Appendix D 795
Appendix D: Answers to Odd-Numbered
Section Exercises, plus Answers to All
End-of-Chapter Statistical Literacy
and Critical Thinking Exercises, Chapter
Quick Quizzes, Review Exercises,
Cumulative Review Exercises
Chapter 1 Answers
Section 1-2
1. A voluntary response sample is one in which the subjects them-
selves decide whether to be included in the study.
3. Statistical significance is indicated when methods of statistics are
used to reach a conclusion that some treatment or finding is ef-
fective, but common sense might suggest that the treatment or
finding does not make enough of a difference to justify its use or
to be practical.
5. Although the program appears to have statistical significance, it
does not have practical significance because the mean loss of
3.0 lb after one year does not seem to justify the program. The
mean weight loss is not large enough.
7. Possible, but very unlikely
9. Possible and likely
11. Possible, but very unlikely
13. Impossible
15. The x values are not matched with the y values, so it does not
make sense to use the differences between each x value and the
y value that is in the same column.
17. Given the context of the data, we could address the issue of
whether the two types of cigarette provide the same levels of
nicotine, or whether there is a difference between the two types
of cigarettes.
19. The x values are matched with the y values. It does not make
sense to use the difference between each x value and the y value
that is in the same column. The x values are weights (in pounds)
and the y values are fuel consumption amounts (in mi/gal), so
the differences are meaningless.
21. Consumers know that some cars are more expensive to operate be-
cause they consume more fuel, so consumers might be inclined to
purchase cars with better fuel efficiency. Car manufacturers can
then profit by selling cars that appear to have high levels of fuel ef-
ficiency, so there would be an incentive to make the fuel consump-
tion amounts appear to be as favorable as possible. In this case, the
source of the data would be suspect with a potential for bias.
23. The Ornish weight loss program has statistical significance, be-
cause the results are so unlikely (3 chances in 1000) to occur by
chance. It does not have practical significance because the
amount of lost weight (3.3 lb) is so small.
25. Yes. If the claimed rate of 50% is correct, there is a very small like-
lihood that the survey would obtain a result of 85%. The survey
result of 85% is substantially greater than the claimed rate of 50%.
27. a. Yes. All of the highway fuel consumption amounts are con-
siderably greater than the corresponding city fuel consump-
tion amounts.
b. Yes. The differences appear to be substantial.
c. The difference does have practical implications. For example,
we can reduce fuel costs by driving at times when there is not
heavy traffic. If we must enter and pass through a large city,
we might fill the tank if it is low, so that we don’t run out of
fuel due to the increased consumption.
Section 1-3
1. A parameter is a numerical measurement describing some char-
acteristic of a population, whereas a statistic is a numerical mea-
surement describing some characteristic of a sample.
3. Discrete data result when the number of possible values is either
a finite number or a “countable” number (where the number
of possible values is 0 or 1 or 2 and so on), but continuous data
result from infinitely many possible values that correspond to
some continuous scale that covers a range of values without
gaps, interruptions or jumps.
5. Statistic
7. Parameter
9. Parameter
11. Statistic
13. Discrete
15. Continuous
17. Discrete
19. Continuous
21. Ratio
23. Ordinal
25. Nominal
27. Interval
29. Sample: The readers who returned the completed survey. Popu-
lation: Answer varies, but all readers of USA Today is a good
answer. The sample is not likely to be representative of the
population because it is a voluntary response sample.
31. Sample: The people who responded. Population: The popula-
tion presumably consisted of all adults at least 18 years of age.
The sample is not likely to be representative of the population
because those with strong opinions about abortion are more
likely to respond.
33. With no natural starting point, temperatures are at the interval
level of measurement, so ratios such as “twice” are meaningless.
35. Either ordinal or interval are reasonable answers, but ordinal
makes more sense because differences between values are not
likely to be meaningful. For example, the difference between a
food rated 1 and a food rated 2 is not likely to be the same as
the difference between a food rated 9 and a food rated 10.
Section 1-4
1. A voluntary response sample (or self-selected sample) is one in
which the respondents themselves decide whether to be in-
cluded. It is unsuitable because people with special interests are
more likely to respond, so that the sample is likely to be biased.
3. No, because a correlation between two variables does not imply
that one is the cause of the other. No, a decrease in registered auto-
matic weapons will not necessarily result in a reduced murder rate.
5. There may be a relationship between studying and living longer,
but that does not mean that one causes the other. College
796 Appendix D

graduates are likely to have higher incomes and can afford better 11. Cluster
health care, so they are likely to live longer. 13. Stratified
7. Perhaps police officers are more inclined to stop minorities than 15. Systematic
whites, so minorities are given more tickets. 17. Random
9. When people are asked whether they wash their hands in a pub- 19. Cluster
lic restroom, they are inclined to answer yes, because that re- 21. Yes; yes. All pills have the same chance of being chosen. All
sponse reflects better personal hygiene and a better positive samples of size 30 have the same chance of being chosen.
image. The 82% rate is likely to be more accurate because it is 23. The sample is random, because all subjects have the same
based on actual behavior instead of what people said. chance of being selected. It is not a simple random sample, be-
11. Because the study was funded by a candy company and the cause some samples are not possible, such as a sample that in-
Chocolate Manufacturers Association, there is a real possibility cludes subjects from precincts that were not selected.
that researchers were somehow encouraged to obtain results fa- 25. No, no. Not everyone has the same chance of being selected,
vorable to the consumption of chocolate. and some samples have no chance of being chosen.
13. The voluntary response sample cannot be used to conclude any- 27. Retrospective
thing about the general population. In fact, Ted Koppel re- 29. Cross-sectional
ported that a “scientific” poll of 500 people showed that 72% of 31. Blinding is a method whereby a subject (or researcher) in an ex-
us want the United Nations to stay in the United States. In this periment doesn’t know whether the subject is given the treatment
poll of 500 people, respondents were randomly selected by the or a placebo. It is important to use blinding so that results are
pollster so that the results are much more likely to reflect the not distorted because of a placebo effect, whereby subjects think
true opinion of the general population. that they experience improvements because they were treated.
15. Motorcyclists who were killed 33. Answers vary.
17. No. States with small population sizes were counted equally
with large states, but there should be some compensation Chapter 1: Statistical Literacy
whereby incomes from small population states should be and Critical Thinking
counted less. The population sizes should be taken into account.
1. No, because the sample is a voluntary response sample.
19. Nothing. The sample is a voluntary response sample, so it is very
2. a. Quantitative
possible that those with strong interests in the topics replied. The
b. Continuous
sample is not necessarily representative of the population of all
c. Observational study
women.
d. Nominal
21. a. 62.5%
e. Ratio
b. 0.234
3. The subjects must be randomly selected in such a way that all
c. 185
samples of the same size have the same chance of being selected.
d. 12.7%
4. No. The results should be weighted to account for the different
23. a. 360
numbers of workers in the different states.
b. 44%
25. a. 1921
b. 47% Chapter 1 Quick Quiz
c. No. Because the sample is a voluntary-response sample, people 1. True
with strong interest in the topic are more likely to respond, so 2. Continuous data
the sample is not likely to be representative of the population. 3. False
27. If foreign investment fell by 100%, it would be totally elimi- 4. False
nated, so it is not possible for it to fall by more than 100%. 5. Experiment
29. All percentages of success should be multiples of 5. The given 6. False
percentages cannot be correct. 7. Population
8. Quantitative
Section 1-5 9. Nominal
10. No
1. With a random sample, each individual has the same chance of
being selected. With a simple random sample, all samples of the
same size have the same chance of being selected. Chapter 1 Review Exercises
3. No, the convenience sample of friends is not likely to represent 1. a. It uses a voluntary response sample, and those with special
the adult population, no matter how a sample is selected from interests are more likely to respond, so it is very possible that
the friends. The sample is likely to represent only the population the sample is not representative of the population.
of friends. b. Because the statement refers to 72% of all Americans, it is a
5. Observational study parameter (but it is probably based on a 72% rate from the
7. Experiment sample, and the sample percentage is a statistic).
9. Convenience c. Observational study
 Appendix D 797

2. No, it is very possible that those who agree to respond constitute 10. 30,517,578,125 (Most calculators won’t show the last few digits,
a population that is fundamentally different from those who so a result such as 30,517,578,000 is acceptable.)
refuse to respond. When encountering someone who refuses to 11. 31,381,059,609 (Most calculators won’t show the last few digits,
respond, the polling organization should make further attempts so a result such as 31,381,059,000 is acceptable.)
to gain cooperation. They might call back again with a strategy 12. 0.000244140625
designed to gain cooperation.
3. a. Ratio
Chapter 2 Answers
b. Nominal
c. Interval Section 2-2
d. Ordinal 1. No. For each class, the frequency tells us how many values fall
4. a. Nominal within the given range of values, but there is no way to deter-
b. Ratio mine the exact weights represented in the class.
c. Ordinal 3. No. The table does not show the distribution of a data set
d. Interval among all of several different categories. The sum of the fre-
5. a. Discrete quencies is 81%, not 100%. There must be a missing category,
b. Ratio such as “none,” and it is not clear whether respondents could se-
c. Stratified lect more than one category.
d. Statistic 5. Class width: 4. Class midpoints: 11.5, 15.5, 19.5, 23.5, 27.5.
e. The mailed responses would be a voluntary response Class boundaries: 9.5, 13.5, 17.5, 21.5, 25.5, 29.5.
sample; those with strong opinions about the topic would 7. Class width: 1.00. Class midpoints: 0.495, 1.495, 2.495, 3.495,
be more likely to respond, so the results would not be likely 4.495. Class boundaries: -0.005, 0.995, 1.995, 2.995, 3.995,
to reflect the true opinions of the population of all 4.995.
stockholders. 9. If the criteria are interpreted very strictly, the distribution
6. a. Systematic; representative does not appear to be normal (mainly due to the lack of sym-
b. Convenience; not representative metry between the frequencies of 0 and 7), but if the criteria
c. Cluster; representative are very loosely interpreted, the distribution does appear to be
d. Random; representative normal.
e. Stratified; not representative 11. The nonfiltered cigarettes have much more tar. The filtered cig-
7. a. 34% arettes are associated with lower amounts of tar, so the filters do
b. 818 appear to be effective.
8. a. Parameter Relative Relative
b. Discrete Frequency Frequency
c. 34,226,000 Tar (mg) (Nonfiltered) (Filtered)
9. a. If they have no fat at all, they have 100% less than any other
2–5 0% 8%
amount with fat, so the 125% figure cannot be correct.
6–9 0% 8%
b. 686
10–13 4% 24%
c. 28% 14–17 0% 60%
10. The Gallup poll used randomly selected respondents, but the 18–21 60% 0%
AOL poll used a voluntary-response sample. Respondents in 22–25 28% 0%
the AOL poll are more likely to participate if they have 26–29 8% 0%
strong feelings about the candidates, and this group is not
necessarily representative of the population. The results from
13. Tar (mg) in
the Gallup poll are more likely to reflect the true opinions of
Nonfiltered Cumulative
American voters. Cigarettes Frequency

Less than 14 1
Less than 18 1
Chapter 1 Cumulative Review Exercises
Less than 22 16
1. 1.256 mg Less than 26 23
2. 119.9 min Less than 30 25
3. 1.52
4. 5.31 15. Category Relative Frequency
5. 2401
6. 0.9581 Male Survivors 16.2%
7. 0.12 Males Who Died 62.8%
8. 0.346 Female Survivors 15.5%
Females Who Died 5.5%
9. 0.000016777216
798 Appendix D

17. Because there are disproportionately more 0s and 5s, it appears 27. Weight (g) Frequency
that the heights were reported instead of measured. Conse-
6.0000–6.0499 2
quently, it is likely that the results are not very accurate.
6.0500–6.0999 3
x Frequency 6.1000–6.1499 10
6.1500–6.1999 8
0 9 6.2000–6.2499 6
1 2 6.2500–6.2999 7
2 1 6.3000–6.3499 3
3 3 6.3500–6.3999 1
4 1
5 15
6 2 29. Blood Group Frequency
7 0 O 22
8 3 A 20
9 1 B 5
AB 3
19. Nicotine (mg) Frequency
31. An outlier can dramatically affect the frequency table.
1.0–1.1 14
1.2–1.3 4
1.4–1.5 3 Weight (lb) With Outlier Without Outlier
1.6–1.7 3
200–219 6 6
1.8–1.9 1
220–239 5 5
21. The distribution does not appear to have a normal distribution. 240–259 12 12
The frequencies do not start low, reach a maximum, then de- 260–279 36 36
280–299 87 87
crease. The frequencies do not appear to be symmetric.
300–319 28 28
Voltage (volts) Frequency 320–339 0
340–359 0
123.3–123.4 10 360–379 0
123.5–123.6 9 380–399 0
123.7–123.8 10 400–419 0
123.9–124.0 10 420–439 0
124.1–124.2 1 440–459 0
23. The frequency distribution appears to be consistent with the la- 460–479 0
480–499 0
bel of 3/4 in. The frequency distribution appears to be centered
500–519 1
around the length of 0.75 in.
Length (in.) Frequency

0.720–0.729 5 Section 2-3

0.730–0.739 10 1. It is easier to see the shape of the distribution of the data by ex-
0.740–0.749 11 amining the graph of the histogram than by examining the
0.750–0.759 17
numbers in a frequency distribution.
0.760–0.769 7
3. With a data set that is so small, the true nature of the distribu-
25. The distribution does not appear to be normal because the fre- tion cannot be seen with a histogram.
quencies are not symmetric about the maximum frequency. 5. 60; 20
7. 2500 miles; 42,500 miles
FICO Score Frequency 9. The digits of 0 and 5 occur disproportionately more often than the
others, so it appears that the heights were reported, not measured.
400–449 1
450–499 1
500–549 5
550–599 8
600–649 12
650–699 16
700–749 19
750–799 27
800–849 10
850–899 1
 Appendix D 799

11. 21. Pulse Rate

129.5
119.5
109.5
99.5
89.5
79.5
69.5
59.5
49.5

0%

0%

%
%
%
%

%
%

%
%
%

20

40
50
10
20

30
30

13. The distribution does not appear to have a normal distribution.
The frequencies do not start low, reach a maximum, then de-
crease. The frequencies do not appear to be symmetric.
10
50
40
Women Men
Section 2-4
1. The dotplot allows you to identify all of the original data values.
The dotplot is simpler, and easier to construct.
3. By using relative frequencies consisting of proportions or per-
centages, both data sets use comparable measures. If frequency
polygons are constructed with two samples having very different
numbers of elements, the comparison is made difficult because
the frequencies will be very different.
5. The dotplot suggests that the amounts have a distribution that is ap-
proximately normal, and it is centered around 150 millibecquerels.

15. The lengths appear to be centered around the length of 0.75 in.
The frequency distribution appears to be consistent with the
label of 3/4 in. 7.

9. The stemplot suggests that the weights of discarded plastic have

17. The distribution does not appear to be normal because the a distribution that is skewed to the right, but it is not too far
graph is not symmetric. from being a normal distribution.

0. 12356677888999
1. 11234444444445556678
2. 001111113334668888999
3. 0345
4. 36
5. 2

11. 59 of the weights are below 4 lb.

19.
800 Appendix D

13. 23.

25. Total numbers in both categories have increased due to increas-

15.
ing population, so the rates give better information about
trends. The marriage rate has remained somewhat steady, but
the divorce rate grew more during the period from 1900 to
1980, and it has declined slightly since then.

17.

27. The pulse rates of men appear to be generally lower than the
pulse rates of women.

Women Stem (tens) Men

5 666666
888884444000 6 00000004444444888
66666622222222 7 22222266
19. 88888000000 8 44448888
6 9 6
4 10
11
4 12

Section 2-5
21. In king-size cigarettes, there does not appear to be a relationship
between tar and carbon monoxide.
1. The illustration uses images of dollar bills, which are objects of
area, but the amounts of purchasing power are one-dimensional,
so the graph is misleading.
3. No. The graph should be created in a way that is fair and ob-
jective. The readers should be allowed to judge the informa-
tion on their own, instead of being manipulated by a mislead-
ing graph.
5. No. The weights are one-dimensional measurements, but the
graph uses areas, so the average weight of men appears to be
much larger than it actually is. (Based on the values given, men
weigh about 25% more than women, but by using areas, the
graph makes it appear that men weigh about 58% more than
women.)
7. No. By using areas, the graph creates the impression that the
average income for men is roughly twice that of women, but the
 Appendix D 801

actual values show that the average income for men is about 4. False
43% more than that of women. 5. Variation
6. 52, 52, 59
$60,000 7. Scatterplot
$50,000 8. True

Average Annual Income

9. Distribution of the data
$40,000 10. Pareto chart
$30,000
Chapter 2 Review Exercises
$20,000
1. The pulse rates of men are generally lower than those of women.
$10,000
Pulse Rate Frequency
0
Men Women 50–59 6
60–69 17
9. By starting the horizontal scale at 100 ft, the graph cuts off the
70–79 8
left portions of the bars so that perceptions are distorted. The 80–89 8
graph creates the false impression that the Acura RL has a brak- 90–99 1
ing distance that is more than twice that of the Volvo S80, but a
comparison of the actual numbers shows that the braking dis- 2. The distribution shape is roughly the same, but the pulse rates
tance of the Acura RL is roughly 40% more than that of the of males appear to be generally lower than those of females.
Volvo S80, not 100% more.

Acura RL
Honda Accord
Volvo S80
0
0
0
50
0

10
15
20

Braking Distance (feet)

11. The graph is misleading because it is essentially a pie chart that
is distorted by overlaying it atop the head of a stereotypical 3. The dotplot shows pulse rates that are generally lower than the
CEO. The graph could be better represented by a standard pie pulse rates of females.
chart, and an even better graph would be a bar graph with a
horizontal axis representing the successive categories of the age
groups listed in order from youngest to oldest. The vertical axis
should represent the percentages, and it should start at 0%.

Chapter 2 Statistical Literacy

and Critical Thinking 4. The pulse rates of males are generally lower than the pulse rates
1. The histogram is more effective because it provides a picture, of females.
which is much easier to understand than the table of numbers. 5 666666
2. The time-series graph would be better because it would show 6 00000004444444888
changes over time, whereas the histogram would hide the 7 22222266
changes over time. 8 44448888
3. By using images with heights and widths, the graph tends to cre- 9 6
ate a distorted perception. The number of men is roughly twice
5. There does not appear to be a relationship between weight and
the number of women, but the graph exaggerates that difference
braking distance.
by using objects with areas.
4. The bars of the histogram start relatively low, then they increase
to some maximum height, then they decrease. The histogram
should be symmetric with the left half being roughly a mirror
image of the right half.

Chapter 2 Quick Quiz

1. 10
2. -0.5 and 9.5
3. No
802 Appendix D

6. There is a trend that consists of a repeating cycle. 7. The data have a normal distribution if the histogram is approxi-
mately bell-shaped. Using a fairly strict interpretation of nor-
mality, the histogram does not represent data with a normal dis-
tribution.

7. The graph is misleading. The differences are exaggerated because

the vertical scale does not start at zero.
8. Statistic. In general, a statistic is a measurement describing some
characteristic of a sample, whereas a parameter is a numerical
10
Time for Acceleration from 0 to 60 mph

9 measurement describing some characteristic of a population.

8
7 Chapter 3 Answers
(in seconds)

6 Section 3-2
5
4 1. They use different approaches for providing a value (or values)
3 of the center or middle of a set of data values.
2 3. No. The price exactly in between the highest and lowest is the
1 midrange, not the median.
0 5. x = 53.3; median = 52.0; mode: none; midrange = 56.5. Us-
Volvo Audi VW BMW ing the mean of 53.3 words per page, the estimate of the total
XC-90 Q7 Passat 3 Series number of words is 77,765. Because the mean is based on a
small sample and because the numbers of defined words on
8. a. 25
pages appear to vary by a large amount, the estimate is not likely
b. 100 and 124
to be very accurate.
c. 99.5 and 124.5
7. x = $6412.2; median = $6374.0; mode: none; midrange =
d. No
$6664.0. The different measures of center do not differ by very
large amounts.
Chapter 2 Cumulative Review Exercises 9. x = $20.4 million; median = $14.0 million; mode: none;
1. Yes. All of the responses are summarized in the table, and the midrange = $23.2 million. Because the sample values are the
sum of the percentages is 100%. 10 highest, they give us almost no information about the salaries
2. Nominal, because the responses consist of “yes,” “no,” or “maybe.” of TV personalities in general. Such top 10 lists are not valuable
Those responses do not measure or count anything, and they can- for gaining insight into the larger population.
not be arranged in order. The responses consist of names only. 11. x = 217.3 hours; median = 235.0 hours; mode = 235.0
3. Yes: 884; no: 433; maybe: 416 hours; midrange = 190.5 hours. The duration time of 0 hours is
4. Voluntary response sample (or self-selected sample). The volun- very unusual. It could represent a flight that was aborted. (It actu-
tary response sample is not likely to be representative of the ally represents the duration time of the Challenger on the flight
population, because those with special interests or strong feel- that resulted in a catastrophic explosion shortly after takeoff.)
ings about the topic are more likely to respond, and they might 13. x = 2.4 mi/gal; median = 2.0 mi/gal; mode = 2 mi/gal;
be very different from the general population. midrange = 2.5 mi/gal. Because the amounts of decrease do
5. a. In a random sample, every subject has the same chance of be- not appear to vary much, there would not be a large error.
ing selected. 15. x = 259.6 sec; median = 246.0 sec; mode: 213 sec and 246
b. In a simple random sample of subjects, every sample with sec; midrange = 330.5 sec. The time of 448 sec is very different
the same number of subjects has the same chance of being from the others.
selected. 17. x = 6.5 years; median = 4.5 years; mode: 4 and 4.5; midrange =
c. Yes; no 9.5 years. It is common to earn a bachelor’s degree in four years,
6. a. 100 but the typical college student requires more than four years.
b. -0.5 and 99.5 19. x = 110.0; median = 97.5; mode: none; midrange = 192.5.
c. 0.275 or 27.5% Stricter federal bankruptcy laws went into effect after October,
d. Ratio so the bankruptcy filings increased up to the time of the new
e. Quantitative laws, then decreased substantially after that.

21. 30 Days in Advance: x = $272.6; median = $264.0. 1 Day in
Advance: x = $690.3; median = $614.0. There is a substantial
savings by purchasing the tickets 30 days in advance instead of 1
day in advance.
23. Nonfiltered: x = 1.26 mg; median = 1.10 mg. Filtered:
x = 0.92 mg; median = 1.00 mg. The filtered cigarettes do
appear to have less nicotine. It appears that cigarettes with filters
are associated with lower amounts of nicotine.
25. x = 98.20°F; median = 98.40°F. These results suggest that the
mean is less than 98.6°F.
27. Home: x = 123.66 volts; median = 123.70 volts. Generator:
x = 124.66 volts; median = 124.70 volts. UPS: x = 123.59
volts; median = 123.70 volts. The three different groups appear
to have about the same voltage.
29. x = 20.9 mg. The mean from the frequency table is close to the
mean of 21.1 mg for the original list of data values.
31. 46.8 mi/h, which is very close to the value found from the origi-
nal list of values. The speeds are well above the speed limit of
30 mi/h (probably because the police ticketed only those who
were traveling well above the posted speed limit).
33. 2.93; no
35. a. 642.51 micrograms
b. n - 1
37. x = 703.1; 10% trimmed mean: 709.7; 20% trimmed mean:
713.7. The results are not dramatically different. The results
appear to show a trend of increasing values as the percentage
of trim increases, so the distribution of the data appears to be
skewed to the left.
39. 1.056; 5.6%. The result is not the same as the mean of 10%,
5%, and 2%, which is 5.7%.
41. 20.9 mg; the value of 20.0 mg is better because it is based on
the original data and does not involve interpolation.
Section 3-3
1. Variation is a general descriptive term that refers to the amount
of dispersion or spread among the data values, but the variance
refers specifically to the square of the standard deviation.
3. The incomes of adults from the general population should have
more variation, because the statistics teachers are a much more ho-
mogeneous group with incomes that are likely to be closer together.
5. Range = 45.0 words; s 2 = 245.1 words2; s = 15.7 words. The
numbers of words appear to have much variation, so we do not
have too much confidence in the accuracy of the estimate.
7. Range = $4774.0; s 2 = 3,712,571.7 dollars2; s = $1926.8.
Damage of $10,000 is not unusual because it is within two stan-
dard deviations of the mean.
9. Range = $29.6 million; s 2 = 148.1 (million dollars)2;
s = $12.2 million. Because the sample values are the 10 high-
est, they give us almost no information about the variation of
salaries of TV personalities in general.
11. Range = 381.0 hours; s 2 = 11,219.1 hours2;s = 105.9 hours.
The duration time of 0 hours is unusual because it is more than
two standard deviations below the mean.
13. Range = 3.0 mi/gal; s 2 = 0.6 (mi/gal)2; s = 0.7 mi/gal. The
decrease of 4 mi/gal is unusual because it is more than two stan-
dard deviations above the mean.
Appendix D 803
15. Range = 235.0 sec; s 2 = 2975.6 sec2; s = 54.5 sec. Yes, it
changes from 54.5 sec to 22.0 sec.
17. Range = 11.0 years; s 2 = 12.3 year2; s = 3.5 years. No, be-
cause 12 years is within two standard deviations of the mean.
19. Range = 357.0; s 2 = 8307.3; s = 91.1. The only unusual
value is 371.
21. 30 Days in Advance: 8.5% (Tech: 8.6%). 1 Day in Advance:
33.8% (Tech: 33.7%). The costs for the tickets purchased 30 days
in advance have much less variation than the costs for the tick-
ets purchased 1 day in advance. There is a substantial savings
when purchasing the tickets 30 days in advance instead of
1 day in advance.
23. Nonfiltered: 18.3% (Tech: 18.5%). Filtered: 27.2% (Tech:
27.1%). The nonfiltered cigarettes have less variation.
25. Range = 3.10°F; s 2 = 0.39 (degrees Fahrenheit)2; s = 0.62°F.
27. Home: Range = 0.90 volt; s 2 = 0.06 volt2; s = 0.24 volt.
Generator: Range = 1.20 volt; s 2 = 0.08 volt2; s = 0.29 volt.
UPS: Range = 1.00 volt; s 2 = 0.09 volt 2; s = 0.31 volt. The
three different groups appear to have about the same amount of
variation, which is very small in each case. (The coefficients of
variation are 0.19%, 0.23%, and 0.25%.)
29. s = 3.2 mg. The standard deviation is the same as for the origi-
nal list of data values.
31. s L 1.5 years
33. a. 68%
b. 95%
35. At least 75% of the heights are within 2 standard deviations of
the mean. The minimum is 147 cm and the maximum is 175 cm.
37. a. 32.7
b. 32.7
c. 16.3
d. Part (b), because repeated samples result in variances that target
the same value (32.7) as the population variance. Use n - 1.
e. No. The mean of the sample variances (32.7) equals the pop-
ulation variance (32.7), but the mean of the sample standard
deviations (4.1) does not equal the mean of the population
standard deviation (5.7).
Section 3-4
1. Below the mean. Her age is 0.61 standard deviations below the
mean.
3. 0 hours is the length of the shortest flight, the first quartile Q 1
is 166 hours, the second quartile Q 2 (or median) is 215 hours, the
third quartile Q 3 is 269 hours, and the maximum is 423 hours.
5. a. 25.2 years
b. 2.23
c. 2.23
d. Unusual
7. a. 135 sec
b. 3.71
c. -3.71
d. Unusual
9. a. 4.52; unusual
b. -2.10; unusual
c. -1.97; usual
11. 2.67; unusual
804 Appendix D

13. The score of 1840 converts to z = 0.99, and the score of 26.0 2. Because the ZIP codes are not measurements or counts of any-
converts to z = 1.02, so the score of 26.0 is relatively better be- thing and there isn’t a consistent ordering, the ZIP codes are at
cause it has the larger z score. the nominal level of measurement, so the mean and standard
15. 38th percentile deviation are meaningless in this context.
17. 50th percentile 3. The mean, standard deviation, and range will change by a fairly
19. 39 large amount, but the median will not change very much.
21. 60 4. Trend or pattern over time
23. 53.5
25. 42 Chapter 3 Quick Quiz
27.
1. 4.0 cm
2. 3.0 cm
3. 2 cm
4. 25.0 ft2
29.
5. -3
6. Variation
7. Sample: s; population: s
8. Sample: x; population: m
31. The weights of diet Coke appear to be generally less than those of
9. 75
regular Coke, probably due to the sugar in cans of regular Coke.
10. True

Chapter 3 Review Exercises

1. a. 19.3 oz
b. 19.5 oz
c. 20 oz
d. 19.0 oz
e. 4.0 oz
33. The weights of the pre-1964 quarters appear to be considerably f. 1.3 oz
greater than those of the post-1964 quarters. g. 1.8 oz2
h. 18 oz
i. 20 oz
2.

