Sample Final Exam 1 – Part A
1. We would like to estimate the true mean amount (in $) consumers spent last year on
Christmas gifts. We record the amount spent for a simple random sample of 30 consumers
and we calculate a 95% confidence interval for µ to be (500, 545), i.e., the length of the
interval is 45. The standard deviation σ of the amount spent by consumers is known.
Suppose we had instead selected a simple random sample of 90 consumers and calculated
a 95% confidence interval for µ. What would be the length of this interval?
(A) 5.00 (B) 12.99 (C) 15.00 (D) 25.98 (E) 77.9
4
2. A statistician conducted a test of H0 : µ = 20 vs. Ha : µ < 20 for the mean µ of some normally distributed population. Based on the gathered data, the statistician concluded that H0 could be rejected at the 5% level of significance. Using the same data, which of the following statements must be true?
(I) A test of H0 : µ = 20 vs. Ha : µ < 20 at the 10% level of significance would also lead to rejecting H0.
(II) A test of H0 : µ = 19 vs. Ha : µ < 19 at the 5% level of significance would also lead to rejecting H0.
(III) A test of H0 : µ = 20 vs. Ha : µ 6= 20 at the 5% level of significance would also
lead to rejecting H0.
(A) I only
(B) II only
(C) I and II only
(D) I and III only
(E) I, II and III
3. Prior to distributing a large shipment of bottled water, a beverage company would like
to determine whether there is evidence that the true mean fill volume of all bottles differs
from 600 ml, which is the amount printed on the labels. Fill volumes are known to follow
a normal distribution with standard deviation 2.0 ml. A random sample of 25 bottles is
selected. The sample has a mean of 598.8 ml and a standard deviation of 3.0 ml. What
is the value of the test statistic for the appropriate test of significance?
(A) t = −0.50 (B) z = −2.00 (C) t = −2.00 (D) z = −3.00 (E) t = −3.00
4. Yields of apples per tree in a large orchard are known to follow a normal distribution
with standard deviation 30 pounds. We will select a random sample of 25 trees and
conduct a hypothesis test at the 1% level of significance to determine whether the true
mean yield per tree is greater than 275 pounds. What is the probability of making a
Type II error if the true mean yield per tree is actually 300 pounds?
(A) 0.0329 (B) 0.0537 (C) 0.0606 (D) 0.0838 (E) 0.0985
5. Nine runners are asked to run a 10-kilometer race on each of two consecutive weeks.
In one of the races, the runners wear one brand of shoe and in the other a different
brand (with the order randomly determined). All runners are timed and are asked to
run their best in each race. The results (in minutes) are given below, with some sample
calculations that may be useful:
Runner 1 2 3 4 5 6 7 8 9 Mean Std. Dev.
Brand 1 31.2 29.3 30.5 32.2 33.1 31.5 30.7 31.1 33.0 31.40 1.2
2
Brand 2 32.0 29.0 30.9 32.7 33.0 31.6 31.3 31.2 33.3 31.67 1.3
1
Difference −0.8 0.3 −0.4 −0.5 0.1 −0.1 −0.6 −0.1 −0.3 −0.27 0.35
We wish to conduct a hypothesis test to determine if there is evidence that running
times for the two brands differ. Suppose it is known that differences in times for the two
brands follow a normal distribution. The P-value for the appropriate test of significance
is:
(A) between 0.01 and 0.02
(B) between 0.02 and 0.04
(C) between 0.04 and 0.05
(D) between 0.05 and 0.10
(E) between 0.10 and 0.20
6. The blood cholesterol levels for randomly selected individuals were compared for two
diets – one low fat and one normal. Some summary statistics are given in the table
below:
Sample Size Sample Mean Sample Std. Dev.
