5
$\begingroup$

I'm new to difference-in-difference method and wanted to familiarize myself with the literature. I'm familiar with Mostly Harmless Econometrics and the papers on DiD discussed in the book. I would appreciate it if some of you could suggest more papers to read.

Thanks,

$\endgroup$
2
  • $\begingroup$ A few clarification questions: Are you applying DiD techniques to any particular type of problem? (Different contexts often present with their own specific applications and challenges.) What level of difficulty are you looking for? Are you more interested in the econometric theory of DiD, how to apply the technique, common problems, examples where DiD is used, or something else? Are you solely interested in DiD, or interested in the broader class of methods (synthetic controls, event study, etc.)? Thanks! $\endgroup$
    – AndrewC
    Aug 10 at 12:41
  • $\begingroup$ I'm applying DiD techniques to estimate impact of interventions on educational performance of children from low income households. However, I'd prefer to acquaint myself with papers using DiD in the broader area of development economics. The focus should be on application of the technique in a variety of problems and not the econometric theory of DiD. I would like a mix of introductory and advanced level papers and would like to focus solely on DiD. thanks $\endgroup$
    – Jan3
    Aug 10 at 13:45
8
$\begingroup$

This is far from exhaustive, but hopefully it can get you started! (I'd recommend skimming most of them, then really focusing on those where the application and/or modeling environment is uniquely relevant or interesting for your research.)

General reviews of empirical approaches:

Angrist and Krueger (1999) looks at several methods (including DiD) in the context of labor economics, while Blundell and MaCurdy (1999) also look at DiD in labor, but more from a natural experiment perspective. Chapter 1 in the textbook the Economics of Education (2020) similarly discusses a couple empirical approaches, including DiD, in the context of the education literature. Roberts and Whited (2012) discuss several methods (including IV and DiD) in corporate finance applications. This working paper in the education space looks at a variety of different works that have used various empirical methods to study international standardized test-related outcomes, and might be of interest (mostly for the works it cites). Similarly, this provides a long list of empirical applications of DiD in economics of education.

Papers describing how to properly apply DiD:

I actually have found this paper from the public health world helpful in framing DiD issues previously, so I'll include it here as well. This paper from 2019 is pretty approachable, and might be of interest as well. This paper goes into a bit more depth (I think) than the others, so might be worth looking at.

Specific examples of DiD in use:

A comprehensive list would probably crash the Stack Exchange servers, but some examples (and their particular focus) are: Medicaid on early childhood (2016), food stamps (2016), information regarding food safety (i.e. health scores) on consumer behavior (2002 version linked, published in 2003), "green" advertising on consumer preferences (2010), high school employment on education (with a propensity score matching element) (2008), Nestle's impact on mortality in low income countries (2018), and tracking and educational outcomes (2020). Particularly well-known examples of DiD applications are this Dube Lesterer and Reich example looking at minimum wage, and this Duflo paper looking at a natural experiment related to education in Indonesia. There's also this paper from 2010 that looks at how exposure to Malaria impacts labor market outcomes, and this one from 2015 that examines the impacts of community health centers.

Other papers of interest:

I know you mentioned wanting to avoid the econometric theory, but a few papers I'd recommend looking into (they also include applications) are this paper by Bertrand, Duflo and Mullainathan (2003 version linked, published in 2004), these two papers that look at DiD when the treatment varies over time, and somewhat relatedly, this one that looks at DiD when the treatment intensity varies over time. Finally, this paper that looks at the importance of properly testing for, and dealing with, the potential for parallel trends.

$\endgroup$
1
  • 1
    $\begingroup$ Thank you for these references! $\endgroup$
    – Jan3
    Aug 12 at 8:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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