Experimental research has been touted as one of the most rigorous research designs, due to a built-in safeguard for internal validity -- randomization. A quasi-experimental design is very similar to an experimental research design, but lacks the key element of randomization. Both designs feature an experimental group and a control group, but the manner of group selection differs. Therefore, the researcher ends up with non-equivalent groups. This design is referred to as a non-equivalent groups design, the most common quasi-experimental design. Quasi-experimental designs offer some advantages and disadvantages.
Limited Ability to Compare
Using a sampling method other than random sampling increases the potential for constructing non-equivalent groups. Ideally, researchers endeavor to obtain experimental and control groups that are alike. This is most effectively achieved and most likely to occur through random selection. Quasi-experimental designs do not use random sampling in constructing experimental and control groups. Using nonuniform comparison groups can limit generalization of the findings because non-controlled variables may have influenced the results.
Beginning research with non-equivalent groups presents a threat to internal validity. Internal validity refers to the degree to which a researcher can be sure that the treatment was responsible for the change in the experimental group. If the researcher does not start with equivalent groups, then the researcher cannot be sure that the treatment was the sole factor causing change. Other confounding factors may have contributed to the change. Therefore, not using random sampling methods to construct the experimental and control groups, increases the potential for low internal validity.
Logistically Easy to Manage
Quasi-experimental designs are commonly utilized in social research. These designs are also used in education to test the effectiveness of a program. In a typical quasi-experimental design, two classes may be selected, a pretest given to both, and then the program -- treatment -- given to the experimental group. A post test is conducted to determine if there was a change in the groups. In education, these groups often come pre-determined, such as in a school or class. Therefore, the researcher is not required to group individuals, as they come pre-grouped.
Generalization Possible with Control Group
Some Quasi-experimental research designs offer the benefit of comparison between groups that can be statistically analyzed. For example, if an experimental group of elderly arthritis sufferers is given treatment and the control group receives no treatment, the findings could potentially reveal a statistically significant difference in pain relief or increased mobility among the treated group. This is a major advantage because it helps the researcher to make inferences about the possible existence of a cause and effect relationship of the treatment.
- dolgachov/iStock/Getty Images