What Is a Control Group? (With Uses, Types and 11 Examples)

By Jamie Birt

Updated October 17, 2022 | Published July 13, 2021

Updated October 17, 2022

Published July 13, 2021

Jamie Birt is a career coach with 5+ years of experience helping job seekers navigate the job search through one-to-one coaching, webinars and events. She’s motivated by the mission to help people find fulfillment and belonging in their careers.

A scientific researcher looks through a microscope while another person makes notes in the foreground. Both are dressed in white coats.

Experimentation helps scientists and other researchers learn information, support hypotheses, formulate new methods and determine the effectiveness of current investigative strategies. Using both control groups and variables, researchers compare results and generate conclusions, which research teams and the scientific community review for accuracy. Learning about the different types of control groups can help you establish best practices for conducting experiments in your research career. 

In this article, we explain what control groups and variables are, describe who uses control groups and offer 11 control group examples to help you prepare a successful experiment.

What is a control group?

A control group is a collection of factors that remains constant throughout an experiment. While some groups in an experiment may experience a particular changing variable, like a new medication, others may not encounter this variable to see if the researchers notice any effect for those that didn't. A control group may also receive alternative treatment. Researchers can then compare the control group with the outcomes of the experiment to collect accurate results. This can tell them if the variable had an effect or if some other factor caused a change in both control and variable groups. 

Read more: What Is a Control in an Experiment? (Definition and Guide)

What are variables?

Variables are the factors in an experiment that are subject to change. When they design the experiment, researchers may control the assignment of variables to different experimental groups to see their effects. From there, they can compare variable groups with the control group to draw conclusions. 

For example, if a medical team is researching the efficacy of a new drug in treating cancerous cells, they might use a common cancer drug as a control. This can help them determine how much more effective the new version of the drug, or the variable, is. Comparing the common drug to the new drug can help the research team find conclusive evidence to support or deny the original hypothesis.

Related: Extraneous Variables: Types, Examples and How To Control Them

Who uses control groups?

Here are some industries that use control groups in experiments:

  • Medicine: Control groups in medicine help researchers determine the efficacy of new drugs, potential side effects and how certain illnesses respond to specific treatments.

  • Psychology: Psychologists can use control groups to examine mental health and behavior in specific groups of people.

  • Marketing: Marketers may use control groups to study the effectiveness of different ad campaigns and determine who their ideal customers are.

Related: 10 Careers You Can Pursue in Medical Research

11 control group examples

Here are 11 examples of common control groups that different industries may use in their research:

1. Positive control group

A positive control group is one that receives samples or treatments that the researchers running the study already know work. For example, if researchers are testing new body armor for police officers, they may use industry-standard body armor as the positive control and conduct experiments to evaluate the protection level of the new body armor. The standard body armor is the experiment's control group because the researchers know how it performs.

2. Negative control group

Negative control groups are ones that researchers don't expect to influence the results of an experiment. This type of control group allows researchers to compare variables against a group they know won't produce different results. For example, if a company produces a new drug to treat stomach pain, researchers may only give the drug to one set of study participants. Those who don't receive the drug are the negative control group, and they provide a comparison point for the drug's effectiveness in treating stomach pain.

Related: What Is Multivariate Analysis in Data Science? (Plus Techniques)

3. Placebo control group

A placebo control group accounts for the possibility of the placebo effect occurring in an experiment. The placebo effect is when a member of a control group believes they feel an effect from the experiment even though they haven't received treatment. This can create inaccurate results so researchers may add some control factors into a group to account for the possibility of the placebo effect. For example, a company testing a stomach pain drug might give the placebo group a pill that doesn't contain medication but looks identical to the pill given to the experimental group.

Related: 6 Types of Research Studies (Advantages and Disadvantages)

4. Randomized control group

Randomized control groups allow researchers to keep their studies balanced by making control factors unpredictable. By selecting members of the control group randomly, they can make sure that factors they aren't testing,  like gender and age, won't affect results. Researchers could flip a coin for each participant, pick a number or randomly select control group members out of a lineup. This helps make the groups more even so researchers know that none of these factors caused an unexpected effect. 

Related: Designing an Experiment: 8 Steps Plus Experimental Design Types

5. Untreated control group

An untreated control group creates a baseline or default for researchers. For example, if a medical team is researching a new hand cream to treat eczema, they might create a small group that receives no cream. The research team can verify how well the cream works by comparing the untreated group with the treated group, which is the positive control group. In many cases, researchers can use an untreated control group as a negative control group if they don't expect those in the group to experience any effect. 

6. Double-blind control group

Researchers use double-blind, or double-masked, control groups to help eliminate potential biases from the research team and research groups. In a double-blind study, neither the researchers nor the participants know which group is the control group. This method may produce a more accurate, untainted result for an experiment.

Related: Single-Masked vs. Double-Masked Study: What's the Difference?

7. Historical control group

Researchers create historical control groups based on previous studies with similar variables or control groups. This helps ensure consistency and accuracy in the results and eliminates the need for new experiment requirements. For example, if a medical research team is testing a newer version of a drug, they might create a historical control group based on the requirements of the control group in the drug's original study.

Related: Prospective Study vs. Retrospective Study: What Are the Differences?

8. Waitlist control group

Researchers may receive many applicants for a study. A waitlist control group considers applicants who don't participate in the first round of a trial but aim to serve in a control group in a later experiment. This can help researchers create a large pool of potential control groups and give them the ability to compare results across different groups and time periods.

Related: 10 Types of Variables in Research and Statistics

9. Multiple control groups

Some experiments use multiple control groups to gather more information and improve the accuracy of results. For example, many medical studies include positive, negative and placebo control groups for a better idea of a treatment's or medication's efficacy. In this case, some groups may receive treatment, some may receive a placebo and some may not receive anything. This can help researchers identify exact problems or advantages and create fewer biases during experimentation.

10. Blinded control group

Researchers may create a single-blinded control group so participants aren't aware that they're in the control group. Alternatively, it may be the researchers who are unaware of who the control group is until after the experiment to allow for unbiased testing. For example, a construction technology company might enlist a workplace safety company to develop a new safety harness. The company may provide three groups of people for testing, each from a different construction background, but may not tell the researchers which group is the control. This can help the company establish more accurate results and choose the best product.

11. Natural control group

Natural control groups occur randomly because of natural forces such as weather, gravity or temperature. These groups often provide valuable information and the opportunity for further investigation. For example, a research team analyzing the effects of pollution on fish in a certain body of water might create a natural control group using the same species of fish from a different location.

Explore more articles