,Solution manual for Statistics Unlocking the
Power of Data 3rd Edition Robin H. Lock
Hello all ,
We have all what you need with best price
Our email :
Our website :
testbanks-store.com
,CHAPTER 1 1
Section 1.1 Solutions
1.1 (a) The cases are the people who are asked the question.
(b) The variable is whether each person supports the law or not. It is categorical.
1.2 (a) The cases are the 100 stocks.
(b) The variable is the percentage change, which is a numerical quantity, for each of the stocks. It is
quantitative.
1.3 (a) The cases are the teenagers in the sample.
(b) The variable is the result (yes or no) indicating whether each teenager eats at least five servings a day
of fruits and vegetables. It is categorical.
1.4 (a) The cases are the bunches of bananas in the sample.
(b) The variable is the number of days until the bananas go bad. It is quantitative.
1.5 (a) The 10 beams that were tested.
(b) The force at which each beam broke. It is quantitative.
1.6 (a) The cases are countries of the world.
(b) The variable is whether or not the literacy rate is over 75%. It is categorical.
1.7 Since we expect the number of years smoking cigarettes to impact lung capacity, we think of the number
of years smoking as the explanatory variable and the lung capacity as the response variable.
1.8 Since we expect the amount of fertilizer used to impact the yield (and not the other way around), we
think of the amount of fertilizer as the explanatory variable and the yield of the crop as the response variable.
1.9 Ingesting more alcoholic drinks will cause the level of alcohol in the blood to increase, so the number of
drinks is the explanatory variable and blood alcohol content is the response.
1.10 The world record time will continue to decrease as the years go by so we expect the year to impact
marathon record time. We think of the year as the explanatory variable and the record time as the response
variable.
1.11 (a) Year and HigherSAT are categorical. The other six variables are all quantitative, although Siblings
might be classified as either categorical or quantitative.
(b) There are many possible answers, such as “What proportion of the students are first year students?”
or “What is the average weight of these students?”
(c) There are many possible answers, such as “Do seniors seem to weigh more than first year students?”
or “Do students with high Verbal SAT scores seem to also have high Math SAT scores?”
1.12 (a) In addition to the identification column, Country, there are 24 variables. We see that Developed
is a categorical variable, while the other 23 variables are all quantitative.
(b) There are many possible answers, such as “What is the average life expectancy for all countries of the
world?” or “What proportion of countries are developed?”
,2 CHAPTER 1
(c) There are many possible answers, such as “Do countries with a greater land area have a larger percent
rural?” or “Do countries that spend a relatively large amount on the military spend a relatively small
amount on health care?” or “Do developed countries have a longer life expectancy than developing
countries?”
1.13 (a) The explanatory variable is the type of format in which the story is presented. It is categorical,
with three categories (audio, illustrated, and animated).
(b) The response variable is the measure of brain connectivity. It is quantitative.
(c) The cases are the four-year-olds in the study, so there are 27 cases.
1.14 (a) The cases are female gamers in Great Britain.
(b) There are three variables mentioned: Whether the gamers had received obscene messages (categorical),
how many hours a week they played (quantitative), and whether they felt there were enough strong
female characters in games (categorical).
(c) There are 1151 cases and 3 variables, so the dataset will have 1151 rows and 3 columns.
1.15 (a) The cases are the students in the college physics class.
(b) There are 3 variables: Whether the student was assigned to an active or passive learning class, the
measure of how much the student thought they learned, and the score of actual learning on the test.
The class assignment is categorical, while the other two are quantitative.
(c) The class assignment (active or passive) is the explanatory variable, while the two measures of learning
are response variables.
(d) The students in the physics class are the cases, so there are 154 + 142 = 296 cases. There are three
variables, so the dataset will have 296 rows and 3 columns.
1.16 There are at least two variables. One variable is whether or not the spider engaged in mock-sex. This
variable is categorical and the explanatory variable. Another variable is length of time to reach the point of
real mating once the spider is fully mature. This variable is quantitative and the response variable.
