44) Problem respondents fall into several categories. Which of the following does NOT represent
one of these problem respondent categories?
A) Homogeneous responses
B) Break-offs
C) Item omissions
D) Yea-or nay-saying patterns
45) With online surveys most bothersome, problems during questionnaire screening problems
can usually be prevented by selecting ________ that prevent the occurrence of such errors.
A) proper market research analysts
B) options or requirements in the online questionnaire program
C) a proper survey design software program
D) a proper data analytics software program
46) When a researcher encounters data quality issues, there are several options. Which of the
following is NOT one of the options in addressing data quality issues?
A) Throw out the respondent’s entire data row.
B) Leave the respondent’s entire data row in the dataset.
C) Leave the respondent’s data row but correct the answers.
D) Set the bad data items to blanks or missing data.
47) The counterpart to sampling error is nonsampling error, which consists of all errors in a
survey EXCEPT those attributable to the sample plan and sample size.
48) Data analysis is the phase of the marketing research process during which respondents
provide their answers or information in response to inquiries posed by the researcher.
49) A very small portion of the marketing research industry uses panel companies that maintain
very large numbers of respondents who are encouraged to fill out surveys quickly and accurately.
50) The first general type of possible error in field data collection is fieldworker error, defined as
errors committed by the individuals who administer questionnaires, typically interviewers.
51) Respondent errors can occur regardless of the method of data collection, but some data
collection methods have greater potential for respondent error than others.
52) Intentional fieldworker errors occur whenever a data collection supervisor unintentionally
violates the data collection requirements set forth by the researcher.
53) Interviewer cheating is a concern, especially when compensation is calculated on a per
completed-interview basis.
54) A type of intentional error on the part of an interviewer is encouraging respondent cheating,
or attempting to influence the respondent’s answers through wording, voice inflection, or body
language.
55) Unintentional interviewer errors include misunderstandings and fatigue.
56) Interviewer cheating and interviewer leading are categorized as unintentional errors because
professional interviewers are trained to avoid them, so if they commit such errors, they are
usually not aware of their violations.
57) An unintentional interviewer error occurs whenever an interviewer commits an error while
believing that he or she is performing correctly.
58) There are four general sources of unintentional interviewer errors: interviewer personal
characteristics, interviewer misunderstandings, interviewer cheating, and interviewer fatigue.
59) A questionnaire may include various types of instructions for the interviewer, a variety of
response scales, directions on how to record responses, and other complicated guidelines to
which the interviewer must adhere.
60) Interviewer experience can generally overcome poor questionnaire instructions, which helps
avoid errors.
61) There are at least two major intentional respondent errors: falsehoods and refusals.
62) Nonresponse includes failure on the part of an interviewer to take part in the survey,
premature termination of the interview, and refusal to answer specific questions on the
questionnaire.
63) Business-to–business (B2B) marketing research is even more challenging than consumer
research, presenting additional hurdles that must be cleared (such as negotiating “gatekeepers”)
just to find the right person to take part in the survey.
64) Unintentional respondent errors may result from misunderstanding, guessing, cheating,
falsehoods, attention loss, distractions, or fatigue.
65) Two general strategies-supervision and validation-can be employed to guard against the
interviewer intentionally committing an error.
66) An industry standard is to randomly select 15%-25 % of the completed surveys for purposes
of making a call-back to validate that the interview was conducted.
67) Validation verifies that the interviewer did the work and is aimed at the falsification/cheating
problem.
68) Even with supervised orientation sessions and role playing, unintentional fieldworker errors
cannot be reduced.
69) The more competent the field interviewer becomes through training, supervision, and
development of personal skills, the lower the potential for interviewer error.
70) Tactics such as anonymity, confidentiality, validation checks, and third-person technique
have been used minimize respondent error but have been found to have the opposite effect.
71) Incentives sometimes compel respondents to be more truthful while also discouraging
nonresponse.
