This classical experimental design developed to permit measurement of the effect of the experimental variable alone, may be depicted as follows:
Experimental Group Control Group
Before measurement Yes (x1) Yes (y1)
Experimental variable Yes No
After measurement Yes (x2) Yes (y2)
Effect of experimental variable = (x2 – x1) – (y2 – y1)
The experimental group and the control group are selected in such a way that they are similar, and they are considered interchangeable for test purposes. The control group is measured at the same times as the experimental group, but no experimental variable is introduced. Thus, the difference between the after and before measurements of the control group (y2–y1) is the result of uncontrolled variables. The differences between after and before measurements in the experimental group (x2–x1) is the result of the experimental variables plus the same uncontrolled variables affecting the control group. The effect of the experimental variable alone can be determined by subtracting the difference in the two measurements of the control group from the difference in the two measurements of the experimental group (x2 – x1) – (y2 – y1)
An experiment run by the National Broadcasting Company (NBC) illustrates this design. A carefully selected sample of 2,441 male and female household heads in a medium-sized Midwestern market was interviewed at two different times three months apart. Purchases during the preceding four week period of 22 different brands in 11 different household product categories varying from beer to toothpaste were determined in each of the two interviews. In the first interview the percentage buying one or more of the brands was determined for the entire sample and used as the before measure. In the second interview the sample was separated into two sub-samples – those exposed to 5television and magazine advertising of the products and those not exposed to the advertising.
Purchases by those exposed to advertising during the three month period increased slightly (20.5%-19.4%=1.1%), while purchases by those not exposed to the advertising actually decreased 2.5 percent (16.9%-19.4%). It is assumed that purchases by the experimental group would have also decreased 2.5 percent except for the effect of the advertising. Therefore, it is concluded that the net effect of the advertising was to increase purchasing from 16.9 percent of the household heads to 20.5 percent, a difference of 3.6 percent. This is an 18.6 percent increase in purchases as the results of advertising.
Experimental Group Control Group
In past four weeks ) 19.4% (x1) 19.4 (y1)
(exposed to advertising
of products) Yes No
After measurement (percentage
Purchasing in past four weeks
At time of second interview) 16.9% (y2)
Effect of experimental variable =(20.5%-19.4%) ( 16.9%–19.4%)
= (1.1%) – ( — 2.5% ) = 3.6%
Additional Comments: In this example, the control group is a group of consumers, presumably similar to the experimental group except for not being exposed to the experimental variable. Another type of control could have been purchases of similar, but unadvertised, products by the same consumers. When the experiment is limited to one medium sized Midwestern market, the degree to which the results can be generalized is limited to the one market studied; although strong suggestions may be drawn relative to other medium sized Midwestern cities and even beyond.
This design has definite advantages over the simple before-after design. The effects of the uncontrolled variables, history, maturation, pre-testing, and measurement variability, should be the same for the control group as for the experimental group. In this case, these factors appear to have had a negative effect on purchases of 2.5 percent. If it had not been for the experimental variable, presumably the experimental group would have shown a similar drop in purchases during the period.