Simpson paradox

The Simpson-Paradox refers to a tri-variate statistical problem in which a contingency between two variables, x and y, can be accounted for by a third variable, z. Solving the problem thus calls for a cognitive routine analogous to analysis of covariance. Cognitive research on the Simpson-Paradox addresses the question of whether the human mind can correctly handle these statistical tools.

One example for the Simpson-Paradox: One supermarket may be more expensive on aggregate (correlation between supermarket (x) and price (y)), but only because the latter supermarket sells more high-quality products (correlation between supermarket (x) and the percentage of high-quality products (z), correlation between price (y) and the percentage of high-quality products (z)). Considering high and low-quality products separately, the seemingly more expensive supermarket may turn out to be cheaper at any quality level.

Literature: Fiedler, Walther & Nickel (1997), Simpson (1951)

Entry by: Susanne Haberstroh

June 11, 1999
Direct questions and comments to: Glossary master