Chapter 6: Traditional Survey Research
Multiple Choice
1. Which of the following is not a reason surveys have a high rate of usage in
marketing research?
a. the need to know why certain things are done or not done
b. the need to know who does certain things
c. the need to explain cause and effect relationships
d. the need to know how consumers make decisions
e. All are reasons for the high rate of usage of surveys in marketing research.
2. Collecting information such as consumer demographics and lifestyle
characteristics would be illustrative of which reason for the high usage of surveys
in marketing research?
a. the need to know why certain things are done or not done
b. the need to know who does certain things
c. the need to explain cause and effect relationships
d. the need to know how consumers make decisions
e. All are reasons for the high rate of usage of surveys in marketing research
3. The difference between the sample value and the true value in the population
under study is referred to as _______.
a. sampling error
b. random error
c. measurement error
d. chance variation
e. none of the above
75
4. _____________ results from mistakes or problems in the research design.
a. Systematic error
b. Processing error
c. Interviewer error
d. Random error
e. None of the above
5. When a researcher uses an incomplete sampling frame for a study, this is referred
to as __________.
a. surrogate information error
b. selection error
c. frame error
d. none of the above
6. A manufacturer of desserts tested a new five-minute dessert on working females.
A marketing research firm was employed to measure consumer response to the
dessert. The results of the survey were dismal. The manufacturer was sure the
new quickie dessert would be appealing to working females. A follow-up focus
group revealed that working females without children were not interested in the
new idea, but working females with children were. What kind of error was
evident in the initial survey?
a. sample frame error
b. random error
c. population specification error
d. surrogate information error
e. none of the above
76
7. The fact that not every household has a published phone number can cause which
type of error in telephone surveys?
a. sample frame error
b. random error
c. population specification error
d. surrogate information error
e. none of the above
8. This type of sample design error can occur even when the population is correctly
specified and a proper sample frame is being used.
a. measurement error
b. interviewer error
c. processing error
d. selection error
e. none of the above
9. A researcher hired 20 interviewers to conduct door-to-door interviews in a
particular city. In the process of verifying the responses for one of the
interviewers, the researcher found that the interviewer surveyed consecutive
houses in a particular subdivision, instead of selecting every third house. What
kind of error has occurred here?
a. selection error
b. measurement error
c. processing error
d. interviewer error
e. none of the above
10. The Bureau of the Census reported that a more accurate estimate of the total
population in the U.S. could be accomplished by a carefully selected sample,
rather than with a census. What broad classification of errors is the researcher
indirectly alluding to?
a. random error
b. selection error
c. measurement error
d. frame error
e. none of the above
11. A marketing researcher subcontracts the data collection process to a field services
firm. After the data is collected, the researcher is editing the questionnaires, and
finds a large number of mistakes. In many cases skip patterns were not followed,
and difficult questions were often skipped. The researcher contacts several of the
persons that did the interviewing and finds that they received little, and in some
cases, no training or supervision. What kind of error is evident here?
a. processing error
b. interviewer error (bias)
c. surrogate information error
d. measurement instrument bias
e. none of the above
12. In the pretesting of a questionnaire, a researcher finds that respondents perceive
several of the questions as leading, and others as difficult to understand. If the
researcher does not make the necessary revisions to the questionnaire, there will
probably be a large amount of which of the following?
a. processing error
b. interviewer error
c. surrogate information error
d. measurement instrument bias
e. none of the above
13. A questionnaire that is not computer coded, thereby increasing the complexity of
inputing the questionnaire data into a database, will probably suffer from which of
the following?
a. processing error
b. interviewer error
c. surrogate information error
d. measurement instrument bias
e. none of the above
14. When the researcher achieves a very low response rate, he/she should be
concerned with ______________.
a. surrogate information error
b. refusal rate error
c. nonresponse bias (error)
d. measurement instrument bias (error)
e. none of the above
79
15. At which point do people typically refuse to take a survey?
a. before the survey introduction
b. after the introduction
c. after the warm-up questions
d. once the survey is well underway
e. the rate of refusal is about the same for each of the above
16. When survey respondents deliberately falsify their income this is known as which
type of bias?
a. interviewer bias
b. nonresponse bias
c. measurement instrument bias
d. response bias
e. none of the above
17. Which of the following is currently the least popular type of data collection?
a. Internet surveys
b. mall-intercept interviews
c. door-to-door interviews
d. telephone surveys
18. Which of the following data collection techniques can produce excellent sample
quality?
a. mail surveys
80
b. mall intercept
c. central location telephone
d. self-administered interviews
e. executive interviews
19. Which of the following is an advantage of CATI?
a. automatically handling respondent selection
b. automatically handling skip patterns
c. eliminating selection bias
d. computer tabulation of data can occur at any time
e. All of the above are advantages of CATI.
