Title: Crime and Place: Plenary papers of the 1997 Conference on Criminal Justice Research and Evaluation. Series: Research Forum Author: Ralph B. Taylor; Gordon Bazemore; Barbara Boland; Todd R. Clear and Ronald P. Corbett, Jr.; John Feinblatt, Greg Berman, and Michele Sviridoff; and Christopher Stone Published: July 1998 Subject: Crime patterns and future trends, restorative justice, community- based corrections, court management, crime mapping 151 pages 380,000 bytes ------------------------------- Figures, charts, forms, and tables are not included in this ASCII plain-text file. To view this document in its entirety, download the Adobe Acrobat graphic file available from this Web site or order a print copy from NCJRS at 800-851-3420. ------------------------------- U.S. Department of Justice Office of Justice Programs National Institute of Justice Crime and Place: Plenary Papers of the 1997 Conference on Criminal Justice Research and Evaluation NIJ Research Forum Cosponsored by Office of Justice Programs Bureau of Justice Assistance Office of Juvenile Justice and Delinquency Prevention U.S. Department of Justice Office of Justice Programs 810 Seventh Street N.W. Washington, DC 20531 Janet Reno Attorney General U.S. Department of Justice Raymond C. Fisher Associate Attorney General Laurie Robinson Assistant Attorney General Noel Brennan Deputy Assistant Attorney General Jeremy Travis Director, National Institute of Justice Nancy E. Gist Director, Bureau of Justice Assistance Shay Bilchik Administrator, Office of Juvenile Justice and Delinquency Prevention Department of Justice Response Center 800-421-6770 Office of Justice Programs World Wide Web Site http://www.ojp.usdoj.gov National Institute of Justice World Wide Web Site http://www.ojp.usdoj.gov/nij Crime and Place: Plenary Papers of the 1997 Conference on Criminal Justice Research and Evaluation Ralph B. Taylor, Gordon Bazemore, Barbara Boland, Todd R. Clear and Ronald P. Corbett, Jr., John Feinblatt, Greg Berman, and Michele Sviridoff , Christopher Stone Cosponsored by the Office of Justice Programs, the National Institute of Justice, the Bureau of Justice Assistance, and the Office of Juvenile Justice and Delinquency Prevention July 1998 NCJ 168618 NIJ Jeremy Travis Director John Thomas Program Monitor The Professional Conference Series of the National Institute of Justice supports a variety of live, researcher-practitioner exchanges, such as conferences, workshops, planning and development meetings, and similar support to the criminal justice field. Foremost among these forums is The Annual Conference on Criminal Justice Research and Evaluation. The Research Forum publication series was designed to share information from this and other forums with a larger audience. Opinions or points of view expressed in this document are those of the authors and do not necessarily reflect the official position of the U.S. Department of Justice. ------------------------------- The National Institute of Justice is a component of the Office of Justice Programs, which also includes the Bureau of Justice Assistance, the Bureau of Justice Statistics, the Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime. ------------------------------- Contents Crime and Small-Scale Places: What We Know, What We Can prevent, and What Else We Need to Know Ralph B. Taylor, Ph.D. Evaluating Community Youth Sanctioning Models: Neighborhood Dimensions and Beyond Gordon Bazemore, Ph. D. The Manhattan Experiment: Community Prosecution Barbara Boland The Community Corrections of Place Todd R. Clear, Ph. D., and Ronald P. Corbett, Jr., Ed. D. Neighborhood Justice at the Midtown Community Court John Feinblatt, Greg Berman, and Michele Sviridoff The Lessons of Neighborhood-Focused Public Defense Christopher Stone ------------------------------- Foreword In the course of a year, researchers studying crime and criminal behavior have a number of opportunities to present their work to their peers at conferences and other meetings of professional organizations, in seminars at academic institutions, through publication in scholarly journals, and in other ways. The Annual Conference on Criminal Justice Research and Evaluation is distinctive among these forums. It is the major annual event at which federally sponsored criminal justice research is showcased. Federal sponsorship means recognition at the highest level of government of the need for and value of empirically based, objective research in the field. It also ensures that the projects funded by the National Institute of Justice (NIJ), as well as the Office of Juvenile Justice and Delinquency Prevention, the Bureau of Justice Assistance, and other bureaus of the U.S. Department of Justice Office of Justice Programs will respond to the needs of practitioners--police chiefs and sheriffs, district attorneys, judges, and correctional officials and administrators, among others--as the major "consumers" of research findings. The centrality of practitioner needs in the work of NIJ and the other OJP bureaus and the scope of the conference combine to make it a major annual event for the sponsoring agencies. To ensure the conference reaches an even broader audience, NIJ is publishing the plenary papers. This inaugural volume presents the six plenary session addresses delivered at the 1997 conference. Each paper explores a variation on the 1997 conference theme, "crime and place." For a number of reasons, researchers are revisiting at the microlevel the locations where crime is committed and at the macrolevel community involvement with criminal justice agencies in addressing crime. Not least among these reasons are the advent of computerized mapping as an aid to tracking and studying crime, the trend toward devolution of government to the local level and, perhaps, the need to recover a sense of community. With law enforcement leading the way, other components of the criminal justice system are redefining their visions and missions to better address crime within the context of community. Many of these plenary addresses detail specific programs and experiments that have adopted promising new models. We are confident the plenary papers will whet readers' appetites for more. Abstracts of the other presentations at the 1997 conference can be obtained from the Institute for Law and Justice, which organized the conference on behalf of the sponsoring agencies (Web site: http://www.ilj.org). Jeremy Travis Director National Institute of Justice ------------------------------- Crime and Small-Scale Places: What We Know, What We Can Prevent, and What Else We Need to Know* Ralph B. Taylor, Ph.D., Temple University, and Visiting Fellow (1997), National Institute of Justice *This paper is an updated version of the author's presentation at the 1997 Conference on Criminal Justice Research and Evaluation. The author received support from grant IJ-CX-96-0022 from the National Institute of Justice while preparing this manuscript. Opinions are solely the author's and reflect neither the opinions nor official policies of the U.S. Department of Justice or the National Institute of Justice. Comments on earlier drafts from Alex Piquero, Keith Harries, Darcy Kim Rossmo, Ron Davis, and especially Christy Visher and Lorraine Green Mazerolle, contributed substantially to improving the paper. The author benefited as well from conversations with Philip McGuire, Steve Edwards, and Nancy La Vigne. Address correspondence to Ralph B. Taylor, Department of Criminal Justice, Gladfelter Hall, Temple University, Philadelphia, PA 19122 (e-mail: V1008E@VM.TEMPLE.EDU). Criminal justice researchers and practitioners recently began to shift their focus from people to places--from people who commit offenses to specific places where offenses occur (Weisburd, 1997). Some argue that such a shift in focus will result in more effective crime prevention and suppression policies. My presentation reviews the lessons learned from this new emphasis on specific places.[1] I examined results from a small number of interventions designed to reduce crime or disorder (or both) in "hot spots"; that is, in specific places with high crime rates. These efforts showed some success in reducing crime or disorder at specific sites, but they also showed stronger effects on some crimes than others and greater reductions in disorder than crime. Some of the benefits produced by these targeted interventions were both more modest and shorter lived than expected. It appears that changing from an offender-centered view to a place- centered view has brought police some success and shows potential for other areas of criminal justice practice, such as parole and probation. We may be able to prevent crime or disorder more effectively if we integrate our theorizing and practice in this area with the work of analysts outside criminal justice who have been studying places for some time. More specifically, I propose turning to ecological psychology, changing our units of analysis from hot spots to behavior settings, and using tools in spatial epidemiology to help us describe and understand how crime and disorder cluster in both time and space. These viewpoints may help us predict how crime differences will shift in the future, dramatically enriching what we know about crime and place and what we can do about it. Progressing down the cone of resolution Geographers use a "cone of resolution" to organize knowledge about spatial processes at different levels of analysis (see exhibit 1) (Brantingham et al., 1976; Harries, 1974).