(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

1. (3 points) An insurance company found that 2.5% of male drivers between the ages of 18 and 25 are
involved in serious accidents annually. To simplify the analysis, assume that every such accident costs
the insurance company $65,000 and that a driver can only have one of these accidents in a year.
(a) If the company charges $2,500 for such coverage, what is the chance that it loses money on a single
policy?

(b) Suppose that the company writes 1,000 such policies to a collection of drivers. What is the proba-
bility that the company loses money on these policies? Assume that the drivers don’t run into each
other and behave independently.

(c) Does the difference between the probabilities of parts (a) and (b) explain how insurance companies
stay in business? Large auto insurers are certainly profitable.
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

2. (3 points) List the following characteristics:

(a) Population
(b) Parameter of interest
(c) Sampling frame
(d) Sample size
(e) Sampling design, including whether randomization was employed
(f) Any potential sources of bias or other problems with the survey or sample
on the basis of the brief description of the survey that is shown.

A vendor opens a small booth at a supermarket to offer customers a taste of a new beverage.
The staff at the booth offer the beverage to adults who pass through the soda aisle of the
store near a display of the product. Customers who react favorably receive a discount coupon
toward future purchases.

Page 2
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

3. (3 points) A package of light bulbs promises an average life of more than 750 hours per bulb. A consumer
group did not believe the claim and tested a sample of 40 bulbs. The average lifetime of these 40 bulbs
was 740 hours with s = 30 hours. The manufacturer responded that its claim was based on testing
hundreds of bulbs.
(a) If the consumer group and manufacturer both make 95% confidence intervals for the population’s
average lifetime, whose will probably be shorter? Can you tell for certain?

(b) Given the usual sampling assumptions, is there a 95% probability that 740 lies in the 95% confidence
interval of the manufacturer?

(c) Is the manufacturer’s confidence interval more likely to contain the population mean because it is
based on a larger sample?

Page 3
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

4. (3 points) In a survey of employees, Watson-Wyatt reported that 51% had confidence in the actions of
senior management. To be 95% confident that at least half of all employees have confidence in senior
management, how many would have to be in the survey sample?

5. (3 points) A company that stocks shelves in supermarkets is considering expanding the supply that it
delivers. Items that are not sold must be discarded at the end of the day, so it only wants to schedule
additional deliveries if stores regularly sell out. A break-even analysis indicates that an additional
delivery cycle will be profitable if items are selling out in more than 60% of markets. A survey during
the last week in 45 markets found the shelves bare in 35.
(a) State the null and alternative hypotheses.

Page 4
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

(b) Describe a Type I error and a Type II error in this context.

(c) Find the p-value of the test. Do the data supply enough evidence to reject the null hypothesis if
the α-level is 0.05?

6. (4 points) A variety of stores offer loyalty programs. Participating shoppers swipe a bar-coded tag at
the register when checking out and receive discounts on certain purchases. Stores benefit by gleaning
information about shopping habits and hope to encourage shoppers to spend more. A typical Saturday
morning shopper who does not participate in this program spends $120 on her or his order. In a sample
of 80 shoppers participating in the loyalty program, each shopper spent $130 on average during a recent
Saturday, with standard deviation s = $40. Is this statistical proof that the shoppers participating in
the loyalty program spend more on average than typical shoppers? (Assume that the data meet the
sample size condition.)
(a) State the null and alternative hypotheses. Describe the parameters.

Page 5
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

(b) Describe the Type I and Type II errors.

(c) How large could the kurtosis be without violating the CLT condition?

Page 6
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

(d) Find the p-value of the test. Do the data supply enough evidence to reject the null hypothesis if
α = 0.05?

7. (4 points) Members of a sales force were randomly assigned to two management groups. Each group
employed a different technique for motivating and supporting the sales team. Let’s label these groups A
and B, and let µA and µB denote the mean weekly sales generated by members of the two groups. The
95% confidence interval for µA − µB was found to be [$500, $2, 200].
(a) If profits are 40% of sales, what’s the 95% confidence interval for the difference in profits generated
by the two methods?

