Solomon Four-Group Design

Benefits of Pretest Inclusion

Researchers implement pretests with the goal of obtaining information regarding baseline levels of specific variables of interest (e.g., self-esteem, knowledge) prior to the implementation of an experimental stimuli or intervention. The central benefit of collecting pretest data is that it provides a comparison point for posttest data. Illustrative of the benefits of a pretest, consider a study where two classes received educational interventions. If the average SAT score of Class A was 1,000 and Class B was 800, it is possible that the intervention provided to Class A was more useful than the one provided to Class B. However, if SAT scores of Class A were 1,200 at pretest and Class B’s were 200 at pretest, a very different picture of the intervention impact is presented. In this instance, without knowledge of the pretest scores, researchers could make inaccurate conclusions about the effectiveness of the treatment. On the other hand, if pretest scores indicated that the two classes did not significantly differ from each other at pretest, the investigator can have increased confidence that posttest differences were caused by the treatment.

Costs of Pretest Inclusion

Even though the benefits of a pretest include greater control over an experimental or quasi-experimental design, several drawbacks accompany their use. These shortcomings include increased monetary cost, increased time consumption, and pretest sensitization. As noted, pretest sensitization occurs when the implementation of the pretest leads participants to respond to the stimuli or the posttest assessment differently than they would have otherwise.

The problems associated with pretest sensitization are manifold. Firstly, a pretest can alert participants to the questions that are likely to appear on the posttest. For instance, if students receive a difficult pretest examination prior to a math-based intervention, some of them might take it upon themselves to learn the answers to the pretest’s challenging questions—irrespective of the quality of the intervention that was received. If students in both conditions learn the answers to the posttest as result of the pretest, any possible differences between groups caused by the educational intervention will be obscured as, due to the pretest, both the treatment condition and the control condition will have high scores on the posttest. Another possibility is that the students will focus only on the aspects of the educational intervention that were covered on the pretest, thus resulting in scores that they would not have received if they did not have prior knowledge of questions on the posttest.

...