Chapter 9 Selecting the Sample
1) The highlight of a progressive research company, Peanut Labs®, in the text demonstrates
the potential of:
A) maximizing the impact of the sample frame.
B) reducing sample frame errors.
C) reducing the impact of sample errors.
D) tapping into social media user populations.
E) innovative ways to study the entire population.
Answer: D
2) An entire group under study, as specified by the objectives of the research project, is
known as the:
A) population.
B) sample.
C) sample unit.
D) census.
E) study group.
Answer: A
3) Managers, relative to researchers, define the population:
A) the same.
B) in very specific terms.
C) in less specific terms.
D) not at all; researchers do it for them.
E) none of the above
Answer: C
4) A subset of the population that should represent the entire group, is known as a:
A) population.
B) sample unit.
C) census.
D) sample.
E) population set.
Answer: D
5) A sample unit could be a:
A) household.
B) consumer.
C) purchasing agent.
D) all of the above
E) none of the above
Answer: D
6) An accounting of the complete population is known as:
A) a population.
B) a sample.
C) a sample unit.
D) a census.
E) comprehensive accounting.
Answer: D
7) The target population for the U.S Census is:
A) all households in the United States.
B) as many people as can be convinced to fill out the survey.
C) a certain percentage of the population.
D) an equal number of people from each region of the United States.
E) an equal percentage of people from each region of the United States.
Answer: A
8) The population for a market research project in Iowa to determine the cleaning service
needs in companies would be defined as:
A) all the current clients in Iowa.
B) all the potential clients in Iowa.
C) all the companies in Iowa not interested in cleaning service.
D) individuals responsible for cleaning service in companies located in Iowa.
E) all companies in Iowa.
Answer: D
9) A master list of all the sample units in the population is known as the:
A) sample list.
B) population master list.
C) master frame.
D) sample frame.
E) population frame.
Answer: D
10) Sampling error is caused by two factors; one of them is:
A) the size of the population.
B) the method of the sample unit definition.
C) the method of sample selection.
D) the method of population selection.
E) the education of the researcher.
Answer: C
11) If we define our population as all households in the city of Chicago, Illinois, and we use
the Chicago telephone directory from which to draw our sample units, we would likely have:
A) a survey with sample frame error.
B) a representative survey.
C) a survey containing error.
D) a survey with a poorly defined master list.
E) a census.
Answer: A
12) Reasons for taking a sample instead of a census include:
A) cost.
B) inability of research firms to handle huge amounts of data generated by a census on a large
population.
C) There is no good reason; a census is preferred over a sample.
D) A and B
E) B and C
Answer: D
13) Probability samples are those for which the members of the population:
A) can be found for survey purposes.
B) all have a chance of being included in the sample.
C) have an unknown chance of being included in the sample.
D) whose responses may be predetermined with some known probability.
E) are likely to have been sampled before.
Answer: B
14) Regarding calculating the probability of including a sample unit in a probability sample,
which of the following is most accurate?
A) It can be determined with a high level of accuracy.
B) Practically speaking, it may be impossible to calculate the probability precisely.
C) If it cannot be calculated precisely, then the sample is not to be considered a probability
sample.
D) Precise estimates are known and may be found in statistical tables.
E) Only professional statisticians can determine the accuracy.
Answer: B
15) It is impossible to determine the probability when using nonprobability methods because:
A) the selection technique is subjective.
B) the population size is known.
C) the population size is unknown.
D) the selection technique is objective.
E) there is no true selection technique.
Answer: C
16) Which of the following is NOT true of nonprobability samples?
A) They are sometimes called "haphazard sampling."
B) They can be prone to subconscious biases.
C) They use human intervention.
D) The chances of any one population member being selected into the sample can be
computed.
E) There is no way to determine the probability of a population member being selected into
the sample, even if the population size is known.
Answer: D
17) Which of the following sample plans would not be considered a probability sample?