Page 7
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

(b) Assuming the usual conditions, should we conclude that the approach taken by Group A sells more
than that taken by Group B, or can we dismiss the observed difference as due to chance?

(c) The manager responsible for Group B complained that his group had been assigned the “poor
performers” and that the results were not his fault. How would you respond?

Page 8
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

8. (4 points) The following contingency table breaks down a month of customer complaints received by a
retailer. Some of those complaining are long-term customers, whereas the rest are recent. There are four
types of complaints.

(a) What would it mean to the retailer if customer status and complaint type were dependent?

(b) If many of these complaints come from the same irate customer, do the assumptions needed for a
chi-squared test hold?

Page 9
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

(c) Does it appear from a brief inspection that customer status and the type of complaint are indepen-
dent or dependent? Don’t calculate χ2 ; just skim the table.

(d) If every count in the table increased by a factor of 2, how would the value of x2 change (for example,
if the first cell were to hold 84 rather than 42)? As in (b), think rather than calculate.

Page 10
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

9. (3 points) Refer to the data shown in Question 8.

(a) Compute the value of χ2 and the p-value for the test of the null hypothesis of independence.

(b) Summarize the results of this test for the retailer’s sales manager.

Page 11
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

10. (Optional Question) A tire manufacturer warranties its tires to last at least 20,000 miles or “you get
a new set of tires.” In its experience, a set of these tires lasts on average 26,000 miles with SD 5,000
miles. Assume that the wear is normally distributed. The manufacturer profits $200 on each set sold,
and replacing a set costs the manufacturer $400.
(a) What is the probability that a set of tires wears out before 20,000 miles?

(b) What is the probability that the manufacturer turns a profit on selling a set to one customer?

(c) If the manufacturer sells 500 sets of tires, what is the probability that it earns a profit after paying
for any replacements? Assume that the purchases are made around the country and that the drivers
experience independent amounts of wear.

Page 12
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

11. (Optional Question) Two members of a bank’s research group have proposed different questions to ask
in seeking customers’ opinions.
Question 1: Should United Banks take employees away from their families to open on Saturdays
for a few hours?
Question 2: Should United Banks offer its customers the flexibility of convenient banking over
the weekend?
Do you think responses to these two questions might differ? How?

12. (Optional Question) Many advocates for daylight savings time claim that it saves money by reducing
energy consumption. Generally it would be hard to run an experiment to test this claim, but as fortune
would have it, counties in Indiana provided an opportunity. Until 2006, only 15 of the 92 counties in
Indiana used daylight savings time. The rest had remained on standard time. Now all are on daylight
savings time. The local energy provider has access to utility records in counties that recently moved to

Page 13
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

daylight savings time and counties that did not. How can it exploit these data to test the benefits of
daylight savings time?

13. (Optional Question) Auto manufacturers buy car components from suppliers who deliver them to the
assembly line. A manufacturer can schedule staff time to check a sample of 5 parts a day for defects or
a sample of 25 parts over the weekend when the assembly line is slowed.
(a) If the manufacturer is interested in finding a flaw that shows up rarely, which approach will have a
better chance of finding it?

(b) What if the manufacturer is concerned about a problem that will be evident and persistent?

Page 14
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

14. (Optional Question) The table summarizes the performance of golfers in the 2010 U.S. Open golf tour-
nament. The table shows the number of holes in each round completed in fewer than the allowed number
of strokes (birdies, which includes double birdies), at the allowed number of strokes (par), with one extra
stroke (bogey), or with two extra (double bogey). For example, in the first round players scored one
under par on 341 holes. Only players who score among the lower half after two rounds participate in
the final two rounds.

(a) What would it mean to find a statistically significant value of χ2 for this table? Interpret your
answer in the context of the tournament.

Page 15
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

(b) What is the impact of dropping the higher scoring players from the later rounds on the assumptions
and interpretation of χ2 ?

(c) Do these data meet the conditions required by the chi-squared test of independence?

Page 16
(Fall 2023) STAT 1010 Homework 8 Due: Friday, December 8 at 11:59pm

(d) Regardless of your answer to part (d) compute χ2 with a p-value. How would you interpret these
values, given your answer to part (d)?

Page 17