Chapter Ten: Measurement and Attitude Scaling
Chapter 10
Measurement and Attitude Scaling
AT-A-GLANCE
I. Introduction
II. What Needs to be Measured?
A. Concepts
B. Operational definitions
1. Variables
2. Constructs
III. Levels of Scale Measurement
A. Nominal scale
B. Ordinal scale
C. Interval scale
D. Ratio scale
E. Mathematical and statistical analysis of scales
2. Continuous measures
F. Reliable and valid measures
IV. Indexes and Composites
A. Computing scale values
B. Reliability
1. Internal consistency
2. Test-retest reliability
V. Validity
A. Establishing validity
B. Construct validity
C. Reliability versus validity
VI. What is an Attitude?
VII. Attitude Measures and Scaling
A. Physiological measures
B. Self-report scales
2. Category scales
4. Selecting items for a Likert scale
6. Semantic differentials and meaning
Chapter Ten: Measurement and Attitude Scaling
RESEARCH SNAPSHOTS
➢ Recoding Made Easy
Most survey related software makes scale recoding easy. Software, like SPSS or
Qualtrics, provide easy procedures for recoding variables.
➢ Someone Tell Me
Attitude measurement remains a mainstay of marketing research. What factors influence
consumer attitudes? Given the large amount of time that people spend online, today a wide
range of sources influence consumer attitudes. Although personal sources have always been a
big part of attitude change as reflected in the concept of subjective norms, there is a desire to
conform to what people you like would like you to do. Now, many unfamiliar sources online
TIPS OF THE TRADE
➢ Research questions and hypotheses reveal the concepts to measure during the
research.
➢ When planning to use parametric statistical analysis, scales with at least 5 response
points provide the greatest flexibility.
➢ Coefficient α provides an estimate of scale reliability. It is only applicable to multiple
item scales with three or more items to be formed into a composite. The following
guide can be used in judging the results from estimating coefficient α.
▪ Above 0.8, very good reliability
• Scale can be used as is from a reliability standpoint.
▪ From 0.7 to 0.79, good reliability
• Scale can be used as is from a reliability standpoint.
▪ From 0.6 to 0.69, fair reliability
➢ Whenever there is a question about the number of scale points to use, it is better to
use more rather than fewer scale points
➢ Forced choice questions have their place in marketing research
Chapter Ten: Measurement and Attitude Scaling
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
➢ Attitude scores have no absolute meaning so results for more than one brand are
needed to provide diagnostic recommendations.
OUTLINE
I. INTRODUCTION
A. Good measurement is important in business management.
B. Researchers can measure business and marketing in more than one way.
II. WHAT NEEDS TO BE MEASURED?
A. Managers cannot know if their strategies are creating value and if the firm is
performing well without valid measurement.
B. Measurement is the process of describing some property of a phenomenon of
interest usually by assigning numbers in a reliable and valid way.
2. All measurement systems present the potential for error.
3. Scientific research would be impossible without measurement.
C. Concepts
1. A researcher has to know what to measure before knowing how to measure
something.
2. A concept is a generalized idea that represents something of meaning.
a. Concepts such as age, sex, and number of children are relatively concrete
properties and present few problems in either definition or measurement.
D. Operational Definitions
1. Researchers measure concepts through a process known as operationalization,
3. Scales provide correspondence rules which indicate that a certain value on a
scale corresponds to some true value of a concept, hopefully in a truthful way.
4. Variables
a. Researchers use the variance in concepts to make diagnoses.
5. Constructs
a. Sometimes a single variable cannot capture a concept alone.
b. Using multiple variables to measure one concept can often provide a more
Chapter Ten: Measurement and Attitude Scaling
d. Operational definitions translate conceptual definitions into measurement
scales.
III. LEVELS OF SCALE MEASUREMENT
A. The level of scale measurement is important because it determines the mathematical
comparisons that are allowed.
B. The four levels of scale measurement are:
2. Ordinal
4. Ratio
C. Nominal Scale
2. The value can be but does not have to be a number since no quantities are being
represented.
4. Marketing researchers use nominal scales quite often.
6. Examples of nominal numbering are uniform numbers, airport terminals, and
school bus numbers.
D. Ordinal Scale
1. Ordinal scales have nominal properties, but they also allow things to be
arranged based on how much of some concept they possess.
3. Somewhat arbitrary, but not nearly as much as a nominal scale.
4. Example: “win,” “place,” and “show” in a horse race ► tells which horse was
first, second, and third, but does not tell by how much a horse won.
E. Interval Scale
2. Classic example is the Fahrenheit temperature scale:
26.7° C, and 40° F = 4.4° C.
c. The scale is iconic— it does not exactly represent some phenomenon.
3. Interval scales are very useful because they capture relative quantities in the form
of distances between observations.
F. Ratio Scale
1. Ratio scales represent the highest form of measurement—they have all the
3. Ratio scales provide iconic measurement.
4. Zero represents an absence of some concept.
a. An absolute zero is a defining characteristic in determining between ratio and
Chapter Ten: Measurement and Attitude Scaling
3. The measures of attributes are often combined to represent some less
concrete concept.
IV. INDEXES AND COMPOSITES
A. Multi-item instruments for measuring a construct are called index measures, or
composite measures.
2. An index is often formed by putting several variables together.
4. For most practical applications, composite measures and indexes are computed in
the same way.
