Quasi-experimental design is a technique for performing research studies that has multiple advantages and disadvantages. The disadvantages are easy to see; the quasi-experimental design falls short of the true experimental design both in terms of accurately sampling the population and with respect to the researcher’s degree of control over the intervention. Its advantages may be less obvious, but as one researcher describes the situation, “The use of a quasi-experimental design is suggested when either randomization cannot be achieved or when setting up a control group is not feasible. In this way, the experimenter can approximate experimental procedures for collecting data...” (Riecken and Boruch 87). This shows that the quasi-experimental design has advantages in that it allows researchers to do studies that would not work with any other technique.
Quasi-experimental design is particularly relevant for the field of criminal justice. For example, the population of interest for criminal justice studies can involve those who are in prison or on parole, which can put limits on the range of interventions that would be available to the researcher if a classical experimental design were used. In addition, ethical and legal considerations in criminal justice might prevent researchers from studying criminal activity out in the real world without intervening, which means that many studies end up being retroactive. These works can be considered quasi-experimental in design because the intervention in question—for example, the presence or absence of certain programs at an inmate’s high school—is not under the direct control of the researcher. However, by means of propensity score matching, each subject in one experimental outcome—such as being in prison, for this example—can be matched with a similar subject in a different outcome. This approximates the role of randomness in statistically robust studies, allowing rigorous analyses to be done even in areas that would not otherwise be capable of study.
Riecken, Henry W. and R. F. Boruch, eds. Social Experimentation: A Method for Planning and Evaluating Social Intervention. New York, NY: Academic Press, 1974.