What are Extraneous Variables? Extraneous Variables vs Confounding Variables

“The technology is the independent variable, the social system the dependent variable. Social, systems are therefore determined by systems of technology; as the latter change, so do the former.”

― Leslie White

 

 

What are Extraneous Variables?

 

The first and the foremost thing is to understand the meaning of a variable. In mathematics we use variables, but variables are mainly used in psychological experiments. These variables can be altered and determine the changes. 

 

Researchers use variables to find out the cause and effect relationship in psychological experiments. Now these variables are of two types- independent and dependent variables. The main point of difference between both is the way they manipulate the experiment. 

 

If the manipulation is intentional or direct then we call them independent variables and the variables that are measured to come to conclusion are dependent variables. Now every experiment has more factors that put an impact on it. 

 

One such variable that impacts the experiment is an extraneous variable. These variables affect the relationship between an independent and a dependent variable. Here is an example to understand it better.

 

Suppose a sleep deprivation based experiment is conducted. The objective of this experiment is to find out the impact of less sleep on the performance of a person. In this experiment, sleep deprivation is a straightaway independent variable. 

 

There will be some participants that are sleep deprived while the others are fully rested. In this experiment there are certain other factors that will impact the results. Age, gender, educational background etc are some of those factors. 

 

These are neither independent or dependent but impact the relationship between both. These factors or variables are called extraneous variables. Experimenters always note the impact of these variables on the experiment and the results. 

 

Another example- Suppose researchers want to find out the best teaching method that will ensure that most students pass the math exam. An extraneous variable in this case will be the prior knowledge that students possess related to the same math exam that can affect the present score. 

 

There are lots of challenges associated with extraneous variables and many popular researchers have expressed their concerns over the need to control these variables. Some have also said that the impact of these variables is most in case of social science experiments. But in reality, these variables impact every experiment.

 

Also Read | What is Correlational Research?

 

Ways to Control Extraneous Variables

 

Random sampling is the biggest way with which you can control the extraneous variables. It will completely eliminate the extraneous variables but evenly divide them so that the impact is not too much. 

 

Without random sampling the effect of the extraneous variables will become a matter of concern. It will heavily impact the experiment or the research that is being conducted. 

 

Another way is either inclusion or exclusion. It is more of a direct approach in which you have to either include or exclude any extraneous variable that may impose an impact on the research. 

 

Variables causing threat or can cause potential threat can be excluded from the experiment altogether. Removing means to put a complete control on the extraneous variable that is causing disruption by keeping it constant throughout.

 

In the inclusion method, potential variables are included in the experiment. It might sound weird to add such a variable in the experiment but the main point is that it will become a way to measure the distractions in the research. All potential deviations in the experiment can be measured using these variables and by including them.

 

Here is an example- gender of the researcher is an extraneous variable but this variable should be included in the research and communicated to the panel. It will help when the experiment is conducted as the gender of the counselor can be used to administer the treatment. The relationship between the dependent and independent variables can also be determined.

 

Apart from the ones mentioned above, there are 3 other ways in which extraneous variables can be controlled:

 

1. Consistency in environment

 

It is important that each individual takes part in the exact same situation and exact same environment. The level of noise, temperature and even the lighting should be the same. 

 

It will remove and minimize even the slightest distractions or deviations in the final result. So, to control the extraneous variables, the environmental conditions must be consistent.

 

2. Design of the Experiment

 

While designing the experiment, it is important to randomly assign people to groups. It will also make the researchers clueless as to which person is assigned to which group. So, the experimental extraneous variables can be reduced.

 

3. Randomization

 

Every participant is different and adds a different aspect in the group. So, by randomization, all participants will be divided randomly. This way, all the personalities will be mixed and different abilities of people will also be divided into different groups. It will reduce the participant's extraneous variables.

 

Also Read | Stratified Random Sampling

 

Types of Extraneous Variables

 

“To say that... behaviors have different 'meanings' is only another way of saying that they are controlled by different variables.”

