Control experiment?

What is a control experiment and why do we need that?

Control Experiment, by example

Imagine that your science project is about the effect of heat on evaporation of water. In other words you want to know if heat can cause evaporation of water. (Seems simple, doesn't it?)

Your hypothesis is "If we heat up water, it will evaporate."

As your experiment, you place some water in a beaker and heat it up over an electric stove. All water evaporates in about 5 minutes. You write your results and conclude that heat does in fact cause evaporation of water. 

Is this a reliable result?

Someone may argue that water would probably evaporate by itself even if you would not apply any heat. As a result, there is no clear relation between the heating and evaporation of water.

This is when you need to repeat your experiment; however, this time you will also set aside a second beaker with the same amount of water and do nothing with that. You call it a control experiment. 

You heat up the first beaker until all the water evaporates, then you look back at your control experiment and see that all the water is still there. Now you have shown that water would not evaporate by itself and conclude that heat has caused evaporation.

Control experiment is a parallel (simultaneous) experiment in which you do not change any thing in order to show that changes in your independent variable are not caused by an unknown phenomena, but they are a response to the changes in the independent variable.

Related topics:

Go Back