A quasi-experiment is a way to investigate whether an intervention or event may have caused a change, even when researchers cannot randomly assign participants to groups. This lesson introduces the design and its basic components.
A quasi-experiment is a research design used to estimate the causal impact of an intervention. Researchers compare outcomes for a group that experienced an intervention with outcomes for a group that did not. An intervention may be deliberately introduced, such as a reading program, or may be an event that affects a group, such as an earthquake.
All flashcards
Quasi-experiment
A design that estimates an intervention's effects by comparing conditions without random assignment.
Quasi-independent variable
The intervention, exposure, or condition expected to affect an outcome.
Dependent variable
The outcome measured in a study.
Covariate
An additional variable that may predict group membership, the outcome, or both.
External validity
The extent to which findings apply across relevant people, times, settings, and methods.
Quiz questions
In difference-in-differences analysis, what does the parallel-trends assumption require?Answer: Without the intervention, treatment and comparison groups would have followed similar outcome trends.
In a regression discontinuity design, how is treatment eligibility typically determined?Answer: By whether a unit's assignment-variable value falls above or below a cutoff.
What does internal validity concern in a study of an intervention?Answer: How credible it is that the intervention caused the observed outcome.
Why might researchers use a quasi-experiment instead of randomly assigning participants?Answer: Randomly assigning participants to the exposure could be unethical or impractical.
What characterizes a natural experiment?Answer: An external event or process creates exposure differences that may approximate random assignment.
Based on the Wikipedia article Quasi-experiment, licensed CC BY-SA 4.0. Adapted into a summary, flashcards and quizzes.