Scott Cunningham, "Causal Inference: The Mixtape" (Yale UP, 2021)

Just about everyone knows correlation does not equal causation, and probably that a randomized controlled experiment is the best way to solve that problem, if you can do one. If you’ve been following the economics discipline you will have heard about the Nobel Prize given to Abhijit Banerjee, Esther Duflo, and Michael Kremer for their work applying the experimental method to test real-world policy interventions out in the field. But what if you can’t do this? Are you just stuck with untestable claims? This year’s Nobel Prize to Josh Angrist, David Card, and Guido Imbens for methods of causal inference with observational data confirms that you don't have to give up. Scott Cunningham’s Causal Inference: The Mixtape (Yale UP, 2021) provides an accessible practical introduction to techniques developed by these luminaries and others. Along with the statistical theory, it provides intuitive explanations of these techniques, and examples of the computer code needed to run them. In our conversation we discuss why economists needed these techniques and how they work. Scott Cunningham is a professor of economics at Baylor University. He researches topics including mental healthcare, sex work, abortion and drug policy. He is active on Twitter, has a blog on Substack, and frequently conducts workshops on causal inference methods. A complete web version of his book is available here. Host Peter Lorentzen is an Associate Professor in the Department of Economics at the University of San Francisco, where he leads a new digital economy-focused Master's program in Applied Economics. Learn more about your ad choices. Visit megaphone.fm/adchoices