Control Group

Control Group

Author: Jonathan Osters

This lesson will explain control groups in experiments

See More
Introduction to Statistics

Get a load of these stats.
Our Intro to Stats course is only $329.

Sophia's online courses not only save you money, but also are eligible for credit transfer to over 2,000 colleges and universities.*
Start a free trial now.


Video Transcription

Download PDF

This tutorial is going to talk to you about control groups. And a control group is a group that doesn't receive any kind of treatment. Now, you might be wondering if that's just a waste to have a group that doesn't receive any treatment. Well, it's not a waste, because a group that doesn't receive any treatment helps to establish a baseline for what the effectiveness of nothing would have been. And so this helps us to determine the effectiveness of the treatments, if we can determine what the effectiveness of nothing would have been.

So for example, AIDS is a devastating disease. It's killed many people. And so suppose a company wants to test the effectiveness of a new drug that they've come out for AIDS. So many people might say that well, you should just give the drug to everybody, because it might help. And that's true. It might help, but it might not. If they just gave the drug to everybody, we wouldn't be sure if maybe some people just got better by chance, and it wouldn't have been due to the drug. We can't be sure whether observed differences are due to the drug or some other thing.

So when we create a control group, what we're doing is we're saying, all right, some people are going to get the real drug, some people are going to get a fake drug, and we'll see if the differences are significant enough. So what the company should do is implement a control group, a group that doesn't receive the drug, and a treatment group which does.

We can make comparisons between the two groups if they're as similar as possible, except for the drug they're receiving. And the response of the control group will be compared with the response of the treatment group. And if there's a difference, then we can attribute it to the drug, and not some other factor.

Now sometimes, it's possible to have multiple treatments be compared to a single control group. Suppose I'm doing an experiment on plants, and I want to know if colored lighting is just as good as regular old sunlight for plants. My control group might be randomly assigning certain plants to be outside in the sun, and then randomly assigning other plants to be under blue light or orange light or green light. So these are multiple treatment groups and one control group. But the control group helps to establish a baseline for growth that we can compare against all of these other groups.

And so to recap, the purpose of the control group is to provide a baseline, so we can determine if there's a difference between the treatments and the control by assigning no treatment to the control group. And the terms that we used were control group and treatment group. The control group receives no active treatment. The treatment group does receive an active treatment. Good luck. We'll see you next time.

  • Control Group

    A group which receives no treatment (or an inert treatment) in an experiment. This group's response provides a standard against which the response to the treatment(s) can be compared.

  • Treatment Group

    A group that is manipulated by the treatments in the experiment. There may be many treatment groups in an experiment, but only one control group.