Quasi-experimental Research Designs
True Quasi-experimental Research Designs – in which a treatment or stimulus is administered to only one of two groups whose members were randomly assigned – are considered the gold standard in assessing causal hypotheses. True experiments require researchers to exert a great deal of control over all aspects of the design, which in turn allows strong statements to be made about causal relationships.
In many situations, especially those involving human subjects, it is simply not possible for researchers to exert the level of control necessary for a true experiment. For example, it may be unethical to expose subjects to a stimulus which the researcher knows may cause harm. Researchers often study processes too complex or lengthy for experimental settings. Quasi-experimental designs relax key experimental requirements, making them more practical but weakening causal claims.
True experiments require randomly assigning subjects to treatment or control groups. Random assignment distributes subject characteristics across groups based on probability, reducing bias.
Often it is not possible for researchers to randomly assign subjects to groups, for either practical or ethical reasons. Quasi-experimental research designs therefore use alternative ways of assigning subjects to the treatment and control groups.
The most common subset of quasi-experimental research designs are the nonequivalent control group designs.
In this design, researchers intentionally match control group subjects to treatment group subjects based on relevant characteristics. Matching at the individual level creates a one-to-one pairing between groups.
Aggregate matching creates a control group with similar characteristics to the treatment group, like gender proportion and age distribution. This method is quasi-experimental because group assignment is intentional, not random. Another approach uses existing groups, such as comparing students in two classrooms, with only one receiving the stimulus.
Some quasi-experimental designs lack a control group. Known as before-and-after, pre-test/post-test, or pre-experimental designs, they expose all subjects to the treatment. The comparison comes from measuring outcomes before and after exposure. A significant change in post-treatment values suggests the treatment caused the effect.
A time-series design observes one group repeatedly before and after treatment. It works in controlled or natural settings where researchers analyze existing data. For example, yearly test scores before and after an extended school day serve as time-series data. This approach is stronger than a single pre-test/post-test design, as it shows long-term effects. Adding a control group creates a multiple time-series design, further improving reliability.
Quasi-experimental designs are more practical than true experiments but face greater threats to internal validity. Researchers should take special care to address these threats and use additional data to rule out alternative explanations.