3. a. 918.8 mm
b. 928.0 mm
c. 936 mm
35. Outliers: 27.0 cm, 31.1 cm, 32.1cm, 48.6 cm. d. 907.0 mm
e. 58.0 mm
f. 23.0 mm
g. 530.9 mm2
h. 923 mm
i. 934 mm
4. The z score is -1.77. Because the z score is between -2
and 2, the sitting height of 878 mm is not unusual.
5. Because the median and quartiles are located to the far right
37. The result is 33.75 with interpolation and 35 without interpola- portion of the boxplot, the boxplot suggests that the data are
tion. These two results are not too far apart. from a population with a skewed distribution, not a distribution
that is normal (or bell-shaped).
Chapter 3 Statistical Literacy
and Critical Thinking
1. No. In addition to having a mean of 12 oz, the amounts of cola
must not vary too much. If they have a mean of 12 oz but they
vary too much, some of the cans will be underfilled by unac- 6. The score of 1030 converts to z = -1.50, and the score of 14.0
ceptable amounts, while others will be overfilled, so the produc- converts to z = -1.48, so the score of 14.0 is relatively better
tion process would need to be fixed. because it has the larger z score.
 Appendix D 805

7. a. Answer varies, but an estimate around 4 years is reasonable.
b. Answer varies, but using a minimum of 0 year and a maxi-
mum of 20 years results in s = 5.0 years.
8. a. Answer varies, but assuming that the times are between 15 sec
and 90 sec, a mean of about 50 sec or 60 sec is reasonable.
b. Answer varies, but assuming that the times are between 15
sec and 90 sec, a standard deviation around 18 sec or 20 sec
is reasonable.
9. Minimum: 1110 mm; maximum: 1314 mm. Because the design
should accommodate most women pilots, the grip reach of
1110 mm is more relevant. No instruments or controls should
be located more than 1110 mm from a sitting adult woman.
10. Minimum: 83.7 cm; maximum: 111.3 cm. The height of 87.8 cm
is not unusual, so the physician should not be concerned.
8. A simple random sample of n subjects is selected in such a way
that every possible sample of the same size n has the same chance
of being chosen. A voluntary response sample is one in which
the subjects decide themselves whether to be included in the
study. A simple random sample is generally better than a
voluntary response sample.
9. In an observational study, we observe and measure specific char-
acteristics, but we don’t attempt to modify the subjects being
studied. In an experiment, we apply some treatment and then
proceed to observe its effects on the subjects.
10. By not starting the vertical scale at 0, the graph exaggerates the
differences among the frequencies.

Chapter 3 Cumulative Review Exercises Chapter 4 Answers

1. a. Continuous Section 4-2
b. Ratio 1. The chance of an injury from sports or recreation equipment is
2. very small. Such injuries occur roughly one time in 500. Be-
Sitting Height (mm) Frequency
cause the probability is so small, such an injury is unusual.
870–879 1 3. A denotes the complement of event A, meaning that event A does
880–889 1 not occur. If P (A ) = 0.995, P (A) = 0.005. If P (A) = 0.995,
890–899 0 A is unusual because it is very unlikely to occur.
900–909 0 5. 4>21 or 0.190
910–919 0 7. 1>2 or 0.5
920–929 3 9. 6>36 or 1>6 or 0.167
930–939 4 11. 0
13. 3:1, 5>2, -0.5, 321>123
3. The distribution appears to be skewed. 15. a. 3>8
b. 3>8
c. 1>8
17. a. 98
b. 41
c. 41>98
d. 0.418
19. 0.153. The probability of this error is high, so the test is not
very accurate.
4. 21. 0.16. No, the probability is too far from the value of 0.5 that
would correspond to equal chances for men and women. It
appears that more men are elected as Senators.
5. 87 8 23. 0.000000966. No, the probability of being struck is much greater
88 0 in an open golf course during a thunderstorm. He or she should
89 seek shelter.
90
25. a. 1>365
91
b. Yes
92 348
93 0466 c. He already knew.
d. 0
6. a. The mode is most appropriate because it identifies the 27. 0.0571. No, a firmware malfunction is not unusual among
most common choice. The other measures of center do malfunctions.
not apply to data at the nominal level of measurement. 29. 0.0124; yes, it is unusual. There might be more reluctance to
b. Convenience sentence women to death, but very serious crimes are commit-
c. Cluster ted by many more men than women.
d. The standard deviation should be lowered. 31. 0.0811; not unusual
7. No. Even though the sample is large, it is a voluntary re- 33. a. bb, bg, gb, gg
sponse sample, so the responses cannot be considered to be b. 1>4
representative of the population of the United States. c. 1>2
806 Appendix D

35. a. brown> brown, brown> blue, blue> brown, blue> blue 9. Dependent (same power supply)
b. 1>4 11. Independent
c. 3>4 13. 0.0221. Yes, it is unusual because the probability is very low,
37. 423:77 or roughly 5.5:1 or 11:2 such as less than 0.05.
39. a. $12.20 15. 0.109. The event is not unusual because its probability is not
b. 6.1:1 or roughly 6:1 very small, such as less than 0.05.
c. 443:57 or roughly 8:1 17. a. 0.152
d. $17.54 or roughly $18 b. 0.150
41. Relative risk: 0.939; odds ratio: 0.938; the probability of a 19. a. 0.0000130
headache with Nasonex (0.0124) is slightly less than the proba- b. 0.00000383
bility of a headache with the placebo (0.0132), so Nasonex does 21. a. Yes
not appear to pose a risk of headaches. b. 1>8 = 0.125
43. 1>4 c. No, it would be better to be prepared.
23. 0.959 = 0.630. With nine independent polls, there is a reason-
Section 4-3 ably good chance (0.370) that at least one of them is not within
the claimed margins of error.
1. The two events cannot happen at the same time.
25. 1>8 = 0.125. The likelihood of getting 3 girls in 3 births by
3. P (A and B) denotes the probability that on a single trial, events
chance is high, so the results do not indicate that the method is
A and B both occur at the same time.
effective.
5. Not disjoint
27. a. 0.1
7. Disjoint
b. 0.01
9. Not disjoint
c. 0.99
11. Not disjoint
d. Yes, the probability of not being awakened drops from 1
13. P (M ) denotes the probability of getting a STATDISK user who
chance in 10 to one chance in 100.
is not using a Macintosh computer, and P (M ) = 0.95.
29. a. 0.849
15. 0.21
b. 0.0169. Yes, because there is a very small chance that all 100
17. 89>98 or 0.908
tires are good.
19. 32>98 or 16>49 or 0.327
31. a. 0.9999
21. 512>839 or 0.610
b. 0.9801
23. 615>839 or 0.733
c. The series arrangement provides better protection.
25. 713>839 or 0.850
33. 0.0192
27. 156>1205 = 0.129. Yes. A high refusal rate results in a sample
that is not necessarily representative of the population, because
Section 4-5
those who refuse may well constitute a particular group with
opinions different from others. 1. The exact number of defects is 1 or 2 or 3 or 4 or 5 or 6 or 7 or
29. 1060>1205 = 0.880 8 or 9 or 10.
31. 1102>1205 = 0.915 3. No, the two outcomes are not equally likely. The actual survival
33. a. 300 rate was not considered, but it should have been.
b. 178 5. All 15 players test negative.
c. 178>300 = 0.593 7. At least one of the four males has the X-linked recessive gene.
35. 0.603 9. 0.984. Yes.
37. 27>300 = 0.090. With an error rate of 0.090 (or 9%), the test 11. 0.590. There is a good chance of passing with guesses, but it
does not appear to be highly accurate. isn’t reasonable to expect to pass with guesses.
39. 3>4 = 0.75 13. 0.5. No.
41. P (A or B) = P (A ) + P (B ) - 2P (A and B) 15. 0.179. The four cars are not randomly selected. They are driven
43. a. 1 - P (A ) - P (B ) + P (A and B ) by people in the same family, and they are more likely to have
b. 1 - P (A and B ) similar driving habits, so the probability might not be correct.
c. Different 17. 0.944. Yes, the probability is quite high, so they can be reasonably
confident of getting at least one offspring with vestigial wings.
19. 15>47 = 0.319. This is the probability of unfairly indicating
Section 4-4 that the subject lied when the subject did not lie.
1. Answers vary. 21. 9>41 = 0.220. The results are different.
3. The events are dependent. Because different numbers are used, 23. a. 1>3 or 0.333
the selections are made without replacement, so the probability b. 0.5
of an adult being selected changes as the selections are made. 25. 0.5
5. Independent 27. 0.999. Yes. With only one alarm clock, there is a 0.10 chance of
7. Dependent the alarm clock not working, but with three alarm clocks there
 Appendix D 807

is only a 0.001 chance that they all fail. The likelihood of a 19. 27,907,200
functioning alarm clock increases dramatically with three alarm 21. 10
clocks. The results change if the alarm clocks run on electricity, 23. 256; yes, because the typical keyboard has about 94 characters.
because they are no longer independent. The student should use 25. 120 ways; JUMBO; 1>120
battery-powered alarm clocks to maintain independence. 27. a. 7920
29. 0.271 b. 330
31. a. 0.431 29. a. 16,384
b. 0.569 b. 14
33. 1>12; 35 c. 14>16,384 or 0.000854
d. Yes, because the probability is so small and is so far away
from the 7 girls> 7 boys that is typical.
Section 4-6
31. 2>252 or 1>128. Yes, if everyone treated is of one sex while
1. No. The generated numbers between 2 and 12 would be equally everyone in the placebo group is of the opposite sex, you would
likely, but they are not equally likely with real dice. not know if different reactions are due to the treatment or sex.
3. Yes, it does. Each of the 365 birthdays has the same chance of 33. 1>146,107,962
being selected, and the cards are replaced, so it is possible to 35. 144
select the same birthday more than once. 37. 2,095,681,645,538 (about 2 trillion)
5. Randomly generate 50 random numbers, with each number 39. a. Calculator: 3.0414093 * 1064; approximation:
between 1 and 100. Consider the numbers 1 through 95 to be 3.0363452 * 1064
adults who recognize the brand name of McDonald’s, while the b. 615
numbers 96 through 100 represent adults who do not recognize 41. 293
McDonald’s.
7. Randomly generate a number between 1 and 1000. Consider an
Chapter 4 Statistical Literacy
outcome of 1 through 528 to be a free throw that is made, and
and Critical Thinking
consider an outcome of 529 through 1000 to be a free throw
that is missed. 1. The results are very unlikely to occur by chance. The results
9. a. Answer varies. would occur an average of only 4 times in 1000.
b. It would be unusual to randomly select 50 consumers and 2. No, because they continue to have a common power source. If
find that about half of them recognize McDonald’s. the home loses all of its power, all of the smoke detectors would
11. a. Answer varies. be inoperable.
b. It would be unusual for O’Neal to make all of five free 3. No. The calculation assumes that the events are independent,
throws in a game. but the same detective is handling both cases, and both cases are
13. The exact answer is 0.5, so an answer from a simulation should likely to be in the same general location, so they are not inde-
be around 0.5. Such runs are not unusual. pendent. The probability of clearing both cases depends largely
15. For the simulation, generate 152 random numbers between 0 on her commitment, ability, resources, and support.
and 1 inclusive. Sum the results to find the number of boys. 4. No. The principle of independence applies here. Based on the
Repeating the simulation should show that a result of 127 boys way that lottery numbers are drawn, past results have no effect
is unusual, suggesting that the YSORT method is effective. on the probability of future outcomes, so there is no way to pre-
17. With switching, P ( win ) = 2>3. With sticking, P ( win ) = 1>3. dict which numbers are more likely or less likely to occur.
19. No; no
Chapter 4 Quick Quiz
Section 4-7 1. 0.30
2. 0.49
1. With permutations, different orderings of the same items are
3. Answer varies, but an answer such as 0.01 or lower is reasonable.
counted separately, but they are not counted separately with
4. No, because the results could have easily occurred by chance.
combinations.
5. 0.6
3. Because the order of the first three finishers does make a differ-
6. 427>586 = 0.729
ence, the trifecta involves permutations.
7. 572>586 = 0.976
5. 120
8. 0.00161
7. 1326
9. 10>586 = 0.0171
9. 15,120
10. 10>24 = 0.417
11. 5,245,786
13. 1>25,827,165
15. 1>376,992 Chapter 4 Review Exercises
17. a. 1>1,000,000,000 1. 576>3562 = 0.162
b. 1>100,000. No, but if they could keep trying with more at- 2. 752>3562 = 0.211
tempts, they could eventually get your social security number. 3. 1232>3562 = 0.346
808 Appendix D

4. 3466>3562 = 0.973 A simple random sample is definitely better than a voluntary

5. 96>3562 = 0.0270 response sample.
6. 2330>3562 = 0.654 3. a. Regular: 371.5 g; diet: 355.5 g. Diet Coke appears to weigh
7. 0.0445 (not 0.0446) substantially less.
8. 0.0261 b. Regular: 371.5 g; diet: 356.5 g. Diet Coke appears to weigh
9. 480>576 = 0.833 substantially less.
10. 656>752 = 0.872 c. Regular: 1.5 g; diet: 2.4 g. The weights of diet Coke appear
11. Based on data from J.D. Power and Associates, 16.7% of car colors to vary more.
are black, so any estimate between 0.05 and 0.25 is reasonable. d. Regular: 2.3 g 2; diet: 5.9 g 2
12. a. 0.65 e. No, diet Coke appears to weigh considerably less than regular
b. 0.0150 Coke.
c. Yes. Because the probability is so small (0.0150), it is not 4. a. Yes, because 38 is more than two standard deviations below
likely to happen by chance. the mean.
13. a. 1>365 b. 1>1024 or 0.000977. Because the probability is so small, the
b. 31>365 outcome of a perfect score is unusual.
c. Answer varies, but it is probably small, such as 0.02. 5. a. Convenience sample
d. Yes b. If the students at the college are mostly from a surrounding re-
14. a. 0.000152 gion that includes a large proportion of one ethnic group, the re-
b. 0.0000000231 sults will not reflect the general population of the United States.
c. 0.999696 c. 0.75
15. a. 0.60 d. 0.64
b. No. Because the poll uses a voluntary response sample, its 6. 14,348,907
results apply only to those who chose to respond. The results
do not necessarily apply to the greater population of all
Americans. Chapter 5 Answers
16. 0.0315. It is not likely that such combined samples test positive. Section 5-2
17. 0.8530 = 0.00763. Because the probability of getting 30 De- 1. A random variable is a variable that has a single numerical value,
mocrats by chance is so small, chance is not a reasonable expla- determined by chance, for each outcome of a procedure. The
nation, so it appears that the pollster is lying. random variable x is the number of winning tickets in 52 weeks,
18. a. 0.998856 and its possible values are 0, 1, 2, Á , 52.
b. 0.998 3. The random variable x represents all possible events in the entire
c. 0.999 sample space, so we are certain (with probability 1) that one of
d. Males are more likely to die as a result of military engagement, the events will occur.
car crashes, violent crimes, and other such high-risk activities. 5. a. Discrete
19. 501,942; 1>501,942. It is unusual for a particular person to buy b. Continuous
one ticket and win, but it is not unusual for anyone to win, be- c. Continuous
cause wins often do occur. d. Discrete
20. 10,000,000,000,000 e. Continuous
7. Probability distribution with m = 1.5 and s = 0.9.
9. Not a probability distribution because ©P (x) = 0.984 Z 1.
Chapter 4 Cumulative Review Exercises 11. Probability distribution with m = 2.8 and s = 1.3.
1. a. 19.3 oz 13. m = 6.0 peas; s = 1.2 peas
b. 19.5 oz 15. a. 0.267
c. 1.1 oz b. 0.367
d. 1.2 oz 2 c. Part (b)
e. No. The mean is well below the desired value of 21 oz, and d. No, because the probability of 7 or more peas with green
all of the steaks weigh less than 21 oz before cooking. pods is high. (0.367 is greater than 0.05).
2. a. 58.2% 17. a. Yes
b. 0.582 b. m = 5.8 games and s = 1.1 games.
c. It is a voluntary response sample (or self-selected sample), c. No, because the probability of 4 games is not small (such as
and it is not suitable for making conclusions about any larger less than 0.05), the event of a four-game sweep is not unusual.
population; it is useful for describing the opinions of only 19. a. Yes
those who choose to respond. b. m = 0.4 bumper sticker and s = 1.3 bumper stickers.
d. The poll had 5226 respondents. A simple random sample of c. -2.2 to 3.0
5226 subjects would be selected in such a way that all sam- d. No, because the probability of more than 1 bumper sticker is
ples of 5226 subjects have the same chance of being selected. 0.093, which is not small.
 Appendix D 809

21. m = 1.5; s = 0.9. It is not unusual to get 3 girls, because the 33. 0.244. It is not unusual because the probability of 0.244 is not
probability of 3 girls is high (1> 8), indicating that 3 girls could very small, such as less than 0.05.
easily occur by chance. 35. a. 0.882
23. m = 4.5; s = 2.9. The probability histogram is flat. b. Yes, because the probability of 7 or fewer graduates is small
25. a. 1000 (less than 0.05).
b. 1> 1000 or 0.001 37. a. 0.00695
c. $249.50 b. 0.993
d. -25¢ c. 0.0461
e. Because both games have the same expected value, neither d. Yes. With a 22 share, there is a very small chance of getting at
bet is better than the other. most 1 household tuned to NBC Sunday Night Football, so it
27. a. -39¢ appears that the share is not 22.
b. The bet on the number 13 is better because its expected value 39. a. 0.209
of -26¢ is greater than the expected value of -39¢ for the b. Unlike part (a) the 20 selected subjects are more than 5% of
other bet. the 320 subjects available, so independence cannot be assumed
29. a. -$161 and $99,839 by the 5% guideline. The independence requirement for the
b. -$21 binomial probability formula is not satisfied.
c. Yes. The expected value for the insurance company is $21, 41. 0.662. The probability shows that about 2> 3 of all shipments
which indicates that the company can expect to make an av- will be accepted.
erage of $21 for each such policy. 43. 0.000139. The very low probability value suggests that the low
31. The A bonds are better because the expected value is $49.40, number of hired women is not the result of chance. The low
which is higher than the expected value of $26 for the B bonds. probability value does appear to support the charge of discrimi-
She should select the A bond because the expected value is posi- nation based on gender.
tive, which indicates a likely gain. 45. 0.030. The result is unusual, but with 580 peas, any individual
33. Put 1, 2, 3, 4, 5, 6 on one die and put 0, 0, 0, 6, 6, 6 on the number is unusual. The result does not suggest that 0.75 is wrong.
other die. The probability value of 0.75 would appear to be wrong if 428
peas with green pods is an unusually low number, which it is not.
Section 5-3 47. a. 0.0000000222
b. 0.00000706
1. The given event describes only one way in which 2 people have
c. 0.0104
blue eyes among 5 randomly selected people, but there are other
d. 0.510
ways of getting 2 people with blue eyes among 5 people and their
corresponding probabilities must also be included in the result.
3. Yes. Although the selections are not independent, they can be Section 5-4
treated as being independent by applying the 5% guideline. 1. Yes. Because q = 1 - p, the two expressions are equivalent in
Only 30 people are randomly selected from the group of 1236, the sense that they will always give the same results.
which is less than 5%. 3. 59.3 people2 (or 58.7 people2 if unrounded results are used)
5. Binomial 5. m = 10.0; s = 2.8; minimum = 4.4; maximum = 15.6
7. Not binomial; there are more than two possible outcomes. 7. m = 144.0; s = 8.7; minimum = 126.6; maximum = 161.4
9. Not binomial. Because the Senators are selected without replace- 9. a. m = 37.5 and s = 4.3
ment, the events are not independent, and they cannot be b. It would not be unusual to pass by getting at least 45 correct an-
treated as being independent. swers, because the range of usual numbers of correct answers is
11. Binomial. Although the events are not independent, they can be from 28.9 to 46.1, and 45 is within that range of usual values.
treated as being independent by applying the 5% guideline. 11. a. m = 16.0 and s = 3.7.
13. a. 0.128 b. The result of 19 green M&Ms is not unusual because it is
b. WWC, WCW, CWW; 0.128 for each within the range of usual values, which is from 8.6 to 23.4.
c. 0.384 The claimed rate of 16% does not appear to be wrong.
15. 0.420 13. a. m = 287.0 and s = 12.0
17. 0+ b. The result of 525 girls is unusual because it is outside of the
19. 0.075 range of usual values, which is from 263 to 311. The results
21. 0.232 suggest that the XSORT method is effective.
23. 0.182 15. a. m = 160.0 and s = 8.9
25. 0.950; yes b. 142.2 to 177.8
27. 0.0185. Yes, with a probability that is small (such as less than c. 250, which is unusual because it is outside of the range of
0.05), it is unusual. usual values given in part (b). The results suggest that the
29. 0.387; 0.613 headline is justified.
31. 0.001 or 0.000609. It is unusual because the probability is very d. Because the sample of 320 respondents is a voluntary
small, such as less than 0.05. response sample, all of the results are very questionable.
810 Appendix D

