The purpose of the control group in an experiment is to control the various possible effects of rival explaining a causal finding. We may then be in a position to take actual view that the study is valid internally. The random assignment of subjects to the control and experimental groups and the presence of a control group enable to eliminate rival explanations and eliminate internal validity threats.
The threats include the following:
- The threat of testing implies the possibility that subjects may become sensitive to the objectives of the experiment.
- The threat of history refers to the possibility that events in the experimental environment may have led to the changes.
- The threat of maturation says that people change and the ways they change have implications for the dependent variable
- Another threat is that if differences exist between the two groups because of selection by a non random process, there would be variations between the control group and experimental groups. This could be attributed to pre-existing differences in their membership.
- A direction of causality would be there if there is the notion of an independent and dependent variable. There may be situations when the temporal sequence becomes unclear which makes it difficult to establish which variable affects the other. The presence of a control group helps to clear this feature.