What Are Extraneous Variables?
An extraneous variable is any variable that is not a subject of your experiment but appears in the process and can influence the results. To avoid inadequate conclusions, you should control extraneous variables carefully. Relations between independent and dependent variables should be accurately checked, and any external impacts should be completely excluded.
Why Are Extraneous Variables So Important?
The appearance of extraneous variables and their influence on the results of the research can diminish the internal validity of the study. You may go away from the correct conclusions by finding alternative ones that are not correct.
When you conduct an experiment, you do certain manipulations over the independent variable to see its impact on the dependent variables. For example, you want to test the influence of the uniform color on the mental performance of clinic nurses. You go to the hospital and choose the participants for your study. You divide them into two groups. The experimental group wears white uniforms, while the control group wears blue ones.
Then, you give a medical quiz on first aid and ask all the participants to answer the questions within a limited amount of time. You compare the results of the two groups and see that the group wearing blue uniforms has answered all the questions correctly, while nurses in white gowns have made several bad mistakes. Therefore, you make the conclusion that the blue color in medical workers can stimulate their mental performance.
However, there may be several extraneous variables that have affected the results. They are participants’ experience, major at college, weather conditions, time of day, the environment in which the experiment has been conducted, or the participants’ mood and health condition. These variables can differ from person to person, so you cannot trust the final results. Moreover, it is difficult to make conclusions.
As you see, the outcomes of such an experiment can explain why it is so important to control extraneous variables in the experimental design. If the external variable is under your control, you can name it a control variable.
The Difference Between Extraneous and Confounding Variables
It is believed that a confounding variable is a kind of extraneous variable. Though, it is different from it.
- Extraneous variables influence dependent variables only.
- Confounding variables can affect both independent and dependent variables because they correlate with them.
If you want to imagine it on a diagram, you can draw an arrow from a confounder to an independent variable and to a dependent variable.
As in the example, if you have a nurse with more than 15 years of experience in her job, it is a confounding variable. It implies that this person is more professional and knowledgeable, so not any manipulations with the color of the uniform can influence her perfect results - she would produce them wearing either a white or blue uniform because she has gotten accustomed to wearing uniforms of different colors during all those years of work.
All other variables that do not relate to the subject of your experiment, such as the college major or mood, are extraneous variables because they influence only the mental presentation of the participants, though they never affect the independent variable you manipulate.
How to Control Different Types of Extraneous Variables
Here are various factors that affect the ways of controlling extraneous variables of different types.
Variables Produced by Demand Characteristics
If participants can understand what is expected from them (and they can do it when seeing the materials, questions, or settings of the experiments), they receive important cues that help them behave as it is expected. These cues are called demand characteristics.
✔️ They can make participants behave according to the study hypothesis. Such behavior, in its turn, can bias the results of the experiment and make them invalid. Therefore, generalizability, or external validity reduction, is also possible.
In the experiment described above, the nurses who were asked to wear uniforms of different colors understood that there should be some specific difference between their ways of answering the quiz questions because they were wearing different colors. Even if they did not know what color was better for doing the quiz successfully, they decided to be more focused because of the gown color they were wearing. So, the results obtained in this way cannot either support or reject the hypothesis that the color of the uniform can influence mental performance.
You can avoid providing your participants with the demand characteristics by making it difficult for them to guess the real purpose of the experiment. You can ask them to go to isolated rooms, wear their uniform color, and answer the questions without being shown up to the other group.
Variables Produced by Demand Characteristics
The researcher can unwillingly do some actions that can impact the results of the study. These actions can be of two types:
- The experimenter interacts with participants in two different groups differently, and this behavior prompts them on what they have to do (or not to do).
- The researcher makes errors in measurement, observation, and interpretation.
You encourage the participants wearing white uniforms by saying that they look like serious professionals in this attire, so they need to do their best to prove that they really are. They start feeling more confident and concentrated and, as a result, perform well. For some reason, participants who wear blue uniforms are not encouraged at all. They do not know they have to correspond to their high professional status, do not concentrate, and perform worse than the first group.
Therefore, you need to use masking or blinding to hide the conditions of the assignment from the participants and other experimenters. A double-blind study is an excellent way to do that because no researcher can interpret the results of the experiment as it is needed to support the initial hypothesis.
Situational Variables
These variables can change the reactions and behavior of the participants. The air temperature, lighting, weather conditions, sudden noises, and many other factors are called situational because they do not exist permanently in the experimental environment. They can lead to accidental errors and unexpected variations in the measurements.
The best way to avoid the influence of situational variables is to get rid of the factors that may affect the study results.
A good example of a situational variable in our experiment is the appearance of a sudden noise in the rooms where the experiment is conducted. Suppose there is a sort of emergency pipe repair work outside at the time the participants are doing the quiz. Their attention can be distracted. Therefore, the best way to avoid such influence is to change the room for the experiment and choose one where the participants cannot hear this noise. If it is impossible to make any abrupt changes, you need to account for them while doing the statistical analysis.
Variables Resulting from Participants
Any aspect of participants’ backgrounds or characteristics can influence the study too. Such factors are called participant variables. They are not usually the aim of the experiment. Such variables can include gender, marital status, age, religious beliefs, etc. These individual peculiarities can lead to unexpected results and differences in the outcomes. That is why it is essential to take them into account before the experiment, or measure and analyze them if they are impossible to avoid.
For example, the participants of the nurse experiments are mostly young women aged 25-30. One-third of them is married, while others are single. They do not have strong educational backgrounds because half of them are only undergraduates in specific STEM programs. Their age, gender, and educational background influence their mental performance a lot, and you need to consider this while dividing your participants into experimental and control groups. It is probably more useful to apply a random assignment to make an equal division of your sample into two groups. When divided equally, the participants' individual characteristics will be distributed evenly and will not influence the results of the experiment.
Final Thoughts
You know now what extraneous variables are and why they can influence the outcomes of your study. You can also distinguish between different types of extraneous variables and know how to deal with them. Using this knowledge in your experiments will let you obtain more viable and valid results. If you manage to do it, your final research paper will be successful, and you will make a good advancement in your further academic career.