Measurement in Marketing

Operationalization of Latent Constructs

Azza Frikha

• Publisher: Wiley-ISTE
• Year: 2019
• Print ISBN: 9781786304629
• eBook ISBN: 9781119671282
• Edition: 1
• Topic: Business
• Subtopic: Marketing
• Index: Business

Scientific research uses concepts (or constructs) and requires means to measure them. Often latent, abstract and not directly observable, these concepts demand special attention. When facing problems related to their operationalization, considerable efforts are required to construe measures that effectively represent the phenomena studied. Measurement in Marketing presents a wide range of ideas to help researchers in the selection, design and validation of measurements of constructs. It analyzes the provisions that must be implemented to allow the transition from a latent conceptual construct to an operational level close to reality, and thus to make possible a fluid, reliable and valid reading of the phenomena observed. This instructive book guides readers through all stages of the implementation of a measure. It is intended for a wide audience, offering examples, summary tables and boxes in order to emphasize the primary information.

1
Characteristics and Main Types of Measurement

---

OBJECTIVES OF THE CHAPTER.–

This chapter offers a set of prerequisites for better progress in the operationalization of a construct by means of attitude scales. Its objectives are to:

• – provide a better understanding of the notion of measurement and some connected concepts;
• – present the specificities of the basic scales (nominal, ordinal, interval and proportional);
• – highlight the main characteristics of different scales for measuring attitudes/opinions;
• – present some determinants that allow the choice of an attitude scale format.

---

1.1. Introduction

The issue of measuring variables, such as perceptions, attitudes and human preferences, is of undeniable importance. Indeed, one of the major difficulties of research is to match the measurement to the object to be measured. The questionnaire is often used as an information gathering tool to scientifically apprehend a multitude of objects and abstract concepts. Often it consists of closed-ended questions in order to translate, for example, a theoretical notion, an opinion or the characteristics of a product into an empirical assessment by assigning it a numerical rating and thus proposing a measure. The evaluation of a brand, an attitude, an opinion, etc., can thus be done through a multitude of ways to formulate questions which give rise to a variety of measurement approaches.

In this chapter, we will try to highlight some fundamental concepts associated with measurement. The objective is to clear up any confusion about the primary measurement options that are used when assessing phenomena. Indeed, a good understanding of the characteristics and formats of primary scales is essential for any further reflection on the construction and validation of measurement scales. We will then focus on the presentation of the measurement concept and the various basic measures (qualitative and quantitative) that can be mobilized. After that, we will examine, from different angles, the scales most frequently used to measure attitudes, particularly in the practice of marketing research.

1.2. The basis of the measure

Measurement, scale, metric scale, nominal scale, item, etc., form a set of notions that should be considered as prerequisites before reflecting on the operationalization of a phenomenon. However, they often remain confused. Clarifying these concepts is a recurrent element in the debate surrounding the most appropriate way to carry out a measurement operation. In this section, the aim is to define the measure while distinguishing it from other related concepts and to highlight the different basic measurement formats, each of which gives rise to specificities.

1.2.1. Definition of the measure and connected concepts

Measurement is generally understood to be a question of establishing symbols that can represent the objects evaluated and, more specifically, to operate a number translation relating to the object (a brand, a product, etc.). The symbols used to represent a phenomenon are of a specifically numerical nature. Several definitions of the measure have been proposed. According to Gregory (1995, p. 95), “measurement is precisely the process by which we move from the theoretical world to the world of observation. By measurement, a number is assigned to the characteristic of a concept”. Measuring is therefore a numerical reading of an event, a characteristic, an attribute, etc. or more generally, a construct. It should be noted that a construct “expresses an abstraction, it is a concept deliberately and consciously invented or adopted for a scientific purpose” (Kerlinger 1973, cited by Venkatraman and Grant 1986, p. 73). The measurement operation is crucial when conducting research. This step effectively assigns numbers to the concepts studied, thus facilitating the set of statistical tests necessary to understand realities. Edwards and Bagozzi (2000, p. 156) noted “that a measure refers not to an instrument used to gather data or to the act of collecting data, but to the score generated by these procedures”. Howell et al. (2007, p. 207) defined a measure as “an observed score generated via self-report, observation, or some other means”. Measurement is a kind of springboard from the theoretical world to the empirical test.

Even if the idea conveyed by the measurement is always the same, different measurement approaches are possible, resulting in different methodological and statistical choices. During each measurement operation, rules for assigning numerical values are defined, thereby conditioning the significance of the number and its statistical importance. Indeed, the logic of assigning a number to a phenomenon, or to one of its components, may be very different from the logic of assigning a number to another phenomenon (see section 1.2.2).

It is relevant to note that the notion of measurement is often referred to as the notion of scale. Although these two concepts may, in a way, be interchangeable, they present two distinct aspects¹. Measurement consists of assigning a number to an object (attitude, opinion, etc.) while a scale presents a variety of possibilities (including numb...

Table of contents

• Cover
• Table of Contents
• Foreword
• Introduction
• 1 Characteristics and Main Types of Measurement
• 2 Standardizing or Constructing a Measurement Scale
• 3 Conception of a Measurement Scale
• 4 Construction of a Measurement Scale
• 5 Design of a Measurement Scale
• 6 Quantitative Purification of a Reflective Scale
• 7 Validity of a Measurement Scale
• Conclusion
• Glossary
• References
• Index
• End User License Agreement