17. a. m = 611.2, s = 15.4 4. No. There are many discrete probability distributions that don’t
b. No, because the range of usual values is 580.4 to 642.0. The satisfy the requirements of a binomial distribution or a Poisson
result of 701 is unlikely to occur because it is outside of the distribution.
range of usual values.
c. No, it appears that substantially more people say that they Chapter 5 Quick Quiz
voted than the number of people who actually did vote.
1. No
19. a. m = 10.2 and s = 3.2
2. 0.7
b. 30 is unusual because it is outside of the range of usual val-
3. 200
ues, which is from 3.8 to 16.6.
4. 10
c. Chantrix does appear to be the cause of some nausea, but the
5. Yes
nausea rate with Chantrix is still quite low, so nausea appears
6. Yes
to be an adverse reaction that does not occur very often.
7. 0.5904 or 0.590
21. m = 3.0 and s = 1.3 (not 1.5)
8. 0.0016
9. 0.1792
Section 5-5 10. Yes
1. The random variable x is the number of occurrences of an event
over some interval, the occurrences are random, the occurrences Chapter 5 Review Exercises
are independent of each other, and the occurrences are uni- 1. x p
formly distributed over the interval.
3. The probability is so small that, for all practical purposes, we 0 0.004
can consider it to be zero. 1 0.031
5. 0.180 2 0.109
7. 0.0332 3 0.219
9. a. 0.0970
4 0.273
5 0.219
b. 0.000141
6 0.109
c. Yes, because the probability is very small, such as less than 0.05. 7 0.031
11. a. 62.2 8 0.004
b. 0.0155 (0.0156 using rounded mean)
13. a. 0.728 2. m = 4.0 and s = 1.4. Usual values are from 1.2 to 6.8. The
b. 0.231 occurrence of 8 deaths in the week before Thanksgiving is un-
c. 0.0368 usual because it is outside of the range of usual values (or be-
d. 0.00389 cause its probability of 0.004 is so low).
e. 0.000309 3. a. 0.0370
Using the computed probabilities, the expected frequencies are b. Yes, because the probability of 0.0370 is very small.
266, 84, 13, 1.4, and 0.1, and they agree quite well with the ac- c. No. Determining whether 14 is an unusually high number of
tual frequencies. deaths should be based on the probability of 14 or more
15. a. 2.24 deaths, not the probability of exactly 14 deaths (or the range
b. 0.811 (or 0.812 if binomial is used). No, the probability of a of usual values could be used).
profit is high, but not high enough to be “almost sure.” 4. $315,075. Because the offer is well below her expected value,
17. The Poisson probabilities are not close enough to the binomial she should continue the game (although the guaranteed prize of
probabilities. For example, P (6) = 0.205 with the binomial $193,000 had considerable appeal). (She accepted the offer of
distribution, but P (6) = 0.146 with the Poisson distribution. $193,000, but she would have won $500,000 if she continued
the game and refused all further offers.)
5. a. 1.2¢
Chapter 5 Statistical Literacy b. 1.2¢ minus cost of stamp.
and Critical Thinking 6. No, because ©P (x) = 0.9686, but the sum should be 1. (There
1. A random variable is a variable that has a single numerical out- is a little leeway allowed for rounding errors, but the sum of
come, determined by chance, for each outcome of a procedure. 0.9686 is too far below 1.)
It is possible for a discrete random variable to have an infinite 7. a. 1> 10,000 or 0.0001
number of values. b. x P (x)
2. A discrete random variable has either a finite number of values or a
countable number of values, but a continuous random variable has -$1 0.9999
infinitely many values, and those values can be associated with mea- $4999 0.0001
surements on a continuous scale without gaps or interruptions. c. 0.0365
3. The sum of p and q is 1, or p + q = 1, or p = 1 - q, or d. 0.0352
q = 1 - p. e. - 50¢
 Appendix D 811

8. a. 0.00361 17. 0.0668

b. This company appears to be very different because the event of at 19. 0.8907
least four firings is so unlikely with a probability of only 0.00361. 21. 0.0132
9. a. 236.0 23. 0.9599
b. m = 236.0 and s = 12.8 25. 0.1498 (Tech: 0.1499)
c. 210.4 to 261.6 27. 0.1574 (Tech: 0.1573)
d. Yes, because 0 is well outside of the range of usual values. 29. 0.8593
10. a. 535>576 = 0.929 31. 0.9937 (Tech: 0.9938)
b. 0.395 33. 0.9999 (Tech: 0.9998)
c. 227.5 regions 35. 0.5000
d. The actual result of 229 is very close to the computed result 37. 68.26% (Tech: 68.27%)
of 227.5. 39. 99.74 %(Tech: 99.73%)
41. 1.645
Chapter 5 Cumulative Review Exercises 43. 1.28
45. 0.9500
1. a. $159.674
47. 0.0100
b. $142.94
49. 1.645
c. $102.60
51. -1.96, 1.96
d. $41.985
53. a. 95.44% (Tech: 95.45%)
e. 1762.700 dollars2
b. 31.74% (Tech: 31.73%)
f. $75.704 to $243.644
c. 5.00%
g. No, because all of the sample values are within the range
d. 99.74% (Tech: 99.73%)
of usual values.
e. 0.26% (Tech: 0.27%)
h. Ratio
55. a. 1.75
i. Discrete
b. -2.00
j. Convenience
c. 1.50
k. $21,396.32
d. 0.82
2. a. 0.026
e. 0.12
b. 0.992 (or 0.994)
c. m = 8.0 and s = 1.3.
d. 5.4 to 10.6
Section 6-3
3. a. m = 15.0, s = 3.7 1. The standard normal distribution has a mean of 0 and a stan-
b. Twelve positive cases falls within the range of usual values dard deviation of 1, but a nonstandard distribution has a differ-
(7.6 to 22.4), so 12 is not unusually low. Because 12 positive ent value for one or both of those parameters.
cases could easily occur with an ineffective program, we do not 3. The mean is 0 and the standard deviation is 1.
have enough justification to say that the program is effective. 5. 0.9082 (Tech: 0.9088)
4. a. 706>2223 = 0.318 7. 0.5899 (Tech: 0.5889)
b. 0.101 9. 103.8
c. 0.466 11. 75.3
5. No. The mean should take the state populations into account. 13. 0.8413
The 50 values should be weighted, with the state populations 15. 0.4972 (Tech: 0.4950)
used as weights. 17. 92.2 (Tech: 92.1)
19. 110.0 (Tech: 110.1)
21. a. 85.77% (Tech: 85.80%)
Chapter 6 Answers b. 99.96%
Section 6-2 c. The height is not adequate because 14% of adult men will
1. The word “normal” has a special meaning in statistics. It refers need to bend, so it would be better to have doorways with a
to a specific bell-shaped distribution that can be described by higher opening, but other design considerations probably
Formula 6-1. make that impractical.
3. The mean and standard deviation have the values of m = 0 d. 74.7 in. (Tech: 74.8 in.)
and s = 1. 23. a. 3.67% (Tech: 3.71%)
5. 0.5 b. 0.52%
7. 0.75 c. No. The percentage of eligible men is greater than the per-
9. 0.7734 centage of eligible women.
11. 0.6106 (Tech: 0.6107) 25. a. 98.74% (Tech: 98.75%). No, only about 1% of women are
13. 2.05 not eligible.
15. 1.24 b. Minimum: 57.8 in.; maximum: 68.7 in.
812 Appendix D

27. a. 4.09% 13. a. (56, 56), (56, 49), (56, 58), (56, 46), (49, 56), (49, 49),
b. 2630 g (49, 58), (49, 46), (58, 56), (58, 49), (58, 58), (58, 46),
c. Without a specific cutoff birth weight, there would be no (46, 56), (46, 49), (46, 58), (46, 46)
way to know if a baby is in the bottom 3% until all babies b. x Probability
have been born, but waiting would deny treatment for those
babies who need special treatment. 46 1>16
29. a. 0.01%; yes 47.5 2>16
b. 99.22°
49 1>16
51 2>16
31. a. 0.0038; either a very rare event has occurred or the husband
52 2>16
is not the father.
52.5 2>16
b. 242 days
53.5 2>16
33. a. x = 118.9, s = 10.5, histogram is roughly bell shaped.
56 1>16
b. 101.6, 136.2 57 2>16
35. a. The z scores are real numbers that have no units of 58 1>16
measurement.
b. m = 0; s = 1; distribution is normal. c. The mean of the population is 52.25 and the mean of the
37. a. 75; 5 sample means is also 52.25.
b. No, the conversion should also account for variation. d. The sample means target the population mean. Sample
c. 31.4, 27.6, 22.4, 18.6 means make good estimators of population means because
d. Part (c), because variation is included in the conversion. they target the value of the population mean instead of sys-
39. 0.0074 (Tech: 0.0070) tematically underestimating or overestimating it.
15. a. Same as Exercise 13 part (a).
Section 6-4 b. Range Probability
1. A sampling distribution of a statistic is the distribution of all 0 4>16
values of that statistic when all possible samples of the same size 2 2>16
are taken from the same population. 3 2>16
3. The mean of all of the sample means is equal to the population 7 2>16
mean. 9 2>16
5. The sample is not a simple random sample from the population 10 2>16
of all U.S. college students. It is likely that the students at New 12 2>16
York University do not accurately reflect the behavior of all U.S. c. The range of the population is 12, but the mean of the sam-
college students. ple ranges is 5.375. Those values are not equal.
7. Normal (approximately) d. The sample ranges do not target the population range of 12,
9. a. Sample Median Probability so sample ranges do not make good estimators of population
2 1>9 ranges.
2.5 2>9 17. Proportion
3 1>9 of Odd
6 2>9 Numbers Probability
6.5 2>9 0 4>9
10 1>9 0.5 4>9
b. The population median is 3, but the mean of the sample me- 1 1>9
dians is 5. The values are not equal.
c. The sample medians do not target the population median of
Yes. The proportion of odd numbers in the population is 1>3,
3, so sample medians do not make good estimators of popu-
and the sample proportions also have a mean of 1>3, so the sam-
lation medians.
ple proportions do target the value of the population propor-
11. a. s2 Probability tion. The sample proportion does make a good estimator of the
0 3>9 population proportion.
0.5 2>9 19. a. The proportions of 0, 0.5, 1 have the corresponding proba-
24.5 2>9 bilities of 1>16, 6>16, 9>16.
32 2>9 b. 0.75
c. Yes; yes
b. The population variance is 12.667 and the mean of the sam- 21. The formula yields P (0) = 0.25, P (0.5) = 0.5, and P (1) =
ple variances is also 12.667. The values are equal. 0.25, which does describe the sampling distribution of the
c. The sample variances do target the population variance of proportions. The formula is just a different way of presenting
12.667, so the sample variances do make good estimators of the same information in the table that describes the sampling
the population variance. distribution.

Section 6-5
1. It is the standard deviation of the sample means, which is de-
noted as sx or s> 2n.
3. The notation mx represents the mean of the population consist-
ing of all sample means. The notation sx represents the standard
deviation of the population consisting of all sample means.
5. a. 0.4761 (Tech: 0.4779)
b. 0.2912 (Tech: 0.2898)
7. a. 0.0316 (Tech: 0.0304)
b. 0.1227 (Tech: 0.1210)
c. If the original population is normally distributed, then the
distribution of sample means is normally distributed for any
sample size.
9. a. 0.3897 (Tech: 0.3913)
b. 0.1093 (0.1087)
c. Yes. The probability of exceeding the 3500 lb safe capacity is
0.1093, which is far too high.
11. a. 0.5675 (Tech: 0.5684)
b. 0.7257 (Tech: 0.7248)
c. The gondola is probably designed to safely carry a load some-
what greater than 2004 lb, but the operators would be wise
to avoid a load of 12 men, especially if they appear to have
high weights.
13. a. 0.0274 (Tech: 0.0272)
b. 0.0001
c. Because the original population is normally distributed, the
sampling distribution of sample means will be normally dis-
tributed for any sample size.
d. No, the mean can be less than 140 while individual values are
above 140.
15. a. 0.8577 (Tech: 0.8580)
b. 0.9999 (Tech: Most technologies provide a result of 1, but
the actual probability is very slightly less than 1.)
c. The probability from part (a) is more relevant because it shows
that 85.77% of male passengers will not need to bend. The re-
sult from part (b) gives us information about the mean for a
group of 100 men, but it doesn’t give us useful information
about the comfort and safety of individual male passengers.
d. Because men are generally taller than women, a design that
accommodates a suitable proportion of men will necessarily
accommodate a greater proportion of women.
17. a. 0.5302 (Tech: 0.5317)
b. 0.7323 (Tech: 0.7326)
c. Part (a), because the seats will be occupied by individual
women, not groups of women.
19. 0.0001 (Tech: 0.0000). The results do suggest that the Pepsi
cans are filled with an amount greater than 12.00 oz.
21. a. 73.6 in.
b. 69.5 in.
c. The result from part (a) is more relevant because it indicates
that the doorway height of 73.6 in. is adequate for 95% of
individual men. When designing the doorway height, we are
concerned with the distribution of heights of men, not the
sampling distribution of mean heights of men.
23. a. Yes. Sampling is without replacement (because the 12 passen-
gers are different people) and the sample size of 12 is greater
than 5% of the finite population of size 210.
Appendix D 813
b. 175 lb
c. 0.0918 (Tech: 0.0910). This probability is too high; the ele-
vator could be overloaded about 9% of the time.
d. 10 passengers
Section 6-6
1. The histogram should be approximately normal or bell shaped,
because sample proportions tend to approximate a normal
distribution.
3. No. With n = 4 and p = 0.5, the requirements of np Ú 5 and
nq Ú 5 are not satisfied, so the normal distribution is not a
good approximation.
5. The area to the right of 8.5
7. The area to the left of 4.5
9. The area to the left of 15.5
11. The area between 4.5 and 9.5
13. 0.117; normal approximation: 0.1140 (Tech: 0.1145)
15. 0.962; normal approximation is not suitable
17. 0.0018 (Tech using binomial: 0.0020). Because that probability
is very low, it is not likely that the goal of at least 1300 will be
reached.
19. 0.0001 (Tech: 0.0000). The method appears to be effective, because
the probability of getting at least 525 girls by chance is so small.
21. a. 0.0318 (Tech using normal approximation: 0.0305; tech us-
ing binomial: 0.0301)
b. 0.2676 (Tech using normal approximation: 0.2665; tech us-
ing binomial: 0.2650)
c. Part (b)
d. No, the results could easily occur by chance with a 25% rate.
23. 0.2709 (Tech using normal approximation: 0.2697; tech using
binomial: 0.2726). Media reports appear to be wrong.
25. 0.7704 (Tech using normal approximation: 0.7717; tech using
binomial: 0.7657). There is a 0.7704 probability that the pool
of 200 volunteers is sufficient. Because blood banks are so im-
portant, it would probably be much better to get more donors
so that the probability of reaching the goal of at least 10 univer-
sal donors becomes more likely.
27. 0.2776 (Tech using binomial: 0.2748). No, 27 blue M&Ms is
not unusually high because the probability of 27 or more blue
M&Ms is 0.2776, which is not small.
29. 0.3015 (Tech using normal approximation: 0.3000; tech using
binomial: 0.2900). Lipitor users do not appear to be more likely
to have flu symptoms.
31. 0.4168 (Tech using normal approximation: 0.4153; tech using
binomial: 0.4264). No, the probability is too high, so it is not
wise to accept 236 reservations.
33. a. 6; 0.4602 (Tech using normal approximation: 0.4583; tech
using binomial: 0.4307)
b. 101; 0.3936 (Tech using normal approximation: 0.3933;
tech using binomial: 0.3932)
c. The roulette game provides a better likelihood of making a
profit.
35. a. 0.821
b. 0.9993 (Tech using binomial: 0.9995)
c. 0.0000165
d. 0.552
814 Appendix D

Section 6-7 Chapter 6 Statistical Literacy

1. A normal quantile plot can be used to determine whether sam-
ple data are from a population having a normal distribution.
3. Because the weights are likely to have a normal distribution,the pat-
tern of the points should be approximately a straight-line pattern.
5. Not normal. The points are not reasonably close to a straight-
line pattern.
7. Normal. The points are reasonably close to a straight-line pattern
and there is no other pattern that is not a straight-line pattern.
9. Normal
11. Not normal
13. Normal
and Critical Thinking
1. A normal distribution is the probability distribution of a
continuous random variable with the properties that a graph
of the distribution is symmetric and bell shaped, and it can be
described by Formula 6-1. A standard normal distribution is a
normal distribution that has mean of 0 and a standard
deviation of 1.
2. No. The term “normal” distribution has a special meaning in
statistics that is different from the ordinary usage of the word
“normal.” One property of a normal distribution is that it is bell
shaped and symmetric, so the skewed distribution of incomes is
not a normal distribution.
3. The sample means will tend to have a normal distribution.
4. No. The sample might well be biased. It is a convenience sam-
ple, not a simple random sample. Those with strong feelings
about the issues involved might be more likely to respond, and
that group might be very different from the general population.

Chapter 6 Quick Quiz

1. 1.88
2. Normal
3. m = 0 and s = 1.
15. Not normal 4. 0.1587
5. 0.9270
6. 0.8413
7. 0.1151
8. 0.5762 (Tech: 0.5763)
9. 0.8413
10. 0.0228

Chapter 6 Review Exercises

17. Heights appear to be normal, but cholesterol levels do not ap-
pear to be normal. Cholesterol levels are strongly affected by
diet, and diets might vary in dramatically different ways that do
not yield normally distributed results.
19. Normal; - 1.28, -0.52, 0, 0.52, 1.28.
21. a. Yes
b. Yes
c. No
1. a. 1.62% (Tech: 1.61%)
b. 0.01% (Tech: 0.00%)
c. The day bed length appears to be adequate, because very few
men or women have heights that exceed it.
2. 73.6 in.
3. a. Men: 0.07%; women: 0.01% (Tech: 0.00%)
b. 75.5 in.
4. 12.22%. Yes, because the minimum acceptable height is greater
than the mean height of all women. All of the Rockettes are
taller than the mean height of all women.
5. 0.2296 (Tech using normal approximation: 0.2286; tech using
binomial: 0.2278). The occurrence of 787 offspring plants with
long stems is not unusually low, because its probability is not
small. The results are consistent with Mendel’s claimed propor-
tion of 3>4.
6. Parts (a), (b), (c) are true.
7. a. 0.0222 (Tech: 0.0221)
b. 0.2847 (Tech: 0.2836)
c. 0.6720 (Tech: 0.6715)
d. 254.6 mg> 100 mL
8. 0.0778 (Tech using normal approximation: 0.0774; tech using
binomial: 0.0769). Because the probability is not very small,
such as less than 0.05, the results could easily occur by chance.
 Appendix D 815

There is not strong evidence to charge that the Newport Temp Chapter 7 Answers
Agency is discriminating against women.
Section 7-2
9. a. 0.44
b. -2.65 1. The level of confidence (such as 95%) was not provided.
c. 1.96 3. The point estimate of 43% or 0.43 does not reveal any
10. a. Normal information about the accuracy of the estimate. By providing a
b. 3420 g range of values associated with a probability (confidence level),
c. 53.7 g a confidence interval does reveal important information about
11. a. 0.4602 (Tech: 0.4588) accuracy.
b. 0.0655 (Tech: 0.0656). Yes, because the weight limit will be 5. 2.575 (Tech: 2.5758293)
exceeded about 7% of the time, or about one flight out of every 7. 1.645 (Tech: 1.6448536)
15 flights. 9. 0.350 ; 0.150
12. Yes. A histogram is approximately bell shaped, and a normal 11. 0.483 ; 0.046
quantile plot results in points that are reasonably close to a 13. 0.370; 0.050
straight-line pattern without any other systematic pattern. 15. 0.480; 0.047
17. 0.0304
19. 0.0325
Cumulative Review Exercises 21. 0.145 6 p 6 0.255
1. a. $327,900 23. 0.0674 6 p 6 0.109
b. $223,500 25. 475
c. $250,307 27. 1996 (Tech: 1998)
d. 62,653,600,000 square dollars (Tech: 62,653,655,570 square 29. a. 0.915
dollars) b. 0.892 6 p 6 0.937
e. -0.37 c. Yes. The true proportion of girls with the XSORT method is
f. Ratio substantially above the proportion of (about) 0.5 that is ex-
g. Discrete pected with no method of gender selection.
2. a. A simple random sample is a sample selected in such a way 31. a. 0.505
that every possible sample of the same size has the same b. 0.496 6 p 6 0.514
chance of being selected. c. No, because the proportion could easily equal 0.5. The pro-
b. A voluntary response is a sample obtained in such a way that portion is not substantially less than 0.5 the week before
the respondents themselves decide whether to be included. Thanksgiving.
Because the respondents decide whether to be included, it is 33. a. 0.226 6 p 6 0.298
common that those with strong or special interests are more b. No, the confidence interval includes 0.25, so the true per-
likely to respond, so the sample is biased and not representa- centage could easily equal 25%.
tive of the population. 35. a. 0.0267% 6 p 6 0.0376%
3. a. 0.0000412 b. No, because 0.0340% is included in the confidence interval.
b. 0.1401 (Tech using normal approximation: 0.1399; tech us- 37. 0.714 6 p 6 0.746. Although the confidence interval limits
ing binomial: 0.1406) do not include the value of 3> 4 or 0.75, the statement gives
c. No. The probability of 40 or more viral infections is not the value of 3> 4, which is not likely to be wrong by very
small, such as less than 0.05, so getting as many as 40 or much.
more viral infections could easily occur, and it is not 39. 0.615 6 p 6 0.625. No. The sample is a voluntary response
unusually high. sample that might not be representative of the population.
d. No. We would need evidence to show that the rate of viral 41. a. 4145 (Tech: 4147)
infections among Nasonex users is significantly greater than b. 3268 (Tech: 3270)
the rate of viral infections among those not treated with 43. 1068
Nasonex, but that information was not given here. 45. 11.3% 6 p 6 26.7%. Yes. Because the confidence interval
4. By using a vertical scale that does not begin at 0, the graph dis- limits do contain the claimed percentage of 16%, the true per-
torts the data by exaggerating the differences. centage of green M&Ms could be equal to 16%.
5. a. 0.001 47. Wednesday: 0.178 6 p 6 0.425; Sunday: 0.165 6 p 6 0.412.
b. 0.271 The confidence intervals are not substantially different. Precipi-
c. The requirement that np Ú 5 is not satisfied, indicating that tation does not appear to occur more on either day.
the normal approximation would result in errors that are too 49. 985; the sample size is not too much smaller than the sample of
large. 1068 that would be required for a very large population.
d. 5.0 51. 0.0395 6 p 6 0.710; no
e. 2.1 53. a. The requirement of at least 5 successes and at least 5 failures
f. No, 8 is within two standard deviations of the mean and is is not satisfied, so the normal distribution cannot be used.
within the range of values that could easily occur by chance. b. 0.15
816 Appendix D
Section 7-3
1. A point estimate is a single value used to estimate the popula-
tion parameter. The best point estimate of the population mean
is found by computing the value of the sample mean x.
3. Assuming that the standard deviation of the heights of all
women is 2.5 in., the mean height of all women is estimated to
be 63.20 in., with a margin of error of 0.78 in. In theory, in
95% of such studies, the mean should differ by no more than
0.78 in. in either direction from the mean that would be found
by using the heights of all women.
5. 1.645 (Tech: 1.6448536)
7. 1.28 (Tech: 1.2815516)
9. E = 18.8; 658.2 6 m 6 695.8
11. The margin of error and confidence interval cannot be calcu-
lated by the methods of this section.
13. 1974
15. 2981 (Tech: 2982)
17. 21.120 mg
19. 21.120 mg ; 1.267 mg
21. a. 146.22 lb
b. 136.66 lb 6 m 6 155.78 lb
23. a. 58.3 sec
b. 55.4 sec 6 m 6 61.2 sec
c. Yes, because the confidence interval limits do contain 60 sec.
25. a. 1464 6 m 6 1580
b. 1445 6 m 6 1599
c. The 99% confidence interval is wider. We need a wider range
of values to be more confident that the interval contains the
population mean.
27. 128.7 6 m 6 139.2; ideally, all of the measurements would be
the same, so there would not be an interval estimate.
29. $89.9 million 6 m 6 $156.1 million
31. 35
33. 6907. The sample size is too large to be practical.
35. 38,416
37. 1486 6 m 6 1558. The confidence interval becomes much nar-
rower because the sample is such a large segment of the population.
Section 7-4
1. The amount refers to an average, which is likely to be the mean
or median, but the margin of error is appropriate for a propor-
tion, not a mean or median. The margin of error should be an
amount in dollars, not percentage points.
3. No. The simple random sample is obtained from Californians,
so it would not necessarily yield a good estimate for the United
States population of adults. It is very possible that Californians
are not representative of the United States. A simple random
sample of the population is needed.
5. t a>2 = 2.074
7. Neither normal nor t distribution applies.
9. z a>2 = 1.645
11. t a>2 = 3.106
13. E = $266; $8738 6 m 6 $9270
15. 8.0518 g 6 m 6 8.0903 g. There is 95% confidence that the
limits of 8.0518 g and 8.0903 g contain the mean weight of the
population of all dollar coins.
17. a. 3.2
b. -2.3 6 m 6 8.7. Because the confidence interval limits
contain 0, it is very possible that the mean of the changes in
LDL cholesterol is equal to 0, suggesting that the Garlicin
treatment did not affect the LDL cholesterol levels. It does
not appear that the Garlicin treatment is effective in lowering
LDL cholesterol.
19. a. 98.20°F
b. 98.04°F 6 m 6 98.36°F; because the confidence interval
limits do not contain 98.6°F, the results suggest that the
mean is less than 98.6°F.
21. a. 5.8 days 6 m 6 6.2 days
b. 5.9 days 6 m 6 6.3 days
c. The two confidence intervals are very similar. The echinacea
treatment group does not appear to fare any better than the
placebo group, so the echinacea treatment does not appear to
be effective.
23. a. 3.9 6 m 6 6.1
b. 3.3 6 m 6 6.1
c. The confidence intervals do not differ by very much, so it ap-
pears that the magnet treatment and the sham treatment pro-
duce similar results. The magnet treatment does not appear
to be effective.
25. - 0.471 6 m 6 3.547; the confidence interval is likely to be a
poor estimate because the value of 5.40 appears to be an outlier,
suggesting that the assumption of a normally distributed popu-
lation is not correct.
27. a. $11.7 million 6 m 6 $29.1 million
b. No.
c. Reasonable answers for the population are the salaries of all
TV personalities, or the top 10 TV salaries for different years,
but the given sample is not representative of either of these
populations.
d. No. The nature of the sample makes the confidence interval
meaningless as an estimate of the mean of some larger
population.
29. A normal quantile plot shows that the data appear to be from a
normally distributed population. 4092.2 sec 6 m 6 4594.1 sec
(Tech: 4092.3 sec 6 m 6 4594.1 sec)
31. a. 1.16 6 m 6 1.35
b. 0.81 6 m 6 1.02
c. The filtered cigarettes result in a confidence interval that is com-
pletely less than the confidence interval for the nonfiltered ciga-
rettes. Assuming that the king-size cigarettes and the 100 mm
cigarettes contain about the same amount of tobacco, it appears
that the filters are effective in reducing the amount of nicotine.
33. 40.5 years 6 m 6 86.9 years (Tech: 40.5 years 6 m 6
86.8 years). This result is substantially different from the confi-
dence interval of 52.3 years 6 m 6 57.4 years that was found
in Exercise 30. Confidence intervals are very sensitive to out-
liers. Outliers should be carefully examined and discarded if
they are found to be errors. If an outlier is a correct value, it
might be very helpful to see its effects by constructing the con-
fidence interval with and without the outlier included.
35. 0.8462 g 6 m 6 0.8668 g; 0.8474 g 6 m 6 0.8656 g; the
second confidence interval is narrower, indicating that we have a
 Appendix D 817

