Department of Statistics, UCLA

This report discusses the reversal e ect known as Simpson's paradox from a causal- theoretic viewpoint. It analyzes the reasons why the e ect has been (and still is) considered paradoxical and why its resolution has been so late in coming. The re- port is extracted from a forthcoming book Causality [Pearl, 2000], and assumes some familiarity with causal diagrams and the do( ) (or set( )) notation (e.g., [Pearl, 1995]).