Last updated on June 3, 2018
If you’ve been reading this blog for more than a few months, you may remember that I was invited to teach a course titled Causal Inference with Observational Data by the University of Copenhagen’s Department of Food and Resource Economics. I co-taught the course from May 14 to 18 with my Copenhagen colleague Arne Henningsen–I taught the lectures, and he taught the labs.
I wanted to make the lectures for the course publicly available bit by bit, and so today I am posting the first two lectures: 1. Introduction, and 2. Causality. I will be posting lectures 3 to 8 over the next few weeks.
Speaking of causality, Judea Pearl–a pioneer of causality, with his eponymous textbook–has a new popular-press book out discussing his pioneering work and the Causal Revolution of the last 30 years, titled The Book of Why. I started reading it last week and it is excellent.
Also, a post last month by Matt Bogard drew my attention to this American Journal of Public Health article, in which the author argues that authors (and reviewers and editors) should stop being afraid to use the word “causal,” arguing that “being explicit about the causal objective of a study reduces ambiguity in the scientific question, errors in the analysis, and excesses in the interpretation of the results.”
The article’s clever title aside, I feel as though this might be a public health-specific problem. In the literature in which I work, the causal goal of the typical study, although often implicit, is usually pretty obvious, and by talking of associations and being gun-shy regarding using the word “causality,” we are only exercising the proper caution necessary when speaking to policy makers, who often exhibit wishful thinking.