-- B. F. Skinner

 

Controlling the Extraneous variables is an important task, as it may help in understanding the alternate effects that can arise. In order to find out reliable results, extraneous variables must be controlled. There are 4 types of Extraneous variables and we are going to discuss each of them below.

 

---

4 Types of Extraneous Variables

---

 

1. Situational Variables

 

Situational Variables include those factors of the environment that affect the behavior of a person. These variables have to be controlled in such a way that they remain constant for all the participants. There are some set standards based on which these variables are controlled.

 

Examples of Situational Variables- Noise, temperature, Visual disturbances, lighting etc.

 

2. Person Variables

 

Just like the name shows, person or participant variables means the ways in which each person is different from another person. It is very important as it affects the overall results of the experiments. 

 

For example- if a memory test is conducted and there are 2 participants. One is tired or dyslexic and the other one has great eyesight. Then both these will have different results in the same experiment. So, to prevent such issues and control these person extraneous variables, proper resource allocation can be done.

 

Examples of Person Variables- Natural intelligence, moods, physical abilities, concentration power etc.

 

3. Investigator Variables

 

Investigator Variables are also called Experimental Biases. These variables unconsciously give instructions on how a person should behave. Giving clues unintentionally is an experimental bias. 

 

This way participants get to know about the experiments and what is expected from them. It will definitely affect the person’s behavior. The experimenter remains blank about the clues he is handing over unintentionally but the participants use it nonetheless.

 

Examples of Investigator Variables- Giving hints as to how the researchers want the people to behave, changing tone to positive or negative while explaining something so that people get hints etc.

  

4. Demand Variables

 

Demand Variables or demand characteristics are the clues that are conveyed to the people which tell them the basic purpose of the experiment. These variables affect the experiments a lot. People often behave a certain way or change their outlook so that a specified outcome comes. 

 

Examples of Demand Variables- There are lots of ways in which people get affected-

 

• Surroundings
• Researcher's behavior
• Their own interpretation etc.

 

The overall atmosphere of the experiment must be kept as minimal and natural as possible to avoid any conflicts or discrepancies.

 

Also Read | Types of Research Methods

 

Differences and Similarities between Extraneous and Confounding Variables

 

Confounding Variables are a type of Extraneous variable. They are similar to the extraneous variables also. The main reason for the similarity between extraneous and confounding variables is that they affect two variables that are not related to each other by any means. 

 

Here is an example how - Suppose X and Y are two variables such that X affects Y. Then the relationship between two is not only of casual relationship but they also signal towards a third variable i.e. Z. 

 

It also means that X and Y both are related but this relationship is exaggerated to such a point that the third variable Z impacts both X and Y only because X had an effect on Y. In this case, Z is a confounding variable. 

 

In the above example, Z is also an extraneous variable because of the impact it puts on X and Y. We stated above that confounding variables are a type of Extraneous variables only. This is a similarity between both these variables. 

 

Now let us look at the differences between the extraneous and confounding variables. Extraneous variables affect only the dependent variables whereas confounding variables affect the dependent variables but also correlates with the independent variables.

 

Participants that work in a science lab show a confounding variable. Reason being, it has a direct relation with the people wearing white lab coats and having a scientific reasoning skill. It is believed that people who wear lab coats have higher scientific knowledge than the others. So, any kind of bias or manipulation will not have any effect on the participants.

 

From this case we can derive the following differences between extraneous and confounding variables:

 

• The variables that impact scientific reasoning are called Extraneous variables. It includes the interests of participants.

 

• The variables that are related to both lab coats and other skills are confounding variables.

 

In simple words, People wearing lab coats (independent variables) and having scientific skills (dependent variable) are both confounding variables. While scientific skills (dependent variable) and having interest in science is an extraneous variable. 

 

Interest in science will have an impact on the scientific skills of a person but will not affect lab coats in any way. Hence, Extraneous variables. People that wear lab coats always possess some kind of scientific skills. Hence, a correlation exists between both i.e. confounding variables.

 

It becomes very important to measure the impact of independent variables on dependent variables and vice-versa. Extraneous variables can create a lot of distractions and change the results of the research. Therefore, it is essential to add some control to them.