more accurate estimate when the relatively large sample is from Chapter 7 Quick Quiz
a relatively small finite population.
1. There is 95% confidence that the limits of 10.0 and 20.0 con-
tain the true value of the population mean m. This means that if
Section 7-5 the sampling procedure is repeated many times, in the long run
1. There is 95% confidence that the limits of 0.0455 g and 0.0602 g 95% of the resulting confidence interval limits will contain the
contain the true value of the standard deviation of the population true value of the population mean m.
of all M&Ms. 2. (2) The election is too close to call.
3. No. The outcomes are equally likely and they have a uniform 3. 2.093
distribution, not a normal distribution as required. Because the 4. 1.645
requirement of normality is not satisfied, the resulting confi- 5. 2401
dence interval cannot be expected to provide a good estimate. 6. 0.4
5. 2.180, 17.535 7. 0.361 6 p 6 0.439
7. 51.172, 116.321 8. 36.6 years 6 m 6 43.4 years
9. 265 6 s 6 448 9. 36.7 years 6 m 6 43.3 years
11. 1.148 6 s 6 6.015 (Tech: 1.148 6 s 6 6.016) 10. 2213
13. 19,205; no, the sample size is excessively large for most applications.
15. 101; yes, the sample size is small enough to be practical.
17. 586 g 6 s 6 717 g; no, because the confidence interval limits
Chapter 7 Review Exercises
contain 696 g. 1. 76.1% 6 p 6 82.0%. There is 95% confidence that the limits
19. a. 17.7 min 6 s 6 32.4 min of 76.1% and 82.0% contain the true value of the percentage of
b. 14.7 min 6 s 6 35.3 min all adults who believe that it is morally wrong to not report all
c. The two confidence intervals are not substantially different. income on tax returns.
There does not appear to be a difference between the stan- 2. 4145 (Tech: 4147)
dard deviation of lengths of movies rated PG or PG-13 and 3. 21,976 (Tech: 21,991). The sample size is too large to be practi-
lengths of movies rated R. cal, unless the data can be processed automatically.
21. 253.2 sec 6 s 6 811.8 sec 4. 2.492 g 6 m 6 2.506 g. The confidence interval does contain
23. 1.195 6 s 6 4.695; yes, the confidence interval is likely to be the required mean value of 2.5 g, so the manufacturing process
a poor estimate because the value of 5.40 appears to be an out- appears to be OK.
lier, suggesting that the assumption of a normally distributed 5. 368.4 6 m 6 969.2
population is not correct. 6. 145.0 6 s 6 695.6 (Tech: 145.0 6 s 6 695.3)
25. 80.6 6 s 6 106.5 (Tech: 81.0 6 s 6 107.1) 7. a. 0.890
27. 152.3644 and 228.4771 are close to the STATDISK values. b. 0.871 6 p 6 0.910
c. Yes. Because there is 95% confidence that the limits of 0.871
and 0.910 contain the true value of the population propor-
Chapter 7 Statistical Literacy tion, it appears that there is strong evidence that the popula-
and Critical Thinking tion proportion is greater than 0.5.
1. A point estimate is the single value of the sample proportion, 8. a. 601
and it is the best estimate of the population proportion. A confi- b. 1383
dence interval is a range of values associated with a confidence c. 1383
level, and it consists of the range of values that is likely to con- 9. a. 3.840 lb
tain the value of the population proportion. By including a b. 1.786 lb 6 m 6 5.894 lb
range of values with an associated confidence level, the confi- c. 1.686 lb 6 m 6 5.994 lb
dence interval has the advantage of giving meaningful informa- 10. a. 1.624 lb 6 s 6 5.000 lb
tion about how accurate the estimate of the population propor- b. 2.639 lb 2 6 s2 6 25.002 lb 2 (Tech:
tion is likely to be. 2.639 lb 2 6 s2 6 25.004 lb 2)
2. There is 95% confidence that the limits of 0.0262 and 0.0499
contain the true value of the population proportion.
3. The confidence interval has a confidence level of 95% or 0.95. In Chapter 7 Cumulative Review Exercises
general, the confidence level is the proportion of times that the 1. 118.4 lb, 121.0 lb, 9.2 lb
confidence interval actually does contain the value of the popula- 2. Ratio
tion parameter, assuming that the process of sampling and creat- 3. 111.8 lb 6 m 6 125.0 lb
ing a confidence interval is repeated a large number of times. 4. 55
4. Because the poll uses a voluntary response sample, it is very pos- 5. a. 0.962
sible that the respondents are not representative of the general b. 0.00144
population, so the confidence interval should not be used to c. 0.4522 (Tech using normal approximation: 0.4534; tech us-
form conclusions about the general population. ing binomial probability distribution: 0.4394)
818 Appendix D
6. a. 0.5910 (Tech: 0.5906)
b. 0.8749 (Tech: 0.8741)
c. 27.2 (Tech: 27.3)
7. A sample of n items is a simple random sample if every sample
of the same size has the same chance of being selected. A sample
is a voluntary response sample if the subjects decide themselves
whether to be included.
8. 1
9. With random guesses, the probability of 12 correct answers to
12 true–false questions is 1/4096 or 0.000244. It is possible that
random guesses were made, but it is highly unlikely.
10. It is a convenience sample, and it is not likely to be representa-
tive of the population.
Chapter 8 Answers
Section 8-2
1. Given the large sample size and the sample percentage of 20%,
the sample data appear to support the claim that less than 50%
of people believe that bosses are good communicators. Given
that the survey subjects constitute a voluntary response sample
instead of a simple random sample, the sample results should not
be used to form any conclusions about the general population.
3. The claim that the mean is equal to 325 mg becomes the null hy-
pothesis. The hypothesis testing procedure allows us to reject that
null hypothesis or we can fail to reject it, but we can never sup-
port a null hypothesis. Hypothesis testing can never be used to
support a claim that a parameter is equal to some particular value.
5. There is sufficient evidence to support the claim that the coin
favors heads (because 90 heads in 100 tosses is very unlikely to
occur by chance with a fair coin).
7. There is not sufficient evidence to support the claim that the
mean pulse rate is less than 75.
9. H0: m = $60,000. H1: m 7 $60,000.
11. H0: s = 0.62°F. H1: s Z 0.62°F.
13. H0: s = 40 sec. H1: s 6 40 sec.
15. H0: p = 0.80. H1: p Z 0.80.
17. z = ;2.575 (Tech: ;2.5758)
19. z = 2.05
21. z = ;1.96
23. z = -2.575 (Tech: - 2.5758)
25. z = 0.67
27. z = 2.31
29. 0.1056; fail to reject the null hypothesis.
31. 0.0802 (Tech: 0.0801); fail to reject the null hypothesis.
33. 0.0060; reject the null hypothesis.
35. 0.0107; reject the null hypothesis.
37. There is not sufficient evidence to support the claim that the
percentage of blue M&Ms is greater than 5%.
39. There is not sufficient evidence to warrant rejection of the claim
that the percentage of Americans who know their credit score is
equal to 20%.
41. Type I error: Reject the claim that the percentage of nonsmokers
exposed to secondhand smoke is equal to 41%, when that percent-
age is actually 41%. Type II error: Fail to reject the claim that the
percentage of nonsmokers exposed to secondhand smoke is equal
to 41% when that percentage is actually different from 41%.
43. Type I error: Reject the claim that the percentage of college stu-
dents who use alcohol is equal to 70% when that percentage is
actually 70%. Type II error: Fail to reject the claim that the per-
centage of college students who use alcohol is equal to 70%
when that percentage is actually greater than 70%.
45. a. No, because rejection at the 0.01 significance level requires a
sample statistic that differs from the claimed value of the pa-
rameter by an amount that is more extreme than the differ-
ence required for rejection at the 0.05 significance level. If H0
is rejected at the 0.05 level, it may or may not be rejected at
the 0.01 level.
b. Yes, because rejection at the 0.01 significance level requires
a sample statistic that differs from the claimed value of the
parameter by an amount that is more extreme than the
difference required for rejection at the 0.05 significance level,
so rejection of H0 at the 0.01 level requires that H0 must also
be rejected at the 0.05 level.
47. a. 0.7852 (Tech: 0.7857)
b. Assuming that p = 0.5, as in the null hypothesis, the critical
value of z = 1.645 corresponds to pN = 0.6028125, so any
sample proportion greater than 0.6028125 causes us to reject
the null hypothesis, as shown in the shaded critical region of
the top graph. If p is actually 0.65, then the null hypothesis
of p = 0.5 is false, and the actual probability of rejecting the
null hypothesis is found by finding the area greater than
pN = 0.6028125 in the bottom graph, which is the shaded
area in the bottom graph. That is, the shaded area in the bot-
tom graph represents the probability of rejecting the false
null hypothesis.
Section 8-3
1. pN = 0.230. The symbol pN is used to represent a sample proportion.
3. The sample proportion is pN = 0.439. Because the P-value of
0.9789 is high, we should not reject the null hypothesis of
p = 0.5. We should conclude that there is not sufficient evi-
dence to support the claim that p 7 0.5.
5. a. z = 1.90
b. z = ;2.575 (Tech: ;2.576)
c. 0.0574 (Tech: 0.0578)
d. There is not sufficient evidence to reject the null hypothesis
that the percentage of college applications submitted online
is equal to 50%.
e. No, a hypothesis test cannot be used to prove that a popula-
tion proportion is equal to some claimed value.
7. H0: p = 0.75. H1: p Z 0.75. Test statistic: z = -2.17.
P-value = 0.0301. Reject H0. There is sufficient evidence to
warrant rejection of the claim that the proportion of adults who
use cell phones while driving is equal to 75%.
9. H0: p = 0.75. H1: p Z 0.75. Test statistic: z = 2.56. Critical
values: z = ;2.575. P-value: 0.0104 (Tech: 0.0105). Fail to re-
ject H0. There is not sufficient evidence to warrant rejection of
the claim that 75% of adults say that it is morally wrong to not
report all income on tax returns.
11. H0: p = 1>3. H1: p 7 1>3. Test statistic: z = 3.47. Critical
value: z = 2.33. P-value: 0.0003. Reject H0. There is sufficient
evidence to support the claim that the proportion of challenges

that are successful is greater than 1> 3. When evaluating the
quality of referee calls, it should be noted that only close calls
are likely to be challenged, and 327 successful challenges among
thousands of calls is a very small proportion of wrong calls, so
the quality of the referees is quite good, but they are not perfect.
13. H0: p = 0.06. H1: p 7 0.06. Test statistic: z = 4.47. Critical
value: z = 1.645. P-value: 0.0001 (Tech: 0.000004). Reject H0.
There is sufficient evidence to support the claim that the rate of
nausea is greater than the 6% rate experienced by flu patients
given a placebo. The rate of nausea with the Tamiflu treatment
appears to be greater than the rate for those given a placebo, and
the sample proportion of 0.0994 suggests that about 10% of
those treated experience nausea, so nausea is a concern for those
given the treatment.
15. H0: p = 0.000340. H1: p Z 0.000340. Test statistic: z = -0.66.
Critical values: z = ;2.81. P-value: 0.5092 (Tech: 0.5122).
Fail to reject H0. There is not sufficient evidence to support
the claim that the rate is different from 0.0340%. Cell phone
users should not be concerned about cancer of the brain or
nervous system.
17. H0: p = 0.20. H1: p 6 0.20. Test statistic: z = -2.93. Critical
value: z = -2.33. P-value: 0.0017. Reject H0. There is suffi-
cient evidence to support the claim that less than 20% of Michi-
gan gas pumps are inaccurate. The percentage of inaccurate
pumps should be very low, but this hypothesis test shows only
that it appears to be less than 20%. The sample proportion of
0.186 suggests that too many pumps are not accurate.
19. H0: p = 0.80. H1: p 6 0.80. Test statistic: z = -1.11. Critical
value: z = - 1.645. P-value: 0.1335 (Tech: 0.1332). Fail to re-
ject H0. There is not sufficient evidence to support the claim
that the polygraph results are correct less than 80% of the time.
Based on the sample proportion of correct results in 75.5% of
the 98 cases, polygraph results do not appear to have the high
degree of reliability that would justify the use of polygraph re-
sults in court, so polygraph test results should be prohibited as
evidence in trials.
21. H0: p = 0.20. H1: p 6 0.20. Test statistic: z = -8.84. Critical
value: z = -2.33. P-value: 0.0001 (Tech: 0.0000). Reject H0.
There is sufficient evidence to support the claim that less than
20% of the TV sets in use were tuned to 60 Minutes.
23. H0: p = 0.5. H1: p Z 0.5. Test statistic: z = - 0.66 (using
pN = 0.473) or z = -0.65 (using x = 71). Critical values:
z = ;1.96 (assuming a 0.05 significance level). P-value: 0.5092
or 0.5156 (Tech: 0.5136). Fail to reject H0. There is not suffi-
cient evidence to warrant rejection of the claim that among se-
nior executives, 50% say that the most common job interview
mistake is to have little or no knowledge of the company. The
important lesson is to always prepare for a job interview by
learning about the company.
25. H0: p = 3>4. H1: p Z 3>4. Test statistic: z = -2.53 (using
pN = 0.73) or z = -2.54 (using x = 2198). Critical values:
z = ;1.96 (assuming a 0.05 significance level). P-value: 0.0114 or
0.0110 (Tech: 0.0112). Reject H0. There is sufficient evidence to
warrant rejection of the claim that 3> 4 of all adults use the Internet.
The proportion appears to be different from 3> 4 or 0.75. The re-
porter should not write that 3> 4 of all adults use the Internet.
Appendix D 819
27. H0: p = 0.5. H1: p Z 0.5. Test statistic: z = -1.08 (using
pN = 0.43) or z = - 1.17 (using x = 25). Critical values:
z = ;1.96. P-value: 0.2802 or 0.2420 (Tech: 0.2413). Fail to
reject H0. There is not sufficient evidence to warrant rejection of
the claim that these women have no ability to predict the sex of
their baby. The results for these women with 12 years of educa-
tion or less suggest that their percentage of correct predictions is
not very different from results expected with random guesses.
29. H0: p = 1>4. H1: p 7 1>4. Test statistic: z = 5.73. Critical
value: z = 1.645. P-value: 0.0001 (Tech: 0.0000). Reject H0.
There is sufficient evidence to support the claim that more than
1> 4 of people say that bosses scream at employees. If the sample
is a voluntary response sample, the conclusion about the popu-
lation might not be valid.
31. H0: p = 0.5. H1: p 7 0.5. Test statistic: z = 5.83 (using
pN = 0.61) or z = 5.85 (using x = 429). Critical value:
z = 1.645. P-value: 0.0001 (Tech: 0.0000). Reject H0. There is
sufficient evidence to support the claim that most workers get
their jobs through networking.
33. H0: p = 0.20. H1: p Z 0.20. Test statistic: z = -1.75. Critical
values: z = ;1.96 (assuming a 0.05 significance level). P-value:
0.0802 (Tech: 0.0801). Fail to reject H0. There is not sufficient
evidence to warrant rejection of the claim that 20% of plain
M&M candies are red.
35. H0: p = 0.5. H1: p Z 0.5. Test statistic: z = 2.18. Critical val-
ues: z = ;1.96. P-value: 0.0292 (Tech: 0.0295). Reject H0.
There is sufficient evidence to warrant rejection of the claim
that the percentage of males in the population is equal to 50%.
37. H0: p = 0.55. H1: p Z 0.55. Using the binomial probability
distribution with an assumed value of p = 0.55 and with
n = 35, the probability of 12 or fewer movies with R ratings is
0.0109367, so the P-value is 0.02187. Because that P-value is
greater than the significance level of 0.01, fail to reject H0.
There is not sufficient evidence to warrant rejection of the claim
that the movies in Data Set 9 are from a population in which
55% of the movies have R ratings.
39. H0: p = 0.10. H1: p Z 0.10. Test statistic: z = -2.36. Critical
values: z = ;2.575. P-value: 0.0182 (Tech: 0.0184). Fail to re-
ject H0. Even though no blue candies are obtained, there is not
sufficient evidence to warrant rejection of the claim that 10% of
the candies are blue.
Section 8-4
1. The sample must be a simple random sample, the population stan-
dard deviation must be known, and the population must be nor-
mally distributed (because the sample size is not greater than 30).
3. 98% or 0.98
5. H0: m = 5 cm. H1: m Z 5 cm. Test statistic: z = 1.34. Criti-
cal values: z = ;1.96 (assuming a 0.05 significance level).
P-value: 0.1797. Fail to reject H0. There is not sufficient
evidence to warrant rejection of the claim that women have a
mean wrist breadth equal to 5 cm.
7. H0: m = 210 sec. H1: m 7 210 sec. Test statistic: z = 4.93.
Critical value: z = 1.645. P-value: 0.0001 (Tech: 0.0000). Reject
H0. There is sufficient evidence to support the claim that the sam-
ple is from a population of songs with a mean length greater than
820 Appendix D
210 sec. These results suggest that the advice of writing a song
that must be no longer than 210 seconds is not sound advice.
9. H0: m = 0.8535 g. H1: m Z 0.8535 g. Test statistic: z = 0.77.
Critical values: z = ;1.96. P-value: 0.4412 (Tech: 0.4404).
Fail to reject H0. There is not sufficient evidence to warrant re-
jection of the claim that green M&Ms have a mean weight of
0.8535 g. The green M&Ms appear to have weights consistent
with the package label.
11. H0: m = 0 lb. H1: m 7 0 lb. Test statistic: z = 3.87. Critical
value: z = 2.33. P-value: 0.0001. Reject H0. There is sufficient
evidence to support the claim that the mean weight loss is
greater than 0. Although the diet appears to have statistical
significance, it does not appear to have practical significance, be-
cause the mean weight loss of only 3.0 lb does not seem to be
worth the effort and cost.
13. H0: m = 91.4 cm. H1: m Z 91.4 cm. Test statistic: z = 2.33.
Critical values: z = ;1.96. P-value: 0.0198 (Tech: 0.0196). Re-
ject H0. There is sufficient evidence to support the claim that
males at her college have a mean sitting height different from
91.4 cm. Because the student selected male friends, she used a
convenience sample that might have different characteristics
than the population of males at her college. Because the require-
ment of a simple random sample is not satisfied, the method of
hypothesis testing given in this section does not apply, so the re-
sults are not necessarily valid.
15. H0: m = $500,000. H1: m 6 $500,000. Test statistic:
z = -1.15. Critical value: z = -1.645. P-value: 0.1251 (Tech:
0.1257). Fail to reject H0. There is not sufficient evidence to
support the claim that the mean salary of a football coach in the
NCAA is less than $500,000.
17. H0: m = 235.8 cm. H1: m Z 235.8 cm. Test statistic:
z = -0.56. Critical values: z = ;1.96. P-value: 0.5754 (Tech:
0.5740). Fail to reject H0. There is not sufficient evidence to
warrant rejection of the claim that the new baseballs have a
mean bounce height of 235.8 cm. The new baseballs do not ap-
pear to be different.
19. H0: m = 678. H1: m Z 678. Test statistic: z = 4.93. Critical
values: z = ;1.96. P-value: 0.0002 (Tech: 0.0000). Reject H0.
There is sufficient evidence to warrant rejection of the claim
that these sample FICO scores are from a population with a
mean FICO score equal to 678.
21. H0: m = 3>4 in. H1: m Z 3>4 in. Test statistic: z = -1.87.
Critical values: z = ;1.96. P-value: 0.0614 (Tech: 0.0610). Fail
to reject H0. There is not sufficient evidence to warrant rejection
of the claim that the mean length is equal to 0.75 in. or 3> 4 in.
The lengths appear to be consistent with the package label.
23. a. The power of 0.2296 shows that there is a 22.96% chance of
supporting the claim that m 7 166.3 lb when the true mean
is actually 170 lb. This value of the power is not very high,
and it shows that the hypothesis test is not very effective in
recognizing that the mean is greater than 166.3 lb when the
actual mean is 170 lb.
b. b = 0.7704 (Tech: 0.7719). The probability of a type II er-
ror is 0.7704. That is, there is a 0.7704 probability of making
the mistake of not supporting the claim that m 7 166.3 lb
when in reality the population mean is 170 lb.
Section 8-5
1. Because the sample size is not greater than 30, the sample data
must come from a population with a normal distribution. To de-
termine whether the requirement of a normal distribution is satis-
fied, examine a histogram to determine whether it is approximately
normal, verify that there are no outliers (or at most one outlier), or
examine a normal quantile plot, or use some formal test of nor-
mality, such as the Ryan-Joiner test described in Section 6-7.
3. A t test is a hypothesis test that uses the Student t distribution. It is
called a t test because it involves use of the Student t distribution.
5. Student t
7. Neither normal nor Student t
9. Table A-3: P-value 7 0.10; technology: P-value = 0.3355.
11. Table A-3: 0.05 6 P-value 6 0.10; technology: P-value =
0.0932.
13. H0: m = 210 sec. H1: m 7 210 sec. Test statistic: t = 4.93.
P-value: 0.000. Reject H0. There is sufficient evidence to sup-
port the claim that the sample is from a population of songs
with a mean length greater than 210 sec. These results suggest
that the advice of writing a song that must be no longer than
210 seconds is not sound advice.
15. H0: m = 21.1 mg. H1: m 6 21.1 mg. Test statistic: t =
-10.676. Critical value: t = -1.711. P-value 6 0.005 (Tech:
0.0000). Reject H0. There is sufficient evidence to support the
claim that filtered 100 mm cigarettes have a mean tar amount
less than 21.1 mg. The results suggest that the filters are effec-
tive in reducing the amount of tar.
17. H0: m = 2.5 g. H1: m Z 2.5 g. Test statistic: t = -0.332.
Critical value: t = ;2.028. P-value 7 0.20 (Tech: 0.7417).
Fail to reject H0. There is not sufficient evidence to warrant re-
jection of the claim that pennies have a mean weight equal
to 2.5 g. The pennies appear to conform to the specifications
of the U.S. Mint.
19. H0: m = 4.5 years. H1: m 7 4.5 years. Test statistic: t = 1.227.
Critical value: t = 1.664. P-value 7 0.10 (Tech: 0.1117). Fail
to reject H0. There is not sufficient evidence to support the
claim that the mean time is greater than 4.5 years.
21. H0: m = 49.5 cents. H1: m 6 49.5 cents. Test statistic:
t = - 8.031. Critical value: t = -2.364 (approximately).
P-value 6 0.01 (Tech: 0.0000). Reject H0. There is sufficient ev-
idence to support the claim that the mean is less than 49.5 cents.
The results suggest that the cents portions of check amounts are
such that the values from 0 cents to 99 cents are not equally likely.
23. H0: m = 8.00 tons. H1: m Z 8.00 tons. Test statistic:
t = - 1.152. Critical values: t = ;2.040. P-value 7 0.20
(Tech: 0.2580). Fail to reject H0. There is not sufficient evi-
dence to warrant rejection of the claim that cars have a mean
greenhouse gas emission of 8.00 tons.
25. H0: m = 1000 hic. H1: m 6 1000 hic. Test statistic:
t = -2.661. Critical value: t = -3.365. P-value is between
0.01 and 0.025 (Tech: 0.0224). Fail to reject H0. There is not
sufficient evidence to support the claim that the population
mean is less than 1000 hic. There is not strong evidence that the
mean is less than 1000 hic, and one of the booster seats has a
measurement of 1210 hic, which does not satisfy the specified
requirement of being less than 1000 hic.