Low Fat 9 170 12
Normal 14 193 25
We would like to construct a confidence interval to estimate the difference between the
true mean blood cholesterol level of all people on a low fat diet (µ1) and all people on a
normal diet (µ2). The standard error of X̄1 − X̄2 is:
(A) 2.95 (B) 3.12 (C) 6.08 (D) 7.79 (E) 8.9
8
7. The following are the percentages of alcohol found in samples of two brands of beer,
along with some sample statistics:
x̄ s
Brand A 5.4 5.6 5.7 5.3 5.5 0.18
3
Brand B 4.9 5.4 5.5 5.0 5.2 0.294
A 90% confidence interval for the difference in mean alcohol content for the two brands
is:
(A) (0.015, 0.585)
(B) (0.002, 0.598)
(C) (−0.022, 0.622)
(D) (−0.036, 0.636)
(E) (−0.051, 0.651)
The next two questions (8 and 9) refer to the following:
A service centre for electronic equipment is conducting a study on three of their tech-
nicians: Joe, Bill, and John. The manager of the service centre wishes to assess if the
average service times for their three technicians are equal. Each technician was given a
random sample of disk drives, and the service time (in minutes) for each was recorded.
Some sample statistics are given in the table below:
Technician Sample Size Sample Mean Sample Std. Dev.
Joe 7 18 4
Bill 4 14 3
John 9 20 5
Service times for each of the three technicians are known to follow a
normal distribution.
8. One assumption required in conducting an ANOVA F test is that all population variances
are equal. The estimate of this common population variance is:
(A) 16.00 (B) 16.67 (C) 18.44 (D) 18.65 (E) 19.00
9. The value of the test statistic is calculated to be 2.63. What is the P-value of the
appropriate test of significance?
(A) between 0.001 and 0.01
(B) between 0.01 and 0.025
(C) between 0.025 and 0.05
(D) between 0.05 and 0.10
(E) greater than 0.10
10. The time it takes a student to drive to university in the morning follows a normal distri-
bution with mean 28 minutes and standard deviation 4 minutes. The time it takes the
student to drive home from university in the afternoon follows a normal distribution with
mean 25 minutes and standard deviation 3 minutes. Morning and afternoon commuting
times are known to be independent. What is the probability that it takes the student
longer to drive to university than to drive home?
(A) 0.5517 (B) 0.6664 (C) 0.7257 (D) 0.8707 (E) 0.9987
The next four questions (11 to 14) refer to the following:
A coffee store owner compiles the following information:
• A customer gets a coffee (C) in 35% of the store’s transactions.
• A customer gets a donut (D) in 40% of the store’s transactions.
• In 49% of the store’s transactions, a customer gets a coffee or a muffin (M).
• In 61% of the store’s transactions, a customer gets a donut or a coffee.
• In 7% of the store’s transactions, a customer gets a donut and a muffin.
• In 11% of the store’s transactions, a customer gets a coffee and a muffin.
• In 3% of the store’s transactions, a customer gets a donut, a coffee and a muffin.
11. What is the probability that a randomly selected customer gets a muffin?
(A) 0.29 (B) 0.25 (C) 0.36 (D) 0.18 (E) 0.32
12. If a customer who buys a coffee, what is the probability that he or she also buys a donut
and a muffin?
(A) 0.0492 (B) 0.0857 (C) 0.1139 (D) 0.2154 (E) 0.4286
13. Which of the following statements is true?
(A) The events C and D are independent.
(B) The events C and D are mutually exclusive.
(C) The events C and M are independent.
(D) The events C and M are mutually exclusive.
(E) The events D and M are independent.
14. If we take a random sample of 300 customers, what is the probability that less than 32%
of them buy a coffee?
(A) 0.1292 (B) 0.1379 (C) 0.1469 (D) 0.1587 (E) 0.1685
The next two questions (15 and 16) refer to the following:
A small deck of cards contains five red cards, four blue cards and one green card. We
will shuffle the deck and select three cards without replacement. Let X be the number
of blue cards that are selected. The probability distribution of X is shown below:
x 0 1 2 3
P (X = x) 0.167 0.500 0.300 0.033
15. What is the variance of X?
(A) 0.56 (B) 0.61 (C) 0.68 (D) 0.72 (E) 0.85
16. What would be the variance of X if we had instead selected the three cards with
replacement?
(A) 0.56 (B) 0.61 (C) 0.68 (D) 0.72 (E) 0.85
17. Ten statisticians took separate random samples and each calculated a 90% confidence
interval to estimate the value of some population parameter. What is the probability
that exactly eight of the intervals contain the true value of the parameter?
