Abstract
In The Criminology of Place, David Weisburd, Elizabeth Groff, and Sue-Ming Yang (WGY) explore the distribution of crime in small units of geography in the city of Seattle. As they observe, their perspective does not “begin with the people that commit crime, but rather with the places where crime occurs” (p. 3). A key theoretical contribution of WGY is that the spatial processes “neighborhood” researchers are interested in vis à vis crime may well play out at very micro geographic units, and therefore they explore small units. Although the interest in studying small geographic units is not unique to WGY, one major contribution is their collection of longitudinal data for micro-places in Seattle over a long period of time (16 years from 1989–2004). The collection of this novel dataset was clearly a challenge, and it allows them to explore questions that prior research has left unaddressed.
Whereas the data collection was one large contribution of this study, the second contribution is a theoretical orientation that bridges the sometimes (surprisingly) disparate literatures studying neighborhoods and crime versus those studying crime in micro-locations (“criminology of place”). The theoretical contribution of WGY is to explore whether theories proposing to explain crime in neighborhoods (typically conceptualized as units much larger than the streetblocks that WGY employ) also operate in these much smaller geographic units. For various reasons that no doubt originate in data limitations as well as intellectual “turf wars,” whereas the criminology of place literature has largely gravitated to “opportunity theories” (i.e., routine activities theory), the neighborhoods and crime literature has tended more toward “control theories” (e.g., social disorganization theory). WGY propose moving beyond this “theoretical competition” to ask whether such control theories may also operate at smaller units. It seems that such a test is overdue—given the scientific imperative in which theories should live or die based on empirical tests and not passionate assertions independent of empirical evidence.
After an introductory chapter that sets out the goals of the study as well as a conceptual defense of using streetblocks as a unit of analysis in such research, the second chapter provides a nice review of the literature studying crime in locations. The third chapter provides descriptive analyses of their crime data in streetblocks over the 16-year period. One major theme that emerges in this chapter is the relative stability in the distribution of crime across the streetblocks in the city over time: typically 5 to 6 percent of the streetblocks in any given year contain 50 percent of the crime events, and between 19 and 23 percent of the streetblocks contain 80 percent of the crime events. The fact that these proportions appear so consistent over the years of this period leads WGY to speculate that perhaps a “law of crime concentrations” is present. Although they are well aware that such a statement is speculative at this point given their data for a single city and the lack of a theory from first principles explaining such a process, this provocative idea does suggest an area of future research. They then explore the trajectory of crime across these streetblocks over the 16 years, and their descriptive analysis suggests eight patterns (such as high increasing, high decreasing, low stable, etc.). A key insight that comes from this analysis is that the changes in city-level crime over time do not necessarily occur evenly across all streetblocks of the city, but instead are concentrated in just a few streetblocks.
The fourth chapter spatially situates these streetblocks, using maps to display the location of these eight trajectory patterns. It then explores the spatial clustering of these eight trajectories. The fifth and sixth chapters describe the various measures they were able to collect that capture the social and physical characteristics of these streetblocks and then describe the spatial distribution of these measures. A major challenge for WGY is that whereas social data on blocks is relatively straightforward to obtain, streetblocks are a different story; and such data is very difficult to obtain. Briefly, a streetblock is both sides of a single street between two intersections, whereas a block is (typically) a set of houses surrounded by streets. Therefore, a block will only have the houses on one side of four separate street segments, whereas a streetblock will have the houses on both sides of a single street segment. Although this may seem a somewhat arcane distinction, it is of utmost importance to the analyst and on a larger level suggests a need for different data collection strategies by agencies such as the U.S. Census if researchers are to truly employ streetblocks as a unit of analysis. Instead, WGY had to undertake what they refer to as a “treasure hunt” for data aggregated to streetblocks. This is arguably both a strength and weakness of the study: on the one hand, one cannot help but be impressed with the lengths they had to go to in trying to collect data capturing the social world in streetblocks. On the other hand, some of the measures that they are forced to use as proxies strain credulity—as one example, using a database of registered voters as a proxy for population, which, while correlated .7 with the actual population in the much larger unit of block groups, would have a much lower correlation with the population in these smaller units, implying relatively lower validity. This chapter discusses the measures they were able to collect, and displays their clustering across the streetblocks of the city over time. For those who want more detail on this “treasure hunt,” an appendix describes the various sources they utilized.
The seventh chapter then explores the relationship between these measures of the social and physical environment and the level of crime in streetblocks. Although their results indicate that the “opportunity theory” measures explain more of the variance in crime trajectories of streetblocks, the “control theory” measures nonetheless explain a relatively high proportion of variance as well.
For those who are interested in the spatial location of crime, WGY have provided a nicely researched statement to add to the literature. The data collection part of the project was Herculean, and the analysis reveals new insights that will help push the field forward. They have amassed a large swath of empirical evidence on these processes, and the style of presentation with maps and figures makes the book relatively understandable (with details of the more advanced statistical techniques provided in an appendix for those with such an interest). This empirical evidence is an important statement that will continue to shape the field in the future. And one can hope that the theoretical contribution, along with the empirical evidence, will help to break down the unnecessary divide between scholars in the neighborhoods and crime versus criminology of place subareas.