What are the different types of variables in psychology?

If you've ever conducted an experiment, you've almost definitely encountered a number of parameters that you needed to alter, measure, or control in order to successfully complete your research. Variables are these factors, and there are numerous types of them.

So, what exactly do all these variables designate in experimental research? Let's step right in.

Experimental research is research that takes an empirical approach to investigate a hypothesis.

A variable is a factor that gets measured.

Experimental research thus focuses on testing and analysing two variables: the independent variable (IV) and the dependent variable (DV). This article will use examples to explain what IV and DV are and why they are used in research.

What are experimental variables?

As we have already learned, the purpose of experimental research is to confirm or reject a hypothesis. We use this type of research to understand the cause and effect relationship. This is done by measuring the outcome of a manipulated factor/variable (experimental method).

IV is a factor that the experimenter manipulates to see if it affects the DV. Thus, the IV is what the researcher suspects to be the phenomenon cause, while DV is a variable/factor measured or tested in the experiment.

It follows that the value of the DV can give us an insight into the effect of a relationship. In summary, experimental research involves manipulating hypothesis IV to determine causal relationships of a phenomenon and observing how this affects the DV.

Examples of independent and dependent variables

When we identify variables in psychology, we must operationalise them. Operationalised variables specify how the variable is to be defined and measured.

For example, if we are studying the impact of social media on self-esteem, the IV would be operationalised as the number of hours spent on social media platforms such as Instagram or Facebook. The DV would be the self-esteem score as measured by the Rosenberg Self-Esteem Scale (RSES).

Examples of IVs and DVs in research, Manreet Thind, StudySmarter Originals

As mentioned earlier, the variables we manipulate are called IV. Using them in research is to observe how manipulating variables affect DV. However, other factors may affect the DV that the researcher is not interested in (potential confounding/extraneous variables). These tend to reduce the validity of the results and increase the likelihood that the null hypothesis will be rejected when it would otherwise be accepted. Therefore, researchers need to control these variables (hold them constant or exclude them from the research), hence the name control variables.

Example research scenario

We will now discuss an example of a research scenario where researchers must manipulate and control variables.

One study examined whether caffeine affected participants’ ability to recall memories. The manipulated variable in this study was the amount of caffeine consumed before a memory recall test. In addition, there were potential control variables that needed to be accounted for/restricted before the study:

• Ensuring participants did not drink caffeinated beverages before the study.
• Age.
• Level of education.
• Noise levels.

Let us now consider the different types of variables and their properties. We will also give examples so that we may understand them better.

Continuous variables

A continuous variable is a variable that can potentially have an unlimited number of possible values and is usually determined by measuring or counting a variable. An example of a continuous variable is age.

Extraneous and confounding variables

Extraneous and confounding variables are factors other than the IV and DV that may affect the study’s outcome. The presence of such variables affects the validity of the results. This section will now discuss the definitions of extraneous and confounding variables with examples of how they affect the validity of the results and how the research combats the effects caused by these variables.

Extraneous variables

Extraneous variables are variables/factors that are not the IV but may influence the results (DV). When extraneous variables are present in the research design, IV and DV may be considered causally related, although this is not the case.

These variables may cause the effects of the independent variable on the DV to be underestimated or overestimated, reducing the power of the results.

For example, the noise level can be a potential extraneous variable when investigating studying time and test scores. The noise level could irritate some participants and cause poor performance. Therefore, because of the extraneous variable (noise level that is not controlled), we cannot conclusively say a relationship between IV and DV exists.

Noise level as an extraneous variable when researching studying time and scores, Pixabay

Examples of different types of extraneous variables are:

• Demand characteristics – results are biased by participants responding to ‘cues’ and behaving ‘artificially’, reducing the reliability of the results.
• Experimenter/experimenter effects – the experimenter’s bias that causes participants to behave in a certain way.

Confounding variables

A confounding variable is a factor that has not been considered because it is associated with both IV and DV. Confounding variables affect the DV and are also correlated or causally related to the IV.

The presence of confounding variables means the research design lacks internal validity because the study does not measure the causal relationship between IV and DV.

Consider this example of a research scenario examining exercise and weight loss. The researchers identified IV as randomly dividing participants into two groups: the exercise group and the non-exercise group, and DV as changes in body mass index (BMI). It is known that dietary change is a factor that affects weight changes. If the research design does not account for dietary changes, this may distort the observed results of how much IV affects DV. It is a confounding variable.

How are extraneous and confounding variables combatted?

As mentioned earlier, extraneous and confounding variables affect the validity of results, so researchers can take steps to minimise them. These are discussed in detail below.

Extraneous and confounding variables are controlled by:

• Participants experimented under the same conditions, e.g. temperature and noise level.
• Investigator effects can be countered by using independent researchers who are trained and not part of the research team.
• Demand characteristics are controlled by not informing participants of the true aims of the hypothesis until the end of the study to ensure that this does not affect their performance.
• Statistical control– use of statistical analysis tests that account for confounding variables, such as regression models.
• Sample matching – limiting the sample selection to participants who all share similar characteristics of potential confounding variables, e.g. when studying exercise and BMI, the sample selected must all be in the ‘overweight’ category.
• Randomised sampling– participants are randomly allocated to experimental groups. This increases the likelihood of the confounding variables present being equal between experimental groups.

The golden standard for quality research in psychological research is to operationalise all variables examined in studies.

Operationalisation of variables means that the variables under study are clearly defined with information about how the study will measure them.

This shows that when operationalising variables, researchers need to conceptualise the variables being measured by breaking down the elements of the variables to show how the researchers are measuring them.

For example, we might measure bullying by observing the frequency of kicking, name-calling, or derogatory language.

Operationalisation of variables, e.g., measuring bullying by observing the frequency of name-calling, Pixabay

We will now discuss an example of a research scenario to explain the operationalisation of variables.

The following example uses the research scenario of investigating whether emotions influence problem-solving skills. The researcher would identify emotions as IV and problem-solving skills as DV. The operationalised definition of IV is ‘emotional intelligence as measured and assessed by the Emotional Intelligence Test’. Furthermore, the operationalised definition of DV would be ‘time required to solve a problem-solving test, such as a Sudoku puzzle’.

Why is it important to operationalise variables?

• Clearly defining variables and how they are measured in research makes it easier for researchers to replicate the research and determine the reliability of the results.

• It is easier to ensure that the variables being studied have high internal validity (they measure what they are supposed to measure).

• Reduces the likelihood that subjectivity will influence the research.

• Ensures that the variables being studied are observable. This is an essential prerequisite for the research to be called empirical.

Variables - Key takeaways

• The IV is a factor that the experimenter manipulates to determine if it has an effect on the DV.
• The DV is a variable/factor measured or tested in the experiment.
• Extraneous variables are defined as variables/factors that are not the IV but may affect the results (DV).
• A confounding variable is associated with both the IV and the DV but not included. Confounding variables affect the DV but may also be correlated with the IV.
• Confounding and extraneous variables are controlled for using various techniques to prevent them from affecting the validity of the results.
• Operationalisation of variables means that the variables under study are clearly defined with information on how the study will measure them.