1.17 The individual cases are the lakes from which water samples were taken. For each lake in the sample,
we record the concentration of estrogen in the water and the fertility level of fish. Both are quantitative
variables.
1.18 There are two variables. One variable indicates the presence or absence of the gene variant and the
second variable indicates which of the three ethnic groups the individual belongs to. Both variables are
categorical.
1.19 (a) There are 10 cases, corresponding to the 10 cities. The two variables are population, which is
quantitative, and the hemisphere the city is in, which is categorical.
(b) We need two columns, one for each variable. The columns can be in either order. See the table.
,CHAPTER 1 3
Population Hemisphere
37 Eastern
26 Eastern
23 Eastern
22 Eastern
21 Eastern
21 Eastern
21 Eastern
21 Western
20 Western
19 Western
1.20 (a) There are 7 cases, representing the seven pigeons. There are two variables. One is the sex of the
pigeon, which is categorical, and the other is the speed of the pigeon, which is quantitative.
(b) The dataset will have 7 rows and 2 columns. See the table. (The seven cases can be listed in any
order.)
Sex Speed
Hen 1676
Hen 1452
Hen 1449
Cock 1458
Cock 1435
Cock 1418
Cock 1413
1.21 (a) The cases are the homing pigeons, so there are 1412 of them.
(b) There are 4 variables. Two are categorical (loft, sex) and two are quantitative (distance, speed).
(c) The dataset will have 1412 rows (one for each pigeon) and 4 columns (one for each variable).
1.22 One variable is whether each male was fed a high-fat diet or a normal diet. This is the explanatory
variable and it is categorical. The response variable is whether or not the daughters developed metabolic
syndrome, which is also categorical.
1.23 One variable is whether the young female mice lived in an enriched environment or not. This is the
explanatory variable and it is categorical. The response variable is how fast the offspring learned to navigate
mazes and is quantitative.
1.24 (a) The mode of transportation (e.g., bus, car, walk) gives a categorical value for each student.
(b) The answer about allergies (yes or no) gives a categorical value for each student.
(c) The proportion of students in the sample who are vegetarians is a single number, based on all of the
students, and does not give a value for each individual student. It is not a variable for the dataset.
(d) The number of hours worked at a paid job gives a quantitative value for each student.
(e) The difference in hours of sleep on school nights and non-school nights gives a quantitative value for
each student.
(f) The maximum time to get to school is a single number, based on all of the students, and does not give
a value for each individual student. It is not a variable for the dataset.
,4 CHAPTER 1
(g) The desired super power chosen from a list gives a categorical value for each student.
1.25 (a) The total tuition and fees is a quantitative value for each school.
(b) The number of schools in the Northeast is a single value, based on all of the schools, and does not give
a value for each individual school. It is not a variable for the dataset.
(c) The type of school (public, private, for profit) is a categorical value for each school.
(d) The number of undergraduates is a quantitative value for each school.
(e) The percentage of part-time students is a quantitative value for each school.
(f) The school with the highest average faculty salary is a single value, based on all of the schools, and
does not give a value for each individual school. It is not a variable for the dataset.
1.26 In the first study, the cases are the students. The only variable is whether or not the student has
smoked a hookah. This is a categorical variable.
In the second study, the cases are the people in a hookah bar. The variables are the length of the session,
the frequency of puffing, and the depth of inhalation. All are quantitative.
In the third study, the cases are the smoke samples, and the variables are the amount of tar, nicotine, and
heavy metals. All three variables are quantitative.
1.27 (a) This description of the study mentions six variables: age, nose volume, nose surface area, nose
height, nose width, and gender.
(b) One of the variables (gender) is categorical, and the other five are quantitative.
(c) There are six variables so the dataset will have six columns. The 859 participants are the cases, so the
dataset will have 859 rows.
1.28 (a) The cases are the 47 participants.
(b) The description of the study includes three different variables: the score on the no-distractions test,
the score on the test while texting, and whether or not the student considered him or herself to be
good at multitasking. The two test score variables are quantitative and the multitasking variable is
categorical.
