Explainer: What is a ‘randomized control trial’ and why are a growing number of local governments using them?

  • First, they need to have a way to split groups randomly. Say you’re testing out whether a newly designed tax bill can get people to pay on time. One group of taxpayers would get the old bill, and at least one other group would get a new one.
  • Second, cities need to have data to measure the outcome — the on-time payment rate in this case.
  • Third, they need to be able to track that data back to individuals within the different groups, in order to be able to make comparisons.
  • Finally, cities need large numbers in each group in order to yield results that are statistically meaningful. While there aren’t any hard rules here, Cardon said, “it’s best if you can aim for thousands.” If you were evaluating a small social program that serves just 50 people, she said, you’d be better off using a more qualitative approach to evaluation.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Bloomberg Cities

Bloomberg Cities


Celebrating public sector progress and innovation in cities around the world. Run by @BloombergDotOrg’s Government Innovation program. bloombergcities.org