[2] Spatial patterns observed for crime rates vary as you progress down the cone to increasingly smaller scales of analysis. But even more important than the shifting spatial patterns, as you change levels of analysis "from national to city-block- level[s of] analysis, . . . this changes our perception of the 'where' and the 'what' of the crime problem . . . [and] the questions that can reasonably be asked of the data at each level" (Brantingham et al., 1976, p. 264). The theoretical processes behind the spatial patterns also shift as you move down the cone of resolution. The appropriate theoretical tools for understanding the spatial and temporal differences should be used at each level of display and analysis. Over the last century and a half, criminological researchers have moved progressively down this cone of resolution (Brantingham and Brantingham, 1981; appendix A). In the past 10 years, criminological researchers and police departments have looked closely at the relationship between crime and specific street addresses. The following section describes what is being learned about both crime and prevention at this level. But first we consider why we have moved in this direction. Forces leading research down the cone Three sets of forces have contributed to focusing on progressively more detailed relationships between crime and place: growing frustration, better tools, and increasingly well-supported theories. Frustration. Understanding that crime is higher in some neighborhoods than others has only limited relevance to a broad range of public agencies, including social services and police. Even in a high-crime neighborhood, most blocks will have low crime rates, and most addresses will have no reported crimes. Links between crime and community do not provide the data on specific places needed to guide deployment of police officers. Police hope to be more efficient and productive by focusing on high-crime places (or hot spots), rather than on high-crime communities. Crime data and police service calls that are "geocoded" reveal concentrations of crime in a few hot places (Sherman, 1995). Increasing patrol deployments to higher crime neighborhoods without knowing where and when crimes are likely to occur within those neighborhoods appears to produce only modest gains in crime control (Kelling and Coles, 1996). If we know exactly where and when crimes are taking place, we should be able to control crime more effectively. Tools. In the last few years relatively low-cost, personal computer-based geographic information systems have become increasingly available to criminal justice practitioners, with far-reaching impact on the police. Numerous departments have sought to integrate these capabilities into patrolling, detective, and community policing operations. For example, in 1994 the New York City Police Department started holding monthly crime strategy meetings, allowing precinct commanders to report current conditions to top management and collaboratively form strategies and review results. Initially, precinct pin maps were transported to the monthly meetings; shortly thereafter crime data were downloaded to personal computers, geocoded, and mapped. Practitioners felt that the electronic mapping successfully supported these monthly meetings (McGuire, 1997).[3] Researchers have begun to document how these tools are used or not used, who uses them, for what purposes, and with what impact (Chicago Community Policing Evaluation Consortium, 1996; Maltz et al., 1991; McEwen and Taxman, 1995). Even though mapping led to some extraordinary successes in the Chicago Alliance for Neighborhood Safety (CANS) mapping evaluation, many officers still reacted negatively to it. Negative reactions from officers in other jurisdictions implementing crime mapping tools also have been documented. In Jersey City, researchers found it a challenge to shift narcotics detectives' thinking from people to drug markets (Braga et al., 1997; Mazerolle et al., in press). In addition to helping officers patrol more strategically, crime maps can support police-community interactions (Maltz et al., 1991), focusing attention on key locations and documenting that some locations of concern to residents may not be crime-prone spots. The crime maps also can serve as beat institutional memory. Officers themselves can't do this because they don't serve all tours and often spend on-duty time away from their beats. A decentralized information base such as crime mapping facilitates the local police-community responses to crime and disorder envisioned by advocates of community policing or problem-oriented policing. Theory. The tools for understanding crime-place links have improved dramatically; the empirical support for key theories about crime and place has improved as well. Several theoretical models point toward features of the social and physical environment below the neighborhood level (see appendix A). Several forces have encouraged researchers and practitioners to learn more about the microscale connections between crime and place. Police departments have sought to target their resources in a more spatially focused way. In addition, easily used tools for mapping crime locations have become more widely available. We are just beginning to learn how these tools are implemented in police strategies, community policing, and problem-oriented policing. Finally, several relevant theoretical developments support the trend. Situational crime prevention has gained respect as more studies have accumulated data, as displacement-of-crime effects have proved less than feared, and as diverse benefits have been documented. The evolution of routine activities theory grounds it more firmly in microscale dynamics. Theory, operations, and research regarding crime and places We have arrived at the bottom of the cone of resolution and are examining links between crime and specific addresses or clusters of addresses. How are researchers thinking about places? What theoretical tools are they applying? What difficulties confront researchers and operations personnel? This section discusses the concept of hot spots and the issues related to it, notes operational challenges when using the concept, and reviews some recent field experiments using the hot-spot notion to guide interventions. Hot spots: Defining and operationalizing What are they? Crime and place researchers have adopted the term "hot spot" to describe a location of extremely high crime. (The term is borrowed from geology; hot spots are places where hot magma rises, often causing volcanoes to erupt.) A hot spot may be a single address, a cluster of addresses close to one another, a segment of a streetblock, an entire streetblock or two, or an intersection. Reviewing data on calls for service in Minneapolis (Sherman et al., 1989), researchers discovered crime hot spots, "small places in which the occurrence of crime is so frequent that it is highly predictable, at least over a 1-year period" (Sherman, 1995, p. 36). Many service calls to police came from a relatively small number of addresses. For example, in the case of calls for domestic violence in Minneapolis, had they been spread evenly over all street addresses and intersections throughout the city without any repeat calls coming from the same location, 21 percent of those locations would have generated a domestic disturbance call. But the data showed that because of the repeat calls from the same addresses, only 9 percent of the street addresses and intersections generated these types of calls. The researchers also found that the clustering of crimes at the same address varied as a function of the type of crime. Not surprisingly, for example, the clustering was stronger for domestic disturbance calls than for robbery. For a number of economic, psychological, and practical reasons, a physically abused woman often will remain with the abuser for a long period before moving out--or may not move out at all (Buzawa and Buzawa, 1990). If the victim and offender remain together, it seems likely that calls will continue to be generated from that address. Attention on hot spots has