Crime Concentration - Environmental Criminology

Crime concentration refers to the empirical regularity, now among the most consistently replicated findings in criminology, that a small percentage of street segments and addresses within any given city accounts for a disproportionately large share of total criminal activity, while the overwhelming majority of locations remain largely or entirely crime-free across extended periods of observation. Within the Environmental Criminology silo, this article examines the empirical documentation of crime concentration, David Weisburd’s proposed law of crime concentration at place, the methodological approaches used to measure and map concentration, the competing explanations for why concentration occurs, and the direct implications this concentration carries for the resource-allocation and prevention strategies examined throughout this silo’s remaining companion articles.

Introduction

The finding that crime concentrates heavily at a small number of specific locations, introduced in the companion article on Place and Crime elsewhere in this silo, deserves sustained independent treatment because of both its unusual empirical consistency and its far-reaching implications for how criminologists conceptualize crime causation and how practitioners allocate limited crime-prevention resources. Lawrence Sherman, Patrick Gartin, and Michael Buerger’s 1989 Minneapolis study first quantified this concentration with precision, and subsequent decades of replication across cities, countries, and crime types have transformed an initially surprising empirical observation into one of the discipline’s most theoretically generative regularities (Sherman, Gartin, & Buerger, 1989).

This article proceeds by examining the empirical documentation of crime concentration in greater analytical depth than the introductory treatment in the companion Place and Crime article, detailing Weisburd’s law of crime concentration specifically, the measurement and mapping methodologies concentration research employs, the competing theoretical explanations for why concentration occurs, and the direct implications concentration carries for prevention practice.

Documenting Crime Concentration Across Cities

The Minneapolis Foundation and Subsequent Replication

Sherman, Gartin, and Buerger’s analysis of calls for police service in Minneapolis found that just over three percent of all addresses and intersections generated more than fifty percent of all calls for service across the year examined, a finding that immediately reoriented subsequent criminological research toward the specific physical location of criminal events rather than only their neighborhood context (Sherman, Gartin, & Buerger, 1989). This original Minneapolis finding proved unusually durable under subsequent scrutiny, with researchers replicating comparably extreme concentration patterns across dozens of American cities employing varied data sources, including calls for service, victimization reports, and arrest records, each converging on broadly similar concentration estimates despite considerable variation in city size, geography, and policing practice. Travis Pratt and colleagues’ broader meta-analytic synthesis of place-based and situational crime research found that concentration-related findings ranked among the most consistently replicated results across the accumulated criminological literature their analysis examined, a pattern of replication achieved by relatively few criminological regularities originating from a single foundational study (Pratt, Cullen, Sellers, Winfree, Madensen, Daigle, Fearn, & Gau, 2010).

David Weisburd, Shawn Bushway, Cynthia Lum, and Sue-Ming Yang’s Seattle-based research extended this documentation using a genuinely longitudinal design, tracking crime at the individual street-segment level across a fourteen-year period and confirming that concentration held not merely as a single-year snapshot but as a persistent structural feature of the city’s crime distribution across an extended span of time during which the city itself experienced substantial demographic and economic change (Weisburd, Bushway, Lum, & Yang, 2004). Anthony Braga and Andrew Papachristos’s Boston-based replication similarly documented pronounced, stable concentration across gun-violence incidents specifically, finding that a small number of identified hot spots accounted for a disproportionate share of the city’s shooting incidents across the multi-year period their study examined, extending concentration documentation to a serious violent offense category with particularly high stakes for policy response (Braga, Papachristos, & Hureau, 2010).

Cross-National and Cross-Crime-Type Replication

Weisburd’s subsequent comparative synthesis, drawing together concentration research conducted across American, British, Israeli, and other national contexts, found that the specific proportion of street segments accounting for approximately half of a city’s crime consistently fell within a narrow range, typically between four and six percent, regardless of the specific national policing system, legal tradition, or urban form examined, a consistency Weisburd argued rivaled well-established regularities in the natural sciences in its cross-contextual stability (Weisburd, 2015). This cross-national consistency has proven theoretically significant because it suggests that crime concentration reflects general features of how criminal opportunity distributes across urban space rather than an artifact of any single country’s particular institutional arrangements.

Subsequent research extending concentration analysis beyond the conventional predatory street crime that supplied the original evidentiary base has documented comparable concentration patterns for offense categories including drug distribution, prostitution, and domestic violence calls for service, suggesting that the underlying concentration phenomenon operates with considerable generality across the full range of criminal behavior rather than being confined to any single narrow offense category (Eck & Weisburd, 1995).

