Causal Analysis in Theory and Practice

December 11, 2019

Generalizing Experimental Results by Leveraging Knowledge of Mechanisms

Filed under: Data Fusion,Generalizability,Identification — Judea Pearl @ 8:44 pm

In a recent post (and papers), Anders Huitfeldt and co-authors have discussed ways of achieving external validity in the presence of “effect heterogeneity.” These results are not immediately inferable using a standard (non-parametric) selection diagram, which has led them to conclude that selection diagrams may not be helpful for  “thinking more closely about effect heterogeneity” and, thus, might be “throwing the baby out with the bathwater.”

Taking a closer look at the analysis of Anders and co-authors, and using their very same examples, we came to quite different conclusions. In those cases, transportability is not immediately inferable in a fully nonparametric structural model for a simple reason: it relies on functional constraints on the structural equation of the outcome. Once these constraints are properly incorporated in the analysis, all results flow naturally from the structural model, and selection diagrams prove to be indispensable for thinking about heterogeneity, for extrapolating results across populations, and for protecting analysts from unwarranted generalizations.  See details in the full note.

1 Comment »

  1. Hi Judea,

    I read The Book of Why several months ago and was enamoured by the complexity of AI learning cause and effect, I wrote some thoughts about it on my blog at I decided to throw myself fully into how to solve the problem of causality by considering how humans solve problems. I spent several months doing this. Prior to my work, I spent 4 years (2 of those full time) with the best psychologists and combining that with every piece of literature I could find learning about how humans think. To solve the issue of causality, I felt the best way was to solve the issue of how we actually think, not how it appears to us that we think.

    I created a few youtube videos on the subject, with the one I believe to hold new insights into how to solve causality, the video is titled “What is a Problem”. I have contacted several academics about my theories on how to solve causality.

    My youtube channel is here –, the video I refer to above is here –

    I apologise for the quality of my videos, creating these videos was confronting as I’ve discovered that how we really think likes to stay hidden.

    Do you believe I am on a pathway that could add value to your very valuable work on this subject? If you answer is yes, I would like to find other people who would like to work with me to flesh these theories out further.

    Thank you for your time.

    Comment by Nathan Aherne — December 12, 2019 @ 5:27 am

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress