Variables

In practice, the term variable is used as a synonym for construct, or the property being studied. In this context, a variable is a symbol of an event, act, characteristic, trait, or attribute that can be measured and to which we assign values. For purposes of data entry and analysis, we assign numerical value to a variable based on the variable’s properties. For example, some variables, said to be dichotomous, have only two values, reflecting the presence or absence of a property: employed–unemployed or male–female have two values, generally 0 and 1. Variables also take on values representing added categories, such as the demographic variables of race or religion. All such variables that produce data that fit into categories are said to be discrete, because only certain values are possible. An automotive variable, for example, where “Chevrolet” is assigned a 5 and “Honda” is assigned a 6, provides no option for a 5.5. Income, temperature, age, and a test score are examples of continuous variables. These variables may take on values within a given range or, in some cases, an infinite set. Your test score may range from 0 to 100, your age may be 23.5, and your present income could be $35,000.

Independent and Dependent Variables

Researchers are most interested in relationships among variables. For example, does a newspaper coupon (independent variable) influence product purchase (dependent variable), or can a salesperson’s ethical standards influence her ability to maintain customer relationships?

As one writer notes: There’s nothing very tricky about the notion of independence and dependence. But there is something tricky about the fact that the relationship of independence and dependence is a figment of the researcher’s imagination until demonstrated convincingly. Researchers hypothesize relationships of independence and dependence: They invent them, and then they try by reality testing to see if the relationships actually work out that way.

Many textbooks use the term predictor variable as a synonym for independent variable (IV). This variable is manipulated by the researcher, and the manipulation causes an effect on the dependent variable. We recognize that there are often several independent variables and that they are probably at least somewhat “correlated” and therefore not independent among themselves. Similarly, the term criterion variable is used synonymously with dependent variable (DV). This variable is measured, predicted, or otherwise monitored and is expected to be affected by manipulation of an independent variable. Exhibit 2.2 lists some terms that have become synonyms for independent variable and dependent variable. In each relationship, there is at least one independent variable (IV) and one dependent variable (DV). It is normally hypothesized that, in some way, the IV “causes” the DV to occur. It should be noted, however, that although it is easy to establish whether an IV influences a DV, it is much harder to show that the relationship between an IV and DV is a causal relationship.

SELF-CHECK ACTIVITY

1. What is the purpose of operational definitions?
2. Describe what is the variables?
3. Why IV is also addresses as predicator variables? 

Moderating or Interaction Variables

In actual study situations, however, such a simple one-to-one relationship needs to be conditioned or revised to take other variables into account. Often, we can use another type of explanatory variable that is of value here: the moderating variable (MV). A moderating or interaction variable is a second independent variable that is included because it is believed to have a significant contributory or contingent effect on the original IV–DV relationship. The arrow pointing from the moderating variable to the arrow between the IV and DV in Exhibit 2.3 shows the difference between an IV directly impacting the DV and an MV affecting the relationship between an IV and the DV. For example, one might hypothesize that in an office situation:

The introduction of a four-day working week (IV) will lead to higher productivity (DV), especially among younger workers (MV).

Exhibit 2.3 Relationships among Types of Variables

In this case, there is a differential pattern of relationship between the four-day week and productivity that results from age differences among the workers. Hence, after introduction of a four-day working week, the productivity gain for younger workers is higher than that for older workers. It should be noted that the effect of the moderating or interaction variable is the “surplus” of the combined occurrence of introducing a four-day working week and being a younger worker. For example, let’s assume that the productivity of younger workers is 12 percentage points higher than that for older workers, and that the productivity of workers having a four-day working week is 6 percentage points higher than those of workers having a five-day working week. If the productivity of a younger worker having a four-day working week is only 18 percentage points higher than the productivity of an older worker with a five-day working week, there is no interaction effect, because the 18 percentage points are the sum of the main effects. There would be an interaction effect if the productivity of the younger worker on a four-day week was, say, 25 percentage points higher than the productivity of the older worker on a five-day week. Whether a given variable is treated as an independent or moderating variable depends on the hypothesis under investigation. If you were interested in studying the impact of the length of the working week, you would make the length of week the IV. If you were focusing on the relationship between age of worker and productivity, you might use working week length as an MV.

Extraneous Variables

An almost infinite number of extraneous variables (EVs) exists that might conceivably affect a given relationship. Some can be treated as IVs or MVs, but most must either be assumed or excluded from the study. Fortunately, an infinite number of variables has little or no effect on a given situation. Most can safely be ignored because their impact occurs in such a random fashion as to have little effect. Others might influence the DV, but their effect is not at the core of the problem we investigate. Still, we want to check whether our results are influenced by them. Therefore, we include them as control variables (CVs) in our investigation to ensure that our results are not biased by not including them. Taking the example of the effect of the four-day working week again, one would normally think that weather conditions, the imposition of a local sales tax, the election of a new mayor, and thousands of similar events and conditions would have little effect on working week and office productivity. Extraneous variables can also be confounding variables (CFVs) to our hypothesized IV–DV relationship, similar to moderating variables. You may consider that the kind of work being done might have an effect on the impact of working week length on office productivity. This might lead you to introducing time spent in a meeting to coordinate the work as a confounding variable (CFV). In our office example, we would attempt to control for type of work by studying the effect of the four-day working week within groups attending meetings with different intensity. In Exhibit 2.4, weather is shown as an extraneous variable; the broken line indicates that we included it in our research because it might influence the DV, but we consider the CV as irrelevant for the investigation of our research problem. Similarly we included the type of work as a CFV.

Exhibit 2.4 Relationships among Types of Variables

Intervening Variables

The variables mentioned with regard to causal relationships are concrete and clearly measurable—that is, they can be seen, counted, or observed in some way. Sometimes, however, one may not be completely satisfied by the explanations they give. Thus, while we may recognize that a four-day working week results in higher productivity, we might think that this is not the whole story—that working week length affects some intervening variable (IVV) that, in turn, results in higher productivity. An IVV is a conceptual mechanism through which the IV and MV might affect the DV. The IVV can be defined as a factor that theoretically affects the DV but cannot be observed or has not been measured; its effect must be inferred from the effects of the independent and moderator variables on the observed phenomenon. In the case of the working week hypothesis, one might view the intervening variable (IVV) to be job satisfaction, giving a hypothesis such as:

The introduction of a four-day working week (IV) will lead to higher productivity (DV) by increasing job satisfaction (IVV).

Here we assume that a four-day working week increases job satisfaction; similarly, we can assume that attending internal meetings is an indicator negatively related to the routine character of work. Exhibit 2.5 illustrates how theoretical constructs, which are not directly observed, fit into our model.

Exhibit 2.5 Relationships among Types of Variables