A) simple random sample
B) systematic sample
C) cluster sample
D) judgment sample
E) stratified sample
Answer: D
18) Using a table of random numbers to draw a simple random sample requires that every
member of the ________ must be uniquely identified (i.e., numbered).
A) population
B) sample
C) numbers
D) tables
E) sample unit
Answer: A
19) If you perform a blind draw in sampling, what type of simple random sampling method
are you using?
A) random digit
B) random numbers
C) random units
D) random device
E) random skip
Answer: D
20) Calculation of the skip interval is important in which sampling plan?
A) simple random sample
B) systematic sample
C) "skip" sample
D) cluster sample
E) interval sample
Answer: B
21) Simple random sampling is an appealing method because:
A) it meets the requirements to obtain a probability sample.
B) it can be used to derive unbiased estimates of population characteristics.
C) it guarantees every population member has an equal chance of being selected.
D) it results in a valid representation of the population.
E) all of the above
Answer: E
22) In practice, simple random sampling should be strongly considered:
A) if the population is small and can be numbered easily.
B) in telephone surveys where random digit dialing programs may be used to generate
numbers.
C) when population units are stored in an electronic file allowing computer programs to
randomly select the sample.
D) All of the above are situations that allow for the practical application of simple random
sampling.
E) Simple random samples are not practical in the real world.
Answer: D
23) Which of the following may be used to overcome the problems of unlisted and new
telephone numbers in a telephone directory?
A) systematic sampling
B) random digit dialing
C) CATI systems
D) a different format
E) an online phone directory
Answer: B
24) Which technique allows you to achieve the benefits of random digit dialing by adding
some randomly determined number to the telephone number drawn from the telephone
directory?
A) plus-one dialing
B) plus random number dialing
C) random "1" digit dialing
D) minus one dialing
E) additive dialing
Answer: A
25) Simple random sampling is not a wise choice when you have:
A) a large population.
B) a small population.
C) a sample frame in electronic format.
D) a large population that is not in electronic format.
E) a small population that is not in electronic format.
Answer: D
26) A sampling company that has worldwide capabilities and many different sampling
options, such as Internet samples, business-to-business samples, and so on, that was
highlighted in the textbook is:
A) Samples-R-Us.
B) Survey Sampling International.
C) World Wide Survey Samples, Inc.
D) Sample Surveys.
E) Sampling Options.
Answer: B
27) One of the advantages of systematic sampling is that:
A) it is more representative in the final analysis.
B) the listing of the population may contain hidden periodicities.
C) the listing of the population can have sample frame error.
D) it has "economic efficiency"; it is less time consuming and easier to select.
E) it has very little problem with sample frame error.
Answer: D
28) One of the reasons systematic sampling is faster than simple random sampling is that the
population size:
A) needs to be counted and numbered.
B) does not need to be known.
C) may be estimated and each element not numbered.
D) cannot be known.
E) It is not faster than simple random sampling.
Answer: C
29) A sampling method that separates the population into different subgroups and then
samples one or some of these subgroups is:
A) cluster sampling.
B) simple sampling.
C) stratified sampling.
D) systematic sampling.
E) grouped sampling.
Answer: A
30) Which of the following is the best definition of cluster sampling?
A) The population is divided into subgroups and a simple random sample is taken from each
subgroup.
B) The population is divided into subgroups; subgroups are combined into clusters, and
subsets of clusters are sampled.
C) The population is divided into subgroups and a single/or several subgroup(s) is/are then
selected from which either a census or a sample is drawn.
D) The population is divided into subgroups and a single/or several subgroup(s) is/are then
selected from which a sample is drawn.
E) The population is divided into subgroups and a systematic sample is taken from each
subgroup.
Answer: C
31) Sampling a geographical area using area sampling is a form of:
A) systematic, geographic sampling.
B) simple random sampling.
C) nonprobability sampling.
D) cluster sampling.
E) geosynchronous sampling.
Answer: D
32) A sampling method that separates the population into different subgroups, and then
samples all of these subgroups, is the definition for:
A) stratified sampling.
