This paper compares and contrasts experimental and quasi-experimental research designs, examining their defining characteristics, strengths, and limitations. It explains how experimental designs achieve strong internal validity through random assignment, using the two-group posttest-only randomized design as an example. The paper then explores why researchers sometimes opt for quasi-experimental designs despite their weaker internal validity, focusing on real-world feasibility, social threats, and situations where random assignment is impossible. The discussion concludes that neither design is inherently superior; the appropriate choice depends on the research context, available resources, and the specific question being investigated.
In research, different types of problems require different ways of observation and testing. Often limiting a researcher's view of the problem are situational factors that can skew the results of an experiment — effects of pretesting, social threats, and group differences (Trochim, 2008, p. 188). External factors, such as possible sample size, can limit even the type of testing available to the researcher. As such, researchers have developed a number of different types of designs over the years. This essay compares and contrasts two of these: experimental and quasi-experimental designs.
"Experimental designs are often touted as the most rigorous of all research designs" (Trochim, 2008, p. 186). What makes them so rigorous is that they represent the strongest type of design with regard to internal validity (Trochim, 2008, p. 186). This is because the basic form of the experimental design uses random assignment — that is, chance — to place participants into groups. In effect, this makes the two groups, if selected from the same sample, essentially equivalent; the only thing that separates them is chance. This enables the researcher to actually "calculate the chance that the two groups will differ just because of the random assignment (that is, by chance alone)" (Trochim, 2008, p. 189).
The way that experimental designs control internal validity so effectively is a function of their structure. A simple experimental design called a two-group, posttest-only, randomized design uses random assignment to create two groups: one in which a treatment is offered, for example, and one that receives no treatment at all. Posttesting is then performed. If the treatment group shows a different outcome than the no-treatment group, we may be able to conclude that the difference in outcome was a result of the treatment. We can be confident in the experiment's internal validity because, since groups were assigned randomly, we know that the difference in outcome was not the result of pre-existing differences between the groups.
Contrast this with the quasi-experimental design, which has significant difficulties with internal validity. A quasi-experimental design "is one that looks a bit like an experimental design but lacks the key ingredient — random assignment" (Trochim, 2008, p. 210). Why would a researcher who values internal validity choose a quasi-experimental design that specifically lacks it? The reason is that experimental designs are not always the most effective option. They are subject to social threats to internal validity and have difficulties with external validity (Trochim, 2008, p. 188). An experimental design is "intrusive and difficult to carry out in most real-world contexts" and is essentially an "artificial situation" created to "assess [a] causal relationship with high internal validity." Consequently, there are difficulties in generalizing the findings to the real world.
"Real-world feasibility favors quasi-experimental approaches"
"Context and purpose determine the better research design"
You’re 66% through this paper. Sign up to read the remaining 2 sections.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.