Applying computer simulation to forecast homicide rates

Abstract

The purpose of this article is to expose readers to a relatively new and virtually unused analytical tool—computer simulation of dynamic social trends. For purposes of illustrating this technique, several variables assumed to contribute to homicide rates were examined. Prior research has pointed to structural explanations for homecide. Specifically, “anomie” and a “subculture of violence” have repeatedly been found to be correlated with homecide rates. However, neither adequately explains the inordinately high number of homecides in the United States. Therefore, an alternative explanation, focusing on the interactive and reciprocal nature of these two constructs, is considered in this article. Through the use of simulation technology, a computer model was built that tested the ability of these constructs and the associated variables to predict homecide rates. Findings indicated that simultaneously increasing several variables (e.g., population density, availability of handguns, etc.) produced a sharp rise in the number of homecides. However, after a period of several years, societal responses (such as enhanced social programs and improved rehabilitative efforts) resulted in homecide rates stabilizing at a level only slightly greater than that recorded prior to the changes.