Quasi-experimental designs share the same purpose with true experiments that attempt to test the causal impact of an independent variable such as an intervention by manipulating the intervention and observing the outcome. However, a quasi-experimental design does not utilize random selection and/or random assignment of participants. Rather, participants are assigned nonrandomly to the experimental and control groups.

There are three basic forms of quasi-experimental designs. The first type is nonequivalent groups design (NEGD), in which researchers often use intact groups (expected to be as similar as possible before the intervention) as experimental and control groups. The second form is time series design, in which multiple consecutive observations on an outcome are measured before and after the intervention over time. Discontinuity in behavior in the time series ...