(A) 0.1937
(B) 0.2166
(C) 0.2408
(D) 0.2691
(E) depends on which parameter they are trying to estimate
18. Harvey the clumsy waiter is in trouble at work because he has been breaking too many
dishes. Suppose that the number of dishes he breaks follows a Poisson distribution with
a rate of 0.11 per hour. Harvey works an 8-hour shift today, and his boss has warned him
that if he breaks any dishes during the shift, he will be fired. What is the probability
that Harvey will be fired today?
(A) 0.324 (B) 0.415 (C) 0.585 (D) 0.676 (E) 0.896
19. A random variable X follows a Poisson distribution with parameter λ. We are conducting
a hypothesis test of H0: λ = 5 vs. Ha: λ < 5. We decide to reject the null hypothesis if
X ≤ 2. What is the power of the test if λ = 1?
(A) 0.1912 (B) 0.3679 (C) 0.5518 (D) 0.7358 (E) 0.9197
20. The number of undergraduate students at the University of Winnipeg is approximately
9,000, while the University of Manitoba has approximately 27,000 undergraduate stu-
dents. Suppose that, at each university, a simple random sample of 3% of the undergrad-
uate students is selected and the following question is asked: “Do you approve of the
provincial government’s decision to lift the tuition freeze?” Suppose that, within each
university, approximately 20% of undergraduate students favour this decision. What can
be said about the sampling variability associated with the two sample proportions?
(A) The sample proportion from U of W has less sampling variability than that from U
of M.
(B) The sample proportion from U of W has more sampling variability that that from
U of M.
(C) The sample proportion from U of W has approximately the same sampling variabil-
ity as that from U of M.
(D) It is impossible to make any statements about the sampling variability of the two
sample proportions without taking many samples.
(E) It is impossible to make any statements about the sampling variability of the two
sample proportions because the population sizes are different.
21. A local newspaper would like to estimate the true proportion of Winnipeggers who
approve of the job being done by the mayor. A random sample of Winnipeggers is
selected and each respondent is asked whether they approve of the job being done by
the mayor. When the newspaper reports the results of the poll, it states, “results are
accurate to within ± 3%, 19 times out of 20”. What sample size was used for this poll?
(A) 672 (B) 720 (C) 894 (D) 936 (E) 1068
22. A Tim Horton’s manager claims that more than 75% of customers purchase a coffee
when they visit the store. We take a random sample of 225 customers and find that 180
of them purchased a coffee. We would like to conduct a hypothesis test to determine
whether there is significant evidence to support the manager’s claim. The P-value for
the appropriate hypothesis test is:
(A) 0.0179 (B) 0.0256 (C) 0.0301 (D) 0.0359 (E) 0.0418
23. A random sample of voters is selected from each of Canada’s three prairie provinces
(Manitoba, Saskatchewan and Alberta) and respondents are asked which federal political
party they support (Conservative, Green, Liberal or NDP). We would like to conduct
a test at the 10% level of significance to determine whether voters in Canada’s prairie
provinces are homogeneous with respect to which political party they support. What is
the critical value for the appropriate test of significance?
(A) 9.24 (B) 10.64 (C) 12.59 (D) 14.68 (E) 18.55
The next two questions (24 and 25) refer to the following:
A game of bowling consists of ten frames. A bowler scores a strike in a frame if he or she
knocks down all the pins on the first shot. A bowler counted the number of strikes he
has scored in his last 120 games. We would like to conduct a chi-square goodness-of-fit
test at the 5% level of significance to determine whether the number of strikes per game
for this bowler follows a binomial distribution. The data are shown in the table below,
together with some expected cell counts:
# of strikes 0 1 2 3 4 5 6 7 8 9 10
# of games 6 10 16 28 27 18 11 3 0 1 0
Expected Count 1.62 8.70 21.08 30.26 ??? ??? 8.27 2.54 0.52 0.06 0.00
24. Under the null hypothesis, what is the expected number of games in which the bowler
scores five strikes?
