No abstract

The recent upturn in U.S. homicide rates may have resulted from increases in the number of serious violent incidents, growth in the percentage involving firearms, or increases in lethality. Data from the National Crime Victimization Survey (NCVS) and Supplementary Homicide Reports (SHR) suggest that the national homicide increase was not driven by increases in serious violence or the proportion of these crimes involving firearms, but may be related to a longer-term increase in lethality. In St. Louis, the increase was concentrated in communities that previously had the highest violence rates, but changes in the structural conditions and population composition were unrelated to the upsurge in homicide.

No abstract

Jails are a critical part of the criminal justice system but, until recently, have been omitted from discussions of reform. The goal of the current study is to describe and evaluate a collaborative, problem-solving initiative designed to reduce the jail population in St. Louis County, Missouri. The initiative was implemented as part of a research-practitioner partnership and is designed around a case review model, deemed the Population Review Team (PRT) commonly used in epidemiological analysis and problem-solving policing models. The outcome analysis suggests that the implementation of the PRT was associated with a significant decline in the total jail population, the number of individuals held on nonviolent felonies, and those held for over 100 days. The collaborative project provides one model for criminal justice systems reform and a tool for decarceration.

Although much of the crime and place literature seeks to explain crime frequency in relation to environmental conditions, few studies have examined crime diversity in space. This article reexamines a study of crime diversity in relation to a neutral model assuming environmental conditions have minimal influence on crime patterns. The original study results show that the variety of crime types in a given area (i.e., crime richness) increases regularly across spatial scales, and is largely consistent with a neutral or random process. This conclusion makes no appeal to the crimeenvironment dependencies often believed to influence crime occurrence, making the study a worthy candidate for additional scrutiny. The current study first verifies the original study results in Los Angeles, CA, and demonstrates their robustness with alternative crime classification schemes. Next, two alternative methods are used to check whether results differ when (a) locations are sampled randomly from the entire city rather than observed crime locations, and (b) when the unit of analysis is grid cells rather than "point-buffers." Finally, all analyses are replicated in St. Louis, MO, as a first look at generalizability. Conclusions are largely consistent with the original study, but important differences arise when alternative sampling techniques and units of analysis are used. Future directions for crime diversity research are discussed.