Chapter 10: The Concept of Measurement
Multiple Choice
1. Ranking questions generally produces ___________ level of measurement.
a. nominal
b. ordinal
c. interval
d. ratio
e. none of the above
2. An example of nominal scale data would be which of the following?
a. preference data
b. index numbers
c. sex of the respondent
d. unit sales
e. none of the above
3. Quality grades such as “good,” “better,” and “best” represent which of the
following levels of measurement?
a. nominal
b. ordinal
c. interval
d. ratio
e. none of the above
118
Learning Objective 10.3: To understand the four levels of scales and their typical
usage.
4. When few differences in scores are found between the first and second
administration of a test, the test is said to have _________.
a. validity
b. stability
c. sensitivity
d. none of the above
5. The Cronbach alpha technique gives researchers an idea of a scale’s
__________.
a. validity
b. reliability
c. sensitivity
d. none of the above
6. A scale that gives you a different weight every time you step on suffers from
poor:
a. face validity.
b. construct validity.
c. split-half validity.
d. Reliability.
e. none of the above
7. Which is the weakest form of validity?
a. content
b. face
c. construct
d. criterion related
e. none of the above
8. If a restaurant customer satisfaction questionnaire lacked a question concerning
the quality of the food being served, we’d say the questionnaire lacked______
validity.
a. content
b. face
c. construct
d. criterion-related
e. none of the above
9. Which step(s) in the measurement process must be completed before evaluating
the reliability and validity of the scale?
a. Identify the concept of interest.
b. Develop a construct.
c. Develop an operational definition.
d. Create a measurement scale.
e. all of the above
10. Before a measurement scale can be shown to be valid, it must be _______.
a. stable
b. reliable
c. consistent
d. convergent
11. When a researcher describes how a construct will be measured, this is referred to
as what type of definition?
a. constitutive
b. theoretical
c. conceptual
d. operational
12. Brand loyalty has been defined as “A strongly motivated and long standing
decision to purchase a particular product or service.” This is an example of which
type of definition?
a. constitutive
b. operational
c. reliable
d. valid
13. ______________ is a guide, a method, or a command that tells a research what to
do.
a. Construct
b. Rule
c. Operational definition
d. Valid concept
e. Reliable concept
14. _______________ specifies which observable characteristics will be measured
and the process for assigning a value to the concept.
a. Operational definition
b. Constitutive definition
c. Concept
d. Rule
e. Valid concept
15. If the same instrument gives consistent results every time it is used for a
measurement, then it indicates:
a. test-retest reliability.
b. equivalent-forms reliability.
c. internal consistency.
d. internal validity.
e. external validity.
16. Which of the following does not apply to an operational definition?
a. There can be many different potential operational definitions for a single
concept.
b. It defines which observable characteristics will be measured.
c. It defines a concept and establishes boundaries for the concept.
d. Instruments are developed on the basis of it.
e. All of the above pertain to an operational definition.
17. When a researcher wants to show the differences separating two objects, the ideal
type of scale to use is _________________.
a. nominal
b. ordinal
c. interval
d. none of the above
18. _______________ reveals the lack of or low correlation among constructs that are
supposed to be different.
a. Predictive validity
b. Predictive reliability
c. Concurrent validity
d. Convergent validity
e. Discriminant validity
19. If the researcher wishes to measure something that could have a value of zero,
what would be the appropriate type of scale?
a. ordinal
b. qualitative
c. nominal
d. ratio
e. none of the above
20. This level of measurement would have the fewest possibilities regarding the types
of statistical techniques that can be applied to its data.
a. nominal
b. ratio
c. interval
d. reliable
e. ordinal
21. What level of measurement would the following question produce?
Please indicate the environmental surroundings in which you live using the
following choices:
(1) urban (2) rural (3) suburban
a. nominal
b. ratio
c. interval
d. reliable
e. ordinal
124
Learning Objective 10.3: To understand the four levels of scales and their typical
usage.
22. What level of measurement would the following question produce?
Please indicate your approximate age by checking the appropriate age category.
___(1) 0 to 18 ___(2) 19 to 34 ___(3) 35 and over
a. nominal
b. ratio
c. interval
d. reliable
e. ordinal
23. In marketing research, this level of measurement is the preferred measure when
measuring purchase likelihood and attitude assessment.
a. nominal
b. ratio
c. interval
d. reliable
e. ordinal
24. Which of the following types of scales would allow the researcher to compute
arithmetic means?
a. nominal
b. ordinal
c. ratio
d. interval
e. both (c) and (d)
125
Difficulty: Medium
Response: See page 282-283
Ref: Step Five: Develop a Measurement Scale
Learning Objective 10.3: To understand the four levels of scales and their typical
usage.
