A control group (sometimes referred to as “control” for short) is a group of study participants that does not receive the treatment of interest, but remains the same in all other aspects. Controls are used in studies for multiple reasons, including to decrease study variables, to establish a baseline, to increase internal validity, and to determine causality.

A control group is used by many studies use to limit confounding variables. Confounding variables are factors that can confuse, or “confound,” results of a study. In research, we want to limit confounding variables, so that we know the study’s outcomes were due to the experimental treatment and not due to an unrelated factor, like participants changing their behavior because they’re in a study, or receiving extra care by participating in the study, and so on. Studies compare the results of the experimental treatment to those of the control to try to eliminate confounding variables, in order to accurately evaluate the effects of the experimental intervention on its own.

Controls not only help limit confounding variables, they can also help compare the intervention to a variety of other interventions, or to no intervention at all. If researchers want to determine whether an intervention they’re testing works at all, they may choose to use no treatment or a placebo as a control, since the placebo should have no effect. However, if researchers want to see whether a new intervention is better than an old one, they may use the old therapy as an active control in order to see whether the new intervention outperforms it.

Perhaps most importantly, controls can help determine causality: researchers can determine whether the treatment is responsible for the effects on the outcome(s) by comparing the outcomes of the treatment group to those of the control group. People’s health and symptoms change over time for many reasons unrelated to treatment: people get better naturally on their own, the seasons change, study participation prompts behavioral change, and more. However, these effects should be the same in both the treatment group and the control group. By making everything but the treatment the same between the treatment and control groups, then comparing the change over time in the treatment group to that in the control group, researchers can be more confident that the changes in the treatment group are due to the treatment alone, and not to other factors. This is why single-arm before-after trials can’t reliably determine causality.