Week 4

Date: Sep 21st - Sep 23rd

What we will cover

This week, we will finish our chapter on potential outcomes, and we will be talking all about Randomized Controlled Trials (RCTs). Why is this study design is considered sometimes the “gold standard”? How can this identification strategy allow us to make causal claims? Are there shortcomings related to RCTs?

  • Angrist & Pischke. (2015). “Mastering Metrics”. Chapter 1: Randomized Trials. Pg. 12-33.

  • Marginal Revolution University. (2020). “Randomized Trials: The Ideal Weapon”. Video materials from Mastering Metrics.

  • Deaton, A. (2020). “The Problem with Randomized Controlled Trials”. CEPR & VideoVox.
  • Athey, S. (2013). “The Dangers of Becoming Too Data-Driven”. Stanford Graduate School of Business


Complete before Sunday Sep 19th (11:59 pm) (if your section is on Tue) or Tuesday Sep 21th (11:59 pm) (if your section is on Thu). You can find the assignment here.


New window Download

New window Download


Here is the code for “Beauty in the classroom” example Download

Here is the code for RCTs Download


  • Here is a video explaining another example of collider bias. I also found this article that I think explain collider bias in a very intuitive way as well.
  • Here is a video of the last example we see in this class (CV experiment). I go through the code more slowly, so feel free to review it if you have any questions about the code or the example in general after class!

© Magdalena Bennett - licensed under Creative Commons.