27. H0: m = $5000. H1: m Z $5000. Test statistic: t = 1.639.
Critical values: t = ;2.776. P-value is between 0.10 and 0.20
(Tech: 0.1766). Fail to reject H0. There is not sufficient evi-
dence to reject the claim that the mean damage cost is $5000.
29. H0: m = 3>4 in. H1: m Z 3>4 in. Test statistic: t = -1.825.
Critical values: t = ;2.009 (approximately). P-value is between
0.05 and 0.10 (Tech: 0.0741). Fail to reject H0. There is not
sufficient evidence to warrant rejection of the claim that the
mean length is equal to 0.75 in. or 3> 4 in. The lengths appear
to be consistent with the package label.
31. H0: m = 98.6°F. H1: m Z 98.6°F. Test statistic: t = -6.611.
Critical values: t = ;1.984 (approximately, assuming a 0.05
significance level). P-value 6 0.01 (Tech: 0.0077). Reject H0.
There is sufficient evidence to warrant rejection of the claim
that the sample is from a population with a mean equal to
98.6°F. The common belief appears to be wrong.
33. H0: m = 100. H1: m Z 100. Test statistic: t = 1.999. Critical
values: t = ;2.040. 0.05 6 P-value 6 0.10 (Tech: 0.0545).
Fail to reject H0. There is not sufficient evidence to warrant re-
jection of the claim that the mean equals 100. Using the alterna-
tive method with 15.0 used for s, we get test statistic z = 2.00,
critical values z = ;1.96, P-value = 0.0456, so we reject H0
and conclude that there is sufficient evidence to warrant rejec-
tion of the claim that the mean equals 100. The conclusions are
different. The alternative method does not always yield the same
conclusion as the t test.
35. The same value of 1.666 is obtained.
37. Using Table A-3, the power is between 0.90 and 0.95 (Tech:
0.9440). b is between 0.05 and 0.10 (Tech: 0.0560). Because the
power is high, the test is very effective in supporting the claim
that m 7 166.3 lb when the true population mean is 180 lb.
Section 8-6
1. The normality requirement for a hypothesis test of a claim
about a standard deviation is much more strict, meaning that
the distribution of the population must be much closer to a nor-
mal distribution.
3. No. Unlike the situation with sample means, the use of large
samples does not compensate for the lack of normality. The
methods of this section cannot be used with sample data from
a population with a distribution that is far from normal.
5. x2 = 20.612. Critical values: x2 = 12.401 and 39.364.
P-value 7 0.20 (and also less than 0.80). There is not sufficient
evidence to support the alternative hypothesis.
7. x2 = 26.331. Critical value:
x2 = 29.141. 0.01 6 P-value 6 0.025. There is not sufficient
evidence to support the alternative hypothesis.
9. H0: s = 0.0230 g. H1: s 7 0.0230 g. Test statistic:
x2 = 98.260. Critical value of x2 is between 43.773 and
55.758. P-value: 0.0000. Reject H0. There is sufficient evidence
to support the claim that pre-1983 pennies have a standard de-
viation greater than 0.0230 g. Weights of pre-1983 pennies ap-
pear to vary more than those of post-1983 pennies.
11. H0: s = 3.2 mg. H1: s Z 3.2 mg. Test statistic: x2 = 32.086.
Critical values: x2 = 12.401 and x2 = 39.364. P-value: 0.2498.
Fail to reject H0. There is not sufficient evidence to support the
Appendix D 821
claim that the tar content of filtered 100 mm cigarettes has a
standard deviation different from 3.2 mg.
13. H0: s = 2.5 in. H1: s 6 2.5 in. Test statistic: x2 = 2.880.
Critical value: x2 = 2.733. P-value: 0.0583. Fail to reject H0.
There is not sufficient evidence to support the claim that super-
models have heights that vary less than women in the general
population. The supermodels might well have heights that vary
less than women in general, but the sample evidence is not
strong enough to justify that conclusion.
15. H0: s = 10. H1: s Z 10. Test statistic: x2 = 60.938. Critical
values: x2 = 24.433 (approximately) and 59.342 (approxi-
mately). P-value: 0.0277. Reject H0. There is sufficient evidence
to warrant rejection of the claim that pulse rates of women have
a standard deviation equal to 10.
17. H0: s = 1.34. H1: s Z 1.34. Test statistic: x2 = 7.053. Criti-
cal values: x2 = 1.735 and 23.589. P-value: 0.7368. Fail to re-
ject H0. There is not sufficient evidence to support the claim
that the recent winners have BMI values with variation different
from that of the 1920s and 1930s.
19. H0: s = 32.2 ft. H1: s 7 32.2 ft. Test statistic: x2 = 29.176.
Critical value: x2 = 19.675. P-value: 0.0021. Reject H0. There
is sufficient evidence to support the claim that the new produc-
tion method has errors with a standard deviation greater than
32.2 ft. The variation appears to be greater than in the past, so
the new method appears to be worse, because there will be more
altimeters that have larger errors. The company should take im-
mediate action to reduce the variation.
21. 65.673 and 137.957
Chapter 8 Statistical Literacy
and Critical Thinking
1. There is sufficient evidence to support the claim that the mean
is greater than 0 in. In general, a small P-value indicates that the
null hypothesis should be rejected. If the P-value is large (such
as greater than 0.05), there is not sufficient evidence to reject
the null hypothesis.
2. The low P-value indicates that there is sufficient evidence to
support the given claim, so there is statistical significance. There
is not practical significance, because the mean difference is only
0.019 in., which is roughly 1> 50th in. That tiny amount does
not have practical significance.
3. A voluntary response sample (or self-selected sample) is one in
which the respondents themselves decide whether to be in-
cluded. In general, a voluntary response sample cannot be used
with a hypothesis test to make a valid conclusion about the
larger population.
4. A test is robust against departures from normality if it works
quite well even when the sample data are from a population
with a distribution that is somewhat different from a normal
distribution. The t test is robust against departures from nor-
mality, but the x2 test is not.
Chapter 8 Quick Quiz
1. H0: p = 0.5. H1: p 7 0.5.
2. t distribution
3. Chi-square distribution
822 Appendix D

4. True evidence to warrant rejection of the claim that the standard de-
5. 0.1336 viation of such generated values is equal to 12.5.
6. z = -2.04 10. H0: s = 520 lb. H1: s 6 520 lb. Test statistic: x2 = 28.856.
7. t = -2.093 and t = 2.093 Critical value: x2 = 14.954 (approximately). P-value: 0.4233.
8. 0.2302 (Tech: 0.2301) Fail to reject H0. There is not sufficient evidence to support the
9. There is not sufficient evidence to support the claim that the claim that the standard deviation of the weights of cars is less
population proportion is greater than 0.25. than 520 lb.
10. False

Chapter 8 Cumulative Review Exercises

Chapter 8 Review Exercises 1. a. 10.907 sec
1. H0: p = 1>4. H1: p 6 1>4. Test statistic: z = -0.77. Critical b. 10.940 sec
value: z = -1.645. P-value: 0.2206. Fail to reject H0. There is c. 0.178 sec
not sufficient evidence to support the claim that less than 1> 4 of d. 0.032 sec2
adults between the ages of 18 and 44 smoke. e. 0.540 sec
2. H0: p = 0.5. H1: p 7 0.5. Test statistic: z = 3.06. Critical 2. a. Ratio
value: z = 2.33. P-value: 0.0011. Reject H0. There is suffi- b. Continuous
cient evidence to support the claim that among college stu- c. No
dents seeking bachelor’s degrees, most earn that degree within d. Changing pattern over time
five years. e. Time series plot
3. H0: m = 3700 lb. H1: m 6 3700 lb. Test statistic: t = -1.068. 3. 10.770 sec 6 m 6 11.043 sec. The confidence interval should
Critical value: t = -2.453. P-value 7 0.10 (Tech: 0.1469). not be used to estimate winning times, because the listed values
Fail to reject H0. There is not sufficient evidence to support the show a pattern of decreasing times, so there isn’t a fixed popula-
claim that the mean weight of cars is less than 3700 lb. When tion mean.
considering weights of cars for the purpose of constructing a 4. H0: m = 11 sec. H1: m 6 11 sec. Test statistic: t = - 1.576.
road that is strong enough, the weight of the heaviest car is Critical value: t = -1.860. 0.05 6 P-value 6 0.10 (Tech:
more relevant than the mean weight of all cars. 0.0768). Fail to reject H0. There is not sufficient evidence to
4. H0: m = 3700 lb. H1: m 6 3700 lb. Test statistic: z = -1.03. support the claim that the mean winning time is less than
Critical value: z = -2.33. P-value = 0.1515. Fail to reject H0. 11 sec. Because of the changing pattern over time, which shows
There is not sufficient evidence to support the claim that the a trend of decreasing times, future winning times are likely to be
mean weight of cars is less than 3700 lb. less than 11 sec.
5. H0: p = 0.20. H1: p 6 0.20. Test statistic: z = -4.81. Critical 5. a. Yes
value: z = -2.33. P-value: 0.0001 (Tech: 0.0000). Reject H0. b. 100
There is sufficient evidence to support the claim that fewer than c. 0.02 gram
20% of adults consumed herbs within the past 12 months. d. 0.86 gram
6. H0: m = 281.8 lb. H1: m 6 281.8 lb. Test statistic: e. No
t = -8.799. Critical value: t = -2.345 (approximately). 6. By using a vertical scale that does not begin with 0, differences
P-value 6 0.005 (Tech: 0.0000). Reject H0. There is sufficient among the outcomes are exaggerated.
evidence to support the claim that the thinner cans have a 7. The mean is 3.47.
mean axial load less than 281.8 lb. Given the values of the sam-
Outcome Frequency
ple mean and standard deviation, the thinner cans appear to
have axial loads that can easily withstand the top pressure of 1 16
158 lb to 165 lb. 2 13
7. H0: m = 74. H1: m Z 74. Test statistic: z = 0.32. Critical val- 3 22
ues: z = ;1.96. P-value: 0.7490. Fail to reject H0. There is not 4 21
sufficient evidence to warrant rejection of the claim that the 5 13
6 15
mean is equal to 74. The calculator appears to be working
correctly. 8. 0.0025
8. H0: m = 74. H1: m Z 74. Test statistic: t = 0.342. Critical 9. a. 0.1190 (Tech: 0.1186)
values: t = ;1.984 (approximately). P-value 7 0.20 (Tech: b. 37.07% (Tech: 36.94%)
0.7332). Fail to reject H0. There is not sufficient evidence to c. 0.0268 (Tech: 0.0269)
warrant rejection of the claim that the mean is equal to 74. The d. 781.2 mm (Tech: 781.3 mm)
calculator appears to be working correctly. 10. The sample appears to be from a normally distributed popula-
9. H0: s = 12.5. H1: s Z 12.5. Test statistics: x2 = 86.734. tion. A histogram is roughly bell-shaped. A normal quantile plot
Critical values: x2 = 74.222 and 129.561 (approximately). has 20 points that are close to a straight-line pattern, and there
P-value: 0.3883. Fail to reject H0. There is not sufficient is no other pattern. There are no outliers.

Chapter 9 Answers
Section 9-2
1. The sample is a convenience sample, not a simple random sam-
ple. For the sample of males, the requirement of at least 5 suc-
cesses and at least 5 failures is not satisfied.
3. pN 1 = 15>1583 or 0.00948; pN 2 = 8>157 or 0.0510; p =
23>1740 or 0.0132. The symbol p1 denotes the proportion of
all Zocor users who experience headaches. The symbol p2 repre-
sents the proportion of headaches among all people treated with
a placebo.
5. 1396
7. a. p = 27>63 = 0.429
b. z = - 1.25
c. ;1.96
d. 0.2112 (Tech: 0.2115)
9. E = 0.245; -0.403 6 p 1 - p 2 6 0.0878
11. H0: p 1 = p 2. H1: p 1 Z p 2. Test statistic: z = -0.73. Critical
values: z = ;1.96. P-value: 0.4638. Fail to reject H0. There is
not sufficient evidence to warrant rejection of the claim that the
rate of infections is the same for those treated with Lipitor and
those given a placebo.
13. H0: p 1 = p 2. H1: p 1 6 p 2. Test statistic: z = -2.25. Critical
value: z = - 1.645. P-value: 0.0122 (Tech: 0.0123). Reject H0.
There is sufficient evidence to support the claim that the pro-
portion of college students using illegal drugs in 1993 was less
than it is now.
15. 0.00558 6 p 1 - p 2 6 0.0123. Because the confidence inter-
val limits do not include 0, it appears that the two fatality rates
are not equal. Because the confidence interval limits include
only positive values, it appears that the fatality rate is higher for
those not wearing seat belts. The use of seat belts appears to be
effective in saving lives.
17. H0: p 1 = p 2. H1: p 1 Z p 2. Test statistic: z = 2.04. Critical val-
ues: z = ;1.96. P-value: 0.0414 (Tech: 0.0416). Reject H0.
There is sufficient evidence to support the claim that the per-
centage of women who agree is different from the percentage of
men who agree. There does appear to be a difference in the way
that women and men feel about the issue.
19. H0: p 1 = p 2. H1: p 1 7 p 2. Test statistic: z = 0.89. Critical
value: z = 1.645. P-value: 0.1867 (Tech: 0.1858). Fail to
reject H0. There is not sufficient evidence to support the
claim that the proportion of wins at home is higher with a
closed roof than with an open roof. The closed roof does not
appear to be a significant advantage. (But Houston lost this
World Series.)
21. -0.0798 6 p 1 - p 2 6 0.149. Because the confidence interval
limits do contain 0, there is not a significant difference between
the two proportions. Echinacea does not appear to have a signif-
icant effect.
23. H0: p 1 = p 2. H1: p 1 6 p 2. Test statistic: z = -8.47. Critical
value: z = -2.33. P-value: 0.0001 (Tech: 0.0000). Reject H0.
There is sufficient evidence to support the claim that the rate of
Norovirus sickness on the Freedom of the Seas is less than the rate
on the Queen Elizabeth II. When a Norovirus outbreak occurs
on cruise ships, the proportions of infected passengers can vary
considerably.
Appendix D 823
25. -0.0364 6 p 1 - p 2 6 0.138 (Tech: -0.0364 6 p 1 -
p 2 6 0.139). Because the confidence interval limits do contain
0, there is not a significant difference between the two propor-
tions. There does not appear to be a significant difference be-
tween the success rates of men and women.
27. H0: p 1 = p 2. H1: p 1 Z p 2. Test statistic: z = 11.35. Critical
values: z = ;2.575. P-value: 0.0002 (Tech: 0.0000). Reject H0.
There is sufficient evidence to support the claim that the female
survivors and male survivors have different rates of thyroid
diseases.
29. Using x 1 = 504 and x 2 = 539: -0.0497 6 p 1 - p 2 6 0.0286.
Using pN 1 = 0.69 and pN 2 = 0.70: -0.0491 6 p 1 - p 2 6
0.0291. Because the confidence interval limits do contain 0,
there is not a significant difference between the two propor-
tions. It appears that males and females agree.
31. Using x 1 = 69 and x 2 = 22: H0: p 1 = p 2. H1: p 1 7 p 2. Test
statistic: z = 6.37. Critical value: z = 2.33. P-value: 0.0001
(Tech: 0.0000). Reject H0. There is sufficient evidence to sup-
port the claim that the percentage of returns designating the $3
for the campaign was greater in 1976 than it is now.
33. Using x 1 = 117 and x 2 = 29: H0: p 1 = p 2. H1: p 1 7 p 2. Test
statistic: z = 7.60. Critical value: z = 2.33. P-value: 0.0001
(Tech: 0.0000). Reject H0. There is sufficient evidence to sup-
port the claim that the proportion of headaches is greater for
those treated with Viagra. Headaches do appear to be a concern
for those who take Viagra.
35. 0.0131 6 p 1 - p 2 6 0.0269. Because the confidence interval
limits do not contain 0, there is a significant difference between
the two proportions. Because the confidence interval includes
only positive values, it appears that the proportion of women is
greater than the proportion of men. Because the sample is a vol-
untary response sample, the confidence interval is not necessarily
a good estimate of the general population.
37. a. 0.0227 6 p 1 - p 2 6 0.217; because the confidence inter-
val limits do not contain 0, it appears that p 1 = p 2 can be
rejected.
b. 0.491 6 p 1 6 0.629; 0.371 6 p 2 6 0.509; because the
confidence intervals do overlap, it appears that p 1 = p 2 can-
not be rejected.
c. H0: p 1 = p 2. H1: p 1 Z p 2. Test statistic: z = 2.40. P-value:
0.0164. Critical values: z = ;1.96. Reject H0. There is suffi-
cient evidence to reject p 1 = p 2.
d. Reject p 1 = p 2. Least effective: Using the overlap between
the individual confidence intervals.
39. H0: p 1 - p 2 = 0.15. H1: p 1 - p 2 Z 0.15. Test statistic:
z = 2.36. Critical values: z = ;2.575. P-value: 0.0182. Fail to
reject H0. There is not sufficient evidence to warrant rejection of
the claim that the rate of thyroid disease among female atom
bomb survivors is equal to 15 percentage points more than that
for male atom bomb survivors.
Section 9-3
1. 1.6 6 m1 - m2 6 12.2
3. 98%
5. Independent
7. Dependent
824 Appendix D
9. H0: m1 = m2. H1: m1 Z m2. Test statistic: t = -0.452. Critical
values: t = ;2.014 (Tech: t = ;1.987). P-value 7 0.20
(Tech: 0.6521). Fail to reject H0. There is not sufficient evidence
to warrant rejection of the claim that the two groups are from
populations with the same mean. This result suggests that the
increased humidity does not help in the treatment of croup.
11. 6.2 mg 6 m1 - m2 6 9.6 mg (Tech: 6.3 mg 6 m1 - m2 6
9.5 mg). The confidence interval limits do not include 0, which
suggests that the mean tar content of unfiltered king size ciga-
rettes is greater than the mean for filtered 100 mm cigarettes.
13. H0: m1 = m2. H1: m1 6 m2. Test statistic: t = - 5.137. Critical
value: t = -1.660 approximately (Tech: t = -1.653).
P-value 6 0.005 (Tech: 0.0000). Reject H0. There is sufficient
evidence to support the claim that the cents portions of the
check amounts have a mean that is less than the mean of the
cents portions of the credit card charges. One reason for the dif-
ference might be that many check amounts are for payments
rounded to the nearest dollar, so there are disproportionately
more cents portions consisting of 0 cents.
15. H0: m1 = m2. H1: m1 7 m2. Test statistic: t = 10.343. Critical
value: t = 2.896 (Tech: t = 2.512). P-value 6 0.005 (Tech:
0.0000). Reject H0. There is sufficient evidence to support the
claim that supermodels have heights with a mean that is greater
than the mean of women who are not supermodels.
17. -4.6 ft 6 m1 - m2 6 7.0 ft (Tech: -4.4 ft 6 m1 - m2 6
6.8 ft). The confidence interval includes 0 ft, which suggests
that the two population means could be equal. There does not
appear to be a significant difference between the two means.
19. H0: m1 = m2. H1: m1 Z m2. Test statistic: t = -0.721. Criti-
cal values: t = ;2.064 (Tech: t = ;2.011). P-value 7 0.20
(Tech: 0.4742). Fail to reject H0. There is not sufficient evi-
dence to support the claim that menthol cigarettes and non-
menthol cigarettes have different amounts of nicotine. The
menthol content does not appear to have an effect on the
nicotine content.
21. H0: m1 = m2. H1: m1 6 m2. Test statistic: t = - 59.145. Criti-
cal value: t = -2.426 (Tech: t = -2.379). P-value 6 0.005
(Tech: 0.0000). Reject H0. There is sufficient evidence to sup-
port the claim that the mean FICO score of borrowers with
high-interest mortgages is lower than the mean FICO score of
borrowers with low-interest mortgages. It appears that the
higher FICO credit rating scores are associated with lower mort-
gage interest rates.
23. H0: m1 = m2. H1: m1 7 m2. Test statistic: t = 1.678. Critical
value: t = 1.717 (Tech: t = 1.682). P-value 7 0.05 but less
than 0.10 (Tech: 0.0504). Fail to reject H0. There is not suffi-
cient evidence to support the claim that the unsuccessful appli-
cants are from a population with a greater mean age than the
mean age of successful applicants. Based on the result, there
does not appear to be discrimination based on age.
25. H0: m1 = m2. H1: m1 7 m2. Test statistic: t = 2.790. Critical
value: t = 2.390 (approximately). P-value is less than 0.005
(Tech: 0.0031). Reject H0. There is sufficient evidence to sup-
port the claim that heavy users have a lower mean than light
users. Because marijuana appears to adversely affect mental abil-
ities, it should be a serious concern.
27. H0: m1 = m2. H1: m1 7 m2. Test statistic: t = 0.132. Critical
value: t = 1.729. P-value 7 0.10 (Tech: 0.4480.) Fail to reject
H0. There is not sufficient evidence to support the claim that the
magnets are effective in reducing pain. It is valid to argue that the
magnets might appear to be effective if the sample sizes are larger.
29. a. H0: m1 = m2. H1: m1 6 m2. Test statistic: t = -2.335. Crit-
ical value: t = -1.833 (Tech: t = -1.738). P-value is be-
tween 0.01 and 0.025 (Tech: 0.0160). Reject H0. There is
sufficient evidence to support the claim that recent winners
have a lower mean BMI than the mean BMI from the 1920s
and 1930s.
b. -2.50 6 m1 - m2 6 -0.30 (Tech: -2.44 6 m1 - m2 6
-0.36). The confidence interval includes only negative val-
ues, which suggests that recent winners have a lower mean
BMI than the mean BMI from the 1920s and 1930s.
31. a. H0: m1 = m2. H1: m1 6 m2. Test statistic: t = -1.812. Crit-
ical value: t = -2.650 (Tech: t = -2.574). P-value 7 0.025
but less than 0.05 (Tech: 0.0441). Fail to reject H0. There is
not sufficient evidence to support the claim that popes have a
mean longevity after election that is less than the mean for
British monarchs after coronation.
b. -23.6 years 6 m1 - m2 6 4.4 years (Tech: -23.2 years 6
m1 - m2 6 4.0 years). The confidence interval includes 0
years, which suggests that the two population means could be
equal.
33. a. H0: m1 = m2. H1: m1 7 m2. Test statistic: t = 2.746. Critical
value: t = 2.718 (Tech: t = 2.445). P-value 6 0.01 (Tech:
0.0049). Reject H0. There is sufficient evidence to support the
claim that movies with ratings of PG or PG-13 have a higher
mean gross amount than movies with R ratings.
b. 0.8 million dollars 6 m1 - m2 6 153.2 million dollars
(Tech: 8.4 million dollars 6 m1 - m2 6 145.5 million
dollars). The confidence interval includes only positive val-
ues, which suggests that movies with ratings of PG or PG-13
have a higher mean gross amount than movies with ratings
of R. The confidence interval suggests that PG and PG-13
movies gross much more than R movies.
35. H0: m1 = m2. H1: m1 Z m2. Test statistic: t = -16.830. Criti-
cal values: t = ; 2.023 (Tech: t = ;1.992). P-value 6 0.01
(Tech: 0.0000). Reject H0. There is sufficient evidence to war-
rant rejection of the claim that the population of home voltages
has the same mean as the population of generator voltages. Al-
though there is a statistically significant difference, the sample
means of 123.66 V and 124.66 V suggest that the difference
does not have practical significance. The generator could be
used as a substitute when needed.
37. H0: m1 = m2. H1: m1 Z m2. Test statistic: t = -0.452. Critical
values: t = ;1.987. P-value 7 0.20 (Tech: 0.6521). Fail to re-
ject H0. There is not sufficient evidence to warrant rejection of
the claim that the two groups are from populations with the
same mean. This result suggests that the increased humidity
does not help in the treatment of croup. In this case, the results
are basically the same as in Exercise 9.
39. 6.3 mg 6 m1 - m2 6 9.5 mg. The confidence interval limits
are closer together here than in Exercise 11, so the confidence
interval provides an estimate with a smaller range of values.

