Cross-national variation in homicides: Is race a factor?

J. P. Rushton has done research indicating that predominantly black nations have higher rates of violent crime than predominantly white nations, which have higher rates than predominantly Asian nations. He attributes these differences to constitutional, inheritable factors on which the races vary. This article discusses the numerous problems with Rushton's use of international crime data and analysis, foremost of which are his use of one-way analysis of variance and lack of consideration of other causes of variation. Regression analysis-including dummy race variables and other relevant explanatory variables-indicates no significant association between race and cross-national homicide rates, the best and probably only valid indicator of cross-national variation in violent crime. This is the case whether reanalyzing Rushton's data or analyzing properly adjusted inspected, and selected homicide rates.