moved to many different types of calls for service. The concept also has expanded to concerns about the location of drug markets, uncivil or disorderly behaviors such as prostitution (Maltz et al., 1991), and the fear of crime (Fisher and Nasar, 1994; Fisher and Nasar, 1995). Differential utility. At first, it would seem that the identification of hot spots has enormous policy implications. Lawrence Sherman wrote, "If future crime is six times more predictable by the address of the occurrence than by the identity of the offender, why aren't we doing more about it? Why aren't we thinking more about 'wheredunit,' rather than just 'whodunit'?" (Sherman, 1995, pp. 36-37). I agree that for some crimes the policy implications are substantial. Domestic violence, burglary, and perhaps convenience store robberies represent the clearest cases in point. For domestic violence, both the offender and the victim are fixed in place as long as they live together at the same address. Building in part on the expectation of repeat domestic violence and domestic disturbance calls, a number of experiments have shown positive effects of police intervention for some types of households (Sherman, 1992). For burglary, the victim's site is fixed--a continuing source of vulnerability. After analyzing victimization data, researchers concluded that a burglarized household was most likely to be reburglarized soon after the initial burglary (Polvi et al., 1991) and the risk of another burglary increased with each burglary victimization at the household (Ellingworth et al., 1995). They also learned that burglary hot spots were "hot" due in part to the high numbers of repeat victims (Bennett, 1995). (For a comprehensive review of repeat victimization, see Farrell, 1995, and Spelman, 1995b.) These insights have led to projects in Kirkholt and Huddersfeld, United Kingdom, where council households (public housing households) victimized by burglary received immediate assistance from the police following the event. For the two types of crimes most rooted in space--domestic violence and burglary--there are some successful interventions, although important questions remain. In domestic violence the victim and offender are placebound and in burglary the victim is placebound; hence, the concept of hot spots seems to be worthwhile. In the case of general street crimes, such as robbery or assaults, where neither victim nor offender is fixed in place, the relevance of the hot-spot idea may be weaker. Certainly, with some types of these street crimes, place-based interventions also may prove successful. In the case of convenience store robberies (where the victim's location is fixed) or assaults at bars or taverns (where specific facilitating conditions are fixed), localized interventions may prove relevant. Conceptual questions. Numerous questions arise about hot spots (La Vigne, 1997). The underlying mix of person- and place-based factors making a hot spot hot may vary considerably. Two underlying person- based factors may contribute to victimization: state dependence and risk heterogeneity. High victimization risk may be dependent upon the state the person is in at the time of victimization. For example, if a burglary victim has had her front door jimmied, she is more susceptible to another burglary until the lock is repaired; this is called state dependence. By contrast, some people are more likely to be victimized because of their habits, routines, or occupations; this is called risk heterogeneity (Lauritsen and Quinet, 1995; Lauritsen et al., 1991). A bouncer at a popular night spot in a tough part of town is at more risk of assault than an elementary school crossing guard in a quiet neighborhood. These two underlying factors contribute in varying degrees across the population of users at a site and across sites to making hot spots hot (La Vigne, 1997). In regard to street crime, hot spots may vary in their intensity, depending on the mix of land uses and facilities at the site (La Vigne, 1997). Some sites (called crime generators) may attract many potential victims simply because they generate large volumes of pedestrian traffic. Other sites (called crime attractors) may attract many potential offenders because of their reputations. The appropriate policy response at the site may depend substantially on these underlying place- and person-based variations from site to site. Some have even questioned the source of information used to define calls for service, suggesting it may be too limited (Maltz et al., 1991). Community residents may have other high-problem locations but not call the police about them through 911. Community-defined hot spots may be quite different from police-defined hot spots. Reports of these differences emerged from citizen-police interactions in the Chicago CANS project. At meetings, police officers would sometimes find that residents' concerns centered on sites that were neither crime-based nor 911-call-based hot spots. An additional limitation of dependence on calls for service is that the hot spot is "tied to a physical place; thus, events that are precipitated by activity at a neighboring place are not considered" (Maltz et al., 1991, p. 42). In other words, a hot spot may appear hot because of something going on nearby, outside the hot spot. For example, fights in a parking lot at 2 a.m. may occur because a bar down the street closes around that time. In a study of public housing in the Bronx, when violence in the public housing projects increased, violence in surrounding locations also increased (Fagan and Davies, 1997). In the case of assault, crime not only diffused outward from the projects; it also diffused into the projects from adjoining areas. In the case of homicide or robbery, a crime hot spot might appear in a location outside a project, but be "driven" in part by crime from inside the project. Thus, we need to understand how activities in, attitudes toward, and crime in the hot spot itself are shaped by the surrounding context (Rosenbaum and Lavrakas, 1995). Operational questions. Locating and bounding (that is, defining the boundaries of) hot spots are not straightforward processes. In the Minneapolis Hot Spot experiment, for example, if officers left their car parked in the hot spot and walked on foot through the alley behind the hot spot, were they counted as having a presence in the hot spot, even though they were walking outside it part of the time (Buerger et al., 1995)? Reading the research literature reveals similar difficulties. For example, Lawrence Sherman and David Weisburd (1995, p. 630) described how their research began with 5,538 "lukewarm to hot" addresses: We defined hot spots operationally as small clusters of addresses with frequent "hard" crime calls as well as substantial "soft" crime calls for service. . . . We then limited the boundaries of each hot spot conceptually as easily visible from an "epicenter." They next applied a number of restrictions to these sites and procedures for resolving disagreements about hot-spot boundaries. The disagreements were resolved using the following general principles: o No hot spot is larger than one standard linear streetblock (although a few exceptions were allowed on the basis of visual sightings on very short blocks). o No hot spot extends for more than one-half block from either side of an intersection. o No hot spot is within one standard linear block of another hot spot (again, a few exceptions were made). With this final list of 268 reconfigured address clusters, Sherman and Weisburd looked for sites with at least 20 hard-crime and 20 soft-crime calls in the year studied. They also eliminated sites with high variance from year to year: clusters "with greater than 150 percent increases or 75 percent decreases in hard-crime calls from one year to the next" (Sherman and Weisburd, 1995, p. 632). When all these operations were completed, 150 hot spots remained in which the number of calls for service ranged from 6 to 50 per month. In this same study it also appeared that narcotics detectives held very different views of drug market activities; the way they bounded hot spots depended considerably on the types of drugs sold there (Eck and Wartell, 1998). Such complexities suggest that it may not be easy for police to define crime hot spots precisely and allocate patrol resources accordingly. This raises general questions about the clarity of the concept of hot spots. Can crime mapping programs resolve the "fuzziness" of hot-spot boundaries and precisely locate them? Spatial and Temporal Analysis of Crime (STAC), developed by researchers at the Illinois Criminal Justice Authority, is probably one of the most widely used crime mapping programs, in part because it is free. It analyzes geocoded crime data and draws circles or ellipses around the hot