(A) 18.43 (B) 19.78 (C) 20.92 (D) 21.65 (E) 22.37
25. What is the critical value for the appropriate test of significance?
(A) 9.49 (B) 12.59 (C) 14.07 (D) 15.51 (E) 16.92
26. The table below displays the number of accidents recorded at a particular intersection
during each of the four seasons last year:
Season Spring Summer Fall Winter
# of accidents 13 24 18 25
We would like to conduct a chi-square goodness-of-fit test to determine whether accidents
are uniformly distributed over the four seasons. The value of the test statistic for the
appropriate test of significance is:
(A) 3.8 (B) 4.7 (C) 5.9 (D) 6.4 (E) 7.6
27. The height (in feet) and trunk diameter (in inches) are measured for a sample of 14
oak trees. The sample correlation between height and trunk diameter is calculated to
be 0.68. We would like to conduct a test of H0: ρ = 0 vs. Ha: ρ 6= 0 to determine
whether there exists a linear relationship between the two variables. The P-value for the
appropriate test of significance is:
(A) between 0.001 and 0.0025
(B) between 0.0025 and 0.005
(C) between 0.005 and 0.01
(D) between 0.01 and 0.02
(E) between 0.02 and 0.04
The next three questions (28 to 30) refer to the following:
Earned Run Average (ERA) is a common statistic used for pitchers in Major League
Baseball. A pitcher’s ERA is the average number of runs he gives up per game. The
lower a pitcher’s ERA, the better he is, and so we might expect him to be paid a higher
salary. A sample of 25 pitchers is selected and their ERA X and Salary Y (in $millions)
are recorded.
From these data, we calculate x̄ = 4.21, ȳ = 2.56 and
n∑
i=1
(xi − x̄)2 = 20.7.
A least squares regression analysis is conducted. Some JMP output is shown below:
28. What is the value of the sample correlation?
(A) −0.275 (B) −0.384 (C) −0.525 (D) −0.605 (E) −0.724
29. A 95% confidence interval for the mean salary of all Major League Baseball pitchers with
an ERA of 4.00 is:
(A) (2.33, 3.35)
(B) (1.97, 3.71)
(C) (1.42, 4.26)
(D) (1.13, 4.55)
(E) (0.29, 5.39)
30. Which of the following intervals would be wider than the 95% confidence interval in the
previous question?
(I) A 95% prediction interval for the salary of a pitcher with an ERA of 4.00
(II) A 99% confidence interval for the salary of all pitchers with an ERA of 4.00
(III) A 95% confidence interval for the salary of all pitchers with an ERA of 2.50
(A) I only
(B) II only
(C) I and II only
(D) I and III only
(E) I, II and III
Sample Final Exam 1 – Part B
1. The following three games are scheduled to be played at the World Curling Championship
one morning. The values in parentheses are the probabilities of each team winning their
respective game.
Game 1: Finland (0.2) vs. Canada (0.8)
Game 2: USA (0.3) vs. Switzerland (0.7)
Game 3: Germany (0.4) vs. Japan (0.6)
(a) The outcome of interest is the set of winners for each of the three games. List the
complete sample space of outcomes and calculate the probability of each outcome.
(b) Let X be the number of European teams that win their respective games. Find the
probability distribution of X.
(c) Find the expected value and variance of X.
(d) If two European teams win their games, what is the probability that Finland is one
of them?
2. A candidate for political office would like to determine whether his support differs among
male and female voters. In a random sample of 300 male voters, 180 indicated they
support the candidate. In a random sample of 200 female voters, 104 said they support
the candidate.
(a) Calculate a 95% confidence interval for the difference in proportions of male and
female voters who support the candidate.
(b) Provide an interpretation of the confidence interval in (a).
(c) Conduct a two-sample z test to determine whether there is a significant difference
between the proportions of male and female voters who support the candidate. Use
the P-value approach and a 5% level of significance.
(d) Provide an interpretation of the P-value of the test in (c).
3. Suppose we want to instead conduct the test in the previous question using a chi-square
test for homogeneity. Conduct the test using the P-value approach and a 5% level of
significance. Determine an exact P-value.
