Scribd
Upload
462 views

Revision Questions On Regression

The document contains multiple choice and extended response questions about regression analysis and least squares regression lines. Some key points covered are: - The assumptions and components of a least squares regression line, including the relationship between the variables, the slope, intercept, and correlation coefficient. - How to determine the equation of a least squares regression line from sample data and how to interpret the slope and r-squared value. - Using a regression line to make predictions and calculate residuals. - Applying regression analysis to model relationships between variables in sample data sets.

Uploaded by

Naski Kuafni
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
462 views

Revision Questions On Regression

The document contains multiple choice and extended response questions about regression analysis and least squares regression lines. Some key points covered are: - The assumptions and components of a least squares regression line, including the relationship between the variables, the slope, intercept, and correlation coefficient. - How to determine the equation of a least squares regression line from sample data and how to interpret the slope and r-squared value. - Using a regression line to make predictions and calculate residuals. - Applying regression analysis to model relationships between variables in sample data sets.

Uploaded by

Naski Kuafni
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 9

REVISION QUESTIONS ON REGRESSION

MULTIPLE-CHOICE
1 When using a least squares line to model a relationship displayed in a
scatter plot, one key assumption is that:
A there are two variables
B the variables are related
C the variables are linearly related
D r 2 > 0.5
E the correlation coefficient is positive

2 In the least squares regression line y = −1.2 + 0.52x :


A the y-intercept = −0.52 and slope = −1.2
B the y-intercept = 0 and slope = −1.2
C the y-intercept = 0.52 and slope = −1.2
D the y-intercept = −1.2 and slope = 0.52
E the y-intercept 12 and slope = -0.52

3 If the equation of a least squares regression line is y = 8 − 9x and


r2 = 0.25:
A r = −0.5
B r = −0.25
C r = −0.0625
D r = 0.25
E r = 0.50

4 The least squares regression line y = 8 − 9x predicts that, when x = 5, the


value of y is:
A −45
B −37
C 37
D 45
E 53
5 A least squares regression line of the form y = a + b x is fitted to the data
set below.

The equation of the line is:


A y = −0.69 + 24.4x
B y = 24.4 − 0.69x
C y = 24.4 + 0.69x
D y = 28.7 − x
E y = 28.7 + x

6 The slope of the least squares regression line y


= a + b x is given by b = r sy
sx
Given, r = 0.733, sx = 1.871, sy = 3.391, then b is closest to:
A 0.41
B 0.45
C 1.33
D 1.87
E 2.49

7 Using a least squares regression line, the predicted value of a data value is
78.6. The residual value is −5.4.The actual data value is:
A 73.2
B 84.0
C 88.6
D 94.6
E 424.4

8 The equation of the least squares line plotted on the scatter plot opposite is
closest to:
A y = 8.7 − 0.9x
B y = 8.7 + 0.9x
C y = 0.9 − 8.7x
D y = 0.9 − 8.7x
E y = 8.7 − 0.1x

The following information


relates to Questions 9 to 12
Weight (in kg) can be predicted from height (in cm) from the least squares
regression line:
Weight = −96 + 0.95 × Height, with r = 0.79

9 Which of the following statements that relate to the regression line are
false?
A The slope of the regression line is 0.95.
B The independent variable in the regression equation is height.
C The least squares line does not pass through the origin.
D The intercept is 96.
E The equation predicts that a person who is 180 cm tall will weigh 75 kg.

10 This regression line predicts that, on average, weight:


A decreases by 96 kg for each one centimetre increase in height
B increases by 96 kg for each one centimetre increase in height
C decreases by 0.79 kg for each one centimetre increase in height
D decreases by 0.95 kg for each one centimetre increase in height
E increases by 0.95 kg for each one centimetre increase in height

11 Noting that the value of the correlation coefficient is r = 0.79, we can say
that:
A 62% of the variation in weight can be explained by the variation in height
B 79% of the variation in weight can be explained by the variation in height
C 88% of the variation in weight can be explained by the variation in height
D 79% of the variation in height can be explained by the variation in weight
E 95% of the variation in height can be explained by the variation in weight