Concentration and Its Relationship to Neighborhood-Level Crime

Reconciling Micro-Place and Neighborhood Findings

Crime concentration research’s emphasis on street-segment and address-level analysis has at times appeared to compete with the longer-established neighborhood-level tradition of social disorganization and collective efficacy research, examined elsewhere in this encyclopedia’s broader treatment of criminological theory, since neighborhood-level research typically explains crime variation through structural factors including poverty and residential instability operating at a considerably coarser spatial scale than the individual street segment (Sampson & Groves, 1989). Robert Sampson and W. Byron Groves’s foundational neighborhood-level research found that structural disadvantage and weak collective efficacy substantially predicted neighborhood crime rates, a finding operating at a spatial scale that concentration research’s street-segment focus does not directly contradict but instead complements, since neighborhood-level structural conditions plausibly shape the base rate of favorable conditions for crime concentration to develop within a given neighborhood’s specific street segments.

David Weisburd’s own reconciliation of these two research traditions proposed that neighborhood-level structural conditions and street-segment-level place characteristics operate as distinct but interacting levels of explanation, with neighborhood disadvantage shaping the overall probability that any given street segment within that neighborhood will develop into a concentrated crime hot spot, while street-segment-specific physical and management characteristics determine which particular segments within a disadvantaged neighborhood actually realize that heightened probability (Weisburd, 2015).

Within-Neighborhood Variation as Evidence for Place-Specific Mechanisms

The most theoretically decisive evidence for treating place-specific mechanisms as genuinely distinct from neighborhood-level structural explanation comes from research documenting substantial crime variation among street segments located within the very same disadvantaged neighborhoods, variation that purely neighborhood-level structural theories, applied without attention to place-specific mechanisms, cannot readily explain (Groff, Weisburd, & Yang, 2010). This within-neighborhood variation finding has proven particularly important for justifying place-based intervention strategies targeted at specific high-crime street segments rather than uniform neighborhood-wide intervention, since if crime concentrated purely as a function of neighborhood-level structural disadvantage, targeting specific segments within a disadvantaged neighborhood would offer little advantage over broader neighborhood-level intervention.

This within-neighborhood variation evidence directly supports the resource-allocation logic examined later in this article, confirming that place-specific characteristics, including the physical and management factors examined in the companion article on Crime and the Physical Environment elsewhere in this silo, contribute genuine, independent explanatory power beyond what neighborhood-level structural conditions alone would predict. Wesley Skogan’s earlier disorder research anticipated this within-neighborhood variation finding at a somewhat coarser spatial scale, documenting that physical and social disorder varied considerably even within neighborhoods researchers had previously treated as relatively uniform units, an early indication that neighborhood-level aggregation obscures meaningful sub-neighborhood variation later concentration research would document with considerably greater spatial precision (Skogan, 1990).

Weisburd’s Law of Crime Concentration at Place

Formulating the Law

Weisburd’s proposal that crime concentration constitutes a genuine law of criminology, rather than merely a recurring empirical pattern, rested on the observation that the concentration ratio held with a degree of cross-contextual consistency rarely achieved by criminological findings, a consistency Weisburd argued warranted elevating the finding to the status of a fundamental empirical regularity comparable to well-established laws in fields including physics and biology (Weisburd, 2015). This law-status proposal proved controversial within criminology, since the discipline has historically been more comfortable describing probabilistic tendencies and correlations than asserting law-like regularities, but Weisburd’s proposal has nonetheless substantially shaped how subsequent researchers frame and report concentration findings, with many now explicitly testing whether their own data conform to the specific four-to-six percent range Weisburd’s law specifies.

Elizabeth Groff, Weisburd, and Sue-Ming Yang’s companion research addressed a related question the law-status proposal raised: whether the specific street segments accounting for disproportionate crime remain stable over time or instead shift unpredictably from year to year, finding that while city-wide crime rates fluctuated considerably, the relative concentration ranking of individual street segments remained substantially stable across their sixteen-year observation period, reinforcing the claim that concentration reflects durable place-level properties rather than transient statistical noise (Groff, Weisburd, & Yang, 2010).

