In the scientific method, a control experiment is an experiment where the variable that is being investigated or tested is kept constant. This allows for a comparison to the experiment where the variable is changed to see if there is a different result.

As an example, suppose a scientist wanted to test the hypothesis that "Brand X" fertilizer improves the growth of tomatoes. He might set up an experiment where he grows a hundred tomato plants and fertilizes them with Brand X, monitoring them over a period of time to see how large they grow.

The scientist must make sure that there were no other influences on his experiment, such as the season or the particular genetic strain of tomatoes involved. So he sets up a control experiment to run at the same time. He takes another hundred plants which are as similar as he can find to the first group, and grows them in exactly the same way as the first group, but without Brand X. He then monitors them as he monitored the first group.

The scientist can now compare the results of the first group (experimental group) with the results of the second group (control group) and come to a meaningful conclusion.