spots. Although the program has proved extremely useful in a number of contexts and criminal justice practitioners strongly embrace it, its limitations probably prevent it from identifying clear boundaries of hot spots. Two restrictions illustrate the point: The shape of the hot spots produced is limited to circles and ellipses, and the hot spots defined by the program depend in part on the specific area searched (Block, 1993). Also, officers may encounter difficulties in decoding what is happening at a hot spot and deciding who should be policed (Braga, 1997; Green, 1996). This may be a particular challenge when a hot spot extends beyond a single address to cover a corner or block. Onsite officers need to distinguish between legitimate users, such as teens waiting at a bus stop after school lets out, and illegitimate users, such as teens trying to sell drugs at a bus stop. Narcotics officers, used to chasing dealers rather than cleaning up drug markets, may have little patience for this kind of onsite detail work (Braga, 1997; Braga et al., 1997; Eck and Wartell, 1998). In summary, conceptual definitions of hot spots are understandably broad, resulting in wide variation from study to study, since researchers and practitioners want to search for hot spots for different types of crimes and disorders. The broad scope not only results in a conceptual jumble, but also creates considerable operational difficulties--both for field practitioners deciding on deployment strategies and for evaluators. Hot spots research: Examples Researchers have used the hot-spot idea to guide interventions for reducing drug market activity, disorderly street behavior, dilapidated physical conditions, and crime. Some of the interventions have been traditional or community-oriented policing initiatives; others used multiagency responses to the sites. This section briefly reviews a small number of experiments or quasi-experiments. City park. Police in Oslo, Norway, learned through observation and citizen surveys that a particular downtown city park caused concern among pedestrians and store personnel (Knutsson, 1997). Detailed behavioral observations allowed researchers to target the specific, problematic locations. The park was used by addicts to gather, buy, sell, and use drugs. A park redesign, combined with more patrols by narcotics officers and regular officers, resulted in increased legitimate use of the park, improved resident perceptions of the locale, and more satisfaction among nearby business owners. The nontraditional portion of the program--the park redesign, which facilitated "natural surveillance" by legitimate users-- "prolonged the effort of the intervention . . . [and] discouraged offenders from returning" (Knutsson, 1997, p. 138). Multiagency response in Oakland. In Oakland, California, a coordinated multiagency team sought to improve physical conditions at drug nuisance properties and remove tenants causing problems. The methods used at the experimental sites were complex. Police accompanied housing inspectors, helping them gain entry. Inspectors cited owners for violations, and court action was taken against property owners failing to comply with civil law citations. Police arrested dealers and patrolled more intensively at the sites. Trained raters then judged photographs of the sites before and after the intervention. Comparison of preintervention to postintervention scores showed a dramatic impact on physical disorder and a more modest impact on crime. For example, the number of blighted properties dropped from 134 to 15. However, drug problem calls declined only 4 percent from the pretest period. Commercial and owner-occupied properties showed the greatest drops in narcotics activity (Green, 1996). In part because business owners were more likely than residential owners or occupants to cooperate with police efforts, the greatest impact was found at highly visible commercial establishments (Green, 1996). Another study of the same intervention examined such behavioral outcomes as drug selling and found that the program itself had an impact on block activity. So too did prior block differences in local informal social control (Mazerolle et al., 1997). Formal and informal controls apparently can work side by side to influence drug-selling activity at specific problem locations. Police contribute in several ways to these multi-agency interventions (Green, 1996). Not only are they carrying out traditional arrests and making contacts with suspects in the field, but they are also bolstering citizen confidence with their presence and empowering personnel from other agencies by accompanying them onsite. When police team up with housing inspectors, for instance, it legitimizes the inspectors' visits and helps them gain access. In addition to modest crime reductions at the targeted sites, the adjoining two-block buffer zone around each site also benefited (Green, 1995). The diffusion of crime-reduction benefits slightly outweighed the amount of crime displaced out of the target sites.[4] Two other studies of code enforcement are somewhat similar in spirit to the Oakland initiative. In Cook County, Illinois, researchers studied residents' views of drug activity, comparing those of people living on blocks where a drug property had been targeted for civil abatement with those of people living on blocks not targeted (control blocks) (Lurigio et al., 1998). The study generally found little positive impact of the abatement procedures. In a randomized experiment in San Diego, California, rental addresses targeted for drug enforcement in the previous 6-month period received either no treatment, a letter, or a meeting with the landlord, a police officer, and a code officer to develop remediation strategies (Eck and Wartell, 1998). Researchers found significantly reduced drug activity in the followup period for the addresses in the "meeting" group. The authors concluded that code enforcement can work effectively at rental properties used for drug activities. At the same time the study details the many reasons landlords do little to address these problems. Minneapolis crime and disorder hot spots. In an experiment using random assignment in Minneapolis, Minnesota, hot spots received the normal amount of patrolling or increased police presence (Sherman and Weisburd, 1995). (The procedure used to define hot spots is described above.) A typical hot spot was a group of attached two- and three-story buildings clustered around an epicenter, usually a street corner. These intersections often consisted of a mix of commercial services, usually including food and drink and open until late at night. Exceptions to this pattern included low-rise multifamily housing developments and convenience stores. Bus stops, pay telephones, and intensive street lighting were common features of hot spots (Sherman and Weisburd, 1995). During the first 6 months of the program, experimental sites received two to six times as much police patrol time as did the control sites. As in Oakland, crime decreased only slightly in the experimental site. Although the percentage reduction in calls for service was substantial, it translated into one crime call fewer per month in the experimental location than in the control locations.[5] When measured in percentages, disorder reductions were sizable. For the period when the experiment maintained the best treatment integrity, "half as much disorder was observed in the experimental group as in the control [group]," according to the researchers (Sherman and Weisburd, 1995, p. 643). They concluded that "substantial increases in police patrol presence can indeed cause modest reductions in crime and more impressive reductions in disorder within high-crime locations" (Sherman and Weisburd, 1995, abstract). But in absolute terms, these differences were not substantial, in part because of the extremely low rate of social disorder observed. Observers saw about 2 disorderly minutes per 100 minutes in the experimental sites and about 4 disorderly minutes per 100 minutes in the control sites. Since observers were stationed in the most stable high- crime sites in the city and observed from 7 p.m. to 3 a.m., this seems an extremely low rate of disorder. Other onsite researchers have noted similar low rates (Giacomazzi et al., n.d.). If such low rates of social disorder apply to most other areas, we can wonder about the practicality of initiatives geared to reducing those conditions. A review of calls received after the patrols left the area showed that a police stop of 11-15 minutes was most useful because it represents the shortest time that created the most residual deterrence (Koper, 1995). This latter investigation of the timing of the police stops addressed the question of the most effective "dosage" at a hot spot. Kansas City crackhouse blocks. In