Causal Analysis in Theory and Practice

Eric Neufeld (University of Saskatchewan/Canada) asks:

I have been interested in giving a lecture to lay people about your work, but would like to compare/contrast it with Rubin’s work. I understand how strongly you disagree! But I would appreciate it if you could point me to a couple of your articles that lay the argument out in an accessible way.

Judea Pearl replies:

I have written a fairly decent section on Rubin’s model in Statistics Surveys:
http://ftp.cs.ucla.edu/pub/stat_ser/r350.pdf
and a less technical one in:
http://ftp.cs.ucla.edu/pub/stat_ser/r348-warning.pdf
(unpublished but highly recommended.)

Important to stress: I do not disagree with Rubin’s work (it is subsumed by structural modeling), I am merely mused by the tunnel-visioned culture that this work has engendered.