B) area sampling.
C) cluster sampling.
D) simple sampling.
E) comprehensive sampling.
Answer: A
33) If a population were not distributed symmetrically, you would use:
A) simple random sampling.
B) systematic sampling.
C) cluster sampling.
D) stratified sampling.
E) unbalanced sampling.
Answer: D
34) If we were to ask college students, "To what extent do you value a college degree?",
which type of sample plan would be most appropriate?
A) Any probability sampling plan would be useful.
B) a stratified sample because we would expect the answers to vary by strata: freshmen,
sophomore, junior, and senior
C) a cluster sample because we would expect the answers to vary by cluster: freshmen,
sophomore, junior, and senior
D) an area sample because we would expect the answers to vary by area: freshmen,
sophomore, junior, and senior
E) an area sample because we would expect the answers to vary by area: freshmen,
sophomore, junior, and senior, with seniors more heavily sampled
Answer: B
35) In stratified sampling, we should always ensure that the sample drawn for each stratum is:
A) equal.
B) proportionate to the stratum's proportion of the total population.
C) of the appropriate size to ensure statistical efficiency.
D) proportionate to the stratum's proportion of the total census.
E) proportionate to the stratum's proportion of the total sample.
Answer: C
36) The purpose of the weighted mean in stratified sampling is:
A) to provide an overall sample mean that represents all the strata.
B) to provide more weight to those strata whose means are the lowest.
C) to provide more weight to those strata whose means are the highest.
D) to provide an overall sample mean that represents the majority of the strata.
E) to provide more weight to those strata whose means are equal.
Answer: A
37) Which of the following should be most helpful in dividing the sample into different
strata?
A) a basis that will result in like responses across strata
B) a basis that will result in different responses across strata
C) a basis that is easy to identify
D) a basis that uses some surrogate indicator of identification such as SIC/NAICS codes
E) a basis that is the least expensive
Answer: B
38) When a proportionate stratified sampling plan is used, the weighted mean formula:
A) should be used to ensure statistical efficiency.
B) should be calculated using the weights represented by the proportions.
C) should not be used.
D) should be used if the sample size is greater than 100.
E) should be used only if time constraints permit.
Answer: C
39) One of the nonprobability methods used for sampling is:
A) cluster sampling.
B) systematic sampling.
C) convenience sampling.
D) stratified sampling.
E) simple random sample.
Answer: C
40) A survey administered to obtain 40 percent Hispanics and 60 percent Europeans is:
A) referral sampling.
B) quota sampling.
C) convenience sampling.
D) judgment sampling.
E) racial sampling.
Answer: B
41) An "educated guess" as to who should represent the population is required for:
A) convenience samples.
B) purposive samples.
C) quota samples.
D) referral samples.
E) hypothetical samples.
Answer: B
42) Snowball sampling is another name for:
A) referral sampling.
B) quota sampling.
C) convenience sampling.
D) purposive sampling.
E) simple random sample.
Answer: A
43) The first step in developing a sample plan is to:
A) obtain a listing of the population.
B) define the relevant population.
C) design the method.
D) access the population.
E) ask the manager how many people the sample should include.
Answer: B
44) When assessing sample frame error, which two factors are to be considered?
A) the cost of the frame versus the accuracy of the frame
B) are people in the frame different from the population, and are people in the population not
on the frame's list
C) the coincidence rate and the bi-incidence rate
D) the cost versus the timeliness of the frame
E) the cost versus the manager's preferences
Answer: B
45) You should obtain a listing of the population after:
A) defining the relevant population.
B) accessing the population.
C) designing the sample plan.
D) drawing the sample.
E) determining the method.
Answer: A
46) Which of the following best illustrates what is meant by validating the sample?
A) buying samples only from a well-known and respected sampling firm
B) using samples that have, in the past, provided reliable and valid results
C) comparing the sample's profile with a known profile such as the census
D) confirming the manager's approval
E) There is no way to "validate" a sample. This is why probability sampling methods are so
important.