72) Sometimes the opportunity arises to use the third-person technique, in which the interviewer
directly and indirectly quizzes the respondent about a subject that might be embarrassing.
73) Ways to combat unintentional respondent error include well-drafted questionnaire
instructions and examples, reversals of scale endpoints, use of scale endpoints, and use of
prompters.
74) Long questionnaires often include opportunities for the respondent to discontinue the survey
with phrases such as “We are almost finished,” “That was the most difficult section of questions
to answer,” or other statements.
75) Researchers understand the true value of professional field data collection companies whose
existence depends on how well they can control interviewer and respondent error.
76) Nonresponse has been labeled the marketing research industry’s biggest problem, and it is
multinational in scope.
77) There are four types of nonresponse error: refusals to participate in the survey, break-offs
during the interview, survey cheating, and refusals to answer specific questions.
78) Some industry observers believe that nonresponse is caused by fears of invasion of privacy,
skepticism of consumers regarding the benefits of participating in research, and the use of
research as a guise for telemarketing
79) The identification, control, and adjustments necessary for nonresponse are not as critical to
the success of a survey as once believed.
80) A way to combat unintentional respondent error is to include reversals of scale endpoints in
questionnaires.
81) Some believe it is not wise to put “prefer not to answer” on self-administered questionnaires,
because respondents may use this option simply as a cop-out, when they might have provided
accurate answers if the designation had not been there.
82) If an interviewer does not have all of the data for primary questions and secondary questions,
he or she cannot count the survey as completed.
83) The American Association for Public Opinion Research offers these guidelines regarding
what constitutes a completed survey: complete survey: 80%-100% of applicable questions
answered, partial completion: 50%-80% answered, and break-off: less than 50% answered.
84) The marketing research industry does not currently have an acceptable way to calculate a
survey’s response rate.
85) In most surveys, eligible units are respondents determined by screening or qualifying
questions.
86) Panel companies normally collect a wealth of information (demographic, life style,
possessions, etc.) on each panel member as part of the recruitment and registration process, and
these data may be purchased along with the survey questions, thus eliminating the need to ask for
this information in the survey.
87) The widespread popularity of online panels has given rise to a great many panel companies,
some of which have been praised for superiority in data quality controls.
88) With a code book that describes the data file, any analyst can work on the dataset, regardless
of whether that analyst was involved in the research project during its earlier stages or not.
89) Few statistical analysis programs have features that enable users to identify the number codes
and the associated word labels, which presents an obstacle to analysis.
90) Data quality is a concern for marketing researchers, which means the researcher must
examine the responses for data quality problems prior to analysis.
91) The purpose of completed data inspection and analysis is to determine the presence of “bad”
respondents and to throw out the ones with severe problems.
92) Problem respondents fall into the following five categories: incomplete responses,
nonresponses to specific questions, yea– or nay-saying patterns, and middle-of-the-road patterns.
93) Data collection and nonsampling error go hand-in–hand. Define nonsampling error and what
nonsampling error includes.
94) It is important for researchers to know both the sources of nonsampling errors as well as
ways to minimize its effect. Describe intentional and unintentional errors as well as the causes.
95) Precautions and procedures can be implemented to minimize the effects of the various types
of errors. Discuss controls that may be implemented to minimize errors on the part of both
interviewer and respondent.
96) Precautions and procedures can be implemented to minimize the effects of the various
intentional fieldworker errors. What are the general strategies used to minimize intentional
fieldworker errors? Describe the industry standard used as part of this process?
97) Unintentional fieldworker errors present issues as well for researchers. Precautions and
procedures should be developed to minimize the effects of these errors. What are the general
strategies used to minimize unintentional fieldworker errors?
98) Nonresponse errors of various types occur through the data collection phase. What types of
nonresponse errors exist? How can nonresponse error be determined or calculated?
99) Skyrocketing nonresponse rates and the popularity of online surveys have prompted a
number of firms to use a panel company. How do online panel companies ensure quality and
control error?
100) Describe datasets, coding data, and the use of a data code book.