Glossary A–ZLetter C
Control group
A control group is a comparison group that was not exposed to the measure being studied and is otherwise as similar as possible to the test group.
The difference between the two groups shows what the measure achieved. Without a control group, an effect can hardly be separated from chance and circumstances.
In screen networks, the control group can consist of stores without screen advertising, periods without a campaign or people who did not pass a screen. An example: 20 stores show an offer on screen and 20 comparable stores do not. If sales rise equally in both groups, the cause was more likely the season or the price than the screen.
What matters is that the test and control groups are comparable, for example in size, location, customers and product range. A comparison only with the period before the campaign is weaker, because much can change between the periods. When looking at effectiveness studies, operators should ask whether there was a control group, how it was selected and how large it was. Small groups easily produce chance results.