12 A person of height 179 cm weighs 82 kg. If the regression equation is


used to predict their weight, then the residual will be closest to:
A −8 kg
B 3 kg
C 8 kg
D 9 kg
E 74 kg

13 If a three median line is fitted to the scatter plot shown, then its slope is
closest to:
A 0.2
B 0.4
C 0.6
D 0.8
E 1.03

EXTENDED RESPONSE
1 In an investigation of the relationship between the hours of sunshine (per
year) and days of rain (per year) for 25 cities, the least squares regression
line was found to be:
Hours of sunshine = 2847 − 6.88 × Days of rain, with r2 = 0.4838
Use this information to complete the following sentences.
a In this regression equation, the independent variable is ______ .
b The slope is ______ and the intercept is ________ .
c The regression equation predicts that a city that has 120 days of rain per
year will have ________ hours of sunshine per year.
d The slope of the regression line predicts that the hours of sunshine per year
will _________by ___________ hours for each additional day of rain.
e r = ______ , correct to three decimal places.
f _______% of the variation in sunshine hours can be explained by the
variation in_________ .
g One of the cities used to determine the regression equation had 142 days of
rain and 1390 hours of sunshine.
i The regression equation predicts its hours of sunshine to be ______ hours.
ii The residual value for this city is _________ hours.
h Using a regression line to make predictions within the range of data used
to determine the regression equation is called .

2 We wish to find the equation of the least squares regression line that will
enable height (in cm) to be predicted from femur (thigh bone) length
(in cm).
a Which is the DV and which is the IV?
b Use the following summary statistics to determine the equation of the least
squares regression line that will enable height (y) to be predicted from femur
length (x).

Write the equation in terms of height and femur length.


c Interpret the slope of the regression equation in terms of height and femur
length.
d Determine the value of the coefficient of determination and interpret in
terms of height and femur length.

3 The scatterplot below shows female workplace accident rates plotted


against male workplace accident rates in a selection of countries.
a Explain why it would be questionable to use a least squares regression line
in this situation.
b Fit a three median line to the scatterplot and find its equation.

4 Can the weight of a mouse’s heart be reliably predicted from its body
weight? The body weights (in g) and the heart weights (in mg) of a random
sample of 12 laboratory mice are shown below.

a In this investigation, which would be the dependent variable and which


would be the independent variable?
b Perform a regression analysis to investigate the nature of the relationship
between heart weight and body weight in mice. As part of this analysis, use
your calculator to:
i generate a scatterplot
ii determine the equation of the least squares regression line, the coefficient
of determination and the value of Pearson’s correlation coefficient, r
iii construct a residual plot
c Use the information generated in b to write a report on your investigation.
Use the format below:
ANSWERS
Multiple-choice questions
1 C 2 D 3 A 4 B 5 B 6 C 7 A 8 A 9 D 10 E 11 A 12 C 13 B

Extended-response questions

1 a Days of rain
b −6.88, 2847
c 2021
d decrease, 6.88
e −0.696
f 48.4, days of rain
g i 1870 ii −480
h interpolation Male rate

2 a DV: Height; IV: Femur length


b Height = 36.3 + 5.35 × Femur length
c On average, height increases by 5.35 cm for each centimeter increase in
femur length
d r 2 = 0.988; that is, 98.8% of the variation in height is explained by the
variation in femur length.
3a

b female rate = −3 + 0.5 male rate

7 a DV: Heart weight; IV: Body weight


bi

ii Heart weight = 49.5 + 2.72 Body weight


r 2 = 0.8314, r = 0.9118
iii
c The scatterplot shows that there is a strong positive linear relationship
between heart weight and body weight, r = 0.9118. There are no outliers.
The equation of the least squares regression line is:
Heart weight = 49.5 + 2.72 × Body weight.
The slope of the regression line predicts that, on average, heart weight
increases by 2.72 mg for each one gram increase in body weight.
The coefficient of determination indicates that 83.1% of the variation in
heart weight can be explained by the variation in body weight.
The residual plot shows no clear pattern, supporting the assumption that the
relationship between heart weight and body weight is linear.

You might also like