Chapter 12 – Asking Good Questions
I. Learning Objectives:
Upon completing this chapter, the student should be able to:
2. List the four scales (levels) of measurement.
• Nominal
3. Name some widely used attitude scaling techniques in marketing research.
4. List some other key decisions to be made when designing scales.
5. Explain the concept of validity as it relates to measuring instruments.
II. Chapter Outline:
A. Scales of Measurement
Manager’s Focus
Manager’s Focus
Exhibit 12.1: Scales of Measurement
1. Nominal Scale
2. Ordinal Scale
4. Ratio Scale
B. Measuring Attitudes and Other Unobservable Concepts
1. Itemized-Ratings Scale
Chapter 12 – Asking Good Questions
2
Exhibit 12.3: Using Rating Scales to Assess Unobservable Concepts
Manager’s Focus
a. Summated-Ratings (Likert) Scale
Exhibit 12.4: Example of Likert Summated-Ratings Scale
3. Comparative-Ratings Scale
C. Other Considerations in Designing Scales
1. Number of Items in a Scale
3. Including a “Don’t Know” or “Not Applicable” Response Category
D. Establishing the Validity and Reliability of Measures
Exhibit 12.9: Components of an Observed Response
1. Reliability
Exhibit 12.9: Illustration of Difference between Random and Systematic
Error
Manager’s Focus
E. Summary
F. Key Terms
III. Answers to Review Questions:
1. The four scales of measurement are nominal, ordinal, interval, and ratio scales. For
2. The Likert scale (summated-ratings scale) and the semantic-differential scale are
commonly used attitude scaling techniques in marketing research. Researchers also
3. Factors that may produce systematic errors include some personality traits or other
4. Reliability is the ability of a measure to obtain similar scores for the same object,
5. Validity is the extent to which differences in scores on a measuring instrument
reflect true differences among groups, individuals, or situations in the characteristic
IV. Instruction Suggestions:
1. This chapter contains content that is centrally important to conducting solid
marketing research. Unfortunately, students often view the chapter as a (boring)
set of definitions to be memorized, without ever really grasping the importance of
2. Next discuss the distinctions between nominal, ordinal, interval, and ratio scales.
Stress the properties of each, the possible transformations, and the measures of
to their academic role, e.g., student number for nominal, class standing for
ordinal, grade point average as interval, and number of credits completed as ratio.
3. Turn to a discussion of psychological measurement and the central problem in
marketing research of measuring invisible and intangible qualities such as attitudes
and purchase intentions. Introduce the essential elements of a theory, namely
hypothetical constructs, linkages among and between the constructs, and data
which connect the constructs with the empirical world. Emphasize the distinction
between conceptual and operational definitions, and point out the need for
operational definitions if we are ever going to investigate the frameworks we
develop. This is as true for the marketing manager as it is for the social scientist
investigating basic marketing phenomenon. It is helpful here to present some
4. It is essential that the students appreciate the central role of attitude measurement
5. Next turn to a discussion of the various ways of measuring attitudes. Point out that
measurement is always an inference and that the different methods are subject to
their own particular errors, e.g., halo effects, socially acceptable responses, yea
6. Turn to a discussion of self-report scales and the general categories of self-report
rating scales. Excellent examples of graphic rating scales, itemized ratings scales,
7. Because undergraduate students often have difficulty developing an effective
presentation of research results, instructors may want to prompt students about
presenting a series of semantic differential scale scores as applied to multiple
objects.
8. Overview a number of important considerations that arise when designing rating
scales. Each issue is addressed in the text. Further, each of these issues is of
9. The issue of how best to interpret a rating scale score is one that is too often
ignored by instructors (and practicing marketing researchers). In the absence of
10. Then, turn to a discussion of validity and reliability and how they are used to
support this inference. It is helpful to explore here the basic types of validity and