another experiment, researchers examined the impact of police raids on blocks with crackhouses in Kansas City, Missouri (Sherman and Rogan, 1995). The criteria for including the blocks in the study were receiving a large volume of calls for service, being the site of a successful undercover buy, and receiving at least five calls for service in the month before the undercover buy. Eligible blocks were randomly assigned either to be raided or not to be raided. The raids showed a marginally significant impact on all calls for service, with calls decreasing 18 percent on the raid blocks and 10 percent on the nonraid blocks. The raids had no significant impact on the number of calls for violent or property offenses. Furthermore, the benefits of the raids evaporated quickly--in about 12 days. The benefits also depended on the season; effects on disorder "found in the winter disappeared in the spring" (Sherman and Rogan, 1995, p. 776). The authors concluded, "Raids seem to provide afflicted blocks with nothing more than a brief respite from crackhouses" (Sherman and Rogan, 1995, p. 777). Jersey City drug markets. In this study a complex web of policing services--including increased surveillance, crackdowns of varying scope and intensity, and a maintenance observation period--were directed at a randomly assigned set of drug hot spots (Sherman and Weisburd, 1995). Problem-oriented policing strategies helped officers develop detailed information about the problem profiles at each site (Braga et al., 1997). Using calls to police to construct outcome measures, researchers compared call volumes 7 months preceding and 7 months following the intervention. They found no effects on calls for violent crime or property crime and no reliable effects on narcotics calls. But some categories of disorder calls-- suspicious persons, public morals, and police assistance--did show impacts. Researchers commented, with some surprise, that drug activity did not appear to be affected by the intervention, but some types of disorder calls were affected. They suggested that drug activity and other activities around disorder were more independent of one another than previously suspected. They also noted that the bulk of the treatment effects were "evident primarily in very large changes in a few of the most active hot spots included in the study" (Sherman and Weisburd, 1995, p. 727). In other words, some features of the hot spots themselves, or of the surrounding context, made the treatment more effective in some places than others. Reducing repeat burglary in U.K. public housing developments. In the Kirkholt (United Kingdom) burglary reduction project, a targeted police response to disadvantaged households victimized by burglary produced a 75-percent drop in this crime. Through similar programs in Huddersfield, victims received increasingly comprehensive assistance from the police immediately after the burglary; those with more burglaries received more assistance.[6] Those burgled more than twice received sensor-triggered alarms connected to the police station. The Huddersfield program design produced another benefit: It did not encourage offenders to increase their search space and move beyond the area covered by the program (Anderson and Pease, 1997). Neither study was a true experiment, however, and other initiatives at the Kirkholt site may have contributed to program success (Hope, 1995). Conclusions from examples. The foregoing represent some of the most carefully designed experiments and quasi-experiments available on the impact of place-based, microscale interventions. The small number of studies examined suggests that we draw general conclusions cautiously. Place-based interventions have effects on crime and disorder. The impacts on crime calls are more modest and less enduring over time than had been expected. All else being equal, high-crime addresses or clusters of addresses are most likely located in areas with generally higher crime rates. As Lawrence Sherman and David Weisburd have stated, "Substantial increases in police patrol presence can typically cause modest reductions in crime and more impressive reductions in disorder within high-crime locations" (Sherman and Weisburd, 1995, p. 104). Many hot spots are probably surrounded by and nested within hardened criminal subcultures. Many offenders, as well as previous offenders returned from supervision to the community, are likely to live in or near such locations. The resistance of criminal activity to crime control efforts is not surprising given such surroundings. If we look only at the impact of such interventions on crime, questions arise about their cost-effectiveness. The substantial impact observed on disorder suggests that we know more about reducing disorder than about reducing crime. Interventions like the Specialized Multi Agency Response Team (SMART) program in Oakland show that police can substantially reduce physical disorder by working closely with other regulatory agencies such as those responsible for housing, zoning, and public works. Creating and maintaining a productive partnership with other agencies appear key to an effective intervention, but "perhaps the greatest challenge for SMART-like interventions rests in the ability of the police to develop good working relationships with other city agencies" (Green, 1995, p. 99). The relative independence of changes in crime and changes in disorder in several studies raises questions (Giacomazzi et al., n.d.; Popkin et al., 1997; Taylor, 1996). How do interventions that succeed in reducing disorder more than crime fit into the broken-window thesis (Kelling and Coles, 1996)? What factors contribute to the relative independence of crime and disorder shifts? We know how to reduce disorder in small locations. Other short- and long-term studies also suggest that shifts in crime and disorder are relatively independent of one another (Giacomazzi et al., n.d.; Popkin et al., 1997; Taylor, 1996). Rousting panhandlers, citing landlords for nuisance tenants or substandard conditions, or using other strategies can change local conditions. Perhaps we know less about preventing crime on the microscale. Are the two processes linked more loosely than anticipated, especially on the small scale of the units targeted here--addresses, address clusters, intersections, and blocks? Do researchers expect a strong linkage between crime and disorder in small units only because that linkage exists in large units (Hannan, 1971)? Might the two processes link with considerable strength, but fail to shift over time with comparable speed? Each may cause the other, but one may shift more rapidly than the other. Finally, presuming the prevention effects attributed to the Kirkholt burglary prevention programs have not been overstated, are place-based interventions most effective when either the victim (of burglary) or the victim and offender (in the case of domestic violence) is fixed in place? When neither is fixed in place, are prevention impacts more modest? Issues ahead Studies have clearly demonstrated that context-focused crime control is useful. The gains have not been as substantial as initially promised, but they have been noticeable. To make more progress in this area, we need to pay attention to a number of issues. How we think about places Thus far the development of concepts about crime and small places has been guided by the hot-spot idea or analogies based on individual offender dynamics. Neither line of thinking may be the most helpful one to pursue. The hot-spot analogy with geology suggests one underlying cause "bubbling up" to create disorder. But several factors may be responsible. Features of the site itself and the surrounding area jointly contribute to the high crime rate. In the case of large, open-air drug markets, adjacency to high-volume traffic arteries and vacant housing is important (Rengert, 1996). In the case of tavern crime in the north side of Chicago, tavern density and proximity to mass-transit stops played roles in creating hot spots (Block and Block, 1995). The number of contributing factors may be so great that the emergence of the hot spot is "overdetermined"; removing some of the factors may have little or no effect. Several authors, following up on the theory that neighborhoods can have "criminal careers" (Reiss, 1986), have suggested this might be true for specific places as well (Weisburd, 1997). Places, however, may be too fundamentally different from people to warrant pursuing this notion. The career theory is grounded in offender-centered criminological theory. Although some dimensions of career theory may have some relevance to places, it remains unproven and its relevance is likely to be far weaker than the relevance of place-centered theories. Place-centered theories will be applied as appropriate