Part A Answer Key
1. D 16. D
2. A 17. A
3. D 18. C
4. A 19. E
5. C 20. B
6. D 21. E
7. D 22. E
8. E 23. B
9. E 24. A
10. C 25. A
11. B 26. B
12. B 27. C
13. A 28. E
14. B 29. A
15. A 30. E
Part B Answers
1. (b) P (X = 0) = 0.144, P (X = 1) = 0.468, P (X = 2) = 0.332, P (X = 3) = 0.056)
(c) E(X) = 1.3, V ar(X) = 0.61
(d) 0.3253
2. (a) (−0.0087, 0.1687)
(c) z = 1.77, P-value = 0.0768
3. χ2 = 3.13, P-value = 0.0768
Sample Final Exam 2 – Part A
1. A statistical test of significance is designed to:
(A) assess the strength of evidence in favour of H0.
(B) assess the strength of evidence in favour of Ha.
(C) prove that H0 is true.
(D) find the probability that H0 is true.
(E) find the probability that Ha is true.
2. We would like to conduct a hypothesis test at the 2% level of significance to determine
whether the true mean pH level in a lake differs from 7.0. Lake pH levels are known to
follow a normal distribution. We take 10 water samples from random locations in the
lake. For these samples, the mean pH level is 7.3 and the standard deviation is 0.37.
Using the critical value approach, the decision rule would be to reject H0 if the test
statistic is:
(A) less than −2.326 or greater than 2.326
(B) less than −2.398 or greater than 2.398
(C) less than −2.564 or greater than 2.564
(D) less than −2.764 or greater than 2.764
(E) less than −2.821 or greater than 2.821
3. We would like to conduct a hypothesis test to determine whether the true mean rent
amount for all one-bedroom apartments in Winnipeg differs from $850. We take a
random sample of 50 one-bedroom apartments and calculate the sample mean to be
$900. A 98% confidence interval for µ is calculated to be (830, 970). The conclusion
for our test would be to:
(A) fail to reject H0 at the 2% level of significance since the value 850 is contained in
the 98% confidence interval.
(B) fail to reject H0 at the 1% level of significance since the value 900 is contained in
the 98% confidence interval.
(C) fail to reject H0 at the 2% level of significance since the value 900 is contained in
the 98% confidence interval.
(D) reject H0 at the 2% level of significance since the value 850 is contained in the 98%
confidence interval.
(E) reject H0 at the 1% level of significance since the value 900 is contained in the 98%
confidence interval.
The next two questions (4 and 5) refer to the following:
The GPAs of samples of students from two universities are recorded. Some summary
statistics are shown in the table below:
Sample Size Sample Mean Sample Variance
University 1 10 3.57 0.25
University 2 15 2.99 0.09
GPAs for students at both universities are known to follow normal distributions.
4. We would like to conduct a hypothesis test to determine whether the true mean GPA
of all students at University 1 differs from 3.00. What is the P-value of the appropriate
test of significance?
(A) between 0.001 and 0.0025
(B) between 0.0025 and 0.005
(C) between 0.005 and 0.01
(D) between 0.01 and 0.02
(E) between 0.02 and 0.04
5. We would like to conduct a hypothesis test to determine whether the true mean GPA
for students at University 1 is greater than that for students at University 2. The value
of the test statistic for the appropriate test of significance is:
(A) 3.29 (B) 3.64 (C) 7.04 (D) 7.57 (E) 8.29
6. We would like to conduct a hypothesis test to determine whether the mean final exam
score µ1 for section A01 in a large course is greater than the mean final exam score µ2
for section A02. We will make a Type I Error if we conclude that:
(A) µ1 = µ2 when in fact µ1 > µ2.
(B) µ2 > µ1 when in fact µ1 > µ2.
(C) µ2 > µ1 when in fact µ1 = µ2.
(D) µ1 > µ2 when in fact µ2 > µ1.
(E) µ1 > µ2 when in fact µ1 = µ2.
7. We would like to conduct a matched pairs t test to determine whether premium gasoline
is more efficient than regular gasoline. We take a sample of ten different models of car.
Each car drives for one tank on premium gasoline and one tank on regular gasoline
(with the order randomly determined) and the mileage is recorded for each. Which of
the following statements about the matched pairs t test are true?
(I) For each car, mileage on premium gas and mileage on regular gas are independent.
(II) For each car, mileage on premium gas and mileage on regular gas are dependent.
(III) For any two cars, mileages on premium gas are independent.
(IV) We must assume that mileage on premium gas and mileage on regular gas are
both normally distributed.
