Control Group

Random Assignment

In true experiments, subjects are assigned randomly to either a control or an experimental group. If the groups are alike in all ways except for the treatment administered, then the effects of that treatment can be tested without ambiguity. Although no two groups are exactly alike, random assignment, especially with large numbers, evens differences out. By and large, randomized group assignment results in groups that are equivalent, with the exception of the independent variable, or treatment of interest.

Placebo Control Groups

A placebo is a substance that appears to have an effect but is actually inert. When individuals are part of a placebo control group, they believe that they are receiving an effective treatment, when it is in fact a placebo. An example of a placebo study might be found in medical research in which researchers are interested in the effects of a new medication for cancer patients. The experimental group would receive the treatment under investigation, and the control group might receive an inert substance. In a double-blind placebo study, neither the participants nor the experimenters know who received the placebo until the observations are complete. Double-blind procedures are used to prevent experimenter expectations or experimenter bias from influencing observations and measurements.

Another type of control group is called a waiting list control. This type of control group is often used to assess the effectiveness of a treatment. In this design, all participants may experience an independent variable, but not at the same time. The experimental group receives a treatment and is then contrasted, in terms of the effects of the treatment, with a group awaiting treatment. For example, the effects of a new treatment for depression may be assessed by comparing a treated group, that is, the experimental group, with individuals who are on a wait list for treatment. Individuals on the wait list control may be treated subsequently.

When participants in an experimental group experience various types of events or participate for varying times in a study, the control group is called a yoked control group. Each participant of the control group is “yoked” to a member of the experimental group. As an illustration, suppose a study is interested in assessing the effects of students’ setting their own learning goals. Participants in this study might be yoked on the instructional time that they receive on a computer. Each participant in the yoked control (no goal setting) would be yoked to a corresponding student in the experimental group on the amount of time he or she spent on learning from the computer. In this way, [Page 252]the amount of instruction experienced is held constant between the groups.

...