Implications of Treating Concentration as Law-Like

Treating crime concentration as a genuine law rather than a merely recurring pattern carries direct implications for how criminologists should approach theory construction, since a genuine law demands theoretical explanation with the same rigor that physical laws demand from physical theory, rather than being treated as an incidental finding that any number of loosely specified theories might happen to accommodate. This law-status framing has motivated increasingly precise theoretical work attempting to specify exactly which combination of routine activity, rational choice, and place-management mechanisms, examined in the companion article on Place and Crime elsewhere in this silo, jointly produce the specific concentration ratios Weisburd’s research has documented, rather than treating any single mechanism as a sufficient standalone explanation. Ross Matsueda’s broader theoretical assessment of place-based criminology noted that this law-status proposal, whatever its ultimate merit as a strict philosophical claim, has proven productive in a more modest sense, prompting researchers to test concentration findings against more precisely specified quantitative benchmarks than the discipline’s more qualitative theoretical traditions typically demand (Matsueda, 1988).

This theoretical precision-seeking has also generated productive methodological debate regarding whether the specific four-to-six percent concentration range Weisburd identified reflects a genuine universal constant or instead an artifact of the particular street-segment definitions and crime-counting methodologies most concentration researchers have employed, a debate examined further in the discussion of measurement methodology below.

Measuring and Mapping Concentration

Spatial Unit Selection and the Modifiable Areal Unit Problem

Measuring crime concentration accurately requires careful attention to the modifiable areal unit problem, the methodological challenge that concentration estimates can vary considerably depending on whether researchers define their spatial units as street segments, census blocks, or some other geographic division, since more finely divided units mechanically tend to produce more extreme concentration ratios than more coarsely aggregated units covering the same underlying geographic area (Weisburd, 2015). Most contemporary concentration research has converged on the street segment, defined as the stretch of roadway between two intersections, as the theoretically preferred unit of analysis, on the grounds that street segments correspond reasonably well to the scale at which place-management and guardianship, examined in the companion article on Place and Crime elsewhere in this silo, actually operate.

Jerry Ratcliffe’s methodological research on hot spot identification cautioned that different formal statistical techniques for identifying concentration, including simple point-pattern visual inspection, kernel density estimation, and nearest-neighbor clustering statistics, can generate meaningfully different conclusions about the precise boundaries and even the existence of any given concentration pattern, underscoring that concentration measurement, despite its apparent empirical straightforwardness, involves genuine methodological choices with substantive consequences for research conclusions (Ratcliffe, 2004).

Geographic Information Systems and Visualization

Contemporary concentration research relies heavily on geographic information systems to identify, visualize, and statistically test concentration patterns, a methodological infrastructure examined in dedicated depth in the companion article on GIS and Crime Mapping elsewhere in this silo, since these systems allow researchers to overlay crime incident data onto precise geographic coordinates and apply increasingly sophisticated spatial statistical techniques to distinguish genuine clustering from what would be expected under a purely random spatial distribution of the same total crime volume (Ratcliffe, 2004). This GIS-based methodological infrastructure has proven essential to concentration research’s credibility, since visual inspection of crime maps alone cannot reliably distinguish genuine statistical clustering from apparent clustering that a purely random process might have generated by chance alone.

The development of increasingly accessible and powerful GIS software over the past two decades has substantially lowered the technical barrier to conducting rigorous concentration analysis, extending this research capability from specialized academic research centers to routine practice within many municipal police department crime-analysis units, a practical translation examined further in the companion article on CompStat and Crime Analysis elsewhere in this silo.

Theoretical Explanations for Concentration

Routine Activity and Opportunity Convergence

Lawrence Cohen and Marcus Felson’s routine activity approach supplies the dominant theoretical explanation for why crime concentrates so heavily at specific locations, proposing that criminal events require the convergence of a motivated offender, a suitable target, and the absence of capable guardianship, a convergence that certain physical and social configurations of place host with far greater regularity than others (Cohen & Felson, 1979). Marcus Felson’s subsequent elaboration of crime generators and crime attractors, examined in dedicated depth in the companion article on Crime Generators and Crime Attractors elsewhere in this silo, further specified how certain locations draw large volumes of people together, whether or not those people harbor criminal intent, thereby mechanically increasing the statistical probability that the routine activity convergence Cohen and Felson specified will occur at those locations (Felson, 1995).

This routine activity explanation has proven attractive to concentration researchers because it directly predicts the kind of extreme concentration Sherman, Gartin, and Buerger and subsequent researchers have documented, since only a small subset of any city’s total locations combine the specific physical and social features, including transit access, commercial density, and population flow, that reliably generate the necessary convergence of offenders, targets, and absent guardianship. Derek Cornish and Ronald Clarke’s rational choice perspective supplies a compatible decision-theoretic account operating alongside routine activity theory, proposing that offenders assess the effort, risk, and reward a specific location presents before deciding whether to offend there, an assessment that concentrated, opportunity-rich locations systematically favor relative to more dispersed, lower-opportunity locations (Cornish & Clarke, 1986).