(V) We must assume that the differences in mileage (premium − regular) follow a
normal distribution.
(A) I and IV only
(B) II and V only
(C) I, III and IV only
(D) I, III and V only
(E) II, III and V only
8. We conduct an experiment to compare the effectiveness of four different headache med-
ications. Each medication is randomly assigned to three patients, and the time (in
minutes) until patients experience relief is recorded. Suppose it is known that relief
times for the four medications are normally distributed. The test statistic is calculated
to be 3.97. What is the P-value of the appropriate test of significance?
(A) between 0.001 and 0.01
(B) between 0.01 and 0.025
(C) between 0.025 and 0.05
(D) between 0.05 and 0.10
(E) greater than 0.10
The next two questions (9 and 10) refer to the following:
We would like to conduct an analysis of variance at the 5% level of significance to
compare the mean percentage grades for students in five sections of a first-year university
course. We take a simple random sample of students from each section. We assume that
percentage scores follow normal distributions for each of the five sections. Some summary
statistics are shown below:
Section Sample Size Sample Mean Sample Std. Dev.
A01 3 74 8.9
A02 5 80 8.8
A03 6 59 13.6
A04 2 78 11.3
A05 4 66 15.3
9. What is the critical value for the appropriate test of significance?
(A) 2.71 (B) 2.90 (C) 3.06 (D) 4.56 (E) 5.86
10. One assumption required in conducting an ANOVA F test is that all population standard
deviations are equal. The estimate of this common population standard deviation is:
(A) 11.58 (B) 11.72 (C) 11.88 (D) 12.10 (E) 12.17
The next two questions (11 and 12) refer to the following:
The following six games are being played in the National Hockey League one night. The
values in parentheses are the probabilities of each team winning their respective game:
Game 1: Tampa Bay Lightning (0.47) vs. New York Rangers (0.53)
Game 2: Carolina Hurricanes (0.58) vs. New York Islanders (0.42)
Game 3: Winnipeg Jets (0.27) vs. Washington Capitals (0.73)
Game 4: Chicago Blackhawks (0.60) vs. Dallas Stars (0.40)
Game 5: Colorado Avalanche (0.31) vs. Detroit Red Wings (0.69)
Game 6: San Jose Sharks (0.46) vs. Vancouver Canucks (0.54)
11. The outcome of interest is the set of winners for each of the six games. How many
outcomes are contained in the sample space?
(A) 12 (B) 32 (C) 36 (D) 64 (E) 128
12. What is the probability that Winnipeg wins or that both New York teams win?
(A) 0.4325 (B) 0.4522 (C) 0.4724 (D) 0.4926 (E) 0.5123
13. A hockey player scores a goal in 43% of his games. His team wins 65% of their games.
In 31% of the team’s games, the player scores a goal and the team wins. What is the
probability that the team wins if the player does not score a goal?
(A) 0.3595 (B) 0.4625 (C) 0.4985 (D) 0.5965 (E) 0.6345
14. In a game of poker, you are dealt five cards from a deck of 52 cards. What is the
probability that you get a flush (five cards of the same suit)?
(A) 0.0005 (B) 0.0010 (C) 0.0020 (D) 0.0030 (E) 0.0040
15. The time it takes a student to drive to university in the morning follows a normal distri-
bution with mean 28 minutes and standard deviation 4 minutes. The time it takes the
student to drive home from university in the afternoon follows a normal distribution with
mean 25 minutes and standard deviation 3 minutes. Morning and afternoon commuting
times are known to be independent. What is the probability that the student’s total
travel time to and from school one day exceeds one hour?
(A) 0.0040 (B) 0.0808 (C) 0.1587 (D) 0.2296 (E) 0.3897
16. A random variable X has a binomial distribution with parameter n = 3. What must be
the value of the parameter p in order for P (X = 2) = P (X = 3)?
(A)
1
4
(B)
1
3
(C)
1
2
(D)
2
3
(E)
3
4
17. A car company reports that the number of breakdowns per 8-hour shift on its machine-
operated assembly line follows a Poisson distribution with a mean of 1.5. Assuming
that the machine operates independently across shifts, what is the probability of no
breakdowns during three consecutive 8-hour shifts?