(c) The dataset would have 47 rows (one for each participant) and three columns (one for each of the three
variables).
1.29 (a) The cases are the 40 people with insomnia who were included in the study.
(b) There are two variables. One is which group the person is assigned to, either therapy or not, and the
other is whether or not the person reported sleep improvements. Both are categorical.
(c) The dataset would have two columns, one for each of the two variables, and 40 rows, one for each of
the people in the study.
1.30 If we simply record age in years and income in dollars, the variables are quantitative. Often, however,
in a survey, we don’t ask for the exact age but rather what age category the participant falls in (20 − 29,
30 − 39, etc.). Similarly, we often don’t ask for exact income but for an income category (less than $10,000,
between $10,000 and $25,000, etc.). If we ask participants what category they are in for each variable, then
the variables are categorical.
1.31 We could sample people eligible to vote and ask them each their political party and whether they voted
in the last election. The cases would be people eligible to vote that we collect data from. The variables
would be political party and whether or not the person voted in the last election. Alternatively, we could
ask whether each person plans to vote in an upcoming election.
,CHAPTER 1 5
1.32 We could survey a sample of people and ask their household income and measure happiness in some
way, such as asking how happy they are on a scale of 1–10. The cases would be the people we collect data
from. The variables in this case would be household income and happiness rating, although any two variables
measuring wealth and happiness are possible.
1.33 Answers will vary.
,6 CHAPTER 1
Section 1.2 Solutions
1.34 This is a sample, because only a subset of fish are measured.
1.35 This is a population, because all customers are accounted for.
1.36 This is a population, because all registered vehicles are accounted for.
1.37 This is a sample, because only a subset of college students were sent the questionnaire.
1.38 The sample is the 120 people interviewed. The population might be all people in that town or all
people that go to the mall in that town or a variety of other groups larger than and containing the 120
people in the sample.
1.39 The sample is the five hundred Canadian adults that were asked the question; the population is all
Canadian adults.
1.40 The sample is the 100 customers surveyed; the population is all customers of the cell phone carrier.
1.41 The sample is the 1000 households which have databoxes attached to the televisions. The population
is all US households with televisions.
1.42 (a) The sample is the 100 college students who were asked the question.
(b) The population we are interested in is all Americans.
(c) A population we can generalize to, given our sample, is college students.
1.43 (a) The sample is the 10 selected twitter accounts.
(b) The target population is all twitter accounts.
(c) The population we can generalize to, given the sample, is only twitter accounts of this author’s followers,
since this is the population from which the sample was drawn.
1.44 (a) The sample is the 1500 people who were contacted.
(b) The population we are interested in is all residents of the US.
(c) A population we can generalize to, given our sample, is residents of Minnesota.
1.45 (a) The sample is the girls who are on the selected basketball teams.
(b) The population we are interested in is all female high school students.
(c) A population we can generalize to, given our sample, is female high school students who are on a
basketball team.
1.46 Yes, this is a random sample from the population.
1.47 Yes, this is random sample from the population.
1.48 No, this is not a random sample, because some employees may be more likely than others to actually
complete the survey.
1.49 No, this is not a random sample, because certain segments of the population (e.g., those not attending
college) cannot be selected.
,CHAPTER 1 7
1.50 No, this is not a random sample. We might think we can pick out a “representative sample”, but we
probably can’t. We need to let a random number generator do it for us.
1.51 No, this is not a random sample, this is a volunteer sample, since the only people in the sample are
those that self-select to respond to the online poll.
1.52 This sample is definitely biased because only students who are at the library on a Friday night can be
selected. The random sample should be from all students.
1.53 This is biased because the way the question is worded is not at all objective. Although the sample is
a random sample, the wording bias may distort the results.
1.54 This sample is biased because taking 10 apples off the top is not a random sample. The apples on the
bottom of the truckload are probably more likely to be bruised.
1.55 From the description, it appears that this method of data collection is not biased.
1.56 This sample is biased because it is a volunteer survey in which people choose to participate or not.
Most likely, the people taking the time to respond to the email will have stronger opinions than the rest of
the student body.
