Thinking Like a Researcher
3 Concepts, constructs, and variables
Using concepts, constructs and variables
Concepts and constructs
In a previous chapter, we discussed that research can be focused on describing, explaining and predicting, and that for many academics, understanding and explaining are the ultimate goal of doing research. Explanations require development of concepts. Concepts are generalisable properties or characteristics associated with objects, events, or people. While objects such as a person, a firm, or a car are not concepts, their specific characteristics or behaviour such as a person’s attitude toward immigrants, a firm’s capacity for innovation, and a car’s weight can be viewed as concepts.
Knowingly or unknowingly, we use different kinds of concepts in our everyday conversations. Some of these concepts have been developed over time through our shared language. Sometimes, we borrow concepts from other disciplines or languages to explain a phenomenon of interest. For instance, the idea of gravitation borrowed from physics can be used in business to describe why people tend to ‘gravitate’ to their preferred shopping destinations. Likewise, the concept of distance can be used to explain the degree of social separation between two otherwise collocated individuals. Concepts may have progressive levels of abstraction. Some concepts such as a person’s weight are precise and objective, while other concepts such as a person’s personality may be more abstract and difficult to visualise.
A construct is an abstract concept that is specifically chosen (or created) to explain a given phenomenon. A construct may be a simple concept, such as a person’s weight, or a combination of a set of related concepts such as a person’s communication skill, which may consist of several underlying concepts such as the person’s vocabulary, syntax, and spelling. The former instance (weight) is a unidimensional construct, while the latter (communication skill) is a multidimensional construct because it consists of multiple underlying concepts. The distinctions between constructs and concepts are clearer in multi-dimensional constructs, where the higher order abstraction is called a construct and the lower order abstractions are called concepts. However, this distinction tends to blur in the case of unidimensional constructs.
Constructs used for scientific research must have precise and clear definitions that others can use to understand exactly what it means and what it does not mean. For instance, a seemingly simple construct such as income may refer to monthly or annual income, before-tax or after-tax income, and personal or family income, and is therefore neither precise nor clear. When discussing constructs, there are two types of definitions: conceptual and operational definitions. Studying specific constructs requires defining what they mean, as different academics can disagree on the nature of a construct.
Conceptual versus operational definitions
A conceptual definition involves defining a construct on an abstract and theoretical level. For example, a conceptual definition of perceived service quality could be “the degree and direction of discrepancy between consumers’ perceptions and expectations [of a company’s service performance]” (Parasuraman, Zeithaml & Berry 1988, p.17).[1]
In contrast, operational definitions are necessary when it comes to actually collecting and analyzing data. In constructing an operational definition, you have to be very specific on how you will actually measure your construct. In the aforementioned example of service quality, Parasuraman, Zeithaml & Berry (1988) have constructed a survey scale to measure service quality as a construct. Their operational definition of perceived service quality is very specific: a customer’s rating of a company on a 26-item survey scale that consists of five underlying service quality dimensions: a company’s (service) reliability, their responsiveness, their (perceived) empathy, assurance and their performance on tangibles.
The key thing to remember about operational definitions is that you have to be very specific about how you will measure a construct. For instance, the operational definition of a construct such as temperature must specify whether we plan to measure temperature in Celsius, Fahrenheit, or Kelvin scale. A construct such as income should be defined in terms of whether we are interested in monthly or annual income, before-tax or after-tax income, and personal or family income. You can imagine that constructs such as learning, personality, and intelligence can be quite hard to define operationally. In practice, you may welll encounter studies that have different findings even if they use the same conceptual definitions. Differences in operational definitions could be one possible reason for such divergent findings.
Variables
A term frequently associated with, and sometimes used interchangeably with, a construct is a variable. Etymologically speaking, a variable is a quantity that can vary (e.g., from low to high, negative to positive, etc.), in contrast to constants that do not vary (i.e., remain constant). However, in scientific research, a variable is a measurable representation of an abstract construct. As abstract entities, constructs are not directly measurable, and hence, we look for proxy measures called variables. For instance, a person’s intelligence is often measured as his or her IQ (intelligence quotient) score, which is an index generated from an analytical and pattern-matching test administered to people. In this case, intelligence is a construct, and IQ score is a variable that measures the intelligence construct. Whether IQ scores truly measure one’s intelligence is a matter of ongoing debate. Depending on whether or how well it measures intelligence, the IQ score may be a good or a poor measure of the intelligence construct.
