An Application of Agent Based Modeling to the Age-Crime Relationship

• In Event: Posters II

Abstract

The connection between age and crime is arguably one of the strongest within the field of criminology. It has been rigorously tested within the framework of the life course perspective, among others. The chances that an individual engages in crime tends to decrease with age due to turning points in their lives such as obtaining a college degree, getting married, having children, serving in the military, or maintaining steady employment. A relatively new method of analysis within criminology is agent based modeling (ABM). ABMs instantiate rules of behavior for individual agents, who interact with one another in a simulated environment. In this study, we model the relationship between age and crime within individuals using an ABM. The use of the ABM enables us to gain insight into how characteristics of offenders develop overtime and ultimately manifest themselves in a criminal act. The model and our analysis is validated using state of the art techniques. The ensuing results help criminologists explore and explain the factors and dynamics that cause an individual to be less likely to engage in criminal acts as they age. Furthermore, our work addresses the gap in criminological literature with regard to the use of computer simulation and modeling.

Authors

• Caitlin V.M. Cornelius, Virginia Modeling Analysis and Simulation Center

• Ross Gore, Virginia Modeling Analysis and Simulation Center