1.57 Because this was a random sample of parents in Kansas City, the result can be generalized to all parents
in Kansas City.
1.58 (a) No, the sample is almost certainly not representative, since it is a volunteer sample and only
includes people who visit that website and who chose to participate in the poll.
(b) No, it is not appropriate to generalize since the sample is not representative.
1.59 (a) Yes, the sample is likely to be representative since it is a random sample.
(b) Yes, since the sample is a random sample, we can generalize to the population of all Canadian con-
sumers.
1.60 (a) The sample is the 1, 000 US adults that were contacted. The intended population is all US adults.
(b) Yes, it is reasonable to generalize since the sample was selected randomly.
1.61 (a) The sample is the 800 people who participated in the survey. The intended population is all US
smartphone users.
(b) The cases are the people who participated in the survey. There are two variables: Whether or not a
food delivery app was used in the last month, and which app (if any) was used. Both are categorical.
1.62 (a) The 457 students in the sample is not a larger population. We would know the exact answers for
this group so no need to generalize.
(b) The sample was randomly selected from among all Pennsylvania high school seniors who participated
in the Census at School project, so that would be a reasonable population to generalize results to.
(c) There might be something special about schools (or students) who participate in the Census at School
project, so it might not be reasonable to generalize the results from that sample to all Pennsylvania
high school seniors.
(d) It would not be reasonable to generalize from a sample of only Pennsylvania students to students from
all states.
, 8 CHAPTER 1
1.63 (a) There is no sampling bias here, since we are told that a random sample was used. The results
are being manipulated based only on the order of the options (the wording of the question) so this is
an example of wording bias.
(b) We see that people are more likely to select the first option given, so you should ask your friend to
choose between “Option W and Option Q,“ with Option W presented first.
1.64 (a) Yes, there is probably sampling bias since the group has not shared the method. No, we cannot
know if it is appropriate to generalize, and it is probably not appropriate to do so. The survey might
have been asked of children, for example. This would certainly create sampling bias.
(b) Yes, the way the question was worded almost certainly biased the results. A better way might be to
ask an open-ended question such as describe how chocolate milk is made,
1.65 (a) The individual cases are the over 6000 restroom patrons who were observed. The description
makes it clear that at least three variables are recorded. One is whether or not the person washed
their hands, another is the gender of the individual, and a third is the location of the observation. All
three are categorical.
(b) In a phone survey, people are likely to represent themselves in the best light and not always give
completely honest answers. That is why it is important to also find other ways of collecting data, such
as this method of observing people’s actual habits in the restroom.
1.66 (a) The sample is the survey participants, the population is all professors at the University of Ne-
braska.
(b) No, we cannot conclude that the sample of survey responders is not representative of professors at the
University of Nebraska since we are not given enough information to decide one way or the other.
(c) No, the 94% is based on self descriptions, which can be (and in this case, probably are) biased.
1.67 No. This is a volunteer sample, and there is reason to believe the participants are not representative of
the population. For example, some may choose to participate because they LIKE alcohol and/or marijuana,
and those in the sample may tend to have more experience with these substances than the overall population.
In addition, the advertisements for the study were aired on rock radio stations in Sydney, so only those people
who listen to rock radio stations in Sydney would hear about the option to participate.
1.68 Yes! The sample is a random sample so we can be quite confident that it is probably a representative
sample.
1.69 (a) This is not a simple random sample from the population, since only those who saw and wanted
to click and complete the survey were included.
(b) These results could also have been biased by how the survey was constructed. The wording of the
questions might also introduce bias.
1.70 The study given found a relationship in a sample of rats. This relationship may not generalize to the
human population.
1.71 The sample of planes that return from bombing missions was biased. More bullet holes were found in
the wings and tail because planes that were shot in other regions were more likely to crash and not return.
1.72 (a) The population in the CPS is all US residents. (Also acceptable: US citizens, US households...)
(b) The population in the CES survey is all non-farm businesses and government agencies in the US.