Place Management and Differential Guardianship Capacity

John Eck and David Weisburd’s place-management framework supplies a complementary explanation emphasizing variation in guardianship capacity across locations, proposing that property owners, employees, and other place managers vary considerably in their willingness and capacity to monitor and maintain orderly conditions at a given location, with weak place management producing conditions conducive to crime concentration even at locations that might not otherwise be expected to generate disproportionate criminal opportunity based on routine activity considerations alone (Eck & Weisburd, 1995). This place-management explanation helps account for an empirical pattern routine activity theory alone does not fully explain: why two physically similar locations experiencing comparable pedestrian and vehicular traffic volume can nonetheless display sharply different crime rates depending on the specific quality of ownership, management, and maintenance each location receives. Danielle Reynald’s observational research on guardianship in action extended this place-management framework by directly measuring residents’ actual monitoring and intervention behavior, finding considerable variation in guardianship capacity even among residents occupying structurally similar housing, a finding that helps explain why crime concentration persists even within blocks that appear demographically homogeneous (Reynald, 2011).

Ralph Taylor’s territorial functioning research supplies a further refinement of this guardianship-based explanation, documenting how residents’ and users’ psychological sense of ownership over a specific physical space independently predicts their willingness to intervene when disorder or crime occurs there, adding a social-psychological dimension to the more structurally focused place-management framework Eck and Weisburd proposed (Taylor, 1988).

Policy and Practice Implications of Concentration Research

Resource Allocation and Hot Spots Policing

The empirical concentration of crime at a small, identifiable number of locations carries an immediately practical implication for police resource allocation, since concentrating patrol and investigative resources at documented hot spots should, according to the theoretical mechanisms this article has examined, produce considerably greater crime-reduction return than distributing the same resources evenly across an entire jurisdiction, an implication examined in dedicated depth in the companion article on Hot Spots Policing elsewhere in this silo (Braga, 2001). Anthony Braga’s meta-analytic synthesis of hot spots policing evaluations found consistent evidence supporting this resource-allocation logic, documenting meaningful crime reductions at targeted hot spots across a substantial body of experimental and quasi-experimental evaluation research. Braga’s subsequent updated meta-analysis, incorporating a considerably expanded body of evaluation research accumulated over the following decade, confirmed and strengthened this finding, reporting consistent effects across studies employing randomized designs specifically, addressing earlier concerns that weaker, non-randomized studies might have driven the original meta-analytic conclusions (Braga, Papachristos, & Hureau, 2014).

This resource-allocation implication extends beyond conventional policing to inform the broader place-based crime prevention strategies examined in the companion article on Place-Based Crime Prevention elsewhere in this silo, since municipal code enforcement, business licensing review, and physical environmental modification resources can similarly be concentrated at documented high-crime locations rather than applied uniformly across a jurisdiction’s full geographic extent.

Predictive Applications and Contemporary Analytics

Crime concentration’s temporal stability, documented in Groff, Weisburd, and Yang’s longitudinal research, supplies the empirical foundation for contemporary predictive policing approaches, examined in the companion article on Predictive Policing in America elsewhere in this silo, since predictive algorithms rely fundamentally on the assumption that past concentration patterns will persist with sufficient stability to inform forward-looking resource-allocation decisions (Groff, Weisburd, & Yang, 2010). This predictive application illustrates how concentration research has moved beyond purely descriptive documentation toward directly informing forward-looking operational decision-making within contemporary American policing.

This translation from descriptive concentration research to predictive operational application has not proceeded without controversy, since critics have raised concerns regarding the potential for predictive systems built on historical concentration data to perpetuate or amplify existing patterns of disproportionate police attention toward specific communities, a concern examined in fuller depth in the companion article on Predictive Policing in America and connected to the broader public policy debates examined in the companion article on Environmental Criminology and Public Policy elsewhere in this silo. Cynthia Lum’s broader assessment of evidence-based policing cautioned that translating concentration research into predictive operational tools requires careful attention to data quality and equity safeguards, since concentration patterns derived from historical enforcement data can reflect prior enforcement patterns themselves rather than purely underlying crime distribution, a data-quality caution that has shaped contemporary best-practice guidance for predictive policing implementation (Lum & Koper, 2017).