to particular levels of analysis (Brantingham et al., 1976). To apply place-centered theories, person-centered criminologists need either to learn these theories or to collaborate with those who know them. Behavior settings theory It may prove more profitable to rely on empirically validated constructs found within an extensive volume of available research on places (Felson, 1995). Ecological psychologists have been trying to understand how places work since the late 1940s (Barker, 1968; Wicker, 1979). According to ecological psychology (Barker et al., 1943; Fox, 1983; Fox, 1984a; Fox, 1984b; Wicker, 1972; Wicker, 1979; Wicker, 1987), behavior settings are freestanding, natural units of the everyday environment with a recurring pattern of behaviors and a surrounding and supporting physical milieu. These units organize community life. As Allan Wicker has noted, "Roger Barker [the psychologist who originated ecological psychology] views behavior settings as small-scale social systems whose components include people and inanimate objects. . . . The various components interact in an orderly, established fashion to carry out the setting's essential functions" (Wicker, 1987, p. 614). Analyzing all behavior settings in a small Midwestern town for a year, Roger Barker and his colleagues found different types of behavior settings, such as billiard parlors, taverns, bus stops, parking lots, parks and playgrounds, street fairs, variety stores, and welfare offices (Barker, 1968). In some urban areas, some of these behavior setting types can be high- crime locations. "Streetblocks"--the two sides of a street between two cross streets--qualify as behavior settings for the following reasons (Taylor, 1997): o People get to know others as they pass by and observe their routines. At certain times or on certain days, they know what others are going to do. They consequently develop positive or negative sentiments toward others. o Associated role obligations such as neighborliness go along with being a group member (Mann, 1954). Role differentiation also occurs, with some residents playing more central roles, such as the block organizer or block busybody, and others being more peripheral in the ongoing life of the block. o Unless there is extremely high turnover or heterogeneity, norms about acceptable and unacceptable behavior are generally shared. People generally agree about what is and is not acceptable at various times. The specific points of agreement--and their clarity--vary as a function of location, structure, and social psychological factors. So too, norms, ranging from clear to diffuse, may be more or less widely shared. o Blocks exhibit regularly recurring rhythms of activity (Jacobs, 1961; Jacobs, 1968). In ecological psychology these are called standing patterns of behavior. People go to work and come home, children go to school and come home, mail carriers and paper carriers make their rounds certain times of the day, people engage in weekly activities (like car washing) and seasonal activities (like leaf raking or lawn mowing or gardening). Each block has a regular standing pattern of behavior composed of overlapping cycles, although the pattern may evolve noticeably over a substantial period or, in changing neighborhoods, over a short period. o The surrounding physical milieu supports and contains the behavior program. A streetblock is physically bounded by the fronts of houses or the alleys or fences behind the houses and the cross streets. What happens one block over or behind the streetblock has much less impact on the block than activities occurring within it. This is particularly evident when a fire or large snowstorm occurs. The block is a major container, partitioning residents from what is happening elsewhere. The behavior setting can no longer exist if the physical container is removed (e.g., urban renewal). o Behavior settings and streetblocks evolve over time. "Settings are continually constructed and reconstructed as new personnel and equipment are added or exchanged for exiting components" (Wicker, 1987, p. 616). Similarly, on streetblocks families move in and out and houses may be converted to apartments or stores or abandoned and torn down. Small stores may come and go or be converted back to apartments. As the streetblock changes physically over time or its population shifts, so too may the standing patterns of behavior change. The daily and weekly rhythms on a streetblock depend not only on the residents of the block and its layout but also on conditions on the surrounding blocks and in the broader neighborhood--and how that block and surrounding blocks may shift over time. For example, a block with a corner store will have different numbers and types of people on the sidewalk at different times of day compared with a block that has no store (Baum et al., 1978). Who comes to the store depends on the makeup of the broader neighborhood as well as the block, and the arriving foot traffic will shift as the neighborhood shifts. What will we gain by drawing on ideas about behavior settings? If we understand how specific types of settings work and what their standing patterns of behavior are, we should be able to identify which critical pieces are missing when such a site becomes a high-crime location. An isolated bar may become a high-crime location. A pizza parlor at an intersection with several other stores may likewise become a trouble spot. But the route by which each site became a high-crime site may be quite different. What is needed to lower crime also may be quite different. The evaluation of Oakland's SMART intervention hints that different remedies are needed for commercial and residential sites (Green, 1996, p. 68). Police helped initiate multiagency civil remedies at problem addresses in both types of sites. The tactics most effectively reduced drug activity at the commercial sites, "where business owners risk more than residential tenants do by not cooperating with police." Location-specific remedies are required. Understanding how the setting went awry helps gauge what remediation tactics are most effective. Routine activity theory is beginning to incorporate behavior setting ideas. Marcus Felson (see appendix A) has suggested differentiating guardians of specific targets in a setting--for example, differentiating site managers, who are responsible for the overall functioning of a place, from those with lesser responsibilities (Felson, 1995). This closely parallels Roger Barker's (1968) observation that people participate in behavior settings at different levels (see exhibit 2). A particular type of behavior setting may become a high-crime location because setting participants needed to maintain the standing pattern of behavior are missing from some parts of that setting. Alternatively, involved personnel may lack the experience needed to maintain the standing pattern of behavior. To solve the crime problems in a location, we need to "unpack" the dynamics of the site. Detectives or patrol officers can intervene most effectively if they can discriminate between legitimate and illegitimate participants in the setting (Green, 1996; Rosenbaum and Lavrakas, 1995; Weisburd and Green, 1996). Although behavior setting theory provides an extremely efficient lens for focusing on the most relevant dynamics, the process may prove awkward and unfamiliar to detectives and uniformed patrol officers (Braga, 1997; Braga et al., 1997). A final advantage of behavior setting theory is its explicit recognition that the standing pattern of behavior is timebound. Several of the interventions mentioned earlier point to timebound effects. For example, residual deterrence from crackdowns may be seasonally dependent (Sherman and Rogan, 1995). Understanding how block behavior settings vary by season would help to interpret this dependency. What police personnel do in these locations How do police decide what to do after arriving at a high-crime location? The most effective and minimally intrusive response is for officers to disrupt the illegal activities and avoid disrupting legal activities (Green, 1996). An understanding of setting dynamics would provide guidance on achieving such a goal. Lacking this understanding, the officers may simply want to aggressively police the entire setting (Braga et al., 1997). What information sources are needed To intervene effectively in high-crime sites, detailed knowledge of the setting is needed. Beat officers have such detailed knowledge. The challenge is to systematically elicit, share, and archive that information. Sometimes locations become high-crime sites because of surrounding conditions, including movement patterns of potential offenders and victims (Maltz et al., 1991). Detailed information about these conditions can help to make these connections (Mazerolle and Terrill, 