(A) 0.0111 (B) 0.0498 (C) 0.0744 (D) 0.1923 (E) 0.2065
18. A random variable X follows a Poisson distribution with parameter λ. If we know that
P (X = 1) = P (X = 3), then what is the value of λ?
(A)
√
2 (B)
√
3 (C)
√
5 (D)
√
6 (E)
√
8
19. Suppose it is known that 83% of motorists wear a seatbelt while driving. The police
stop a random sample of 200 drivers. What is the probability that more than 80% of
them are wearing a seat belt?
(A) 0.8023 (B) 0.8212 (C) 0.8554 (D) 0.8708 (E) 0.8997
20. We would like to estimate the true proportion of students at a large university who are
female. What sample size do we require in order to estimate the true proportion to
within 0.05 with 98% confidence?
(A) 136 (B) 379 (C) 542 (D) 1127 (E) 2165
21. In a random sample of 250 Winnipeggers, 80 of them said they use transit. A 94%
confidence interval for the true proportion of all Winnipeggers who use transit is:
(A) (0.260, 0.380)
(B) (0.265, 0.375)
(C) (0.270, 0.370)
(D) (0.275, 0.365)
(E) (0.280, 0.360)
22. The Catholic Church does not allow women to become priests. We will take a random
sample of 300 Catholics and ask them if they believe women should be allowed to become
priests. We would like to conduct a hypothesis test at the 10% level of significance to
determine whether the majority of Catholics support the idea. What is the power of the
test if the true proportion of Catholics who favour the idea is actually 0.56?
(A) 0.7357 (B) 0.7881 (C) 0.8025 (D) 0.8238 (E) 0.8599
23. A clinical trial of Nasonex was conducted, in which 400 randomly selected pediatric
patients (ages 3 to 11 years old) were randomly divided into two groups. The patients in
Group 1 (experimental group) received 200 mcg of Nasonex, while the patients in Group
2 (control group) received a placebo. We conduct a test of H0: p1 = p2 vs. Ha: p1 6= p2
to compare the proportions of subjects in the two groups who experienced headaches as
a side effect. The test statistic is calculted to be 2.25 and the P-value is 0.0244. Suppose
we had instead compared the two proportions using a chi-square test for homogeneity.
The value of the test statistic and the P-value would be, respectively:
(A) 5.06 and 0.0244
(B) 2.25 and 0.0006
(C) 1.50 and 0.0244
(D) 5.06 and 0.0006
(E) 2.25 and 0.0244
The next three questions (24 to 26) refer to the following:
We would like to conduct a test of significance at the 10% level of significance to deter-
mine whether smoking behaviour of university students is independent of their parents’
smoking behaviour. The data is displayed in the table below, as well as some expected
cell counts and cell chi-square values:
Observed Student Student Row
Expected Smokes Doesn’t Smoke Total
Cell Chi-Square
Neither Parent 17 62 79
Smokes 21.13 57.87
0.81 0.29
One Parent 11 40 51
Smokes 13.64 37.36
0.51 0.19
Both Parents 14 13 27
Smoke ??? ???
6.37 ???
Column 42 115 157
Total
24. What is the critical value for the appropriate test of significance?
(A) 4.61 (B) 5.99 (C) 7.78 (D) 9.24 (E) 10.64
25. What is the expected count for the number of non-smoking university students who have
two parents who smoke?
(A) 12.37 (B) 14.22 (C) 16.45 (D) 19.78 (E) 21.09
26. What is the value of the test statistic for the appropriate test of significance?
(A) 4.92 (B) 6.54 (C) 8.17 (D) 10.49 (E) 12.31
The next two questions (27 and 28) refer to the following:
We would like to determine whether the number of errors on income tax forms processed
by an accounting firm has a Poisson distribution. An employee selects a random sample
of 100 tax returns and determines the number of errors on each. The data are shown in
the table below, as well as some expected cell counts:
# of errors 0 1 2 3 4 5
# of forms 36 28 23 8 3 2
Expected 30.12 ??? ??? 8.67 2.60 0.62
27. Under the null hypothesis, what is the expected number of forms with two errors?
(A) 11.53 (B) 17.46 (C) 21.69 (D) 24.22 (E) 26.01
28. What are the degrees of freedom for the appropriate test statistic?
(A) 2 (B) 3 (C) 4 (D) 5 (E) 6
The next two questions (29 and 30) refer to the following:
A study examined the relationship between the sepal width and the sepal length for a
certain variety of tropical plant. Some JMP output is shown below:
29. One plant in the sample had a sepal width of 10.7 and a sepal length of 1.2. What is
the value of the residual for this plant?
