This study examines the structural correlates of homicide in Jamaica, a developing nation with one of the world’s highest homicide rates. We extend existing theories on the ecology of lethal violence to account for the influence of electoral politics on homicide. Controlling for structural factors thought to be associated with violence, we examine the effect of voter turnout on homicides in urban communities located in the Kingston Metropolitan Area. We test the model using negative binomial regression. The findings help clarify the factors that explain variations in homicide across communities in urban Jamaica. The study also reveals how theories, concepts and measures used regularly in research in developed nations may require adaptation for use in developing nations.