1997). Crime mapping tools become even more powerful if crimes are combined with other geocoded community information. When certain rules are followed for the contextual information, it becomes a "GeoArchive" (Block, 1996): A GeoArchive is a particular kind of geographic information system [GIS] database. . . . Like all GIS, a GeoArchive is especially organized for spatial data, and contains a digitized map and data geocoded to that map. It can be seen as a large set of map transparencies that can be overlaid on each other. But a GeoArchive has several characteristics that distinguish it from other GIS databases. . . . A GeoArchive links (1) address-based local- level data from (2) a variety of law enforcement and community sources, and (3) is organized so that it can be updated, maintained, mapped, analyzed, and used by those who are developing and implementing strategies of crime reduction in the local community. Such a system includes information on housing code violations and locations of facilities like bus and subway stops. The geographic display of local features and data on crime and calls for service quickly links local features to various crime problems. For example, proximity to mass-transit stops was an important factor in tavern crime in the north side of Chicago (Block and Block, 1995). Effectively reducing crime in high-crime locations requires knowing much more about the site than the crime profile. Deciding how much the site needs to be fixed Crime varies from place to place and over time, and numerous factors affect those variations. Random influences and temporal patterns, for example, can contribute to a high crime rate at a site. Deciding how much of that crime profile emerges from factors police can address is a challenge for those allocating police resources. Some analytic techniques can answer such questions (Spelman, 1995a). For example, differences in high school crime rates may arise from random sources, changes over time, and persistent differences between the sites. Persistent differences may arise from the demographics of students attending the various schools, the administrations running the schools, and the surrounding neighborhoods. Changes over time may arise from seasonal variation or long-term regional or national trends. But it is only the persistent site differences that are an appropriate target of police interventions. Consequently, in planning an intervention and deciding what the criterion for success is, police planners might profitably isolate those persistent site differences so they know how much of a reduction they seek. Therefore, the intervention goal is tied to the analysis of the crime variation or the calls for service variation. Otherwise, agencies risk "overmedicating" a hot spot that is only "warm," or "under-medicating" a "red-hot" hot spot. Picking sites for intervention The process of identifying high-crime locations for intervention is still in a rudimentary stage, especially compared with the work in spatial epidemiology of disease (Giggs, 1990; Thomas, 1990a; Thomas, 1990b; Thomas, 1992; Wartenberg and Greenberg, 1990). Whereas we look for crime clusters in space, they look for disease clusters in space and time. Whereas we look for clusters compared to a background of random variation, they look for clusters compared to a theoretically meaningful indicator of background intensity. Spatial epidemiologists can readily determine where and when there is more disease than there should be, given a range of factors. For criminologists, the theoretical tools, understanding of etiology, and data available rarely permit determining where there is more crime than there should be. Some would argue the above approach is inappropriate, stating that wherever crime or calls for service are sizable, intervention is needed. On the other hand, if we know how much site features, local population features, and characteristics of adjoining locales--factors that cannot be changed--contribute to the crime rate, we have a better idea of how effective we can expect to be at these sites. Granted, spatial epidemiologists work with causal frameworks far different from those used by criminologists. Given strong enough space- time clustering, they can assume one disease agent is at work. Within our field, pinning down causes seems much more arduous. Nevertheless, despite these differences, spatial epidemiology's simultaneous attention to time and place clustering and concern with finding the appropriate control rate should be incorporated into spatial criminology. Further expanding the concept Policymakers and practitioners alike have lavished considerable attention on connections between crime and small places in the past few years. They have added issues of disorder and fear of crime to the mix as well. In the middle to late 1990s, police operations have been most affected by the small-scale focus. Nonetheless, it is easy to imagine expanding further and, for example, using information about parolee or probationer residence linked to a GeoArchive to help decide supervision levels (Buckley and Kane, 1997; Van Dine, 1997). In short, the current focus may expand the outcomes of interest and the type of place-based processes of interest, proving relevant to criminal justice practitioners beyond the police. Appendix: Overview of Theories Relevant to Crime-Place Links Shifting from an offender-based to a place-based criminological theory--or more accurately, developing the place-based theory as a complement to the individual focus--requires thinking in new ways (Weisburd, 1997). That conceptual retooling is now under way and is needed to effectively support crime prevention and control initiatives. I have proposed that we spur this development by relying on ecological psychology research. Recent revisions to place-based routine activity theory are already moving in that direction. This field helps us better understand how public places go awry and become high-crime or high-fear sites. It also helps us focus on what officers should do onsite. Quick historical tour French and English researchers in the middle of the last century investigated regional and within-county variations in offenders and crime (Glade, 1856). Regional variations continue to inspire debate in the United States, the debate on the Southern subculture of violence being one case in point (Messner, 1983; Wolfgang and Ferracuti, 1967). For example, most recently Dov Cohen and Richard Nisbett suggested that the higher homicide rates in the South emerged from a culture of honor linked to a history of independent pig farmers in mountainous Southern regions (Cohen and Nisbett, 1994; Nisbett, 1993). Social workers in the latter part of the 19th century pointed to parts of London where misery, drunkenness, and crime flourished and many children were homeless. Some urban renewal plans around that time and later were focused on breaking up those troublesome areas (Morris, 1957). In the first half of this century, researchers who were later identified with the Chicago School of Human Ecology documented within-city variations in delinquency (Bursik and Grasmick, 1993; Shaw and McKay, 1969), offender rates (White, 1932), mental illness (Faris, 1948), and other social problems. These and other researchers have linked crime, offending, and delinquency rates with features of community structure like economic status, stability, and racial composition (Bursik and Grasmick, 1993; Sampson and Lauritsen, 1994). Even more recently, with the availability of the British Crime Survey (Sampson and Grove, 1989) and the Project on Human Development in Chicago Neighborhoods (Sampson et al., 1997), researchers can document the contributions of informal social control (Bursik, 1988), independent of community structure or mediating community structural impacts. Recent quantitative and qualitative studies help explain how offending patterns are linked to employment opportunities, unemployment levels, street culture, and ethnic background (Bourgois, 1989; Bourgois, 1996; Sullivan, 1989; Sullivan, 1993; Williams, 1989; Wilson, 1987; Wilson, 1996) and how all these factors contribute to long-term neighborhood deterioration. In short, after several decades of theorizing we have begun to empirically document the individual- and community-level connections and contextual effects that link delinquency, offending, and crime rates to neighborhood features, over time as well as at one point in time. In the past 20 years or so, theorists like Jane Jacobs (Jacobs, 1968) and Don Appleyard (Appleyard, 1981; Craik and Appleyard, 1980) have devoted attention to how streetblocks "work." Researchers have either focused on streetblocks themselves, on both sides of the blockface, between two cross streets, or on census blocks (the four sides of a city block as you walk