31 Variables in Quantitative Research

31.1 Introduction

In quantitative research, a “variable” is a fundamental concept that refers to anything that can vary or change. Think of it as a characteristic or a feature that can take on different values.

For instance, age, income, and temperature are all variables because they can be different for each person or situation.

Variables are crucial in research because they help us understand relationships and patterns by measuring and analysing these differences.

There are different types of variables, such as independent variables, which are thought to influence or cause changes, and dependent variables, which are the effects or outcomes that are being measured. Understanding variables allows us to systematically investigate questions and test hypotheses in a structured way.

31.2 Types of Variables

Independent variables (IV)

An independent variable is a factor that is changed or controlled in an experiment to see how it affects the dependent variable.

For instance, in a study examining the impact of different diets on weight loss, the specific diet followed by each participant is the independent variable.
In an experiment to determine how light affects plant growth, where the amount of light each plant receives is the independent variable.

These are the variables that the researcher changes to observe their effect.

Dependent variables (DV)

A dependent variable, on the other hand, is what you measure in the experiment and what changes as a result of the experiment.

For example, in a study looking at how tutoring might affect test scores, the test scores would be the dependent variable.
In a chemistry experiment observing the rate of a reaction, the rate at which the reaction occurs is the dependent variable.

It’s called ‘dependent’ because it depends on the independent variable.

Confounding variables

Confounding variables are external factors that can influence the outcome of an experiment, affecting both the independent and dependent variables. These variables can compromise the integrity of an experiment if they’re not controlled.

For example, in a study testing the effect of sleep on test performance, the amount of studying done by participants could be a confounding variable. More studying might lead to better test performance, irrespective of the quality of sleep, thus confounding the results.

Control variables

Control variables are crucial in experiments as they help ensure that the results are due to the independent variable and not some other factor.

These are the elements that are kept constant and unchanging throughout the experiment.

For example, if you’re examining the effect of light on plant growth, a control variable could be using the same type of plant for all groups in the experiment. By keeping some variables constant, researchers can more accurately determine the effects of the independent variable on the dependent variable.

31.3 Operationalising variables

Definition

Operationalisation is the process of defining a vague or complex concept in a way that makes it clearly distinguishable, measurable, and understandable in terms of actual observations.

For example, intelligence might be ‘operationalised’ by the score on an IQ test.

This process is crucial because it allows researchers to measure variables and conduct research effectively. By operationalising, concepts that are initially abstract or unclear become tangible and measurable.

Importance

Operationalising variables is important for several reasons.

it enhances clarity in research by providing clear guidelines on how variables are measured. This ensures consistency and clarity throughout the research process.
it facilitates communication among researchers. By operationalising complex concepts, we ensure that our studies are understood and can be replicated by others with the same understanding.
it improves the quality of research. Accurate and reliable data collection, made possible through proper operationalisation, enhances the overall validity and reliability of research findings.

Examples

Attitude might be operationalised by counting the number of positive statements made about an object.
Socioeconomic status could be operationalised through indicators like income levels, educational attainment, and occupational prestige.
Stress could be operationalised by measuring physiological indicators like cortisol levels or heart rate, or through self-reported stress questionnaires.

Challenges

Operationalising variables is not without challenges.

One major issue is the potential loss of nuance; simplifying complex variables into measurable units can sometimes strip them of their complexity and depth.
Another challenge is ensuring validity, which is about making sure that the operationalised variable truly reflects the concept it’s supposed to measure.
Maintaining consistency in measurement across different contexts and studies can be difficult, but it’s crucial for the reliability and comparability of research findings.

31.4 Scales of measurement

Nominal scale

A nominal scale is a basic type of measurement scale used for categorizing variables without assigning any numerical value to them. It’s essentially about labeling.

For example, gender is typically categorised as male or female, making it a nominal variable.
Similarly, types of plants or animals can be categorised using a nominal scale. This scale is about naming or classifying items without any order or level.

Ordinal scale

An ordinal scale is used for ranking variables in a specific order. However, the key point to note is that the intervals between the ranks are not necessarily equal.

A classic example is socioeconomic status, which can be classified as low, middle, or high. This ranking implies an order - high is above low - but the difference between each category is not defined.
Another example is a satisfaction survey where responses range from “very unsatisfied” to “very satisfied.” Here, the order is clear, but the exact difference between each level is not.

Interval scale

An interval scale provides not just order but also equal intervals between values, making the differences between them meaningful. However, it lacks a true zero point.

Temperature is a typical example, measured in Celsius or Fahrenheit. The difference between 10°C and 20°C is the same as between 20°C and 30°C, but there is no true ‘zero’ temperature where there is no temperature at all.
IQ scores are another example of an interval scale, where the difference between scores is consistent, but there is no absolute zero IQ.

Ratio scale

A ratio scale is a more advanced scale of measurement that includes a defined zero point and equal intervals. This scale is used for variables where both differences and proportions are meaningful.

Height and weight are classic examples of ratio variables. Zero height or weight is a meaningful concept (indicating none), and the intervals are consistent.
Age and income are also measured on a ratio scale, where zero has a real meaning (no age or no income) and the intervals are equal and meaningful.

31.5 Reliability and validity of variables

Importance

‘Reliability’ in research refers to the consistency of a measurement tool or method. Reliable variables are those that yield consistent results over time, indicating stability in the measurements.
Reliability is a cornerstone in research as it establishes trust in both the measurement tools and the results they produce. Without reliability, the findings of a study can be questioned and may not be useful.

Methods

There are several methods to establish the reliability of a measurement.

The test-retest method involves administering the same test to the same group of people at two different points in time to assess the stability of the test over time.
Inter-rater reliability ensures consistency in subjective assessments by having different individuals rate or assess the same item.
Internal consistency involves assessing the consistency of results across items within a test, ensuring that all parts of the test contribute equally to what is being measured.

Importance of Validity

Validity, on the other hand, refers to the accuracy of a measurement tool or method. It ensures that the research is truly measuring what it is intended to measure.

Validity is crucial for the generalisability of the study results to the larger population and enhances the overall credibility of the research findings.
Validity supports the decision-making process based on the research, ensuring that the conclusions drawn are based on accurate and relevant data.

We will cover validity in greater detail later in the module.

31.6 Reflect

In your own words, how would you differentiate between independent and dependent variables? Can you provide an example from a study or experiment you are familiar with to illustrate these concepts?
Reflect on the process of operationalising variables. Why is it crucial in research to clearly define how a variable is measured? Think about any potential challenges you might encounter when operationalising variables in a hypothetical study.
Considering the different scales of measurement (nominal, ordinal, interval, and ratio), how would you decide which scale to use for a specific variable in a study? Additionally, how does the choice of scale impact the validity and reliability of the research findings?