Thinking like a researcher is translating between planes
As the examples above show, scientific research proceeds along two planes: a theoretical plane in which we discuss constructs, conceptual definitions and (as you will see in the next chapter) propositions in an abstract sense. There’s also the empirical plane. Here, constructs have operational definitions, are measured by variables, about which we may have hypotheses (again, see next chapter). Thinking like a researcher involves the ability to move back and forth between the more abstract theoretical plane and the more specific and tangible empirical plane. Figure 4.1 illustrates this.
Translating between both planes: an example
One thing you may encounter during this course is that your tutorial teacher asks you to think about how your constructs are defined, or to explain your thinking about why your ideas work on both these planes. For example, I recall that in a previous edition of the course, one group of students proposed that if a firm offered their employees a discount on a bicycle, their employees would become happier as a result. This may well be true. However, a researcher would see that this idea is arguably still somewhat on the empirical plane. To move to the theoretical plane, the students needed to explain why, on an abstract level, they think that offering discounts on bikes would make employees happier.
One obvious idea is that discounts on bikes are essentially another form of reward, and greater rewards make for happier employees. This line of thought could suggest ‘reward size’ as a construct, which in turn causes employee happiness, which would be the second construct. Subsequently, it may be possible to find literature on the effect of rewards. The advantage of abstracting away from the specific context (bicycle discounts) is that this helps you find more literature. There’s bound to be lots of research on rewards, but there’s probably much less (if any) research about bike discounts specifically. Moreover, such abstraction may increase the impact of your findings: perhaps rewards increase employee happiness in other contexts than just discounts on bikes.
However, you may have other ideas about possible explanations as well. For example, bikes are a rather sustainable mode of transportation, and perhaps employees were mainly happy not because of the discount itself, but because this discount was a clear signal that their company was working on sustainability, which employees might have deemed important. This would translate to a rather different conceptual model in which not reward size but perhaps a company’s perceived sustainability would be a construct. A third explanation is that biking is exercise, so giving discounts on bikes may signal that the firm cares about its employees, again an altogether different construct. All of these examples show that the exact same empirical observation (cheaper bikes may cause happier employees) might lead to many different ideas regarding how and why something occurs or works on the theoretical plane.
Types of Variables and Conceptual Models
Types of Variables
Depending on their intended use, variables may be classified as independent, dependent, moderating, mediating, or control variables.
Variables that explain other variables are called independent variables, those that are explained by other variables are dependent variables. Variables that are explained by independent variables but explain dependent variables are mediating variables (or intermediate variables). You can view mediating variables as variables that help us understand the mechanism/process through which something works. For example, suppose I have service quality as an independent variable. I might hypothesize that service quality influences customer loyalty (the dependent variable). I might have a very specific idea however, on how service quality influences customer loyalty. My idea could be that service quality doesn’t influence customer loyalty directly, but works ‘through’ customer satisfaction: a higher service quality makes customers more satisfied, and more satisfied customers -in turn- become more loyal. Here, customer satisfaction would be the mediating variable.
A moderating variable (or: moderator) is a variable that strengthens or weakens the effect of another variable, and can sometimes even change the direction of the effect. For example, suppose we do research on how intelligence (independent variable) affects academic success (dependent variable). We may expect that greater amounts of intelligence would be related to greater amounts of academic success. However, there’s also effort to consider, which could work as a moderator: effort could ‘strengthen’ the effect of someone’s intelligence on academic success, perhaps measured by their GPA (grade point average).
Specifically, I might expect that on average, greater intelligence is related to greater academic success. Furthermore, I might expect that the effect of intelligence on academic success is much stronger for individuals who put in a lot of effort compared to those who don’t. Said another way: for individuals who don’t put in much effort, intelligence may have little to do with academic success. For individuals who work hard, intelligence may have lots to do with success. That is, between two equally intelligent students, the student who puts in more effort achieves greater academic success. Note that moderators are not the same as independent variables. We’re not stating that students who put in more effort have a higher IQ for example. Rather, the point is that the effect of intelligence on academic success depends on a third variable (here: effort).
Finally, there’s control variables, which are variables that one has to ‘control for’, that is: take into account in a scientific study, but are otherwise not of main interest.
Conceptual Models
The overall network of relationships between a set of related constructs is called a conceptual model (see Figure 4.2). Thinking like a researcher requires not only being able to abstract constructs away from observations, but also being able to visualise a conceptual model which links these abstract constructs. In fact, as part of this course, you will be asked to come up with a conceptual model in one of the assignments.
- Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1988). SERVQUAL: A multiple-item scale for measuring consumer perceptions of service quality. Journal of Retailing, 64(1), 12–40. ↵