Answer: B
47) Marketing researchers make use of samples because obtaining information from every
single person in a market is impractical.
Answer: True
48) A population is defined as the entire group under study as specified by the objectives of
the research project.
Answer: True
49) A census is defined as a subset of the population that somehow represents that entire
group.
Answer: False
50) The sample unit refers to the basic level of investigation; this could be a person,
household, or supermarket.
Answer: True
51) Sampling error is defined as the difference between results obtained from a poll versus a
sample survey.
Answer: False
52) A sample frame is a master list of all the sample units in the population.
Answer: True
53) One of the reasons for taking a sample is that the typical researcher cannot analyze the
huge amounts of data generated by a census.
Answer: True
54) Probability samples are those for which members of the population have a known chance
of completing and actually turning in a survey. It has nothing to do with probability of being
selected into the sample.
Answer: False
55) Nonprobability samples are those for which members of the population do not have a
known chance of completing and turning in a survey. It has nothing to do with the probability
of being selected into the sample.
Answer: False
56) With simple random sampling, the probability of being selected into the sample is found
by dividing the population size by the sample size.
Answer: False
57) The blind draw method, or using a table of random numbers, is most appropriate for
quota samples.
Answer: False
58) Random digit dialing is an example in which simple random sampling is employed quite
successfully.
Answer: True
59) Plus-one dialing procedure is a popular variation of random digit dialing that increases
the number of calls to additional sample members (hence "plus") who may participate in the
survey.
Answer: False
60) Systematic sampling is one of the most prevalent types of sampling techniques used in
place of simple random sampling and its use has been increasing.
Answer: False
61) The skip interval used in systematic sampling is computed by dividing the population list
size by the sample size.
Answer: True
62) One of the differences between systematic sampling and simple random sampling is that
the first one works its way through the entire population from beginning to end. The second
one guarantees that the complete population will be covered, but without a systematic pattern.
Answer: True
63) In area sampling, the one-step approach is applied if the researcher believes that the
various geographic areas (or clusters) are sufficiently identical.
Answer: True
64) In area sampling, the two-step approach is more costly than the one-step approach
because more areas and time are involved.
Answer: True
65) In stratified sampling, the population is separated into different subgroups, and then
samples are taken from all of the subgroups.
Answer: True
66) Stratified samples are most appropriate when the population is known to be "skewed."
Answer: True
67) In stratified sampling, subgroups are referred to as clusters.
Answer: False
68) Stratified samples may use weighted means so that estimates of the overall sample may
be made, as opposed to estimates from each stratum.
Answer: True
69) If the strata sample sizes are faithful to their relative sizes in the population, you have
what is called a proportionate stratified sample design.
Answer: True
70) Mall-intercept companies often use convenience sampling to recruit respondents.
Answer: True
71) If a researcher examined the trading area for a shopping mall and felt it was a good match
with the desired population, the subsequent mall-intercept sample could be called a purposive
sample.
Answer: True
72) Purposive samples require a judgment as to who should be in the sample.
Answer: True
73) A convenience sample may also be called an "exemplar sample."
Answer: False
74) Andy was a senior in college, majoring in marketing. He received a call to participate in a
survey on behalf of a company that was targeting graduating seniors. At the end of the
survey, Andy was asked for the name of a friend who was also a senior in marketing. This is
an example of a referral sample.
Answer: True
75) Quota samples refer to samples obtained by bid.
Answer: False
76) A research study conducted in India for Tata Motors found that the automobile
manufacturer has a high vote of consumer satisfaction in total and across three owner age
groups. However, the authors concluded that because the sample was a convenience sample
taken in a specific city in India, this conclusion cannot be defended as more than 99.8% of
Tata owners were omitted from the sample frame.
Answer: True
77) One of the problems with online survey samples is the mechanics of the sample selection
process.
Answer: True
78) Online panel samples are popular, but there are lingering concerns about the true
representativeness of samples provided by panel companies.