41. In the hypothesis test, the test statistic changes from - 1.812
to - 1.083, which is a dramatic change. The P-value changes
from a value between 0.025 and 0.05 to a value greater than
0.10, which is a dramatic change. (Using technology, the
P-value changes from 0.0441 to 0.1493, which is a dramatic
change). The conclusion remains the same in this case, but in-
troduction of an outlier could cause the conclusion to change.
The confidence interval limits change to - 447.5 years and
187.9 years, and they are dramatically different from the origi-
nal confidence interval limits. The outlier causes a large increase
in variation, and that is reflected in the confidence interval that
becomes much wider.
43. H0: m1 = m2. H1: m1 Z m2. Test statistic: t = 15.322. Critical
values: t = ;2.080. P-value| 6 0.01 (Tech: 0.0000). Reject H0.
There is sufficient evidence to warrant rejection of the claim
that the two populations have the same mean.
Section 9-4
1. d = -0.2 min; sd = 8.0 min. md represents the mean of the
differences from the paired data in the population.
3. The methods of this section will produce numerical results, but
they will make no sense. The differences between pulse rates and
cholesterol levels themselves make no sense, because they are
different types of quantities. The methods of this section should
not be used.
5. a. -8.3
b. 0.5
c. -33.000
d. ;3.182
7. -9.0 6 md 6 -7.5
9. H0: md = 0. H1: md Z 0. Test statistic: t = 0.029. Critical val-
ues: t = ;2.776. P-value 7 0.20 (Tech: 0.9783). Fail to reject
H0. There is not sufficient evidence to warrant rejection of the
claim that for students in their freshman year, the mean change
in BMI is equal to 0. The BMI does not appear to change dur-
ing freshman year.
11. H0: md = 0. H1: md 6 0. Test statistic: t = -4.712. Critical
value: t = -1.761. P-value 6 0.005 (Tech: 0.0002). Reject
H0. There is sufficient evidence to support the claim that best
actresses are younger than best actors. Yes.
13. -2.32°F 6 md 6 0.52°F. The confidence interval includes 0oF,
which suggests that the mean of the differences could be 0.
Based on the available data, body temperatures appear to be
about the same at both times.
15. H0: md = 0. H1: md Z 0. Test statistic: t = - 2.712. Critical val-
ues: t = ;2.571. P-value is between 0.02 and 0.05 (Tech: 0.0422).
Reject H0. There is sufficient evidence to warrant rejection of the
claim of no effect. Hospital admissions appear to be affected.
17. -$647.9 6 md 6 $2327.0. The confidence interval includes $0,
which suggests that the mean of the differences could be 0, so there
does not appear to be a significant difference between the costs of
repairing the front end and the costs of repairing the rear end.
19. H0: md = 0. H1: md 7 0. Test statistic: t = 14.104. Critical
value: t = 2.539. P-value 6 0.005 (Tech: 0.0000). Reject H0.
There is sufficient evidence to support the claim that the old
ratings are higher than the new ratings.
Appendix D 825
21. a. H0: md = 0. H1: md Z 0. Test statistic: t = 1.054. Critical
values: t = ;2.023. P-value 7 0.20 (Tech: 0.2983). Fail to
reject H0. There is not sufficient evidence to support the
claim that these paired sample values have differences that are
from a population with a mean of 0 volts. The home voltage
and the UPS voltage appear to be about the same.
b. The gasoline-powered generator is completely independent
of the voltage supplied by the power company to the home,
so the home voltage amounts and the generator voltage
amounts are two independent samples.
23. 6.610 lb 6 md 6 8.424 lb (Tech: 6.611 lb 6 md 6 8.424 lb).
Discarded paper appears to weigh more.
25. The five pairs of data are (97, 171), (116, 196), (116, 191), (165,
207), (191, 224). The answers may differ somewhat due to the
accuracy that the graph provides. H0: md = 0. H1: md Z 0. Test
statistic: t = - 6.286. Critical values: t = ;2.776. P-value 6
0.01 (Tech: 0.0033). Reject H0. There is sufficient evidence to
warrant rejection of the claim that there is no difference be-
tween the right hand and left hand reaction times. There does
appear to be a difference.
Section 9-5
1. Because F is so close to 1, the two sample standard deviations
are very close in value.
3. The F test is sensitive to departures from normal distributions,
which means that it performs poorly by giving bad results when
either or both of the populations does not have a normal distri-
bution. Alternatives are the count five test and the Levene-
Brown-Forsyth test.
5. H0: s1 = s2. H1: s1 7 s2. Test statistic: F = 1.1842. Critical
F value: 2.4034. P-value: 0.3738. Fail to reject H0. There is not
sufficient evidence to support the claim that weights of babies
born to mothers given placebos vary more than weights of ba-
bies born to mothers given zinc supplements.
7. H0: s1 = s2. H1: s1 Z s2. Test statistic: F = 1.5777.
P-value = 0.0028, which is less than the significance level of
0.01, so reject H0. There is sufficient evidence to warrant rejec-
tion of the claim that the two samples are from populations
with the same standard deviation.
9. H0: s1 = s2. H1: s1 Z s2. Test statistic: F = 1.7778. Upper
critical F value is between 1.8752 and 2.0739 (Tech: 1.8907).
P-value: 0.0762. Fail to reject H0. There is not sufficient evi-
dence to warrant rejection of the claim that subjects from both
treatment groups have ages with the same amount of variation.
If treatment groups have different characteristics, comparisons
of treatments become unclear, because differences might be due
to the treatments or they might be due to the different group
characteristics.
11. H0: s1 = s2. H1: s1 7 s2. Test statistic: F = 9.3364. Critical
F value is between 2.0540 and 2.0960 (Tech: 2.0842). P-value:
0.0000. Reject H0. There is sufficient evidence to support the
claim that the treatment group has scores that vary more than
the scores of the placebo group.
13. H0: s21 = s22. H1: s21 7 s22. Test statistic: F = 2.1267. The
critical F value is between 2.1555 and 2.2341. (Tech: P-value =
0.0543.) Fail to reject H0. There is not sufficient evidence to
826 Appendix D
support the claim that the pain reductions for the sham treat-
ment group vary more than the pain reductions for the magnet
treatment group.
15. H0: s1 = s2. H1: s1 7 s2. Test statistic: F = 4.1648. Criti-
cal F value is between 2.7876 and 2.8536 (Tech: 2.8179).
P-value: 0.0130. Reject H0. There is sufficient evidence to sup-
port the claim that amounts of strontium-90 from Pennsylvania
vary more than amounts from New York.
17. H0: s1 = s2. H1: s1 Z s2. Test statistic: F = 1.5061. Upper
critical F value is between 2.1540 and 2.2131 (Tech: 2.1812).
P-value: 0.2990. Fail to reject H0. There is not sufficient evi-
dence to warrant rejection of the claim that both samples are
from populations having the same standard deviation.
19. H0: s1 = s2. H1: s1 7 s2. Test statistic: F = 3.4381. Critical
value of F is between 1.6928 and 1.8409 (Tech: 1.7890).
P-value: 0.0003. Reject H0. There is sufficient evidence to sup-
port the claim that near the end of the freshman year, weights of
male college students vary more than weights of female college
students.
21. c1 = 4, c2 = 0, critical value is 5. Fail to reject s21 = s22.
23. 0.2727, 2.8365
Chapter 9 Statistical Literacy
and Critical Thinking
1. A method is robust if the requirement of normally distributed
populations is a somewhat loose requirement in the sense that
the method works well even with populations that do not have
normal distributions, provided that the distribution is not radi-
cally far from being normal. The F test for comparing two pop-
ulation standard deviations or variances is not robust against de-
partures from normal distributions.
2. Nothing. The results are from a voluntary response (or self-
selected) sample, so the results apply only to those who chose to
respond. The results may or may not be representative of the
general population, so methods of statistics should not be used
to make inferences about the general population. To do so
would be a ginormous mistake.
3. The samples are independent. The pairing of the cans is arbi-
trary, so a can of cola within a particular pair is not actually re-
lated to the other can in any way.
4. The different states have different population sizes, so weighted
means should be used.
Chapter 9 Quick Quiz
1. H0: p 1 = p 2. H1: p 1 7 p 2.
2. 0.5
3. z = -1.73
4. 0.0404
5. There is sufficient evidence to support the claim that m1 7 m2.
6. H0: md = 0. H1: md Z 0.
7. H0: m1 = m2 . H1: m1 6 m2.
8. F distribution
9. There is sufficient evidence to support the claim that the two
populations have different means.
10. True
Chapter 9 Review Exercises
1. H0: p 1 = p 2. H1: p 1 7 p 2. Test statistic: z = 3.12. Critical
value: z = 2.33. P-value: 0.0009. Reject H0. There is sufficient
evidence to support a claim that the proportion of successes
with surgery is greater than the proportion of successes with
splinting. When treating carpal tunnel syndrome, surgery
should generally be recommended instead of splinting.
2. H0: m1 = m2. H1: m1 6 m2. Test statistic: t = -2.908. Critical
value: t = -1.653 (approximately). P-value 6 0.005 (Tech:
0.0019). Reject H0. There is sufficient evidence to support the
claim that the children exposed to cocaine have a lower mean score.
3. a. H0: md = 0. H1: md Z 0. Test statistic: t = -1.532. Critical
values: t = ;2.228. P-value is between 0.10 and 0.20 (Tech:
0.1565). Fail to reject H0. There is not sufficient evidence to
warrant rejection of the claim that there is no difference.
There does not appear to be a difference.
b. -2.7 6 md 6 0.5
c. No, there is not a significant difference.
4. H0: p 1 = p 2. H1: p 1 Z p 2. Test statistic: z = -4.20. Critical
values: z = ; 2.575. P-value: 0.0002 (Tech: 0.000). Reject H0.
There is sufficient evidence to warrant rejection of the claim that
the acceptance rate is the same with or without blinding. Without
blinding, reviewers know the names and institutions of the ab-
stract authors, and they might be influenced by that knowledge.
5. H0: m1 = m2. H1: m1 7 m2. Test statistic: t = 5.529. Critical
value: t = 1.796. P-value 6 0.005 (Tech: 0.0000). Reject H0.
There is sufficient evidence to support the claim that Harry Potter
is easier to read than War and Peace. The result is as expected,
because Harry Potter was written for children, but War and
Peace was written for adults.
6. a. 9.5 6 md 6 27.6
b. H0: md = 0. H1: md 7 0. Test statistic: t = 6.371. Critical
value: t = 2.718 (assuming a 0.01 significance level).
P-value 6 0.005 (Tech: 0.0000). Reject H0. There is suffi-
cient evidence to support the claim that the blood pressure
levels are lower after taking captopril.
7. H0: p 1 = p 2. H1: p 1 7 p 2. Test statistic: z = 1.59. Critical
value: z = 1.645. P-value: 0.0559 (Tech: 0.0557). Fail to reject
H0. There is not sufficient evidence to support the claim that
the proportion of men who smoke is greater than the propor-
tion of women who smoke.
8. H0: m1 = m2. H1: m1 6 m2. Test statistic: t = -9.567. Critical
value: t = -2.429. P-value 6 0.005 (Tech: 0.0000). Reject
H0. There is sufficient evidence to support the claim that work-
ers with a high school diploma have a lower mean annual in-
come than workers with a bachelor’s degree. Solving this exercise
contributes to a bachelor’s degree and to a higher income.
9. Using the sample of incomes from workers with bachelor’s de-
grees as sample 1: H0: s1 = s2. H1: s1 Z s2. Test statistic:
F = 2.9460. Upper critical F value is between 1.6141 and
1.8152 (Tech: 1.6935). P-value: 0.0001. Reject H0. There is suf-
ficient evidence to warrant rejection of the claim that the two
samples are from populations with the same standard deviation.
10. H0: s1 = s2. H1: s1 Z s2. Test statistic: F = 1.2874. Upper
critical F value is between 1.4327 and 1.6668 (Tech: 1.5874).
P-value: 0.2825. Fail to reject H0. There is not sufficient
 Appendix D 827

evidence to warrant rejection of the claim that the two samples 9. a.

are from populations with the same amount of variation.

Chapter 9 Cumulative Review Exercises

1. a. The samples are dependent, because the data are from cou-
ples that are paired.
b. x = 14.9 thousand words, median: 14.5 thousand words,
mode: 8 thousand words and 14 thousand words, range: 17.0
thousand words, s = 5.7 thousand words.
c. ratio
2. H0: md = 0. H1: md 6 0. Test statistic: t = -2.163. Critical b. r = 0.816. Critical values: r = ;0.602. P-value = 0.002.
value: t = -1.833. P-value is between 0.025 and 0.05, but less There is sufficient evidence to support the claim of a linear
than 0.05 (Tech: 0.0294). Reject H0. There is sufficient evi- correlation between the two variables.
dence to support the claim that among couples, females are c. The scatterplot reveals a distinct pattern that is not a straight-
more talkative than males. line pattern.
3. H0: m1 = m2. H1: m1 Z m2. Test statistic: t = - 1.990. Critical 11. a. There appears to be a linear correlation.
values: t = ;2.262 (Tech: t = ;2.132). P-value 7 0.05 but b. r = 0.906. Critical values: r = ;0.632 (for a 0.05 signifi-
less than 0.10 (Tech: 0.0652). Fail to reject H0. There is not suf- cance level). There is a linear correlation.
ficient evidence to reject the claim that the two samples are from c. r = 0. Critical values: r = ;0.666 (for a 0.05 significance
populations with the same mean. level).There does not appear to be a linear correlation.
4. 10.8 thousand words 6 m 6 19.0 thousand words or 10,800 d. The effect from a single pair of values can be very substantial,
words 6 m 6 19,000 words and it can change the conclusion.
5. 13. r = 0.985. Critical values: r = ;0.811. P-value = 0.000.
Word Count Frequency
There is sufficient evidence to support the claim of a linear cor-
6–9 3 relation between CPI and cost of a slice of pizza.
10 – 13 0 15. r = 0.867. Critical values: r = ;0.878. P-value = 0.057.
14 – 17 4 There is not sufficient evidence to support the claim of a linear
18 – 21 2
correlation between the systolic blood pressure measurements of
22 – 25 1
the right and left arm.
6. a. 0.3707 (Tech: 0.3694) 17. r = 0.948. Critical values: r = ;0.811. P-value = 0.004.
b. 0.1587 There is sufficient evidence to support the claim of a linear cor-
c. 22,680 (Tech: 22,689) relation between the overhead width of a seal in a photograph
7. 1083 and the weight of a seal.
8. 32.8% 6 p 6 43.3% 19. r = 0.709. Critical values: r = ;0.754. P-value = 0.075.
9. a. 25.0 There is not sufficient evidence to support the claim of a linear
b. 3.5 correlation between the costs of tickets purchased 30 days in ad-
c. 0.9406 (Tech using normal approximation: 0.9401; tech us- vance and those purchased one day in advance.
ing binomial: 0.9405). 21. r = -0.283. Critical values: r = ;0.754. P-value = 0.539.
10. 0.98420 = 0.724. Such an event is not unusual, because the There is not sufficient evidence to support the claim of a linear
probability is not low (such as less than 0.05). correlation between the repair costs from full-front crashes and
full-rear crashes.
23. r = 0.892. Critical values: r = ;0.632. P-value = 0.001.
Chapter 10 Answers
There is sufficient evidence to support the claim of a linear
Section 10-2 correlation between global temperature and the concentration
1. r represents the value of the linear correlation coefficient com- of CO2.
puted by using the paired sample data. r represents the value of 25. r = 0.872. Critical values: r = ;0.754. P-value = 0.011.
the linear correlation coefficient that would be computed by There is sufficient evidence to support the claim of a linear cor-
using all of the paired data in the population. The value of r is es- relation between the given differences (between runs scored and
timated to be 0 (because there is no correlation between points runs allowed) and the proportions of wins.
scored in a football game and the numbers of new cars sold). 27. r = 0.179. Critical values: r = ;0.632. P-value = 0.620.
3. No. The presence of a linear correlation between two variables There is not sufficient evidence to support the claim of a
does not imply that one of the variables is the cause of the other linear correlation between brain size and intelligence.
variable. It does not appear that people with larger brains are more
5. Yes. The value of ƒ 0.758 ƒ is greater than the critical value of intelligent.
0.254 (approximately). 29. r = 0.744. Critical values: r = ;0.335. P-value = 0.000.
7. No. The value of ƒ 0.202 ƒ is less than the critical value 0.312. There is sufficient evidence to support the claim of a linear
828 Appendix D

correlation between movie budget amounts and the amounts RESIDUAL PLOT
that the movies grossed.
31. r = 0.319. Critical values: r = ;0.254 (approximately).
P-value = 0.017. There is sufficient evidence to support the
claim of a linear correlation between the numbers of words spo-
ken by men and women who are in couple relationships.
33. A linear correlation between two variables does not necessarily
mean that one of the variables is the cause of (or directly affects)
the other variable.
35. Averages tend to suppress variation among individuals, so a lin-
ear correlation among averages does not necessarily mean that
there is a linear correlation among individuals.
37. a. 0.942 Section 10-4
b. 0.839
1. It is the standard error of estimate, which is a measure of the dif-
c. 0.9995 (largest)
ferences between the observed weights and the weights predicted
d. 0.983
from the regression equation.
e. -0.958
3. Unlike the predicted value of 180 lb, the prediction interval
provides us with a range of likely weights, so that we have a
sense of how accurate the prediction is likely to be. The termi-
Section 10-3
nology of prediction interval is used for an interval estimate of a
1. yN represents the predicted value of cholesterol. The predictor variable, whereas the terminology of confidence interval is used
variable represents weight. The response variable represents cho- for an interval estimate of a parameter.
lesterol level. 5. 0.762; 76.2%
3. If r is positive, the regression line has a positive slope and it rises 7. 0.748; 74.8%
from left to right. If r is negative, the slope of the regression line 9. r = -0.806 (r is negative because the highway fuel consump-
is negative and it falls from left to right. tion amounts decrease as the weight of the cars increase, as shown
5. 6.40 people by the fact that the slope of the regression line is negative). The
7. 76.3 beats per minute critical value of r is between 0.361 and 0.335 (assuming a 0.05
9. yN = 3.00 + 0.500x. The data have a pattern that is not a significance level). P-value = 0.000. There is sufficient evidence
straight line. to support a claim of a linear correlation between the weights of
11. a. yN = 0.264 + 0.906x cars and their highway fuel consumption amounts.
b. yN = 2 + 0x (or yN = 2) 11. 27.028 mi> gal
c. The results are very different, indicating that one point can 13. a. 2.64829
dramatically affect the regression equation. b. 0.0800433
13. yN = -0.162 + 0.0101x; $1.68 (which might be rounded up c. 2.728333
to the more convenient value of $1.75) d. 0.9706622
15. yN = 43.6 + 1.31x ; 163.2 mm Hg e. 0.1414596
17. yN = -157 + 40.2x ; 205 kg 15. a. 8880.182
19. yN = -1240 + 7.07x ; $690 b. 991.1515
21. yN = 2060 - 0.186x ; $1615. The predicted cost of $1615 is c. 9871.333
very different from the actual cost of $982. d. 0.8995929
23. yN = 10.5 + 0.0109x ; 14.5°. Yes, the predicted temperature is e. 15.74128
the same as the actual temperature. 17. a. $1.72
25. yN = 0.494 + 0.000490x; 0.519. The predicted proportion of b. $1.27 6 y 6 $2.17
wins is reasonably close to the actual proportion of 0.543. 19. a. 205 kg
27. yN = 71.8 + 0.0286x; 103 (or the known mean IQ score of 100) b. 157 kg 6 y 6 253 kg
29. yN = 20.6 + 1.38x; $186 million 21. $1.32 6 y 6 $2.72
31. yN = 13,400 + 0.302x; 15,200 words in a day 23. $0.12 6 y 6 $0.90
33. With b 1 = 0, the regression line is horizontal so that different 25. -0.229 6 b 0 6 0.298; 0.738 6 b 1 6 1.15
values of x result in the same y value, and there is no correlation
between x and y.
35. The residual plot does not suggest that the regression equation is Section 10-5
a bad model, because there is no obvious pattern and the resid- 1. b1 is the coefficient of x1 and b 1 is the coefficient of x1, but b 1 is
ual plot does not become thicker (or thinner). The scatterplot a population parameter obtained by using all of the data in the
suggests that the regression equation is a bad model because the population, whereas b1 is a sample statistic that is an estimate of
points do not fit the pattern of a straight line very well. b 1. The value of b1 is found from the available sample data.

3. No, because the data are categorical (or qualitative). The meth-
ods of this section require quantitative data. (Exceptions can oc-
cur with logistic regression.)
5. Nicotine = 1.59 + 0.0231 Tar - 0.0525 CO, or yN =
1.59 + 0.0231x1 - 0.0525x2, where yN represents the predicted
amount of nicotine, x1 represents the amount of tar, and x2 rep-
resents the amount of carbon monoxide.
7. No. The adjusted R 2 of 0.017 is too small and the P-value of
0.317 is too high. The multiple regression equation is not a
good model for predicting the amount of nicotine.
9. LP (list price), because it has the best combination of lowest
P-value (0.000) and highest adjusted R 2 (0.990).
11. yN = 99.2 + 0.979 LP, because it has the lowest P-value of
0.000 and the highest adjusted R 2 of 0.990.
13. The best regression equation is yN = 0.127 + 0.0878x 1 -
0.0250x 2, where x1 represents tar and x2 represents carbon
monoxide. It is best because it has the highest adjusted R 2 value
of 0.927 and the lowest P-value of 0.000. It is a good regression
equation for predicting nicotine content because it has a high
value of adjusted R 2 and a low P-value.
15. The regression equation yN = 47.9 - 0.00440x 1 - 0.823x 2
(where x1 = weight and x2 = engine displacement) has the
highest adjusted R 2 of 0.644 and the lowest P-value of 0.000.
It would be better to use the regression equation yN = 50.5 -
0.00587x (where x represents weight), because it also has a
P-value of 0.000 and its adjusted R 2 value of 0.639 is only a lit-
tle less than 0.644, but it is better to use a regression equation
with fewer predictor variables.
17. For H0: b 1 = 0, the test statistic is t = 5.486, the P-value is
0.000, and the critical values are t = ;2.110, so reject H0 and
conclude that the regression coefficient of b1 = 0.707 should
be kept. For H0: b 2 = 0, the test statistic is t = -1.292, the
P-value is 0.213, and the critical values are t = ;2.110, so fail
to reject H0 and conclude that the regression coefficient of
b 2 = 0.164 should be omitted. It appears that the regression
equation should include the height of the mother as a predictor
variable, but the height of the father should be omitted.
19. yN = 3.06 + 82.4x 1 + 2.91x 2, where x1 represents sex and x2
represents age. Female: 61 lb; male: 144 lb. The sex of the bear
does appear to have an effect on its weight. The regression
equation indicates that the predicted weight of a male bear is
about 82 lb more than the predicted weight of a female bear
with other characteristics being the same.
Section 10-6
1. The value of R 2 = 1 indicates that the model fits the data per-
fectly. Given the inherent variation and changing economic con-
ditions, it is very unlikely that the model will fit the data per-
fectly, so the claim is likely to be wrong.
3. 8.2% of the variation in Super Bowl points can be explained
by the quadratic model that relates the variable of year and
the variable of points scored. Because such a small percentage
of such variation is explained by the model, the model is not
very useful.
5. Linear: y = 8 + 2x
Appendix D 829
7. Quadratic: d = -16t 2 + 500
9. Exponential: y = 0.158270(1.05935x), where 1960 is coded
as 1. With the high R 2 value of 0.970, the model appears to be
good. The projected value for 2020 is $5.33.
11. Quadratic: y = 0.00802808x 2 + 2.45538x + 15.3223, where
1980 is coded as 1. The quadratic model is not much better
than the linear, exponential, and power models. The projected
number of 87 isn’t too far from the actual number of 92.
13. Quadratic: y = 4.90x 2 - 0.0286x + 0.00476. The model
yields a distance of 705 m, but the building is only 50 m tall, so
the distance cannot exceed 50 m.
15. Quadratic: y = 0.00447053x 2 + 0.0198551x + 13.8205,
where 1950 is coded as 1, 1955 is coded as 2, and so on. The
predicted temperature for 2010 is 14.8°C.
2 (x - 1)
17. a. Exponential: y = 2 3 (or y = (0.629961)(1.587401)x
for an initial value of 1 that doubles every 1.5 years)
b. Exponential: y = (1.33197)(1.43035)x, where 1971 is
coded as 1.
c. Moore’s law does appear to be working reasonably well. With
R 2 = 0.988, the model appears to be very good.
19. a. 6641.8
b. 73.2
c. The quadratic sum of squares of residuals (73.2) is less than
the sum of squares of residuals from the linear model (6641.8).
Chapter 10 Statistical Literacy
and Critical Thinking
1. With the method of correlation discussed in Section 10-2, the
objective is to determine whether there is an association between
two variables. With the methods of Section 9-4, the objective is
to make inferences about the mean of the differences between
the values in the population of matched pairs.
2. There is sufficient evidence to support the claim of a linear cor-
relation between chest size and weight. Although there is a lin-
ear correlation, we cannot conclude that a larger chest size is the
cause of a larger weight.
3. She is not correct. The value of r = 1 indicates that there is suffi-
cient evidence to support a claim of a linear correlation between
the two sets of prices, but that does not necessarily mean that
both sets of prices are the same. For example, if each ring at the
discount company is priced at 50% of the Tiffany price, the value
of r will be 1, but the corresponding prices are very different.
4. The conclusion is not necessarily correct. The value of r = 0
suggests that there is no linear correlation between the two vari-
ables, but there might be some other nonlinear correlation.
Chapter 10 Quick Quiz
1. An error was made in the calculations, because r must always be
between - 1 and 1.
2. There is sufficient evidence to support the claim of a linear cor-
relation between the two variables.
3. True
4. There is not sufficient evidence to support the claim of a linear
correlation between the two variables.
830 Appendix D

5. False. There might be a nonlinear relationship between the two 4. a. The scatterplot suggests that there is not a linear correlation
variables. between upper leg lengths and heights of males.
6. ;0.514
7. r = -1
8. 15
9. 0.160
10. False

Chapter 10 Review Exercises

1. a. The scatterplot suggests that there is not a linear correlation
between the two variables.

b. r = 0.723. Critical values: r = ;0.878 (assuming a 0.05

b. r = -0.254. Critical values: r = ;0.811 (assuming a 0.05
significance level). P-value = 0.627. There is not sufficient
evidence to support a claim of a linear correlation between
8:00 AM temperatures and midnight temperatures.
c. yN = 126 - 0.285x
d. 98.45oF (the value of y)
2. a. r = 0.522, the critical values are r = ;0.312 (assuming a
0.05 significance level), and the P-value is 0.001, so there is
sufficient evidence to support a claim of a linear correlation
between heights and weights of males.
b. 27.2%
c. yN = -139 + 4.55x
d. 189 lb
3. a. The scatterplot shows that as length increases, weight also
tends to increase, so there does appear to be a correlation be-
tween length and weight.
b. r = 0.964. Critical values: r = ;0.878 (assuming a 0.05
significance level). P-value = 0.008. There is sufficient evi-
dence to support a claim of a linear correlation between
lengths of bears and their weights.
c. yN = -468 + 12.3x
d. 418 lb
significance level). P-value = 0.168. There is not sufficient
evidence to support a claim of a linear correlation between
upper leg lengths and heights of males.
c. yN = 97.3 + 1.74x
d. 170.2 cm (the value of y)
5. yN = 140 + 2.50x 1 - 2.27x 2; R 2 = 0.977; adjusted R 2 =
0.954; P-value = 0.023. Using a 0.05 significance level, the
multiple regression equation is a good model and can be used to
predict the height of a male when given his upper leg length and
arm circumference.
Chapter 10 Cumulative Review Exercises
1. From 1877: x = 66.2 in.; median: 66 in.; s = 2.5 in. From re-
cent results: x = 68.1 in.; median: 68.5 in.; s = 3.6 in.
2. H0: m1 = m2. H1: m1 6 m2. Test statistic: t = -1.372. Critical
value: t = -1.833. P-value 7 0.05 (Tech: 0.094). Fail to reject
H0. There is not sufficient evidence to support the claim that
males in 1877 had a mean height that is less than the mean
height of males today.
3. H0: m = 69.1. H1: m 6 69.1. Test statistic: t = -3.690. Criti-
cal value: t = -1.833. P- value 6 0.005 (Tech: 0.0025). There
is sufficient evidence to support the claim that heights of men
from 1877 have a mean less than 69.1 in.
4. 64.4 in. 6 m 6 68.0 in.
5. - 1.2 in. 6 m1 - m2 6 5.0 in. (Tech: - 1.0 in. 6 m1 - m2
6 4.8 in.). The confidence interval limits do include 0, which in-
dicates that the two population means could be the same, so there
is not a significant difference between the two population means.
6. The two sets of sample data are not matched as required, so the
value of the linear correlation coefficient is meaningless in this
situation.
7. a. A statistic is a numerical measurement describing some char-
acteristic of a sample, but a parameter is a numerical measure-
ment describing some characteristic of a population.
b. A simple random sample is one chosen in such a way that
every possible sample of the same size has the same chance of
being chosen.
c. A voluntary response sample is one in which the respondents
themselves decide whether to be included. Results from such
samples are generally unsuitable because people with strong
interests in the topic are more likely to respond, and the re-
sult is a sample that is not representative of the population.

