CAREER: Crime Victimization Patterns in American Cities

Intellectual Merit: Homicide is the third leading cause of death for African American men, trailing only cardiovascular disease and cancer, and accounts for nearly 18 percent of the seven-year black-white gap in male life expectancy [1, 2]. Conventional explanations of homicide typically take a “risk factors” approach that considers differential rates of violence as the byproduct of individual traits (e.g., self-control or criminal propensity), over-exposure to violence-conducive stimuli (e.g., guns, crack cocaine, or gangs), or the under-exposure to violence mitigating conditions (e.g., educational and economic opportunities). Although such an approach succeeds in explaining aggregate variations in the level of homicide, it fails to explain which individuals become involved in violence. Indeed, the vast majority of people with such risk factors never become involved in a homicide. Risk factor approaches fail for many of the same reasons that early accounts of HIV failed: they ignore how the ways in which people are connected influence the propensity for “infection.” [15]. This project explores the importance of social networks for individual victimization by analyzing the ways social networks and an individual’s position within social networks influence the probability of being a victim of homicide or violent crime. The proposal outlines three main objectives: (1) to establish a theoretical framework for understanding how social networks influence patterns of violent behavior and victimization, (2) to analyze the precise temporal and spatial diffusion of homicide and non-fatal violence, and (3) to explain how an individual’s position within social networks and the structure of such networks influence patterns of violence. Together, these objectives map out a career trajectory that spans research in public health, the social and behavioral sciences, legal studies, and the growing cross-disciplinary interest in social network analysis. In so doing, this study advances an approach to violence that is empirically grounded and scientifically fruitful. Broader Impact: Public policy aimed at reducing violence—especially gang violence—is often based not on scientific research, but on stereotypes and back-of-the-envelope empirics [81]. The results are often broad strategies that generate further inequalities among vulnerable populations, as was the case with Chicago’s 1992 gang loitering law deemed unconstitutional by the Supreme Court because it unfairly targeted non-gang youth [82]. In contrast, the network framework advanced in this research program can better inform policy decisions and violence prevention efforts by determining strategic points of intervention. Thus, rather than crafting strategies based on categorical distinctions (such as profiling race or age), the network approach bases intervention on direct observational information. Tracking the precise patterns of violence in a population thus might reduce morbidity and mortality by pinpointing individuals and groups most at risk of victimization and channeling scarce resources accordingly. Currently, this network approach is being incorporated into a violence reduction strategy in Chicago with extremely promising results. A working relationship between the PI and the National Network for Safe Communities will further integrate this network framework into its on-going violence prevention efforts in more than 50 municipalities across the U.S. This research program includes educational components for students, academics, and professionals. Undergraduate and graduate studen