Fallacious use of regression effects in the i.q. controversy

Abstract

Regression to the population mean in I.Q. scores has been taken by H.J. Eysenck, A.R. Jensen, and other writers to provide evidence for genetic determination of individual and racial I.Q. differences. However, regression is a purely statistical phenomenon, and as such provides no evidence for either genetic or environmental determination of I.Q. The statistical basis of regression is explained, the misuse of regression effects in the literature is reviewed, and arguments based on such misuse are criticized. It is suggested that the fallacious use of regression effects is partly a result of the isolation of biometrical from Mendelian genetics.