8. Yes. Different explanations are possible, including these: 40 is
more than two standard deviations away from the mean; only
0.26% of BMI values are 40 or above; 40 converts to a z score of
2.8, indicating that it is 2.8 standard deviations above the mean.
9. a. 0.3446
b. 0.0548
10. 0.000207. Because the probability of getting four subjects with
green eyes is so small, it is likely that the researcher did not ran-
domly select the subjects.
Chapter 11 Answers
Section 11-2
1. The random digits should all be equally likely, so the distribu-
tion is uniform. The test for goodness-of-fit is a hypothesis test
that the sample data agree with or fit the uniform distribution
with all of the digits being equally likely.
3. O represents the observed frequencies, and they are 5, 8, 7, 9,
13, 17, 11, 10, 10, 12, 8, 10. E represents the expected frequen-
cies, and the twelve expected frequencies are each 10.
5. Critical value: x2 = 16.919. P-value 7 0.10 (Tech: 0.516).
There is not sufficient evidence to warrant rejection of the claim
that the observed outcomes agree with the expected frequencies.
The slot machine appears to be functioning as expected.
7. Test statistic: x2 = 70.160. Critical value: x2 = 7.815.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that the four categories are equally
likely. The results appear to support the expectation that the fre-
quency for the first category is disproportionately high.
9. Test statistic: x2 = 6.320. Critical value: x2 = 7.815.
P-value 7 0.05 (Tech: 0.097). There is not sufficient evidence
to warrant rejection of the claim that the four categories are
equally likely. The results do not support the expectation that
the frequency for the first category is disproportionately high.
11. Test statistic: x2 = 5.860. Critical value: x2 = 11.071. (Tech:
P-value = 0.3201.) There is not sufficient evidence to support
the claim that the outcomes are not equally likely. The out-
comes appear to be equally likely, so the loaded die does not ap-
pear to behave differently from a fair die.
13. Test statistic: x2 = 13.193. Critical value: x2 = 16.919.
P-value 7 0.10 (Tech: 0.154). There is not sufficient evidence
to warrant rejection of the claim that the likelihood of winning
is the same for the different post positions. Based on these re-
sults, post position should not be considered when betting on
the Kentucky Derby race.
15. Test statistic: x2 = 1159.820. Critical value: x2 = 19.675.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that UFO sightings occur in the
different months with equal frequency. July and August have the
highest frequencies, and those summer months are times when
people are outdoors more than other months.
17. Test statistic: x2 = 15.822. Critical value: x2 = 7.815.
P-value 6 0.005 (Tech: 0.001). There is sufficient evidence to
warrant rejection of the claim that the the observed frequencies
agree with the proportions that were expected according to prin-
ciples of genetics.
Appendix D 831
19. Test statistic: x2 = 6.682. Critical value: x2 = 11.071 (assum-
ing a 0.05 significance level). P-value 7 0.10 (Tech: 0.245).
There is not sufficient evidence to warrant rejection of the claim
that the color distribution is as claimed.
21. Test statistic: x2 = 3650.251. Critical value: x2 = 20.090.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that the leading digits are from a
population with a distribution that conforms to Benford’s law. It
does appear that the checks are the result of fraud.
23. Test statistic: x2 = 49.689. Critical value: x2 = 20.090.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that the leading digits are from a
population with a distribution that conforms to Benford’s law.
The contribution amounts do not appear to be legitimate.
25. The test statistic changes from 11.250 to 247.280. (Tech: The
P-value changes from 0.259 to 0.000.) The effect of the outlier
is dramatic.
Section 11-3
1. P-value = 0.0000000000173. Because the P-value is so low, we
should reject the claim that getting paralytic polio is indepen-
dent of whether the child was treated with the Salk vaccine or
was given a placebo. The Salk vaccine appears to be effective.
3. The P-value of 0.0000000000173 is the probability of getting
sample results at least as extreme as those given in the contin-
gency table, assuming that getting paralytic polio is independent
of whether a child was treated with the Salk vaccine or was given
a placebo.
5. Fail to reject the null hypothesis that home> visitor wins are in-
dependent of the sport. It appears that the home-field advantage
does not depend on the sport.
7. Test statistic: x2 = 2.235. Critical value: x2 = 3.841.
P-value 7 0.10 (Tech: 0.135). There is not sufficient evidence
to warrant rejection of the claim of independence between suc-
cess in challenges and sex of the player. Neither sex appears to be
more successful.
9. Test statistic: x2 = 25.571. Critical value: x2 = 3.841.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that whether a subjects lies is in-
dependent of the polygraph test indication. The results suggest
that polygraphs are effective in distinguishing between truths
and lies, but there are many false positives and false negatives, so
they are not highly reliable.
11. Test statistic: x2 = 0.751. Critical value: x2 = 3.841.
P-value 7 0.10 (Tech: 0.386). There is not sufficient evidence
to warrant rejection of the claim of independence between the
type of restoration and adverse health conditions. Amalgam
restorations do not appear to affect health conditions.
13. Test statistic: x2 = 42.557. Critical value: x2 = 3.841.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that the sentence is independent
of the plea. The results encourage pleas for guilty defendants.
15. Test statistic: x2 = 9.750. Critical value: x2 = 6.635.
P-value 6 0.005 (Tech: 0.002). There is sufficient evidence to war-
rant rejection of the claim that success is independent of the type
of treatment. The results suggest that the surgery treatment is better.
832 Appendix D
17. Test statistic: x2 = 0.792. Critical value: x2 = 5.991.
P-value 7 0.10 (Tech: 0.673). There is not sufficient evidence
to warrant rejection of the claim that the sex of the respondent
is independent of the choice for the cause of global warming.
Men and women appear to generally agree.
19. Test statistic: x2 = 42.568. Critical value: x2 = 9.210.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that experiencing an adverse reaction
in the digestive system is independent of the treatment group.
Treatments with Campral appear to be associated with a decrease
in adverse effects on the digestive system.
21. Test statistic: x2 = 0.773. Critical value: x2 = 7.815.
P-value 7 0.10 (Tech: 0.856). There is not sufficient evidence
to warrant rejection of the claim that getting an infection is in-
dependent of the treatment. The atorvastatin treatment does not
appear to have an effect on infections.
23. Test statistic: x2 = 51.458. Critical value: x2 = 6.635.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that the proportions of agree> dis-
agree responses are the same for the subjects interviewed by men
and the subjects interviewed by women. It appears that the gen-
der of the interviewer affected the responses of women.
25. Test statistics: x2 = 2.234562954 and z = 1.494845462, so
that z 2 = x2. Critical values: x2 = 3.841 and z = ;1.96, so
that z 2 = x2 (approximately).
Section 11-4
1. Neither format appears to be better. Considering the results
from the discordant cases, we see that the frequencies of 31 and
33 are quite close, indicating that students performed about the
same with both formats.
3. When trying to determine whether there is a significant
difference between the two different formats, we are not helped
by subjects whose results from the two different formats are the
same, as in the mastery> mastery results and the nonmastery>
nonmastery results. We can detect a difference by analyzing
those cases in which one of the formats resulted in mastery
while the other format did not, as in the discordant pairs.
5. 1017
7. 652
9. b, c
11. 6.635
13. Test statistic: x2 = 0.237. Critical value: x2 = 3.841.
P-value 7 0.10 (Tech: 0.626). There is not sufficient evidence
to warrant rejection of the claim that the following two propor-
tions are the same: (1) the proportion of subjects with no hip
fracture on the protected hip and a hip fracture on the unpro-
tected hip; (2) the proportion of subjects with a hip fracture on
the protected hip and no hip fracture on the unprotected hip.
The hip protectors do not appear to be effective.
15. Test statistic: x2 = 2.382. Critical value: x2 = 3.841. (Tech:
P-value = 0.123.) There is not sufficient evidence to warrant re-
jection of the null hypothesis that the following two proportions
are the same: (1) The proportion of subjects with no cure on the
fungicide-treated foot and a cure on the foot treated with a
placebo; (2) the proportion of subjects with a cure on the
fungicide-treated foot and no cure on the foot treated with a
placebo. The fungicide treatment does not appear to be effective.
17. Test statistic: x2 = 6.750. Critical value: x2 = 3.841 (assuming
a 0.05 significance level). (Tech: P-value = 0.009.) Reject the
null hypothesis that the following two proportions are the same:
(1) The proportion of tumors with incorrect staging from MRI
and correct staging from PET> CT; (2) the proportion of tumors
with correct staging from MRI and incorrect staging from PET>
CT. The PET> CT technology appears to be more accurate.
19. The uncorrected test statistic is 11.267. The uncorrected value is
greater than the corrected value of 9.600. The conclusion is the
same in this case. There could be cases in which the uncorrected
test statistic leads to rejection of the null hypothesis while the
corrected test statistic does not.
21. When rounded to three decimal places, the same P-value of
0.289 is obtained. With a P-value of 0.289, fail to reject the null
hypothesis that the following two proportions are the same:
(1) The proportion of subjects with one foot cured with the
Pedacream treatment while the other foot is not cured with the
Fungacream treatment; (2) the proportion of subjects with one
foot not cured with the Pedacream treatment while the other
foot is cured with the Fungacream treatment. There does not
appear to be a significant difference between the two treatments.
Chapter 11 Statistical Literacy
and Critical Thinking
1. The numbers are frequency counts for the six different categories
corresponding to the six different cells in the table. The categories
are nausea with Celebrex, nausea with Ibuprofen, and so on.
2. The frequency counts are categorized according to two different
variables: (1) whether the subject experienced nausea; (2) whether
the subject was treated with Celebrex, Ibuprofen, or a placebo.
3. No, we can only conclude that nausea is associated with the
treatment, but we cannot attribute the cause of the nausea to the
treatment.
4. If we assume that experiencing or not experiencing nausea is in-
dependent of the treatment (Celebrex, Ibuprofen, placebo), the
expected number of subjects who experience nausea with the
Celebrex treatment is 160.490.
Chapter 11 Quick Quiz
1. H0: p 1 = p 2 = p 3 = p 4 = p 5 = p 6 = p 7. H1: At least one of
the proportions is different from the others.
2. Observed: 40; expected: 30.857
3. x2 = 12.592
4. There is not sufficient evidence to warrant rejection of the claim
that fatal DWI crashes occur equally on the different days of the
week.
5. Right-tailed
6. Chi-square
7. H0: Response is independent of whether the person responding
is a worker or senior-level boss.
H1: Response and whether the person is a worker or senior-level
boss are dependent.
8. x2 = 3.841

9. There is sufficient evidence to warrant rejection of the claim
that response is independent of whether the person is a worker
or senior-level boss. Response appears to be somehow associated
with whether the person is a worker or senior-level boss.
10. There is not sufficient evidence to warrant rejection of the claim
that response is independent of whether the person is a worker or
senior-level boss. Response and whether the person is a worker or
senior-level boss appear to be independent of each other.
Chapter 11 Review Exercises
1. Test statistic: x2 = 9.294. Critical value: x2 = 5.991. P-
value 6 0.01 (Tech: 0.0096). There is sufficient evidence to war-
rant rejection of the claim of independence between experiencing
nausea and the type of treatment. The adverse reaction of nausea
does not appear to be about the same for the different treatments.
2. Test statistic: x2 = 36.366. Critical value: x2 = 16.812.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that deaths from lightning occur
on the days of the week with equal frequency. Sunday appears to
have disproportionately more deaths from lightning, and that
might be explained by the fact that many people engage in out-
door recreational activities on Sundays.
3. Test statistic: x2 = 51.270. Critical value: x2 = 9.488.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that that the participants fit the
same distribution as the U.S. population. If study participants
are not representative of the population, the results might be
misleading because some groups might have cancer rates differ-
ent from others, and they might skew the outcomes.
4. Test statistic: x2 = 10.732. Critical value: x2 = 3.841.
P-value 6 0.005 (Tech: 0.001). There is sufficient evidence to
warrant rejection of the claim that whether a subject smokes is
independent of whether the subject was treated with bupropion
hydrochloride or a placebo. It appears that the bupropion hy-
drochloride treatment is effective in the sense that it is better
than a placebo, but that the treatment is not highly effective be-
cause many of those in the treatment group continued to smoke.
5. Test statistic: x2 = 16.264. Critical value: x2 = 3.841.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that the two given proportions are
the same. The results indicate that when reporting about cough-
ing of children, there is substantial disagreement between the
children and their parents.
Chapter 11 Cumulative Review Exercises
1. a. Observational study
b. Discrete
c. Statistics
d. The organizations releasing the data have a special interest in
the topic and those who conducted the research may have
been influenced by them.
2. H0: p 1 = p 2. H1: p 1 Z p 2. Test statistic: z = - 20.35. Critical
values: z = ;1.96. P-value: 0.0002 (Tech: 0.000). Reject H0.
There is sufficient evidence to warrant rejection of the claim
that the proportion of men who wash their hands is equal to the
proportion of women who wash their hands. There is a
significant difference between men and women.
Appendix D 833
3. Test statistic: x2 = 414.230. Critical value: x2 = 3.841.
P-value 6 0.005 (Tech: 0.000). There is sufficient evidence to
warrant rejection of the claim that hand washing is independent
of gender. It appears that whether a person washes their hands is
related to the gender of the person.
4. First round: x = 71.2, median: 71.5, range = 8.0, s = 3.2.
Fourth round: x = 70.3, median: 69.5, range = 4.0, s = 1.8.
The fourth round scores are slightly lower and closer together.
5. r = -0.406. Critical values: r = ;0.811. P-value = 0.425.
There is not sufficient evidence to support the claim of a linear cor-
relation between the first round scores and the fourth round scores.
6. 67.8 6 m 6 74.5. We have 95% confidence that the limits of
67.8 and 74.5 contain the true value of the mean first round
golf score. This means that if we were to repeat the first round
under the same conditions and construct the 95% confidence
interval from the six scores, 95% of the confidence intervals
would contain the true population mean score.
7. 0.828 6 p 6 0.932. Given that such a high percentage of se-
nior executives believe that a thank-you note is helpful, it would
be wise to send a thank-you note after every job interview.
8. H0: p = 0.75. H1: p 7 0.75. Test statistic: z = 3.68. Critical
value: z = 2.33. P-value: 0.0001. Reject H0. There is sufficient
evidence to support the claim that more than 75% of all senior
executives believe that a thank-you note after a job interview in-
creases the chances of the applicant.
9. a. 0.2033 (Tech: 0.2040)
b. 0.0005
c. The result from part (a) is more relevant. The designers need
to consider the upper leg lengths of individual men who will
occupy the cockpit; a cockpit will never be occupied by a
group of 16 men.
10. 0.677. It is not unusual, because the probability is not low, such
as 0.05 or less.
Chapter 12 Answers
Section 12-2
1. a. The data are categorized according to the one characteristic
or factor of epoch.
b. The terminology of analysis of variance refers to the method
used to test for equality of the three means. That method is
based on two different estimates of the population variance.
3. We should reject the null hypothesis that the three epochs have
the same mean skull breadth. There is sufficient evidence to con-
clude that at least one of the means is different from the others.
5. Test statistic: F = 9.4695. Critical value of F is approximately
3.3158 (Tech: 3.2849). P-value: 0.000562. Reject H0: m1 =
m2 = m3. There is sufficient evidence to warrant rejection of the
claim that the three books have the same mean Flesch Reading
Ease score.
7. Test statistic: F = 0.3521. Critical value: F = 2.6626. P-value:
0.7877. Fail to reject H0: m1 = m2 = m3 = m4. There is not
sufficient evidence to warrant rejection of the claim that the
mean weight loss is the same for the diets. The diets do not ap-
pear to be very effective.
834 Appendix D
9. Test statistic: F = 5.313. Critical value of F is approximately
3.0718 (Tech: 3.0738). P-value: 0.010. Reject H0: m1 =
m2 = m3. There is sufficient evidence to warrant rejection of the
claim that PG movies, PG-13 movies, and R movies have the
same mean gross amount.
11. Test statistic: F = 0.3974. Critical value: F = 3.3541. P-value:
0.6759. Fail to reject H0: m1 = m2 = m3. There is not sufficient
evidence to warrant rejection of the claim that the different car
categories have the same mean. These data do not suggest that
larger cars are safer.
13. Test statistic: F = 27.2488 . Critical value: F = 3.8853 .
P-value: 0.0000. Reject H0: m1 = m2 = m3. There is sufficient
evidence to warrant rejection of the claim that the three differ-
ent miles have the same mean time. These data suggest that the
third mile appears to take longer, and a reasonable explanation is
that the third lap has a hill.
15. Test statistic: F = 18.9931. Critical value of F is approximately
3.1504 (Tech: 3.1239). P-value: 0.0000. Reject H0: m1 = m2 = m3.
There is sufficient evidence to warrant rejection of the claim
that the three different types of cigarettes have the same mean
amount of nicotine. Given that the king size cigarettes have
the largest mean of 1.26 mg per cigarette, compared to the
other means of 0.87 mg per cigarette and 0.92 mg per
cigarette, it appears that the filters do make a difference
(although this conclusion is not justified by the results from
analysis of variance).
17. The Tukey test results show that the mean from small cars is
significantly different from the mean from large cars. The con-
clusion is the same as that obtained with the Bonferroni test
results.
Section 12-3
1. The data are categorized using the two different variables of sex
and age bracket.
3. Because each cell contains the same number of sample values
(five), this is a balanced design.
5. Test statistic: F = 0.24. P-value: 0.625. Fail to reject the null
hypothesis of no interaction effect. There is not sufficient evi-
dence to warrant rejection of the claim that heights are not af-
fected by an interaction between sex and age bracket. There does
not appear to be an effect from an interaction between sex and
age bracket.
7. Test statistic: F = 2.95. P-value: 0.091. Fail to reject the null
hypothesis of no effect from age bracket. There is not sufficient
evidence to support the claim that age bracket (under 30, over
30) has an effect on height.
9. Test statistic: F = 2.25. P-value: 0.159. Fail to reject the null
hypothesis of no effect from the type of car. There is not
sufficient evidence to support the claim that whether the
car is foreign or domestic has an effect on head injury
measurements.
11. Test statistic: F = 3.73. P-value: 0.0291. Reject the null hy-
pothesis of no interaction effect. There is sufficient evidence to
warrant rejection of the claim that measurements of self-esteem
are not affected by an interaction between the subject’s self-
esteem and the target’s self-esteem. There does appear to be an
effect from an interaction between the self-esteem of the subject
and their perceived self-esteem of the target.
13. Test statistic: F = 1.5824. P-value: 0.2132. Fail to reject the
null hypothesis of no effect from self-esteem of the subject.
There is not sufficient evidence to support the claim that the
self-esteem of the subject (low, medium, high) has an effect on
the measurements of self-esteem.
15. For interaction, the test statistic is F = 41.38 and the P-value is
0.000, so there is a significant interaction effect. The rating ap-
pears to be affected by an interaction between the use of the
supplement and the amount of whey.
17. Row factor: Test statistic is F = 0.10 and P-value = 0.771, so
fail to reject the null hypothesis of no effect from the row factor of
supplement; there is not sufficient evidence to support the claim
that ratings are affected by the use of the supplement. Column
factor: Test statistic is F = 1.23 and P-value = 0.434, so fail to
reject the null hypothesis of no effect from the column factor of
the amount of whey; there is not sufficient evidence to support
the claim that ratings are affected by the amount of whey.
Chapter 12 Statistical Literacy
and Critical Thinking
1. Because two of the samples are from the same source at the same
times, they are not independent. One-way analysis of variance
should not be used because it has a strict requirement of inde-
pendent data sets.
2. One-way analysis of variance is used with sample data categorized
according to one factor, but two-way analysis of variance is used
with sample data categorized according to two different factors.
3. No, because the MPAA ratings of PG, PG-13, and R are quali-
tative data, but methods of one-way analysis of variance or two-
way analysis of variance require quantitative data.
4. Using the least expensive cars could easily result in biased sam-
ples consisting of cars that include fewer safety features and are
not as well built, so such biased samples should not be used for
making inferences about all cars.
Chapter 12 Quick Quiz
1. Test a null hypothesis that three or more samples are from pop-
ulations with equal means.
2. Right-tailed
3. Smaller P-values
4. F = 8.98
5. There is sufficient evidence to reject the null hypothesis that the
three populations of grade level reading scores have the same
mean.
6. It appears that at least one of the books has a population mean
grade level reading score that is different from the other popula-
tion means.
7. With one-way analysis of variance, the different samples are cat-
egorized using only one factor, but with two-way analysis of
variance, the sample data are categorized into different cells de-
termined by two different factors.
8. Fail to reject the null hypothesis of no interaction. There does not
appear to be an effect due to an interaction between sex and major.
 Appendix D 835

9. There is not sufficient evidence to support a claim that the length Chapter 12 Cumulative Review Exercises
estimates are affected by the sex of the subject.
1. a. 15.5, 13.1, 22.7
10. There is not sufficient evidence to support a claim that the length
b. 9.7, 9.0, 18.6
estimates are affected by the subject’s major.
2. Test statistic: t = -1.383. Critical values: t = ;2.160 (assum-
ing an 0.05 significance level). (Tech: P-value = 0.1860.) Fail
Chapter 12 Review Exercises to reject H0: m1 = m2. There is not sufficient evidence to sup-
1. a. One (type of diet) port the claim that there is a difference between the means for
b. One-way analysis of variance the two groups.
c. Because the P-value is high, it appears that the four samples 3. Normal, because the histogram is approximately bell-shaped.
have means that do not differ by significant amounts. It ap- 4. 12.3 years 6 m 6 18.7 years
pears that the mean ages of the four treatment groups are 5. a. Interval
about the same. b. Continuous
d. A small P-value would indicate that at least one of the treat- c. 97.87°F
ment groups has a mean age that is significantly different d. 98.00°F
from the others, so we would not know if differences from e. 2.40°F
the diet treatments are due to the diets or to differences in f. 0.71°F
age. A small P-value would undermine the effectiveness of g. 0.50 degree 2
the experiment. 6. Rating Frequency
2. Test statistic: F = 54.70. P-value: 0.000. Reject H0: m1 = m2 = 1.0–1.9 1
m3. There is sufficient evidence to warrant rejection of the claim 2.0–2.9 4
that the different car categories have the same mean weight. Be- 3.0–3.9 0
cause the sample means are 3095.0 lb (4 cylinders), 3835.0 lb 4.0–4.9 1
(six cylinders), and 4159.3 (eight cylinders), it does appear that 5.0–5.9 4
cars with more cylinders weigh more. 6.0–6.9 9
3. Test statistic: F = 0.37. P-value: 0.698. Fail to reject the null 7.0–7.9 9
hypothesis of no interaction effect. There is not sufficient evi- 8.0–8.9 7
dence to warrant rejection of the claim that left femur loads are
not affected by an interaction between the type of car (foreign, 7. No. The shape of the histogram is not reasonably close to the
domestic) and size of the car (small, medium, large). There does bell shape of a normal distribution.
not appear to be an effect from an interaction.
4. Test statistic: F = 1.82. P-value: 0.202. Fail to reject the null
hypothesis of no effect from the type of car. There is not sufficient
evidence to support the claim that whether the car is foreign or
domestic has an effect on left femur load measurements.
5. Test statistic: F = 0.48. P-value: 0.632. Fail to reject the null
hypothesis of no effect from the size of the car. There is not suf-
ficient evidence to support the claim that whether the car is
small, medium, or large has an effect on left femur load
measurements. 8. The sample means of 6.63, 6.43, and 6.18 do not appear to
6. Test statistic: F = 0.5010. Critical value of F is approximately be very different. Analysis of variance test: Test statistic is
3.1504 (Tech: 3.1239). P-value: 0.6080. Fail to reject H0: m1 = F = 0.0886 and the P-value is 0.915, so fail to reject H0: m1 =
m2 = m3. There is not sufficient evidence to warrant rejection m2 = m3 and conclude that there is not sufficient evidence to
of the claim that the three different types of cigarettes have the warrant rejection of the claim that PG movies, PG-13 movies,
same mean amount of carbon monoxide. It appears that the fil- and R movies have the same mean viewer rating.
ters do not make a difference in the amount of carbon monoxide. 9. a. r = 0.603. Critical values: r = ;0.878 (assuming a 0.05
7. For interaction, the test statistic is F = 0.8733 and the P-value significance level). P-value = 0.282. There is not sufficient
is 0.3685, so there is no significant interaction effect. For gen- evidence to support the claim of a linear correlation between
der, the test statistic is F = 0.0178 and the P-value is 0.8960, so amounts of discarded paper and plastic.
there is no significant effect from gender. For smoking, the test b. yN = -2.17 + 0.405x
statistic is F = 3.0119 and the P-value is 0.1082, so there is no c. No. Because there is not a linear correlation between the
significant effect from smoking. amounts of discarded paper and plastic, the regression line
8. Test statistic: F = 3.1095. P-value = 0.0506. Using a 0.05 sig- does not fit the data well enough.
nificance level, fail to reject H0: m1 = m2 = m3. There is not 10. Test statistic: x2 = 2.909. Critical value: x2 = 5.991. P-value 7
sufficient evidence to warrant rejection of the claim that the 0.10 (Tech: 0.2335). There is not sufficient evidence to warrant
three groups have the same mean longevity. The survival times rejection of the claim of independence between injury category
appear to have means that are the same. and whether the firearm was a handgun or a rifle or shotgun.
836 Appendix D
Chapter 13 Answers
Section 13-2
1. The sign test does not have a requirement that the sample data
come from a population with a particular distribution.
3. The alternative hypothesis is that the proportion of girls is greater
than 0.5, but the sample proportion of girls is 39> 211, which is
less than 0.5. The test is not necessary because there is no way
that a sample proportion of 39> 211 (or 0.185) could ever support
a claim that the population proportion is greater than 0.5.
5. The test statistic of x = 1 is less than or equal to the critical
value of 2. There is sufficient evidence to warrant rejection of
the claim of no difference. There does appear to be a difference.
7. The test statistic of z = -0.48 is not in the critical region
bounded by z = -1.96 and 1.96. There is not sufficient evi-
dence to warrant rejection of the claim of no difference. There
does not appear to be a difference.
9. The test statistic of x = 2 is less than or equal to the critical
value of 3. There is sufficient evidence to reject the claim of no
difference. Based on the sample data, it appears that when
awards are presented, there is a difference between the age of the
best actress and the best actor.
11. The test statistic of x = 3 is not less than or equal to the critical
value of 1. There is not sufficient evidence to reject the claim of
no difference. There is not sufficient evidence to support the
claim that the time spent depicting the use of alcohol is less than
the time spent depicting the use of tobacco.
13. The test statistic of z = -9.09 is in the critical region bounded
by z = -2.33. There is sufficient evidence to warrant rejection
of the claim of no difference. The YSORT method appears to be
effective in increasing the likelihood of a boy.
15. The test statistic of z = -0.88 is not in the critical region
bounded by z = -1.645. There is not sufficient evidence to war-
rant rejection of the claim of no difference. It appears that women
do not have the ability to predict the sex of their babies.
17. The test statistic of z = -2.37 is not in the critical region bounded
by z = -2.575 and 2.575. There is not sufficient evidence to
warrant rejection of the claim that the median is equal to 5.670 g.
The quarters appear to be minted according to specifications.
19. The test statistic of z = -5.32 is in the critical region bounded
by z = -1.96 and 1.96. There is sufficient evidence to warrant
rejection of the claim that the median amount of Coke is equal
to 12 oz. Consumers are not being cheated because they are
generally getting more than 12 oz of Coke, not less.
21. Second approach: The test statistic of z = -4.29 is in the criti-
cal region bounded by z = -1.645, so the conclusions are the
same as in Example 4. Third approach: The test statistic of
z = -2.82 is in the critical region bounded by z = -1.645, so
the conclusions are the same as in Example 4. The different ap-
proaches can lead to very different results; see the test statistics
of -4.61, -4.29, and -2.82. The conclusions are the same in
this case, but they could be different in other cases.
Section 13-3
1. The sign test converts the sample data to positive and negative
signs, whereas the Wilcoxon signed-ranks test uses ranks, so
conclusions will sometimes be different. Because the Wilcoxon
signed-ranks test uses more information about the data, it is
likely to yield better results.
3. n = 10. The critical value is T = 8.
5. Test statistic: T = 1.5. Critical value: T = 1. Fail to reject the
null hypothesis that the population of differences has a median
of 0. Based on the sample data, it appears that when Friday falls
on the 13th, the numbers of hospital admissions are not affected.
7. Test statistic: T = 3. Critical value: T = 25. Reject the null hy-
pothesis that the population of differences has a median of 0.
Based on the sample data, it appears that when awards are pre-
sented, there is a difference between the age of the best actress
and the best actor. Reject the null hypothesis that the popula-
tion of differences has a median of 0. Based on the sample data,
it appears that when awards are presented, there is a difference
between the age of the best actress and the best actor.
9. Test statistic: T = 13.5. Critical value: T = 8. Fail to reject the
null hypothesis that the population of differences has a median
of 0. There is not sufficient evidence to support the claim that
the time spent depicting the use of alcohol is less than the time
spent depicting the use of tobacco.
11. Convert T = 196 to the test statistic z = -2.88. Critical val-
ues: z = ;2.575. (Tech: P-value = 0.004.) There is sufficient
evidence to warrant rejection of the claim that the median is
equal to 5.670 g. The quarters do not appear to be minted ac-
cording to specifications.
13. Convert T = 15.5 to the test statistic z = -4.82. Critical val-
ues: z = ;1.96. (Tech: P-value = 0.000.) There is sufficient
evidence to warrant rejection of the claim that the median
amount of Coke is equal to 12 oz. Consumers are not being
cheated because they are generally getting more than 12 oz of
Coke, not less.
15. a. 975
n (n + 1)
b. - k
2
Section 13-4
1. Because the samples are small and they are from populations not
having normal distributions, the parametric t test should not be
used.
3. a. 500.5
b. No, we are testing the null hypothesis that the two popula-
tions have the same median, which is not specified.
5. R1 = 692, R2 = 739, mR = 621, sR = 55.723, test statistic:
z = 1.27. Critical values: z = ;1.96. (Tech: P-value = 0.2026.)
Fail to reject the null hypothesis that the populations have the
same median.
7. R1 = 416, R2 = 325, mR = 468, sR = 33.045, test statistic:
z = -1.57. Critical values: z = ;1.96. (Tech: P-value =
0.1156.) Fail to reject the null hypothesis that the populations
have the same median.
9. R1 = 863, R2 = 412, mR = 637.5, sR = 51.539, test statistic:
z = 4.38. Critical value: z = 2.33. (Tech: P-value = 0.0000.)
Reject the null hypothesis that the populations have the same
median. There is sufficient evidence to support the claim that
the nonfiltered king size cigarettes have a median amount of