B. Computing Scale Values
1. Exhibit 10.4 demonstrates how a composite measure can be created from
common rating scales.
3. A researcher may sometimes choose to average the scores rather than summing
4. Reverse coding means that the value assigned for a response is treated
oppositely from the other items.
C. Reliability
1. Reliability is an indicator of a measure’s internal consistency.
3. When the measuring process provides reproducible results, the measuring
instrument is reliable.
4. Internal Consistency
a. Internal consistency represents a measure’s homogeneity.
b. The set of items that make up a measure are referred to as a battery of scale
items.
c. Internal consistency of a multiple-item measure can be measured by
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
5. Test-Retest Reliability
a. The test-retest method of determining reliability involves administering the
same scale or measure to the same respondents at two separate times to test
for stability.
b. If the measure is stable over time, the test, administered under the same
conditions each time, should obtain similar results.
c. Represents a measure’s repeatability.
V. Validity
A. Good measures should be both precise (i.e., reliable) and accurate (i.e., valid).
B. Validity is the accuracy of a measure or the extent to which a score truthfully
represents a concept.
C. Achieving validity is not a simple matter.
D. Addresses the problem of whether a measure indeed measures what it is supposed to
measure.
E. When a measure lacks validity, any conclusions based on that measure are also likely
to be faulty.
F. Establishing Validity
2. Face (content) validity refers to the subjective agreement among professionals
that a scale logically reflects the concept being measured.
3. Criterion validity addresses the question: “Does my measure correlate with
measures of similar concepts or known quantities?”
G. Construct validity exists when a measure reliably measures and truthfully represents
a unique concept and consists of several components:
a. Face or Content validity (see above)
Chapter Ten: Measurement and Attitude Scaling
H. Reliability versus Validity
10.5).
a. A: The shots from the older gun are scattered ► low reliability.
b. B: The shots from the newer gun are closely clustered and on target ► high
V. WHAT IS AN ATTITUDE?
A. An attitude is an enduring disposition to respond consistently in a given manner to
various aspects of the world, including persons, events, and objects.
B. There are three components of attitude:
1. Affective component – expresses how much affinity someone has toward the
relevant matter.
VI. ATTITUDE MEASURES AND SCALING
A. There are a wide variety of choices in measuring attitudinal concepts, stemming in
part from lack of consensus about the exact definition of the concept.
B. In addition, the affective, cognitive, and behavioral components of an attitude may be
measured by different means.
C. Physiological measures
1. Research may assess the affective (emotional) components of attitudes through
2. These measures provide a means of assessing affect without verbally questioning
the respondent.
3. Provide a gross measure of likes or dislikes, but they are not extremely sensitive
to the different gradients of an attitude
D. Self-report scales
2. Researchers can ask respondents to rank, rate, sort, or choose one of multiple
responses as a way of indicating a response.
3. A ranking task requires the respondent to rank order a small number of stores,
5. Rating requires the respondent to estimate the magnitude or the extent to which
some characteristic exists or some choice is preferred.
6. A sorting task might present the respondent with several product concepts on
7. Another type of attitude measurement requires respondents to make a choice
between two or more alternatives.
Chapter Ten: Measurement and Attitude Scaling
a. If a respondent chooses one object over another, the researcher assumes that
the respondent prefers the chosen object, at least in this setting.
8. Scales with better than ordinal measurement quality are needed to discern small
differences between attitudes or to scale different brands along a continuum.
9. Measuring the Affect Component
a. Exhibit 10.6 displays an interesting approach to get respondents to rate their
10. Category Scales
a. Rating scales consisting of several response categories, often providing
11. The Likert Scale
a. A commonly applied scale format in marketing research
b. With the Likert scale, respondents indicate their attitudes by checking how
12. Selecting Items for a Likert Scale
a. Typically, a researcher will use multiple items to represent a single
13. Semantic Differential
a. The semantic differential is a scale type that has respondents describe their
14. Semantic Differentials and Meaning
15. Scoring Semantic Differentials
a. For scoring purposes, a numerical score is assigned to each position on the
rating scale (i.e., 1, 2, 3, 4, 5, 6, 7 or -3, -2, -1, 0, +1, +2, +3).
© 2016 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, in whole or in
part, except for use as permitted in a license distributed with a certain product or service or otherwise on a
password-protected website for classroom use.
16. Constant-Sum Scale
e. With a constant-sum scale, respondents are asked to divide a fixed number
of points (e.g., 100) among several attributes to indicate their relative
17. Graphic Rating Scales
i. A graphic rating scale presents respondents with a graphic continuum.
j. Respondents are allowed to choose any point on the continuum to indicate
their attitude.
F. Ranking
1. Consumers often rank order their preferences.
3. This technique is easily understood by respondents.
4. Like the constant-sum scale, technically the ranking scale also suffers from
inflexibility—if we know how someone ranked five out of six alternatives, we
know the answer to the sixth.
G. Paired Comparisons
2. Ranking objects with respect to one attribute is not difficult if only a few
3. If the number of comparisons is too large, respondents may fatigue and no longer
carefully discriminate among them.
H. Direct Assessment of Consumer Attitudes
1. Attitudes, as hypothetical constructs, cannot be observed directly.
3. A summated rating scale can be made up of three indications of attitude.
I. How Many Scale Categories or Response Positions?
1. This issue is basically a matter of sensitivity.
J. Balanced or Unbalanced Rating Scale?