Answer: True
79) One of the greatest pluses of online panels is high response rate.
Answer: True
80) A river sample is created via the use of banners, pop-ups, or other online devices that
invite website visitors to take part in the survey.
Answer: True
81) When developing a sample plan, the last step in the process is to "draw the sample."
Answer: False
82) Samples may be validated only by obtaining samples from well-known, reputable
sampling firms.
Answer: False
83) The data taken from a sample reveals that 45 percent are males and 55 percent are
females, and 10 percent of the sample members earn over $50,000 per year. Comparing these
figures with known data, such as census data for the geographical area defined by the
population, would be a way of validating the sample.
Answer: True
84) Tom and Mary Nesbitt commissioned a local university to determine if there would be
adequate demand for sailing outings on a large catamaran at a local beach frequented by
vacationers. The beach was about 5 miles in length and was lined with hotels and
condominiums that rented rooms and condos to thousands of visitors each year. At present,
there were few water sports available and were limited to "active participant" rides, such as
parasailing or kite sailing. The Nesbitts felt that if there was adequate demand, they may
invest the required $750,000 needed to purchase the 110-foot-long catamaran that could carry
as many as 80 passengers and a crew of eight on each trip. The Nesbitts felt that few of the
persons living in the adjoining city near the beach would be in their target market. Instead,
they felt 95% of their business would come from persons at the beach who would see the
catamaran sailing back and forth and would walk to one of two locations to take the ride.
"We want a sample taken of persons who are actually on the beach during a single weekday
in June," Mary said. Given that you want to have a probability sample, which of the
following sample methods would likely be most appropriate?
A) simple random sample
B) judgment sample
C) stratified sample
D) cluster, area sample
E) systematic sample
Answer: D
85) Eddie Livingston is president of BankTrust, a banking firm with several bank locations.
Eddie has been exploring the impact of service satisfaction with bank performance and is
convinced that he can improve the bank's RONW if he can improve customer satisfaction. He
has been contemplating a proposal from Tanner Research to begin a program of improving
customer satisfaction. Tanner has recommended a survey of bank customers to establish gaps
in service as perceived by the customers. Eddie knows that the bank has a list of all the
customers and that this information is on an electronic database complete with mailing
addresses and, for those having Internet service, email addresses. At the next meeting, Eddie
expresses concern that sending a questionnaire to all bank customers is an inconvenience to
them and he doesn't want to create more hassles. The researchers from Tanner Research
assure Eddie that they can get the needed information for as few as a couple of hundred
customers and that the results will be representative of all bank customers. Which of the
following sample methods would most likely be appropriate?
A) simple random sample
B) judgment sample
C) stratified sample
D) cluster, area sample
E) systematic sample
Answer: A
86) Paul Solomon is the owner of Solly's, an upscale restaurant in Tampa, Florida. Each year,
Paul spends about $150,000 in advertising. As this year's ad budget decision must be made,
Paul decides that he wants to know just how good his advertising dollars are working for him.
He hires Getty Research Associates who recommend that he establish a baseline of
awareness. Getty recommends a TOMA study which stands for Top-of-Mind-Awareness.
TOMA studies consist of taking a probability sample of the population and asking
respondents to name the first three "insert type of business here." Obviously, for Solly's,
Getty would be asking respondents to name the top three restaurants in the area. The
percentage of respondents that named Solly's would be a baseline measure of awareness, and
future promotions could be evaluated in terms of whether or not they increased awareness.
While Paul was intrigued with online surveys because they were fast and less expensive,
Getty recommended a traditional telephone survey using the local Tampa directory. Getty
also explained to Paul that "Plus One dialing" could be used to ensure that unlisted numbers
were included. Though the directory was quite large and no electronic version was available,
Getty felt they could efficiently draw the sample using a:
A) simple random sample.
B) judgment sample.
C) stratified sample.
D) cluster, area sample.
E) systematic sample.
