Statistics 2593 Review

1. Metal machine parts are often subject to micro-cracks which are difficult to detect but make the
part defective. Suppose that in a certain production process parts are automatically scanned for
micro-cracks using an X-ray device and digital image processing. The probability that the system will
correctly determine that a part has micro-cracks is 0.98. The probability that the system will
incorrectly indicate that a part has micro-cracks when it is in fact defect free, is 0.03. The production
system is of high quality so that the fraction of parts which actually have micro-cracks is 0.001. Let C
be the event that a part has micro-cracks and D be the event that the automatic system indicates
that a part has micro-cracks.

a) What is the value of P (C ∩ D)? That is, what is the probability that a part has micro-cracks and
the automatic detection system indicates that it has micro- cracks?
b) What is the value of P (C’ ∩ D)? That is, what is the probability that a part has no micro-cracks
and the automatic detection system (incorrectly) indicates that it has micro-cracks?
c) What is the value of P(D)? That is, what is the probability that the automatic detection system
indicates that a part has micro-cracks?
d) Are the events C and D independent? Are they mutually exclusive? Explain briefly.
e) Suppose that the automatic detection system indicates that a part has micro- cracks. What is the
probability that it does indeed have micro-cracks?

2. In engineering, redundancy or “backup” is often built into a system to improve reliability. Suppose
that a fire alarm system has a duplicate of a critical component. The probability that the first
component fails is 0.01 and so is the probability that the second component fails; however, the
probability that the second fails given that the first has failed is 0.03. Let A be the event that
component 1 fails, and B be the event that component 2 fails.

a) Are A and B independent? Explain briefly.

b) Are A and B mutually exclusive? Explain briefly.
c) What is the probability that the fire alarm fails, i.e. that both components fail?
d) What is the probability that neither component fails?
e) What is the probability that exactly one component fails?

3. In a major eastern city, 60% of the automobile drivers are 30 years of age or older, and 40% of the
drivers are under 30 years of age. Of all drivers 30 years of age or older, 4% will have a traffic
violation in a 12-month period. Of all drivers under 30 years of age, 10% will have a traffic violation
in a 12-month period. Assume that a driver has just been charged with a traffic violation; what is the
probability that the driver is under 30 years of age?
4. How many ways can a team of three people be chosen from a group of ten? How many ways can a
committee of three people be chosen, where one person is the president, one is the vice-president,
and one is the treasurer?

5. How many ways can you pick four desserts from a menu of ten? How many ways can you form a list
of your four favorite desserts from a menu of ten?
6. How many ways can the letters of OPTICAL be rearranged? How many ways can the letters of
CORPORATION be arranged?
7. Suppose that 80% of trees near a certain busy highway have a lead content in their bark of at least
920 micro grams per gram dry weight.

a) Suppose that a random sample of 10 trees is taken. What is the probability that at most 7 of
these trees have a bark lead content of 920 or more?
b) What is the probability that exactly 7 of these 10 trees have a bark lead content of 920 or more?
c) How many trees with a lead content of at least 920 micro grams per gram dry weight would you
expect to see in a random sample of 10 trees?

8. Suppose that, on average, phone calls arrive at a call centre at a rate of 4 calls per minute. What is
the probability that the call centre receives no calls in the next minute? What is the probability that
the call centre receives more than 15 calls in the next five minutes?
9. The copper thickness of parts coming from a plating operation is approximately normally distributed
with a mean of 2.68 mm and a standard deviation of 0.12 mm.

a) Would a thickness of 2.85 mm be surprising? (Why/Why not?)

b) What proportion of parts have thickness between 2.6 mm and 2.8 mm?
c) Suppose that 5 parts are selected for inspection. What is the probability that at least one of
them has copper thickness less than 2.6 mm?
d) Suppose the thicknesses of 5 parts are measured, and the results averaged. What is the
probability that this average will be between 2.6 mm and 2.8 mm?

10. The engines made by Ford for speedboats have an average power of 220 horsepower (HP) and
standard deviation of 15 HP. A potential buyer intends to take a sample of four engines and will not
place an order if the sample mean of the horsepower is less than 215 HP. What is the probability
that the buyer will not place an order? Assume that the horsepower of the engines is normally
distributed.
11. A sample of Alzheimer's patients is tested to assess the amount of time in stage IV sleep. It has been
hypothesized that individuals suffering from Alzheimer's Disease may spend less time per night in
the deeper stages of sleep. The number of minutes spent in Stage IV sleep is recorded for sixty-one
patients. The sample produced a mean of 48 minutes of stage IV sleep over a 24-hour period.
Assuming a population standard deviation of 14 minutes, compute a 95 percent confidence interval
for these data. What does this information tell you about an individual’s (an Alzheimer's patient)
stage IV sleep? How large must the sample size be if a 95 percent confidence interval with a width of
5 minutes is desired?
12. A sample of students from an introductory psychology class were polled regarding the number of
hours they spent studying for the last exam. All students anonymously submitted the number of
hours on a 3 by 5 card. There were 24 individuals in the one section of the course polled. The data
was used to make inferences regarding the other students taking the course. Their data are below:

4.5 22 7 14.5 9 9 3.5 8 11 7.5 18 20

7.5 9 10.5 15 19 2.5 5 9 8.5 14 20 8

 Compute a 95 percent confidence interval. What does this tell us?

