Thread by @DToshkov

I recently finished 'The Book of Why' by @yudapearl
I would highly recommend it to everyone interested in causality and scientific explanations. /THREAD
amzn.to/2OW9v0Q
#bookofwhy

The book is rather accessible, so you don't need to be a #causalinference nerd to follow the arguments. At the same time, there is a lot to learn even if you have struggled your way through Causality (CUP, 2009) and the technical papers
amzn.to/2OSmzEb

I have been an early converter and have used the 'structural' approach to causality to organize my own book on research design in political science.
Now I am glad I have an accessible book to point students to for more info and background.
amzn.to/2KfX3Iy

I have two major criticisms:
(1) I don't like that the third level in the Ladder of Causation is called 'Counterfactuals', because this seems to suggest that the second level ('Intervention') is not based on counterfactual reasoning. But it is. #bookofwhy

I think @yudapearl actually would agree, so this is just a matter of labels, but the labels are important as they can lead inattentive readers to believe that the assessment of causal interventions escapes a counterfactual und. of causality

My second criticism of the #bookfowhy is about the rhetorical strategy. The book claims that statisticians and social scientists have ignored for decades problems of causality because of lack of proper language to express such ideas in statistics.

Now that might be true for statisticians, but would be very off-putting to social scientists, as they spend most of their careers worrying about inferring causality from empirical data. This is not to deny that the #bookofwhy gives new and better tools to do so...

but social scientists will feel perplexed by the claims that they have ignored causal questions. After all, at least since King, Keohane and Verba (1994) issues of omitted variables, case selection, etc. have been at the heart of methodological training.
amzn.to/2QQFhOp

more recently, the potential outcomes framework has been central to such concerns.
Again, I believe the #bookofwhy offers a better way of thinking about and solving issues of #causalinference. But I am afraid this rhetorical strategy will slow its acceptance by social scientists

I attach the ToC of Gary King, Koehane and Verba's 1994 book, which was central to my generation of pol scientists. It's all about causal inference. You can say it's incomplete or imprecise or not easy to understand or even wrong. But it's there.