Log in from a computer to take this course

You'll need to log in from a computer to start Learn the Basics of Causal Inference with R. But you can practice or keep up your coding streak with the Codecademy Go app. Download the app to get started.

apple storegoogle store
Learn

In this lesson, we showed that the implementation of an employee-sponsored retirement matching program led to an increase in average monthly employee contributions.

We learned a lot about regression discontinuity design along the way, including:

  • RDD is a method used when the treatment assignment is determined by a continuous forcing variable at a specific cutoff point.
  • An RDD is known as sharp if the cutoff is exact and fuzzy if the cutoff is not exact.
  • Individuals within a narrow window on either side of the cutoff are assumed to be similar to each other, except for the treatment group assignment.
  • Local linear regression can be used to determine the local average treatment effect (LATE) among the individuals in this narrow window.
  • Disadvantages of RDD include reduced sample size and potential lack of generalizability of the LATE.

Instructions

Conclusion

Great job performing a regression discontinuity design analysis on your own! In this lesson, you verified that RDD was an appropriate method to use, got causal estimates of the LATE using local linear regression, and examined the reliability of your results by using different bandwidths.

So what did this RDD analysis tell us? Based on the results of the local linear regression modeling (as shown in the learning environment), we can conclude that in our sample, emissions control devices led to a decrease in AQI of 36.45 points. This means that the emissions control device achieved its goal of providing cleaner air in close proximity to the power plants in our dataset.

Take this course for free

Mini Info Outline Icon
By signing up for Codecademy, you agree to Codecademy's Terms of Service & Privacy Policy.

Or sign up using:

Already have an account?