Sources used in this document:

Why would a researcher, who values internal validity, then choose a quasi-experimental design, that specifically lacks in internal validity? The reason is that experimental designs are not always the most effective. They're subject to social threats of internal validity, and have difficulties with external validity (Trochim, 2008, 188). An experimental design is "intrusive and difficult to carry out in most real-world contexts," and is basically an "artificial situation" created to "assess [a] causal relationship with high-internal validity." Resultingly, there are difficulties in generalizing the findings to the real world.

Alternatively, certain types of quasi-experimental designs are some "of the most intuitively sensible designs around" (Trochim, 2008, 210). This is because random selection is often not logical or possible. For example, say we wanted to understand the effects of a certain type of treatment on both developmentally disabled adults and adults that were not developmentally disabled. Random selection, or an experimental design, would be impossible here, because we already have our two groups; they're dictated by the difference.

The basic reasons one might choose an experimental design or a quasi-experimental design has to do with, I think, resources, and what exactly is being tested. I feel that there is no design that is "better" or "worse" -- applying those labels would not make any sense at all here, because every experiment has a different aim. What is good for one experiment may be bad for another. The most important thing, I feel, is simply to have an understanding of both, and the ability to employ each when necessary.