squares line ŷ = 52.5 + 2.5x predicts y = test score from x = hours studied. Predict y when x = 2.
Ways to write your answer
✓ You've got it! Press Next to keep going. Not yet — the answer is . Read the steps above once more.
Ways to write your answer
Not quite. The answer is
💡 Predicted: ŷ = 55 + 4.2 × 7 = 84.4. Residual = actual − predicted = 81.4 − 84.4 = −3.
✓ Correct!
Not quite. The answer is
💡 Predicted: ŷ = 55 + 4.2 × 7 = 84.4. Residual = actual − predicted = 81.4 − 84.4 = −3.
💡 How to solve: Linear Regression & Correlation
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Try a similar one
The least-squares line ŷ = 18 + 1.4x predicts y = free throws made out of 50 from x = hours of practice per week. Predict y when x = 13.
Ways to write your answer
✓ You've got it! Press Next to keep going. Not yet — the answer is . Read the steps above once more.
Ways to write your answer
Not quite. The answer is
💡 The sign of r = 0.88 gives the direction (positive), and |r| = 0.88 is close to 1, so it is strong: strong positive. That is choice (A).
✓ Correct!
Not quite. The answer is
💡 The sign of r = 0.88 gives the direction (positive), and |r| = 0.88 is close to 1, so it is strong: strong positive. That is choice (A).
💡 How to solve: Linear Regression & Correlation
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Try a similar one
The least-squares line ŷ = 5 + 4.1x predicts y = commute time in minutes from x = distance from school in miles. Predict y when x = 9.
Ways to write your answer
✓ You've got it! Press Next to keep going. Not yet — the answer is . Read the steps above once more.
Ways to write your answer
Not quite. The answer is
💡 The slope is the predicted change in y for each 1-unit increase in x. It is 1.8, so each additional hour increases the predicted free throws made out of 50 by 1.8 free throws. That is choice (D).
✓ Correct!
Not quite. The answer is
💡 The slope is the predicted change in y for each 1-unit increase in x. It is 1.8, so each additional hour increases the predicted free throws made out of 50 by 1.8 free throws. That is choice (D).
💡 How to solve: Linear Regression & Correlation
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Try a similar one
The least-squares line ŷ = 8 + 2.8x predicts y = commute time in minutes from x = distance from school in miles. Predict y when x = 12.
Ways to write your answer
✓ You've got it! Press Next to keep going. Not yet — the answer is . Read the steps above once more.
Ways to write your answer
Not quite. The answer is
💡 The y-intercept is the predicted y when x = 0: ŷ = 640 + (−12.5) × 0 = 640. That is choice (C).
✓ Correct!
Not quite. The answer is
💡 The y-intercept is the predicted y when x = 0: ŷ = 640 + (−12.5) × 0 = 640. That is choice (C).
💡 How to solve: Linear Regression & Correlation
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Try a similar one
The least-squares line ŷ = 30.2 - 2.4x predicts y = value of the car in thousands of dollars from x = age of a car in years. Predict y when x = 8.
Ways to write your answer
✓ You've got it! Press Next to keep going. Not yet — the answer is . Read the steps above once more.
Ways to write your answer
Not quite. The answer is
💡 The coefficient of determination is r2 = (−0.9)2 = 0.81 = 81%. So 81% of the variation in y is explained by the line (squaring removes the sign, so a negative r works the same way).
✓ Correct!
Not quite. The answer is
💡 The coefficient of determination is r2 = (−0.9)2 = 0.81 = 81%. So 81% of the variation in y is explained by the line (squaring removes the sign, so a negative r works the same way).
💡 How to solve: Linear Regression & Correlation
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Try a similar one
The least-squares line ŷ = 820 - 15x predicts y = number of tickets sold from x = ticket price in dollars. Predict y when x = 39.
Ways to write your answer
✓ You've got it! Press Next to keep going. Not yet — the answer is . Read the steps above once more.
Ways to write your answer
Not quite. The answer is
💡 The y-intercept is the predicted y when x = 0: ŷ = 5 + 2.8 × 0 = 5. That is choice (B).
✓ Correct!
Not quite. The answer is
💡 The y-intercept is the predicted y when x = 0: ŷ = 5 + 2.8 × 0 = 5. That is choice (B).
💡 How to solve: Linear Regression & Correlation
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Try a similar one
The least-squares line ŷ = 640 - 15x predicts y = number of tickets sold from x = ticket price in dollars. Predict y when x = 23.
Ways to write your answer
✓ You've got it! Press Next to keep going. Not yet — the answer is . Read the steps above once more.
Ways to write your answer
Not quite. The answer is
💡 The slope is the predicted change in y for each 1-unit increase in x. It is 4.1, so each additional mile increases the predicted commute time by 4.1 minutes. That is choice (A).
✓ Correct!
Not quite. The answer is
💡 The slope is the predicted change in y for each 1-unit increase in x. It is 4.1, so each additional mile increases the predicted commute time by 4.1 minutes. That is choice (A).
💡 How to solve: Linear Regression & Correlation
The least-squares line is ŷ = a + bx: b is the slope (the predicted change in y for each 1-unit increase in x) and a is the predicted y when x = 0. Predict by substituting x. A residual is actual y - predicted y: positive means the point is above the line, negative means below. The correlation r (between -1 and 1) measures the direction and strength of a linear pattern; r2 is the fraction of the variation in y explained by the line. From summary statistics: b = r × (sy ÷ sx) and a = (mean of y) - b × (mean of x). Correlation does not prove causation, and predicting far outside the data (extrapolation) is unreliable.
Try a similar one
The least-squares line ŷ = 820 - 15x predicts y = number of tickets sold from x = ticket price in dollars. Predict y when x = 29.
Ways to write your answer
✓ You've got it! Press Next to keep going. Not yet — the answer is . Read the steps above once more.
Log in to save your progress.