I am trying to study the effects of a policy on educational attainment of individuals (years of schooling, primary/secondary school completion, literacy). Since the policy starts in a specific year I decided to use an RDD using birth year as running variable.

The problem is that my sample is made of 5 different countries in which the policy has been implemented in 5 different years. Am I able to pool the effect of the policy across countries?

Because if I build a unique running variable as being 0 for each cutoff year I get individuals born in e.g. 1994 having running variable both 0, -2, +3 depending on the country of origin.

How can I deal with this issue? Should I run a RD for each country alone?

  • $\begingroup$ You could consider generalized dif-in-dif also $\endgroup$
    – 123
    Apr 8, 2017 at 17:34

1 Answer 1


This paper:https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2712957

Is exactly what you need.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.