A METHODOLOGICAL ADDITION TO THE CROSS‐NATIONAL EMPIRICAL LITERATURE ON SOCIAL STRUCTURE AND HOMICIDE: A FIRST TEST OF THE POVERTY‐HOMICIDE THESIS*

Dozens of cross‐national studies of homicide have been published in the last three decades. Although nearly all these studies test for an association between inequality and homicide, no studies test for a poverty—homicide association. This absence is disconcerting given that poverty is one of the most consistent predictors of area homicide rates in the abundant empirical literature on social structure and homicide in the United States. Using a sample that coincides closely with similar recent studies, applying a proxy for poverty (infant mortality) that is commonly employed in noncriminological cross‐national research, and controlling for several common covariates (including inequality), this study provides the first test of the poverty—homicide hypothesis at the cross‐national level. The results reveal a positive and significant association between a nation's level of poverty and its homicide rate. The findings also suggest that we may need to reassess the strong conclusions about an inequality—homicide association drawn from prior studies, as this relationship disappears when poverty is included in the model.