From the course: Program Evaluation for Data Science
Join today to access over 25,500 courses taught by industry experts.
When to apply difference in difference
From the course: Program Evaluation for Data Science
When to apply difference in difference
- Difference in difference, often referred to as DID, is one of the most common methods used when you can't randomize. Difference in difference is most useful in situations where there are two time periods, one before and one after the program start date, and there are at least two groups of subjects. In the first time period, none of the groups receives the program of interest. In the second time period, only one of the groups receives the program of interest. I have found difference in difference most applicable in situations where a program implementation is limited by geography. For example, a new model for fraud prevention was implemented in only one country, and the other countries were used as comparisons. In that case, we had data on the fraud detection and losses, both before the program started as well as after. The country in which we began the program was our test, and the countries where the program was not started acted as comparisons. While it is intuitive that we can…
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.