Power of Data 3rd Edition Robin H. Lock
Hello all ,
We have all what you need with best price
Our email :
Our website :
testbanks-store.com
,CHAPTER 1 1
Section 1.1 Solutions
1.1 (a) The cases are the people who are asked the question.
(b) The variable is whether each person supports the law or not. It is categorical.
1.2 (a) The cases are the 100 stocks.
(b) The variable is the percentage change, which is a numerical quantity, for each of the stocks. It is
quantitative.
1.3 (a) The cases are the teenagers in the sample.
(b) The variable is the result (yes or no) indicating whether each teenager eats at least five servings a day
of fruits and vegetables. It is categorical.
1.4 (a) The cases are the bunches of bananas in the sample.
(b) The variable is the number of days until the bananas go bad. It is quantitative.
1.5 (a) The 10 beams that were tested.
(b) The force at which each beam broke. It is quantitative.
1.6 (a) The cases are countries of the world.
(b) The variable is whether or not the literacy rate is over 75%. It is categorical.
1.7 Since we expect the number of years smoking cigarettes to impact lung capacity, we think of the number
of years smoking as the explanatory variable and the lung capacity as the response variable.
1.8 Since we expect the amount of fertilizer used to impact the yield (and not the other way around), we
think of the amount of fertilizer as the explanatory variable and the yield of the crop as the response variable.
1.9 Ingesting more alcoholic drinks will cause the level of alcohol in the blood to increase, so the number of
drinks is the explanatory variable and blood alcohol content is the response.
1.10 The world record time will continue to decrease as the years go by so we expect the year to impact
marathon record time. We think of the year as the explanatory variable and the record time as the response
variable.
1.11 (a) Year and HigherSAT are categorical. The other six variables are all quantitative, although Siblings
might be classified as either categorical or quantitative.
(b) There are many possible answers, such as “What proportion of the students are first year students?”
or “What is the average weight of these students?”
(c) There are many possible answers, such as “Do seniors seem to weigh more than first year students?”
or “Do students with high Verbal SAT scores seem to also have high Math SAT scores?”
1.12 (a) In addition to the identification column, Country, there are 24 variables. We see that Developed
is a categorical variable, while the other 23 variables are all quantitative.
(b) There are many possible answers, such as “What is the average life expectancy for all countries of the
world?” or “What proportion of countries are developed?”
,2 CHAPTER 1
(c) There are many possible answers, such as “Do countries with a greater land area have a larger percent
rural?” or “Do countries that spend a relatively large amount on the military spend a relatively small
amount on health care?” or “Do developed countries have a longer life expectancy than developing
countries?”
1.13 (a) The explanatory variable is the type of format in which the story is presented. It is categorical,
with three categories (audio, illustrated, and animated).
(b) The response variable is the measure of brain connectivity. It is quantitative.
(c) The cases are the four-year-olds in the study, so there are 27 cases.
1.14 (a) The cases are female gamers in Great Britain.
(b) There are three variables mentioned: Whether the gamers had received obscene messages (categorical),
how many hours a week they played (quantitative), and whether they felt there were enough strong
female characters in games (categorical).
(c) There are 1151 cases and 3 variables, so the dataset will have 1151 rows and 3 columns.
1.15 (a) The cases are the students in the college physics class.
(b) There are 3 variables: Whether the student was assigned to an active or passive learning class, the
measure of how much the student thought they learned, and the score of actual learning on the test.
The class assignment is categorical, while the other two are quantitative.
(c) The class assignment (active or passive) is the explanatory variable, while the two measures of learning
are response variables.
(d) The students in the physics class are the cases, so there are 154 + 142 = 296 cases. There are three
variables, so the dataset will have 296 rows and 3 columns.
1.16 There are at least two variables. One variable is whether or not the spider engaged in mock-sex. This
variable is categorical and the explanatory variable. Another variable is length of time to reach the point of
real mating once the spider is fully mature. This variable is quantitative and the response variable.