13. The quality manager of a fortune cookie company believes that a larger than acceptable proportion
of paper fortunes being used are blank. Suppose she takes a sample of 320 fortune cookies from the
production line, and 15 of the paper fortunes are blank. Calculate a 90% confidence interval for the
proportion of blank paper fortunes.
14. Measurements of the tensile strength (in psi) of a rope are given below.

18.46 18.42 18.47 18.42 18.51 18.45 18.86

a) Assume the population standard deviation of the tensile strength of the rope is 0.4 psi. Test
whether the mean tensile strength of the population sampled is 18.47 psi. Use a significance
level of 0.05.
b) As it turns out, the standard deviation in part (a) was reported incorrectly, and so a new value
must be estimated from the data. Test whether the mean of the population sampled is 18.47
psi. Use a significance level of 0.05.

15. A group of researchers wants to investigate the proportion of Canadians who are right handed.
Most sources cite a proportion of 80%, but this group believes it may be higher. 100 Canadians are
sampled, and it is found that 87 of them are right handed. Is there any statistical evidence to suggest
that the researchers are correct?
16. Two methods for determining the percentage of iron in ore samples were compared by splitting
each of 87 ore samples into two parts. One part was then measured using method 1, and the other
part was measured using method 2. The data were analyzed in Minitab in two ways, as shown on
page 5. One of the analyses is correct and the other is incorrect.
 a) State which of the two analyses is correct, giving a brief reason for your choice.
b) State appropriate null and alternative hypotheses to be tested using these data.
c) Explain briefly your choice of alternative hypothesis.
d) Based on the correct analysis, state your decision about the hypothesis test, at the 0.05
significance level. Explain briefly in words what your decision means.

17. A stress analysis was conducted on epoxy-repaired truss joints in damaged timber structures
constructed from two species of wood, namely Southern Pine, and Ponderosa Pine. Repaired joints
from forty (40) structures made of Southern Pine, and from another forty structures made of
Ponderosa Pine, were selected for the study. The mean shear stress of each joint, in psi, was
determined. The data were analyzed in Minitab in two ways. One of the analyses is correct and the
other is incorrect.
 a) State which of the two analyses is correct, giving a brief reason for your choice.
b) Based upon the correct analysis, write down a 95% confidence interval for the mean
difference, in strength of repaired joints, between the two species of wood. Does it appear
that there is in fact a real difference between the two species of wood regarding joint
strength? Explain, briefly.
c) Suppose that your supervisor tells you that the Southern pine joints are on average 150 psi
stronger than those of Ponderosa pine. Have you any reason to disbelieve her? Explain,
briefly.

18. Time magazine reported the result of a telephone poll of 800 adult Americans. The question posed
to the Americans who were surveyed was: "Should the federal tax on cigarettes be raised to pay for
health care reform?" 605 of the 800 people surveyed were non-smokers, and the rest were
smokers. Of the 605 non-smokers, 351 answered “yes.” Of the smokers, 41 answered “yes.” Is there
any evidence at the 0.05 level that the opinions of smokers and non-smokers differ significantly?

Test and CI for Two Proportions

Method
p₁: proportion where Sample 1 = Event
p₂: proportion where Sample 2 = Event

Difference: p₁ - p₂
Descriptive Statistics
Sample N Event Sample p
Sample 1 605 351 0.580165

Sample 2 195 41 0.210256

Estimation for Difference
95% CI for
Difference Difference
 0.369909 (0.300499, 0.439319)
CI based on normal approximation
Test
Null hypothesis H₀: p₁ - p₂ = 0
Alternative hypothesis H₁: p₁ - p₂ ≠ 0

Method Z-Value P-Value

Normal approximation 10.45 0.000

Fisher's exact 0.000

19. The regional transit authority for a major metropolitan area would like to determine if there is any
relationship between the age of a bus and the annual maintenance cost. A sample of twenty buses
resulted in the following data.

a) Test of the significance of the constant parameter at a significance level of 0.05.

b) Test the significance of the slope parameter.
c) What is the expected maintenance cost of a 3.5 year old bus?
d) Would you be surprised to find a 3.5 year old bus with maintenance costs of $400? (Why/why not?)
e) Is this model useful for predicting maintenance costs? (Why/why not?)