Quasi-Experimental Design

Quasi-Experimental Design Definition

Quasi-Experimental DesignA quasi-experimental design is a research methodology that possesses some, but not all, of the defining characteristics of a true experiment. In most cases, such designs examine the impact of one or more independent variables on dependent variables, but without assigning participants to conditions randomly or maintaining strict control over features of the experimental situation that could influence participants’ responses.

Example of a Quasi-Experimental Design

Quasi-experimental designs are most often used in natural (nonlaboratory) settings over longer periods and usually include an intervention or treatment. Consider, for example, a study of the effect of a motivation intervention on class attendance and enjoyment in students. When an intact group such as a classroom is singled out for an intervention, randomly assigning each person to experimental conditions is not possible. Rather, the researcher gives one classroom the motivational intervention (intervention group) and the other classroom receives no intervention (comparison group). The researcher uses two classrooms that are as similar as possible in background (e.g., same age, racial composition) and that have comparable experiences within the class (e.g., type of class, meeting time) except for the intervention. In addition, the researcher gives participants in both conditions (comparison and motivation intervention) pretest questionnaires to assess attendance, enjoyment, and other related variables before the intervention. After the intervention is administered, the researcher measures attendance and enjoyment of the class. The researcher can then determine if students in the motivation intervention group enjoyed and attended class more than the students in the comparison group did.

Interpreting Results from a Quasi-Experimental Design

How should results from this hypothetical study be interpreted? Investigators, when interpreting the results of quasi-experimental designs that lacked random assignment of participants to conditions, must be cautious drawing conclusions about causality because of potential confounds in the setting. For example, the previous hypothetical example course material in the intervention group might have become more engaging whereas the comparison group started to cover a more mundane topic that led to changes in class enjoyment and attendance. However, if the intervention group and comparison group had similar pretest scores and comparable classroom experiences, then changes on posttest scores suggest that the motivation intervention influenced class attendance and enjoyment.

The Pros and Cons of Using Quasi-Experimental Designs

Quasi-experiments are most useful when conducting research in settings where random assignment is not possible because of ethical considerations or constraining situational factors. In consequence, such designs are more prevalent in studies conducted in natural settings, thereby increasing the real-world applicability of the findings. Such studies are not, however, true experiments, and thus the lack of control over assignment of participants to conditions renders causal conclusions suspect.


  1. Cook, T. D., & Campbell, D. T. (1979). Quasi-experimental: Design and analysis issues for field settings. Boston: Houghton Mifflin.
  2. Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-experimental designs for generalized causal inference. Boston: Houghton Mifflin.

Read More: