Sonoran Desert

What is a Control Group?

by Diane Boudreau

Control groups are an important part of many scientific experiments. Researchers use control groups to find out if a treatment really has an effect. Sometimes, study subjects act differently after treatment just because they know they got treated.

For example, imagine this scene. You go to your doctor and tell her that you can't sleep at night. She gives you a pill to help. You go home, take the pill, and fall right to sleep. The pill did the trick, right?

Maybe. Or, maybe you believed the pill would work, so you quit worrying about falling asleep. You got so relaxed you had no trouble sleeping. It was your state of mind, not the pill itself, which helped you.

The way to find out what really worked is to conduct an experiment using a control group. In this type of experiment, about half of the subjects would receive the real sleeping pill. The rest would receive a fake pill, usually made of sugar. The fake pill is called a placebo. None of the subjects would know which pill they got.

After the experiment, the researcher would compare the results. If the people taking the real pill fell asleep, but the control group didn't, then we know the pill works. If both groups fell asleep, then we would know that the act of taking the pill, not the pill itself, was what worked. This is known as the "placebo effect."

Even animal researchers often need to use control groups. For instance, Sarah Woodley wanted to know if removing a lizard's ovaries would change her behavior. But she needed to be sure that the loss of ovaries, not the experience of capture and surgery, was what caused any change. So she used a control group of lizards. The "control" lizards went through surgery, but kept their ovaries.

Can you think of some reasons why a lizard might be less aggressive after surgery, even if the surgery is not quite the real thing?