20. Which of the following is an example of a self-administered questionnaire?
a. telephone survey
b. door-to-door survey
c. mail survey
d. mall intercept
e. none of the above
21. A researcher who wants to use probing should consider using which type of
questionnaire?
a. mail
b. Internet
c. telephone
d. none of the above
22. To determine how the same group of respondents feels at different points of time,
the researcher must use what kind of study?
a. mail
b. Intenet
c. telephone
d. longitudinal
23. __________ is the industrial equivalent of door-to-door interviewing.
a. Mall-intercept interviewing
b. Telephone interviewing
c. Corporate interview
d. Executive interview
24. What interviews are conducted by running into people in a big shopping area and
interviewing them face-to-face?
a. Mall-intercept interviews
b. Door-to-door interviews
c. Executive interviews
d. Random interviews
82
Response: See page 159
Ref: Types of Surveys
Learning Objective 6.3: To learn about the types of surveys.
25. All of the following are factors that determine the selection of a particular survey
method except:
a. Sampling precision
b. Specific, chosen respondents
c. Budget
d. Incidence rate
26. What refers to the extent to which the questionnaire follows a set sequence or
order, has a predetermined wording of questions, and relies on closed-ended
question?
a. Arrangement
b. Structure
c. Configuration
d. Organization
27. Surveys have a low rate of usage in marketing research compared to other means
of data collection.
83
28. The total amount of error in a particular project involving a survey is referred to
as random error.
29. Most marketing research is completely error free.
30. Measurement error is often a more serious threat to survey accuracy than is
random error.
31. Using a phone book as a sampling frame would not result in sampling frame
error.
32. Most of the refusals to participate in a survey occur after the introduction is read.
84
Ans: False
Difficulty: Hard
Response: See page 156
Ref: Types of Errors in Survey Research
Learning Objective 6.2: To learn about the types of errors in survey research.
33. When interviewers falsify data, the type of error introduced is interviewer error.
34. One of the major causes of surrogate information error is failure to properly
define the research problem.
35. It is possible for respondents to falsify answers unconsciously.
36. Mall-intercept interviewing is being replaced by some researchers with in-store
research.
37. Random-digit dialing enables market researchers to reach households with
unlisted phone numbers.
38. Inability to probe is a major disadvantage of mail and self-administered
interviews.
39. Mail surveys are a type of self-administered questionnaire.
40. Most current telephone research is biased due to the fact that many people have
abandoned land lines in favor of cell phones.
41. More people participate in mail surveys than in any other type of traditional
survey research.
86
Learning Objective 6.3: To learn about the types of surveys.
42. The quality of data is measured in terms of validity and reliability.
43. If the client needs to have survey results quickly, the Internet is not the best
choice.
44. Interviewers are typically the main interface with consumers and are, therefore, a
vital link to consumer cooperation.
Essay Questions
45. What types of error would be most likely in each of the following circumstances?
a. telephone survey using only a telephone book
b. having members of the Dallas Cowboys Cheerleading Squad administer
mall intercept interviews
c. asking respondents how much they spend on eating out in a year
d. mail survey of physicians
e. “Please indicate your annual household income.” $ _________________
87
a. a. sample frame error, especially if the telephone book is more than six
months old;
b. interviewer bias, in that respondents would become distracted with the
appearance of the interviewer;
c. response bias, as the time period is too long for respondents to be able to
accurately estimate how much they spent eating out;
d. nonresponse bias, as professionals are the least likely to respond to a
survey unless it is something that directly affects them, and mail surveys
generally have problems with high nonresponse rates;
e. response bias in the form of deliberate falsification, as the income question
produces high refusal rates even when categorized. Asking for the exact
amount will result in either refusal or falsification of the amount.
Difficulty: Hard
Response: See pages 152-157
Ref: Types of Errors in Survey Research
Learning Objective 6.2: To learn about the types of errors in survey research.
46. What are the major factors that determine the selection of a particular survey
method?
47. A client is contracting with a marketing research supplier concerning a survey that
would be answered multiple times over a period of time. Which of the non–
Internet forms of survey research would be most appropriate?
48. A potential marketing research client contacts a supplier about a possible survey.
Seems that the client is interested in establishing a snow skiing equipment
business in Arkansas surveying and wants to survey Arkie snow skiers to get an
estimate on potential. The consultant tells the potential client that such a survey
may not be feasible, or at best would be very time consuming and expensive.
What is the basis of the consultant’s response.