25. You are creating a survey that is interested in ranking customers’ preferences for
various brands of cola. You would definitely need to use which type of scale?
a. nominal
b. ordinal
c. ratio
d. interval
e. none of the above
26. Cronbach’s alpha is a statistical technique for measuring _______.
a. face validity
b. sensitivity
c. reliability
d. content validity
27. The measurement process begins with ____________.
a. Identifying the concept of interest
b. Developing a construct
c. Defining the construct
d. Evaluating reliability and validity
126
Ref: Step One: Identify the Concept of Interest
Learning Objective 10.2: To learn about the measurement process and how to
develop a good measurement scale.
28. ____________ are specific types of concepts that exist at higher levels of
abstraction than do everyday concepts.
a. Abstract Concepts
b. Constructs
c. Intuitive Constructs
d. None of the above.
29. ____________________ are/can be a source of measurement differences that
produces random or systematic error.
a. Characteristics of individual respondents
b. Short-term personal factors
c. Situational Factors
d. Sampling of items included in the questionnaire
e. All of the above
30. A constitutive definition is similar to a dictionary definition.
127
Learning Objective 10.2: To learn about the measurement process and how to
develop a good measurement scale.
31. If a questionnaire contains questions which were poorly worded, then the
questionnaire could have a serious “reliability” problem.
32. A questionnaire could be reliable without being valid.
33. A questionnaire could be valid without being reliable.
34. A nominal scale provides the researcher with metric data.
35. An ordinal scale can be used when the researcher is interested in knowing which
brand a consumer prefers in a product category.
128
Difficulty: Easy
Response: See page 281
Ref: Step Five: Develop a Measurement Scale
Learning Objective 10.3: To understand the four levels of scales and their typical
usage.
36. Marketers sometimes measure constructs that do not have a true zero point.
37. Marketers are typically interested in measuring observable constructs.
38. Interval scales are preferable to ordinal scales when the researcher wants to know
the magnitude of differences between two brands.
39. Construct equivalence deals with how people see, understand, and develop
measurements of a particular phenomenon.
129
Learning Objective 10.2: To learn about the measurement process and how to
develop a good measurement scale.
40. A construct should be measured before it is defined.
41. A problem often encountered with rules is a lack of clarity or specificity.
42. Physical characteristics of a respondent such as age, weight, & height are not
examples of ratio-scaled variables.
43. Equivalent form reliability is the ability of two very different forms of an
instrument to produce closely correlated results.
44. The split-half technique is a method of assessing the reliability of a scale by
dividing the total set of measurement items in half and correlating the results.
Essay Questions
45. Revise the following assessment of customer satisfaction for Jim Dandy’s
Hamburger Joint so that content validity is achieved.
Please indicate your satisfaction with Jim Dandy’s Hamburger Joint using a scale
of 1=very satisfied, 2=somewhat satisfied, and 3=not satisfied.
____a. location of the restaurant
____b. cleanliness of restrooms
____c. friendliness of staff
____d. speed of service
____e. restaurant cleanliness
____f. availability of condiments
____g. efficiency of drive-up window
____h. hours of operation
46. Revise the following questions so as to achieve interval scale data.
(1) How satisfied are you with your present car?
1=very satisfied 2=somewhat satisfied 3=not satisfied
131
47. Ratio and Interval scale data can both produce data in which arithmetic means and
metric measures of dispersion can be computed. However, there are several
important differences in these two levels of measurement. What are they?
48. Define Validity. Then provide an example of Face Validity and an example of a
question that has a Face Validity problem. Then illustrate a portion of a
questionnaire with a Content Validity problem.
132
Ans: Validity is the degree to which what the researcher is trying to measure is actually
measured, such as the extent to which a question or series of questions in a
questionnaire actually measure a concept. Face Validity is the degree to which a
measure seems to measure what it is supposed to measure. An example would be
as follows:
A question asking college students to respond to an age category question when
the target market is defined as traditional college age students.
Question: Please indicate your age by marketing the appropriate blank below:
_____0-18 _____19-34 _____35-54 ____55 and over
The categories above would not be appropriate for traditional college ages
students and therefore would not adequately measure a population consisting of
college students. A better scale would be as follows:
_____0-18 _____19-21 _____22-25 ____26 and over
Content validity has to do with how well the content of the questions represents
the concept being measured. An example would be for a restaurant customer
satisfaction study: Please indicate your level of satisfaction with XYZ restaurant
for each of the following using the following scale:
Very Somewhat Somewhat Very
Dissatisfied Dissatisfied Satisfied Satisfied
_____ _____ _____ _____
1 2 3 4
Location ____ Cleanliness____ Parking____ Service____
The above 4 attributes are certainly important, but do not adequately measure the
concept of satisfaction for a restaurant, thereby not achieving content validity for
satisfaction construct. One obvious omission in the above list is food quality.
Price would be another important omitted variable.
Difficulty: Hard
Response: See pages 289-290
Ref: Step Six: Evaluate the Reliability and Validity of the Measurement
Learning Objective 10.4: To explore the concepts of validity and reliability.