Describe a quasi-experimental design and give two examples.

• A. Nonequivalent control group design and interrupted time series are examples of quasi-experimental designs.
• B. Randomized controlled trial and factorial design are quasi-experimental.
• C. A pre-post design with no control group is quasi-experimental.
• D. Mixed-methods single-case study with replication.

Answer: (A)

Quasi-experimental designs aim to estimate causal effects when random assignment to groups isn’t possible, so researchers rely on comparison groups or time-series data around an intervention to infer impact while acknowledging potential confounds.

One strong example is the nonequivalent control group design. Here you have a treatment group and a comparison group that are not created by random assignment, so there may be preexisting differences between the groups. By collecting data before and after the intervention in both groups, researchers look for differential changes that can be attributed to the treatment, while using matching or statistical controls to account for baseline differences.

Another solid example is the interrupted time series. This design uses multiple observations before and after the intervention, allowing you to examine whether the intervention produces a noticeable change in the level or trend of the outcome, beyond what would be expected from existing patterns. The strength lies in leveraging the pre-intervention trajectory to help isolate the intervention’s effect.

Why the other options aren’t the best fits here: a randomized controlled trial uses random assignment, which makes it a true experimental design rather than quasi-experimental. A factorial design is also a formal experimental structure that typically involves randomization to explore multiple factors. A simple pre-post design with no control group is a weaker quasi-experimental approach, since it lacks a comparison to help rule out alternative explanations. The combination of nonequivalent control groups and interrupted time series represents two well-established quasi-experimental approaches that illustrate how causal inference can be pursued without randomization.