1.17 The individual cases are the lakes from which water samples were taken. For each lake in the sample,
we record the concentration of estrogen in the water and the fertility level of fish. Both are quantitative
variables.
1.18 There are two variables. One variable indicates the presence or absence of the gene variant and the
second variable indicates which of the three ethnic groups the individual belongs to. Both variables are
categorical.
1.19 (a) There are 10 cases, corresponding to the 10 cities. The two variables are population, which is
quantitative, and the hemisphere the city is in, which is categorical.
(b) We need two columns, one for each variable. The columns can be in either order. See the table.
,CHAPTER 1 3
Population Hemisphere
37 Eastern
26 Eastern
23 Eastern
22 Eastern
21 Eastern
21 Eastern
21 Eastern
21 Western
20 Western
19 Western
1.20 (a) There are 7 cases, representing the seven pigeons. There are two variables. One is the sex of the
pigeon, which is categorical, and the other is the speed of the pigeon, which is quantitative.
(b) The dataset will have 7 rows and 2 columns. See the table. (The seven cases can be listed in any
order.)
Sex Speed
Hen 1676
Hen 1452
Hen 1449
Cock 1458
Cock 1435
Cock 1418
Cock 1413
1.21 (a) The cases are the homing pigeons, so there are 1412 of them.
(b) There are 4 variables. Two are categorical (loft, sex) and two are quantitative (distance, speed).
(c) The dataset will have 1412 rows (one for each pigeon) and 4 columns (one for each variable).
1.22 One variable is whether each male was fed a high-fat diet or a normal diet. This is the explanatory
variable and it is categorical. The response variable is whether or not the daughters developed metabolic
syndrome, which is also categorical.
1.23 One variable is whether the young female mice lived in an enriched environment or not. This is the
explanatory variable and it is categorical. The response variable is how fast the offspring learned to navigate
mazes and is quantitative.
1.24 (a) The mode of transportation (e.g., bus, car, walk) gives a categorical value for each student.
(b) The answer about allergies (yes or no) gives a categorical value for each student.
(c) The proportion of students in the sample who are vegetarians is a single number, based on all of the
students, and does not give a value for each individual student. It is not a variable for the dataset.
(d) The number of hours worked at a paid job gives a quantitative value for each student.
(e) The difference in hours of sleep on school nights and non-school nights gives a quantitative value for
each student.
(f) The maximum time to get to school is a single number, based on all of the students, and does not give
a value for each individual student. It is not a variable for the dataset.
,4 CHAPTER 1
(g) The desired super power chosen from a list gives a categorical value for each student.
1.25 (a) The total tuition and fees is a quantitative value for each school.
(b) The number of schools in the Northeast is a single value, based on all of the schools, and does not give
a value for each individual school. It is not a variable for the dataset.
(c) The type of school (public, private, for profit) is a categorical value for each school.
(d) The number of undergraduates is a quantitative value for each school.
(e) The percentage of part-time students is a quantitative value for each school.
(f) The school with the highest average faculty salary is a single value, based on all of the schools, and
does not give a value for each individual school. It is not a variable for the dataset.
1.26 In the first study, the cases are the students. The only variable is whether or not the student has
smoked a hookah. This is a categorical variable.
In the second study, the cases are the people in a hookah bar. The variables are the length of the session,
the frequency of puffing, and the depth of inhalation. All are quantitative.
In the third study, the cases are the smoke samples, and the variables are the amount of tar, nicotine, and
heavy metals. All three variables are quantitative.
1.27 (a) This description of the study mentions six variables: age, nose volume, nose surface area, nose
height, nose width, and gender.
(b) One of the variables (gender) is categorical, and the other five are quantitative.
(c) There are six variables so the dataset will have six columns. The 859 participants are the cases, so the
dataset will have 859 rows.
1.28 (a) The cases are the 47 participants.
(b) The description of the study includes three different variables: the score on the no-distractions test,
the score on the test while texting, and whether or not the student considered him or herself to be
good at multitasking. The two test score variables are quantitative and the multitasking variable is
categorical.