Conclusion

Crime concentration stands as one of contemporary criminology’s most rigorously documented and theoretically consequential empirical regularities, established through Sherman, Gartin, and Buerger’s foundational Minneapolis research and extended through Weisburd’s proposed law of crime concentration at place, a finding whose cross-national and cross-crime-type replication has substantially strengthened confidence in its status as a genuine structural feature of how criminal opportunity distributes across urban space rather than an artifact specific to any particular research context. Routine activity theory and place-management frameworks together supply compelling, complementary theoretical accounts of why this concentration occurs, while geographic information systems and increasingly sophisticated spatial statistical methods have enabled progressively more precise measurement and mapping of concentration patterns.

This accumulated concentration research carries direct and far-reaching implications for police resource allocation, hot spots policing, and contemporary predictive analytics, confirming that the seemingly simple observation that crime clusters at specific places has generated one of environmental criminology’s most theoretically generative and practically consequential research programs, one whose continued development shapes numerous companion applications examined throughout this encyclopedia’s broader treatment of environmental criminology. David Weisburd and Lawrence Sherman’s own retrospective assessment of the place-based research tradition they helped establish credited the original Minneapolis concentration finding with catalyzing a fundamental reorientation of criminological inquiry that neither researcher fully anticipated at the time of their original 1989 publication, a reorientation whose continued influence this article’s survey of subsequent theoretical and methodological development confirms (Weisburd & Sherman, 1995).

References

Braga, A. A. (2001). The effects of hot spots policing on crime. Annals of the American Academy of Political and Social Science, 578(1), 104–125.
Braga, A. A., Papachristos, A. V., & Hureau, D. M. (2010). The concentration and stability of gun violence at micro places in Boston, 1980–2008. Journal of Quantitative Criminology, 26(1), 33–53.
Braga, A. A., Papachristos, A. V., & Hureau, D. M. (2014). The effects of hot spots policing on crime: An updated systematic review and meta-analysis. Justice Quarterly, 31(4), 633–663.
Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine activity approach. American Sociological Review, 44(4), 588–608.
Cornish, D. B., & Clarke, R. V. (1986). The reasoning criminal: Rational choice perspectives on offending. Springer-Verlag.
Eck, J. E., & Weisburd, D. (1995). Crime places in crime theory. In J. E. Eck & D. Weisburd (Eds.), Crime and place (pp. 1–33). Criminal Justice Press.
Felson, M. (1995). Those who discourage crime. In J. E. Eck & D. Weisburd (Eds.), Crime and place (pp. 53–66). Criminal Justice Press.
Groff, E. R., Weisburd, D., & Yang, S. M. (2010). Is it important to examine crime trends at a local “micro” level? A longitudinal analysis of street to street variability in crime trajectories. Journal of Quantitative Criminology, 26(1), 7–32.
Lum, C., & Koper, C. S. (2017). Evidence-based policing: Translating research into practice. Oxford University Press.
Matsueda, R. L. (1988). The current state of differential association theory. Crime & Delinquency, 34(3), 277–306.
Pratt, T. C., Cullen, F. T., Sellers, C. S., Winfree, L. T., Madensen, T. D., Daigle, L. E., Fearn, N. E., & Gau, J. M. (2010). The empirical status of social learning theory: A meta-analysis. Justice Quarterly, 27(6), 765–802.
Ratcliffe, J. H. (2004). The hotspot matrix: A framework for the spatio-temporal targeting of crime reduction. Police Practice and Research, 5(1), 5–23.
Reynald, D. M. (2011). Guardians on guard: Gauging the willingness of place managers to plan and monitor against crime. Eleven International Publishing.
Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social-disorganization theory. American Journal of Sociology, 94(4), 774–802.
Sherman, L. W., Gartin, P. R., & Buerger, M. E. (1989). Hot spots of predatory crime: Routine activities and the criminology of place. Criminology, 27(1), 27–56.
Skogan, W. G. (1990). Disorder and decline: Crime and the spiral of decay in American neighborhoods. Free Press.
Taylor, R. B. (1988). Human territorial functioning. Cambridge University Press.
Weisburd, D. (2015). The law of crime concentration and the criminology of place. Criminology, 53(2), 133–157.
Weisburd, D., Bushway, S., Lum, C., & Yang, S. M. (2004). Trajectories of crime at places: A longitudinal study of street segments in the city of Seattle. Criminology, 42(2), 283–322.
Weisburd, D., & Sherman, L. W. (1995). General deterrent effects of police patrol in crime “hot spots”: A randomized, controlled trial. Justice Quarterly, 12(4), 625–648.