No announcement yet.
  • Filter
  • Time
  • Show
Clear All
new posts

  • Difference-in-Difference-in-Difference with Continuous Treatment Variable

    I am trying to estimate the effect of 4H participation on student test scores in Virginia using the triple difference model. I have the 4H participation rates, test scores, a set of observed characteristics and to control for unobserved confounders, I am including fixed effects for school districts, grades and years. To account for even more I have included interaction between these three sets of fixed effects as well. Essentially I need to relate the triple differencing of the test scores to the corresponding triple differencing of the extent of 4H participation over districts, grades and years.

    However, I am having trouble coding this in stata. Most examples are for regular DID and use a binary treatment or a policy variable. So I am unsure how to code the triple difference using the continuous treatment 4H variable.

    Any help or suggestions would be very useful. Thank you.
    Last edited by Alejandro Gonzales; 07 Dec 2019, 13:17.