nicotine that is greater than the median amount of nicotine in
the 100 mm filtered cigarettes.
11. R1 = 487.5, R2 = 142.5, mR = 414, sR = 28.775, test statis-
tic: z = 2.55. Critical value: z = 1.645. (Tech: P-value =
0.0053.) Reject the null hypothesis that the populations have
the same median. There is sufficient evidence to support the
claim that movies with ratings of PG or PG-13 have a higher
median gross amount than movies with R ratings.
13. z = -0.63. The test statistic is the same value with opposite sign.
Section 13-5
1. Because the values from each sample are not matched in any
way with values from the other samples, the samples are
independent.
3. 1 + 4 + 6 + 7.5 + 7.5 + 10 + 12 + 21.5 + 21.5 = 91
5. Test statistic: H = 6.6305. Critical value: x2 = 5.991. (Tech:
P-value = 0.0363.) Reject the null hypothesis of equal medi-
ans. Interbreeding is suggested by the data.
7. Test statistic: H = 1.2239. Critical value: x2 = 5.991. (Tech:
P-value = 0.5423.) Fail to reject the null hypothesis of equal
medians. The available data do not provide sufficient evidence
to suggest that larger cars are safer.
9. Test statistic: H = 20.9247. Critical value: x2 = 5.991. (Tech:
P-value: 0.000.) Reject the null hypothesis of equal medians. It
appears that cars with more cylinders produce larger amounts of
greenhouse gases.
11. Test statistic: H = 27.9098. Critical value: x2 = 5.991. (Tech:
P-value: 0.000.) Reject the null hypothesis of equal medians.
There is sufficient evidence to warrant rejection of the claim that
the three different types of cigarettes have the same median
amount of nicotine. It appears that the filters do make a difference.
13. Corrected H is 5.835 (using ©T = 282 and N = 30), which is
not very different from the uncorrected H of 5.774.
Section 13-6
1. The methods of Section 10-3 should not be used for predic-
tions. The regression equation is based on a linear correlation
between the two variables, but the methods of this section do
not require a linear relationship. The methods of this section
could suggest that there is a correlation with paired data associ-
ated by some nonlinear relationship, so the regression equation
would not be a suitable model for making predictions.
3. The symbol rs is used to represent the rank correlation coeffi-
cient computed from the given sample data, and the symbol rs
represents the rank correlation coefficient of the paired data for
the entire population. The subscript s is used so that the rank
correlation coefficient can be distinguished from the linear cor-
relation coefficient r. The subscript does not represent the stan-
dard deviation s. It is used to honor Charles Spearman, who in-
troduced the rank correlation method.
5. rs = 1. Critical values are -0.886 and 0.886. Reject the null
hypothesis of rs = 0. There is sufficient evidence to support a
claim of a correlation between distance and time.
7. a. ;0.521
b. ;0.521
Appendix D 837
c. ;0.197
d. ;0.322
9. rs = -0.929. Critical values: -0.786, 0.786. Reject the null
hypothesis of rs = 0. There is sufficient evidence to support the
claim of a correlation between the two judges. Examination of
the results shows that the first and third judges appear to have
opposite rankings.
11. rs = -0.007. Critical values: -0.587, 0.587. Fail to reject the
null hypothesis of rs = 0. There is not sufficient evidence to
support the claim of a correlation between the conviction rates
and recidivism rates. Conviction rates do not appear to be re-
lated to recidivism rates.
13. rs = 0.664. Critical values: - 0.618, 0.618. Reject the null hy-
pothesis of rs = 0. There is sufficient evidence to support the
claim of a correlation between quality score and cost. It appears
that higher quality is associated with higher cost, so you can
expect to get higher quality by purchasing a more expensive
LCD TV.
15. rs = 1. Critical values: -0.886, 0.886. Reject the null hypothe-
sis of rs = 0. There is sufficient evidence to conclude that there
is a relationship between overhead widths of seals from pho-
tographs and the weights of the seals.
17. rs = 0.231. Critical values: -0.264, 0.264. Fail to reject the
null hypothesis of rs = 0. There is not sufficient evidence to
conclude that there is a correlation between the numbers of
words spoken by men and women who are in couple relation-
ships.
19. Using Formula 10-1 for the case of ties among ranks, rs = 0.109.
Using the formula for the case of no ties among ranks (even
though there are ties), rs = 0.385. There is a substantial differ-
ence between the two results. The first result is better because it is
exact (except for rounding), whereas 0.385 is not exact. Using a
0.05 significance level, the critical values are -0.274 and 0.274,
so the two results lead to different conclusions.
Section 13-7
1. No. The runs test can be used to determine whether the se-
quence of World Series wins by American League teams and Na-
tional League teams is random, but the runs test does not show
whether the proportion of wins by the American League is sig-
nificantly greater than 0.5.
3. No. There might be other problems with the data selection
process. For example, a voluntary response sample might appear
to be random, but it would not be suitable for most statistical
purposes.
5. n 1 = 9, n 2 = 7, G = 4, critical values: 4, 14. Reject random-
ness. There is sufficient evidence to reject the claim that the gen-
ders of the younger winners occur randomly.
7. n 1 = 19, n 2 = 14, G = 15, critical values: 11, 23. Fail to re-
ject randomness. There is not sufficient evidence to warrant re-
jection of the claim that we elect Democrats and Republicans in
a random sequence.
9. n 1 = 20, n 2 = 10, G = 16, critical values: 9, 20. Fail to reject
randomness. There is not sufficient evidence to reject the claim
that the dates before and after July 1 are randomly selected.
838 Appendix D
11. n 1 = 23, n 2 = 18, G = 16, mG = 21.1951, sG = 3.11307.
Test statistic: z = - 1.67. Critical values: z = ;1.96. (Tech:
P-value = 0.0952.) Fail to reject randomness. The sequence
appears to be random. The runs test suggests that the se-
quence appears to be random, but it does not test for dispro-
portionately more occurrences of one of the two categories, so
the runs test does not suggest that either conference is
superior.
13. The median is 1956, and the value of 1956 is deleted to yield
n 1 = 21, n 2 = 21, G = 2, mG = 22, sG = 3.20061. Test
statistic: z = -6.25. Critical values: z = ;1.96. (Tech:
P-value = 0.0000.) Reject randomness. The sequence does not
appear to be random when considering values above and below
the median. There appears to be an upward trend, so the stock
market appears to be a profitable investment.
15. b. The 84 sequences yield 2 runs of 2, 7 runs of 3, 20 runs of 4,
25 runs of 5, 20 runs of 6, and 10 runs of 7.
c. With P (2 runs) = 2>84, P (3 runs) = 7>84, P (4 runs) = 20>84,
P (5 runs) = 25>84, P (6 runs) = 20>84, and P (7 runs) =
10>84, each of the G values of 3, 4, 5, 6, 7 can easily occur
by chance, whereas G = 2 is unlikely because P (2 runs) is
less than 0.025. The lower critical value of G is therefore 2,
and there is no upper critical value that can be equaled or ex-
ceeded.
d. Critical value of G = 2 agrees with Table A-10. The table
lists 8 as the upper critical value, but it is impossible to get 8
runs using the given elements.
Chapter 13 Statistical Literacy
and Critical Thinking
1. A nonparametric test is a hypothesis test that does not require
that sample data come from a population having a normal dis-
tribution or any other specific distribution. A parametric test is
a hypothesis test with a requirement that the sample data are
from a population having a particular distribution, and the nor-
mal distribution is most common.
2. There is no difference. They are different names for the same
category of hypothesis tests that do not require that popula-
tions have normal distributions or any other particular
distribution.
3. A rank is a number assigned to an individual sample item ac-
cording to its order in the sorted list. The first item is as-
signed a rank of 1, the second item is assigned a rank of 2,
and so on. The given values have ranks of 1.5, 1.5, 5, 4, and
3, respectively.
4. Efficiency is a measure of how much stronger sample evidence
must be for the nonparametric test to produce the same results
as a corresponding parametric test. For example, the sign test
has an efficiency of 0.63, meaning that under the same condi-
tions, the sign test requires 100 sample observations to achieve
the same results as 63 sample observations analyzed with a cor-
responding parametric test. We should use a nonparametric test
when requirements of the corresponding parametric test (such
as the requirement of a normally distributed population) are not
satisfied.
Chapter 13 Quick Quiz
1. 2, 1, 3.5, 3.5, 5
2. The runs test for randomness is used to test the null hypothesis
that sample data in a sequence are in a random order. This test
is based on sample data that have two characteristics.
3. Rank correlation can be used in a wider variety of circumstances
than the parametric method of linear correlation. Rank correla-
tion does not require a normal distribution for any population.
Rank correlation can be used to detect some (not all) relation-
ships that are not linear.
4. When compared to the parametric test, the nonparametric test
requires stronger evidence, such as a larger sample or greater dif-
ferences, in order to reject a null hypothesis.
5. Distribution-free test
6. False
7. Four independent samples
8. Two independent samples
9. Matched pairs
10. Four independent samples
Chapter 13 Review Exercises
1. The test statistic of z = -1.77 is not less than or equal to the
critical value of z = -1.96. Fail to reject the null hypothesis of
p = 0.5. There is not sufficient evidence to warrant rejection of
the claim that in each World Series, the American League team
has a 0.5 probability of winning.
2. The test statistic of x = 1 is less than or equal to the critical
value of 1. Reject the null hypothesis that the median equals
98.6°F. There is sufficient evidence to warrant rejection of the
claim that body temperatures have a median equal to 98.6°F.
3. The test statistic of T = 1.5 is less than or equal to the critical
value of 6. Reject the null hypothesis that the median equals
98.6°F. There is sufficient evidence to warrant rejection of the
claim that body temperatures have a median equal to 98.6°F.
4. n 1 = 8, n 2 = 2, and the number of runs is G = 5. The critical
values are 1 and 6. Fail to reject the null hypothesis of random-
ness. There is not sufficient evidence to warrant rejection of the
claim that the sequence of odd and even digits is random.
5. rs = 0.714. Critical values: -0.738, 0.738. Fail to reject the
null hypothesis of rs = 0. There is not sufficient evidence to
support the claim that there is a correlation between the student
ranks and the magazine ranks. When ranking colleges, students
and the magazine do not appear to agree.
6. Test statistic: x = 2 is not less than or equal to the critical value
of 1. Fail to reject the null hypothesis that the population of dif-
ferences has a median of 0. Based on the sample data, it appears
that the predictions are reasonably accurate, because there does
not appear to be a difference between the actual high tempera-
tures and the predicted high temperatures.
7. The test statistic of T = 11.5 is not less than or equal to the
critical value of 6. Fail to reject the null hypothesis that the pop-
ulation of differences has a median of 0. Based on the sample
data, it appears that the predictions are reasonably accurate, be-
cause there does not appear to be a difference between the actual
high temperatures and the predicted high temperatures.

8. Test statistic: H = 0.559. Critical value: x2 = 5.991. (Tech:
P-value = 0.756.) Fail to reject the null hypothesis of equal me-
dians. The available data do not provide sufficient evidence to
suggest that larger cars are safer.
9. R1 = 273, R2 = 192, mR = 232.5, sR = 24.1091, test statistic:
z = 1.68. Critical values: z = ;1.96. (Tech: P-value = 0.0930.)
Fail to reject the null hypothesis that the populations have the
same median. There is not sufficient evidence to support the
claim that the region becomes cooler as the fall season
progresses.
10. n 1 = 17, n 2 = 13, and the number of runs is G = 12. The
critical values are 10 and 22. Fail to reject the null hypothesis of
randomness. There is not sufficient evidence to warrant rejec-
tion of the claim that the sequence of odd and even tempera-
tures is random.
Chapter 13 Cumulative Review Exercises
1. Mean: 213.2 mg; median: 155.5 mg; range: 469.0 mg; standard
deviation: 150.1 mg; variance: 22,529.5 mg2.
2.
3. r = 0.689. Critical values: r = ;0.632. P-value = 0.0277.
There is sufficient evidence to support the claim of a linear cor-
relation between cholesterol and weight.
4. yN = 104 + 0.158x; 120 lb
5. rs = 0.212. Critical values: -0.648, 0.648. Fail to reject the
null hypothesis of rs = 0. There is not sufficient evidence to
support the claim that there is a correlation between cholesterol
and weight.
6. Test statistic: F = 8.9785. Critical value of F is between
3.2317 and 3.3158 (Tech: 3.2849). P-value: 0.0008. Reject
the null hypothesis of equal population means. There is
sufficient evidence to warrant rejection of the claim that the
three books have the same mean Flesch-Kincaid Grade Level.
The three books do not appear to be written with the same
reading level.
7. Test statistic: H = 13.218. Critical value: x2 = 5.991. (Tech:
P-value = 0.0013.) Reject the null hypothesis of equal medi-
ans. The available data provide sufficient evidence to warrant re-
jection of the claim that the three books have the same median
Flesch-Kincaid Grade Level. The three books do not appear to
be written with the same reading level.
8. H0: m1 = m2. H1: m1 Z m2. Test statistic: t = 1.820. Critical
values: t = ;2.201 (Tech: t = ;2.123). P-value 7 0.05
(Tech: 0.0878). Fail to reject H0. There is not sufficient evi-
dence to reject the claim that the samples of Flesch-Kincaid
Appendix D 839
Grade Level scores from Clancy and Rowling have the
same mean.
9. R1 = 175.5, R2 = 124.5, mR = 150, sR = 17.32051, test sta-
tistic: z = 1.47. Critical values: z = ;1.96. (Tech: P-value =
0.1410.) Fail to reject the null hypothesis that the samples of
Flesch-Kincaid Grade Level scores from Clancy and Rowling are
from populations having the same median.
10. There must be an error, because the rates of 13.7% and 10.6%
are not possible with samples of size 100.
Chapter 14 Answers
Section 14-2
1. No. If the process is within statistical control, we know that the
cans are being filled in such a way that the mean and range do
not vary too much, but being within statistical control indicates
nothing about whether the product specifications are being met.
It is possible to be within statistical control while underfilling all
of the cans by a substantial amount.
3. Random variation is the type of variation inherent in the process
that is not capable of filling every can of Coke with exactly 12
ounces. There are different examples of assignable variation, so
answers will vary, but one example is the changing amounts of
Coke that result from machinery that becomes worn through
extensive use.
5. a. Within statistical control
b. Not applicable
c. The variation is too large, so that some cans are overfilled too
much while others are underfilled too much.
7. a. Out of statistical control.
b. There are 8 consecutive points all below the centerline. There
are also 8 consecutive points all above the centerline.
c. The process does not appear to be behaving as it should.
There appears to be a shift in the process values, with all of
the low values occurring at the beginning.
9. x = 319.26, R = 6.69. R chart: LCL = 1.49, UCL = 11.89.
x chart: LCL = 317.20, UCL = 321.32.
11. The process variation is out of statistical control. All three crite-
ria for statistical stability are violated. There is a pattern of in-
creasing values that is not random, there are points lying beyond
the control limits, and there are 8 points lying below the
centerline.
13. x = 2781.71, R = 1729.38. R chart: LCL = 0,
UCL = 3465.67. x chart: LCL = 1946.42, UCL = 3617.00.
840 Appendix D

15. The process mean is within statistical control. 9. The process appears to be within statistical control.

17. x = 123.66, R = 0.375. R chart: LCL = 0, UCL = 0.79. x
chart: LCL = 123.45, UCL = 123.88.
19. There is no pattern suggesting that the process is not within sta-
tistical control. Based on the runs chart, the process appears to
be within statistical control.
11. The process is within statistical control. College enrollments are
fluctuating between 60% and 68%; ideally, the rate of college
enrollments would be much higher.

13. The process is not within statistical control, because there are
points lying beyond the upper and lower control limits, and
21. The s chart is very similar to the R chart. Because the s chart there is a nonrandom pattern of proportions that consistently
uses sample standard deviations instead of ranges, it is much fall, then rise. It appears that higher proportions of people vote
more difficult to construct than the R chart. in years with a presidential election. Overall, the proportions of
voters are inexcusably low. Less than 60% of those eligible actu-
ally vote, and that percentage should be substantially higher.

Section 14-3
15. Except for the vertical scale, the control chart is identical to the
1. No. It is possible that the proportions of defects in repeated
one included with Example 1.
samplings behave in a way that makes the process within statisti-
cal control, but the actual proportions of defects could be very
high, such as 90%, so that almost all of the tablets fail to meet
the manufacturing specifications.
3. The LCL of -0.022 is replaced by 0, because it is impossible to
have negative values of proportions. UCL = 0.062.
5. Within statistical control
7. The process appears to be out of control because there is a pat-
tern of an upward trend, there is a point lying beyond the upper
control limit, and there are 8 consecutive points that are all
above the centerline.
 Appendix D 841

Chapter 14 Statistical Literacy

and Critical Thinking
1. Statistical process control consists of methods used to monitor
data over time and to ensure that the data do not go out of sta-
tistical control by meeting any of the specific out-of-control cri-
teria being used.
2. Due to a wide variety of causes, such as changing employees
and worn machinery, it is possible that the manufacturing
process could result in far too many defective tablets with
too much or too little of the drug atorvastatin. If the process
is not monitored, the manufacturer Pfizer could be forced
2. R = 62; LCL = 4.7; UCL = 119.3
to halt production of Lipitor, with the result that Pfizer
3. The variation of the process is within statistical control because
would lose some or all of the annual $13 billion income
none of the three out-of-control conditions are met.
from that drug.
3. The process could go out of statistical control because of a chang-
ing mean, or increased variation, or both.
4. No. The limits of 9.5 mg and 10.5 mg are specifications that are
independent of upper and lower control limits used for x charts
and R charts. It is possible for the manufacturing process to be
within control with almost all tablets containing an unaccept-
able amount of atorvastatin, such as 2 mg instead of some value
near the desired level of 10 mg.

Chapter 14 Quick Quiz

1. Process data are data arranged according to some time sequence. 4. x = 256.95; LCL = 230.97; UCL = 282.93
They are measurements of a characteristic of goods or services 5. The mean of the process is within statistical control because
that result from some combination of equipment, people, mate- none of the three out-of-control conditions are met.
rials, methods, and conditions.
2. Random variation is due to chance, but assignable variation re-
sults from causes that can be identified, such as defective ma-
chinery or untrained employees.
3. There is a pattern, trend, or cycle that is obviously not random.
There is a point lying outside of the region between the upper
and lower control limits. There are eight consecutive points all
above or all below the centerline.
4. An R chart uses ranges to monitor variation, but an x
chart uses sample means to monitor the center (mean)
of a process.
5. No. The R chart shows a point lying beyond the upper control 6. The proportions do not appear to be within statistical
limit. control because there is a pattern of declining proportions.
6. R = 21.2. In general, a value of R is found by first finding the The chart suggests that in recent years, the homicide rate has
range for the values within each individual subgroup; the mean been declining, which is good, but it has leveled off in the
of those ranges is the value of R. last four years.
7. The process mean is within statistical control.
8. x = 6.45. In general, a value of x is found by first finding the
mean of the values within each individual subgroup; the mean
of those subgroup means is the value of x.
9. A p chart is a control chart of the proportions of some attribute,
such as defective items.
10. False

Chapter 14 Review Exercises

1. Based on the run chart from the first 21 axial loads, there does
not appear to be a pattern suggesting that the process is out of 7. The process is out of statistical control, because there is a pat-
statistical control. tern of increasing variation. If left uncorrected, the process will
842 Appendix D
eventually result in production runs with very high proportions
of defective tablets.
Chapter 14 Cumulative Review Exercises
1. a. r = 0.691. Critical values: r = ±0.632. P-value = 0.0270.
There is sufficient evidence to support the claim of a linear cor-
relation between carbon dioxide concentration and temperature.
b. No. The presence of a correlation between two variables does
not imply that one of the variables is the cause of the other.
c. yN = 9.54 + 0.0135x
d. The predicted value of 13.47°C is somewhat close to the ac-
tual temperature of 13.88°C. Because r = 0.691 is not ex-
tremely high, predicted values are not likely to be extremely
accurate.
2. The process is out of statistical control because there are points
lying beyond the upper control limit. Also, there appears to be a
pattern of increasing variation.
3. 0.0256 6 p 6 0.0414
4. H0: p = 0.03. H1: p 7 0.03. Test statistic: z = 0.92. Critical
value: z = 1.645. P-value: 0.1794. Fail to reject H0. There is
not sufficient evidence to support the claim that the rate of de-
fects is greater than 3%.
5. a. 1> 256
b. 1> 256
c. 1> 128
6. a. 15.87%
b. 4.4 in. and 7.6 in.
7. No. The online poll resulted in a voluntary response sample that
represents only those who chose to respond. It is likely that
those who chose to respond are not representative of the larger
population.
8. The values appear to be from a population with a
distribution that is not normal. The presence of 12 zeros
among the 30 values causes a histogram to be skewed to the
right instead of being bell-shaped. A normal quantile plot shows
a pattern of points that is not reasonably close to a straight-line
pattern.
9. x = 66.0 cents; median = 21.5 cents; s = 108.1 cents
10. H0: p = 0.5. H1: p 7 0.5. Test statistic: z = 1.10 (using
pN = 18>30 = 0.6). Critical value: z = 1.645. P-value: 0.1357
(Tech: 0.1367). Fail to reject H0. There is not sufficient evidence
to support the claim that most students have some change in
their possession.