control group A group used for comparison with another, either because it represents the most common or typical case, or because it illustrates the absence of some phenomenon being studied. Human beings are not inert, lifeless matter, and hence cannot be studied simply by carrying out before-and-after studies around some experimental stimulus applied to them. Many other developments and changes will be occurring spontaneously at the same time, as people go about their lives, and it is difficult to disentangle the effects of one particular stimulus from all the others affecting people, as individuals or as groups, at the same time. The solution is to identify a group, or social aggregate, that serves as a control group in providing information on characteristics or changes in a baseline situation that is as close to ‘ordinary’ as possible, or illustrates social behaviour in the absence of the key factor of interest. Control groups can be created at the sampling stage, through statistical manipulation of the data-set, or during analysis. The most rigorous type is experimental controls, obtained by random allocation of cases. See also CONTROL (EXPERIMENTAL).

Control group

In an experiment that focuses on the effects of a single condition or variable, the group that is exposed to all the conditions or variables except the one being studied.

Scientists often study how a particular condition or factor influences an outcome. In such an experiment, in which there are two groups of subjects, the group that is exposed to the condition or factor is called the experimental group . The other group, which provides a basis for comparison, is called the control group. For example, in a hypothetical study of the influence of the presence of loud music on the test performances of children, the control group would consist of the group of children not exposed to the loud music during the test. Their test scores would be compared with the experimental group, the group of children who were exposed to loud music during the test. In this type of experimental design , subjects would be randomly assigned to each group to ensure a reliable comparison.

Further Reading

Atkinson, Rita L.; Richard C. Atkinson; Edward E. Smith; and Ernest R. Hilgard. Introduction to Psychology. 9th ed. San Diego: Harcourt Brace Jovanovich, 1987.

Zimbardo, Philip G. Psychology and Life. 12th ed. Glenview, IL: Scott, Foresman, 1988.