This page is still in progress.Comment: Understanding Simpson's Paradox. J. Pearl in Am. Stat. 68:8 (2014). What the paper says!?
The text discusses the Simpson paradox, under the prism of causality.
This is a reply to, apparently, Ref. [1] (although the paper it replies to is not, annoyingly, cited!)
The paradox is defined as follows:
Simpson’s paradox refers to a phenomenon whereby the association between a pair of variables (X, Y) reverses sign upon conditioning of a third variable, Z, regardless of the value taken by Z. If we partition the data into subpopulations, each representing a specific value of the third variable, the phenomenon appears as a sign reversal between the associations measured in the disaggregated subpopulations relative to the aggregated data, which describes the population as a whole.