Answer: E
87) Tata Motors of India is marketing a new vehicle that retails for only $2,200. The vehicle
will carry two persons and has a body made of fiberglass. It is designed in component parts
so that if a part is broken, you can easily disassemble the part and replace it with a new one
ordered from Tata. Tata realizes that it will take some time for countries to allow the small
vehicle on public roads due to safety requirements. However, Tata sees an immediate market
among business firms who need inexpensive yet reliable vehicles to transport personnel
around large manufacturing or warehousing facilities. One Tata manager stated: "Ninety
percent of the time, company cars are used to transport only one or two persons and much of
their travel is confined to the plant premises. The new Tata vehicle will not replace the large
company cars but can reduce the number of company cars a firm must buy by a substantial
amount, thus saving tremendous amounts of money." Executives at Tata are interested in a
sales estimate in Great Britain, their first targeted market outside of India. They will construct
a sales forecast based upon a survey of companies and they know that demand will vary
substantially, depending on the size of the company. Small firms will not want any vehicles,
medium-sized firms may want about 5 or 10 vehicles, and large firms may want 20 to 100
vehicles. Secondary data is available that allows Tata to identify all firms in Great Britain and
to know the size of the firm. If a survey is conducted to predict demand for the new Tata
vehicle among business firms, which of the following sampling methods would be most
appropriate?
A) simple random sample
B) judgment sample
C) stratified sample
D) cluster, area sample
E) systematic sample
Answer: C
88) Louise Fulkerson is president of the Student Marketing Association at Baylor University.
Louise and the other officers want to conduct a short survey to determine the type of
programming they should offer for the upcoming year. Their population is defined as
"students in the College of Business." One of the officers suggests that each of them go into
their classes, mostly seniors, and conduct the survey. "We can do it easily and fast, and our
professors will be glad to help the association out by giving us a few minutes to conduct the
survey," says one of the officers. But Louise is worried about the sample. "Most of us are
marketing majors and we are all seniors taking senior classes. I would prefer to ask the
Department Chairperson if we can survey students taking Fundamentals of Marketing
because all majors have to take this course, and most of the future members of the
Association will be coming from these Fundamentals classes." Louise is suggesting which
one of the following sampling methods?
A) quota sample
B) purposive sample
C) stratified sample
D) representative sample
E) systematic sample
Answer: B
89) Leslie Bradley is an entrepreneur with a small business in Little Rock, Arkansas.
Working with a local engineer/inventor, Leslie has developed a new cooking device that she
thinks will revolutionize cooking. The device is a small oven that is portable and uses a gas
injection system to cook meals in a fraction of the time taken in a normal oven. The new
system creates a very high temperature and the gas jets are located 360 degrees around the
food to be cooked, thus ensuring that the food is cooked equally on all sides. Leslie and the
engineer applied for a patent and then immediately sought a relationship with a large
manufacturer to produce and distribute the new oven. To her surprise, when she demonstrated
it to the top three manufacturers, they all had the same reaction: "We are very interested in
this new invention but we would like to see some consumer research that tells us what
consumers think about the device. Will they think the temperature is too high and represents a
safety issue? Do they really think that a roast cooked in 6 minutes will taste the same as one
roasted for 2 hours? How much, if any, will they be willing to pay for the added convenience
of time savings?" Leslie sought the services of Weber Research, Inc. WRI was an established
research firm in the city and they recommended that they use their mall facility in order to
conduct some research that allowed consumers to use the device and taste the food cooked in
the device. Leslie agreed but was concerned about the types of persons in the shopping mall.
"They really won't represent our area since they will likely be mostly female and they will
have higher incomes than the general population." WRI assured Leslie that this could be
overcome by using which of the following sampling methods?
A) quota sample
B) purposive sample
C) stratified sample
D) representative sample
E) systematic sample
Answer: A
Test Bank for Marketing Research
Alvin C. Burns, Ronald F. Bush
9780133074673, 9780134895406, 9780134167404