(c) The dataset would have 47 rows (one for each participant) and three columns (one for each of the three
variables).
1.29 (a) The cases are the 40 people with insomnia who were included in the study.
(b) There are two variables. One is which group the person is assigned to, either therapy or not, and the
other is whether or not the person reported sleep improvements. Both are categorical.
(c) The dataset would have two columns, one for each of the two variables, and 40 rows, one for each of
the people in the study.
1.30 If we simply record age in years and income in dollars, the variables are quantitative. Often, however,
in a survey, we don’t ask for the exact age but rather what age category the participant falls in (20 − 29,
30 − 39, etc.). Similarly, we often don’t ask for exact income but for an income category (less than $10,000,
between $10,000 and $25,000, etc.). If we ask participants what category they are in for each variable, then
the variables are categorical.
1.31 We could sample people eligible to vote and ask them each their political party and whether they voted
in the last election. The cases would be people eligible to vote that we collect data from. The variables
would be political party and whether or not the person voted in the last election. Alternatively, we could
ask whether each person plans to vote in an upcoming election.
,CHAPTER 1 5
1.32 We could survey a sample of people and ask their household income and measure happiness in some
way, such as asking how happy they are on a scale of 1–10. The cases would be the people we collect data
from. The variables in this case would be household income and happiness rating, although any two variables
measuring wealth and happiness are possible.
1.33 Answers will vary.
,6 CHAPTER 1
Section 1.2 Solutions
1.34 This is a sample, because only a subset of fish are measured.
1.35 This is a population, because all customers are accounted for.
1.36 This is a population, because all registered vehicles are accounted for.
1.37 This is a sample, because only a subset of college students were sent the questionnaire.
1.38 The sample is the 120 people interviewed. The population might be all people in that town or all
people that go to the mall in that town or a variety of other groups larger than and containing the 120
people in the sample.
1.39 The sample is the five hundred Canadian adults that were asked the question; the population is all
Canadian adults.
1.40 The sample is the 100 customers surveyed; the population is all customers of the cell phone carrier.
1.41 The sample is the 1000 households which have databoxes attached to the televisions. The population
is all US households with televisions.
1.42 (a) The sample is the 100 college students who were asked the question.
(b) The population we are interested in is all Americans.
(c) A population we can generalize to, given our sample, is college students.
1.43 (a) The sample is the 10 selected twitter accounts.
(b) The target population is all twitter accounts.
(c) The population we can generalize to, given the sample, is only twitter accounts of this author’s followers,
since this is the population from which the sample was drawn.
1.44 (a) The sample is the 1500 people who were contacted.
(b) The population we are interested in is all residents of the US.
(c) A population we can generalize to, given our sample, is residents of Minnesota.
1.45 (a) The sample is the girls who are on the selected basketball teams.
(b) The population we are interested in is all female high school students.
(c) A population we can generalize to, given our sample, is female high school students who are on a
basketball team.
1.46 Yes, this is a random sample from the population.
1.47 Yes, this is random sample from the population.
1.48 No, this is not a random sample, because some employees may be more likely than others to actually
complete the survey.
1.49 No, this is not a random sample, because certain segments of the population (e.g., those not attending
college) cannot be selected.
,CHAPTER 1 7
1.50 No, this is not a random sample. We might think we can pick out a “representative sample”, but we
probably can’t. We need to let a random number generator do it for us.
1.51 No, this is not a random sample, this is a volunteer sample, since the only people in the sample are
those that self-select to respond to the online poll.
1.52 This sample is definitely biased because only students who are at the library on a Friday night can be
selected. The random sample should be from all students.
1.53 This is biased because the way the question is worded is not at all objective. Although the sample is
a random sample, the wording bias may distort the results.
1.54 This sample is biased because taking 10 apples off the top is not a random sample. The apples on the
bottom of the truckload are probably more likely to be bruised.
1.55 From the description, it appears that this method of data collection is not biased.
1.56 This sample is biased because it is a volunteer survey in which people choose to participate or not.
Most likely, the people taking the time to respond to the email will have stronger opinions than the rest of
the student body.
