title	author	date	output
Lecture 11 Kessler and Roth	Nick Huntington-Klein	`r Sys.Date()`	revealjs::revealjs_presentation theme transition self_contained smart fig_caption reveal_options solarized slide true true true slideNumber true

```{r setup, include=FALSE} knitr::opts_chunk$set(echo = FALSE, warning=FALSE, message=FALSE) ``` ## Difference-in-Differences - Today we'll be going over Kessler & Roth (2014) and seeing how they do DID - This is the same paper used as a coding example in the textbook - There's nothing particularly special about this paper - it's not groundbreaking or famous - But it's just a good example of how DID works in action ## Discussion 1 - What is the theoretical background of why they're studying this? - Why do they think that an observational study might do a *better* job than the experiments in this case? - What has the previous literature found? ## Discussion 2 - How do they set up their DID? How do they define treatment *and control*? - What assumptions do they focus on and how do they justify them? - What causal diagram might they have in mind? ## Discussion 3 - What do they find? - How can we interpret each column of Table 2? - Is there anything else they should have done? - How do they make sense of their results in the conclusion?