(A) 0.1087 (B) −1.3087 (C) 0.3087 (D) 1.3087 (E) −0.1087
30. We would like to conduct a test of H0: ρ = 0 vs. Ha: ρ 6= 0 to determine whether there
exists a linear relationship between sepal width and sepal length. The value of the test
statistic for the appropriate test of significance is:
(A) −3.48 (B) −3.14 (C) −2.13 (D) 2.13 (E) 3.48
Sample Final Exam 2 – Part B
1. (a) The owner of a small store suspects that one of his employees is stealing, as the
amount of money in the cash register at the end of the employee’s shift is often less
than expected. The employee agrees to take a polygraph (lie detector) test. During
the test, the employee claims that he has not stolen any money. The polygraph is
essentially testing the hypotheses
H0: The employee is telling the truth. vs. Ha: The employee is lying.
Explain what it would mean in the context of this example to make a Type I error
and a Type II error. Explain the potential consequences of each type of error.
(b) The strengths of prestressing wires manufactured by a steel company have a mean
of 2000 N and a standard deviation of 100 N. By employing a new manufacturing
technique, the company claims that the mean strength will be increased. To verify
this claim, a builder will test a random sample of 36 wires produced by the new
process and will conduct a hypothesis test of H0: µ = 2000 vs. Ha: µ > 2000 at
the 10% level of significance. What would be the power of the test if the true mean
strength of wires produced by the new process was 2050 N?
2. Three contestants compete on each episode of the TV game show Jeopardy! The number
of female contestants for a random sample of 200 shows are shown in the table below:
# of females 0 1 2 3
# of shows 50 94 52 4
Conduct a chi-square goodness-of-fit test at the 5% level of significance to determine
whether the number of female contestants per episode follows a binomial distribution.
Use the P-value method and show all of your steps.
3. We would like to use the weight of a car to predict its fuel economy. Weights (in 1000s
of pounds) and fuel efficiency (in miles per gallon) are shown in the table below for a
sample of ten car models:
Weight 3.1 3.7 2.2 3.4 2.0 3.9 3.0 2.5 3.6 2.7
Fuel Efficiency 25 21 31 22 35 16 24 27 17 29
It can be shown that x̄ = 3.01, sx = 0.65, ȳ = 24.70, sy = 6.02,
n∑
i=1
(xi− x̄)2 = 3.8025 and
n∑
i=1
(yi − ŷ)2 = 24.92.
The equation of the least-squares regression line is calculated to be ŷ = 51.47− 8.89x.
(a) Write out the least squares regression model and define all terms.
(b) The residual plot is shown below:
What does the residual plot tell you about the validity of the regression model in
(a)?
(c) Construct a 95% confidence interval for the slope of the least squares regression
line.
(d) Provide an interpretation of the confidence interval in (c).
(e) Conduct a test of H0: β1 = 0 vs. Ha: β1 6= 0 to determine whether there exists a
linear relationship between the weight of a car and its fuel efficiency.
(f) Provide an interpretation of the P-value of the test in (e).
(g) Calculate a 95% prediction interval for the fuel efficiency of a car that weighs 3200
pounds.
Part A Answer Key
1. B 16. E
2. E 17. A
3. A 18. D
4. C 19. D
5. B 20. C
6. E 21. B
7. E 22. B
8. D 23. A
9. C 24. A
10. E 25. D
11. D 26. D
12. A 27. C
13. D 28. A
14. C 29. E
15. B 30. A
Part B Answers
1. (b) Power = 0.9573
2. p̂ = 0.35, χ2 = 3.56, df = 2, 0.15 < P-value < 0.20
3. (c) (−10.977, −6.803)
(e) t = −9.82, P-value < 0.0005
(g) (18.735, 27.309)