1.57 Because this was a random sample of parents in Kansas City, the result can be generalized to all parents
in Kansas City.
1.58 (a) No, the sample is almost certainly not representative, since it is a volunteer sample and only
includes people who visit that website and who chose to participate in the poll.
(b) No, it is not appropriate to generalize since the sample is not representative.
1.59 (a) Yes, the sample is likely to be representative since it is a random sample.
(b) Yes, since the sample is a random sample, we can generalize to the population of all Canadian con-
sumers.
1.60 (a) The sample is the 1, 000 US adults that were contacted. The intended population is all US adults.
(b) Yes, it is reasonable to generalize since the sample was selected randomly.
1.61 (a) The sample is the 800 people who participated in the survey. The intended population is all US
smartphone users.
(b) The cases are the people who participated in the survey. There are two variables: Whether or not a
food delivery app was used in the last month, and which app (if any) was used. Both are categorical.
1.62 (a) The 457 students in the sample is not a larger population. We would know the exact answers for
this group so no need to generalize.
(b) The sample was randomly selected from among all Pennsylvania high school seniors who participated
in the Census at School project, so that would be a reasonable population to generalize results to.
(c) There might be something special about schools (or students) who participate in the Census at School
project, so it might not be reasonable to generalize the results from that sample to all Pennsylvania
high school seniors.
(d) It would not be reasonable to generalize from a sample of only Pennsylvania students to students from
all states.
, 8 CHAPTER 1
1.63 (a) There is no sampling bias here, since we are told that a random sample was used. The results
are being manipulated based only on the order of the options (the wording of the question) so this is
an example of wording bias.
(b) We see that people are more likely to select the first option given, so you should ask your friend to
choose between “Option W and Option Q,“ with Option W presented first.
1.64 (a) Yes, there is probably sampling bias since the group has not shared the method. No, we cannot
know if it is appropriate to generalize, and it is probably not appropriate to do so. The survey might
have been asked of children, for example. This would certainly create sampling bias.
(b) Yes, the way the question was worded almost certainly biased the results. A better way might be to
ask an open-ended question such as describe how chocolate milk is made,
1.65 (a) The individual cases are the over 6000 restroom patrons who were observed. The description
makes it clear that at least three variables are recorded. One is whether or not the person washed
their hands, another is the gender of the individual, and a third is the location of the observation. All
three are categorical.
(b) In a phone survey, people are likely to represent themselves in the best light and not always give
completely honest answers. That is why it is important to also find other ways of collecting data, such
as this method of observing people’s actual habits in the restroom.
1.66 (a) The sample is the survey participants, the population is all professors at the University of Ne-
braska.
(b) No, we cannot conclude that the sample of survey responders is not representative of professors at the
University of Nebraska since we are not given enough information to decide one way or the other.
(c) No, the 94% is based on self descriptions, which can be (and in this case, probably are) biased.
1.67 No. This is a volunteer sample, and there is reason to believe the participants are not representative of
the population. For example, some may choose to participate because they LIKE alcohol and/or marijuana,
and those in the sample may tend to have more experience with these substances than the overall population.
In addition, the advertisements for the study were aired on rock radio stations in Sydney, so only those people
who listen to rock radio stations in Sydney would hear about the option to participate.
1.68 Yes! The sample is a random sample so we can be quite confident that it is probably a representative
sample.
1.69 (a) This is not a simple random sample from the population, since only those who saw and wanted
to click and complete the survey were included.
(b) These results could also have been biased by how the survey was constructed. The wording of the
questions might also introduce bias.
1.70 The study given found a relationship in a sample of rats. This relationship may not generalize to the
human population.
1.71 The sample of planes that return from bombing missions was biased. More bullet holes were found in
the wings and tail because planes that were shot in other regions were more likely to crash and not return.
1.72 (a) The population in the CPS is all US residents. (Also acceptable: US citizens, US households...)
(b) The population in the CES survey is all non-farm businesses and government agencies in the US.
