1.  A major problem is how to make the comparison groups as similar to the experimental group as possible.

2.  Differences are computed, and the program is deemed a success if the experimental group has improved more than the control group.

3.  For example, instead of receiving vitamin C as the experimental group is doing, the control group receives a placebo.

4.  Sometimes the whole experimental group is matched with a similar group at the start of the program.

5.  In this case a control group would be selected which matches as closely as possible, member for member, the characteristics of the people constituting the experimental group.

6.  The presumed causal factor, normally referred to as the independent variable, is withheld from the control group and administered only to the experimental group.

7.  So, in our example, it would be the experimental group which was exposed to the new teaching style.

8.  The classic experiment requires both a control and an experimental group to which subjects are randomly allocated.