From the course: Program Evaluation for Data Science
Join today to access over 25,300 courses taught by industry experts.
When to apply matching methods
From the course: Program Evaluation for Data Science
When to apply matching methods
- Matching studies are very useful in situations where the program has already been implemented without a randomized control, and you want to understand the program's impact. The implementation could have been due to some non-random assignment to the program and administrative decision about who gets the program and who doesn't, or self-selection. As a result, there's a population that was exposed to the program and a population that wasn't, but they're different in their observable characteristics. When I say observable characteristics, I mean features that you can and did measure like age and spending patterns, and not unobservable features like motivation or unmeasurable attitudes. Because the data already exists, matching is used for retrospective studies. We can contrast this design with a typical A/B testing where the subjects are first assigned and later the data is collected, also known as prospective studies. As I mentioned, matching studies are performed retrospectively when…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.