around it). Streetblock researchers have confirmed the contributions of land use (Baum et al., 1978; Taylor et al., 1995), block design (Taylor et al., 1984), and social and organizational block characteristics (Perkins et al., 1990; Perkins et al., 1992; Taylor et al., 1981) to social withdrawal, crime, and disorder. Census block researchers have linked crime to a range of block features, including high schools, bars, and public housing (Roncek, 1981; Roncek and Bell, 1981; Roncek et al., 1981; Roncek and Faggiani, 1985; Roncek and Maier, 1991; Roncek and Pravatiner, 1989; Snyder, 1995). Available theoretical tools Both the tools for understanding crime-place links and the empirical support for key theories about crime and place have dramatically improved. The following theoretical models point toward features of the social and physical environment below the neighborhood level. Crime prevention through environmental design and territorial functioning theory Ideas about crime prevention through environmental design (CPTED)-- including defensible space and related ideas about territorial functioning dating from the early 1970s--suggested that some places were safer than others partly because of how they were built and, in turn, how people used those spaces (Taylor, 1988). Some of these early views credited architecture and site planning with too strong an influence on human behavior (Taylor et al., 1980), but later works in this area have extensively documented that site features contribute to safety in many instances (Newman and Franck, 1982; Taylor et al., 1984), although not all (Merry, 1981). People in many different neighborhoods have put these ideas to work. Residents in Asylum Hill in Hartford in the 1970s (Fowler and Mangione, 1986), North Miami Shores in the early 1990s (Atlas and LeBlanc, 1994), Five Oaks in Dayton in the mid-1990s, and Guilford in Baltimore in the mid-1990s redesigned traffic circulation patterns through their neighborhoods and closed entrances to reduce crime. Some of these efforts received considerable attention from the news media, even though success has been hard to document. Across the Nation, the increasing construction of gated communities, once considered only for extremely upscale developments, represents one widespread (albeit controversial) application of these ideas. Situational crime prevention theory Situational crime prevention also focuses on specific features of settings that might contribute to crime. A steady stream of studies over the past 15 years or so has documented how specific setting features and changes in those features can deter offenders (Clarke, 1995). This model assumes that offenders are rational, are motivated by potential benefits, are aware of likely crime costs, and recognize opportunities for getting away with a crime--whether that crime is vandalizing a telephone, putting slugs in the subway, or stealing a car. Sometimes reducing opportunities means making it harder for someone to commit a crime ("hardening a target"), but other times it is more than that. A recent study of Washington, D.C.'s Metro station designs and operations represents an example of situational crime prevention that integrated a number of design, management, and operational features (La Vigne, 1996). Some are concerned that situational crime prevention strategies will simply result in displaced crime (Barnes, 1995; Repetto, 1976)--displaced spatially, temporally, or in other ways (Lab, 1992). Researchers have documented that as a result of situational crime prevention initiatives less than one crime is displaced (that is, occurs elsewhere) for each crime prevented. In some instances, adjoining areas may experience a diffusion of benefits (Clarke and Weisburd, 1994), enjoying enhanced safety only because they are near a prevention site. The debate about the volume of displacement, the quality of studies gauging such volume, and even the definition of displacement continues (Barnes, 1995). Nevertheless, most of the studies show that the effects of displacement do not nullify the benefits of prevention--and the benefits of diffusion sometimes outweigh the effects of displacement (Anderson and Pease, 1997; Green, 1995). If practitioners look at outcomes like physical deterioration and fear of crime, instead of at crime, displacement becomes a much less salient issue. As interest has shifted to these noncrime outcomes and evidence has not supported worst-case displacement scenarios, practitioners have become more comfortable with crime prevention or suppression efforts focused on small-scale locations. Behavioral geography/crime pattern theory According to this theory, offenders go to work, visit friends, come home, do their shopping, and carry out other daily activities just like the rest of us. During these activities, motivated offenders search for likely targets for the type of crime they hope to commit. For example, suburban burglars look for worthwhile houses to enter that are not too far off their route between home and work (Rengert and Wasilchick, 1985). Urban, drug- using burglars may choose sites near drug markets (Rengert, 1996). Crime pattern theory integrates ideas about offenders' movements with the geographic distribution of crime targets (Eck and Weisburd, 1995). It links places with desirable targets and the context in which they are found by offenders. Geographic profiling inverts the usual behavioral geography questions (Rossmo, 1995). Instead of asking, "If we know where the offender works and lives, can we predict what targets he or she will select?" it asks, "If we know the locations of a connected series of crimes, what can we say about where the offender most likely resides or works?" D. Kim Rossmo (1995, p. 217, and abstract) explains: The probable spatial behavior of the offender can thus be derived from information contained in the known crime-site locations, their geographic connections, and the characteristics and demography of the surrounding areas. By determining the probability of the offender residing in various areas and displaying those results through the use of isopleth or choropleth maps, police efforts to apprehend criminals can be assisted. This investigative approach is known as geographic profiling. The technique's creator believes it will assist police most substantially in investigations of serial murder, serial rape, and serial arson. It also may prove relevant to other serial crimes for which a set of spatially defined clues is available. Routine activity theory Routine activity theory and the affiliated lifestyle theory (Hindelang et al., 1978; Hindelang et al., 1979; Titus, 1995) relate closely to behavioral geography. In their earlier versions, these theories proposed that crimes occur when three things come together in space and time: a motivated offender, a suitable target, and the absence of a capable guardian (Felson, 1994).[7] The perspective helps us understand why certain crime rates are higher around high schools and taverns, as Dennis Roncek and his colleagues have demonstrated. It also helps explain long-term shifts in some crime rates (Felson, 1986) and differences in victimization rates among groups of people (Kennedy and Forde, 1990a; Kennedy and Forde, 1990b). More recently, theorists have added three more pieces to the model (Felson, 1995). Motivated offenders have "handlers": Relatives, friends, or acquaintances who, without resorting to force, can discourage the motivated offender from committing the crime. Furthermore, in addition to guardians, who are focused on protecting a particular target, settings often have "place managers" who discourage crime by controlling places. A doorman on Fifth Avenue in front of an apartment building, a bus driver on a bus, a private security guard looking out over a parking lot are all looking after a particular place. Their spheres of concern are spatially broader than those of the guardian. Place managers may be differentiated based on the type of responsibility each has for the place in question. The stronger the responsibility, the more likely the place manager will do something about a crime about to happen or that has already taken place. With the introduction of the place manager, routine activity theory develops into a perspective clearly focused on small-scale locations. It tells us about differences across sites, not across communities, over time (Eck, 1995). According to John Eck, "Specific places should be a focus of research . . . for small time increments" (Eck, 1995, pp. 795-796). The theory itself has progressed down the cone of resolution. Notes 1. The current focus on small-scale places may be a continuation of a longer term trend. See the appendix for a brief historical analysis of research on crime and its relationship to place. 2. Criminal geographers differ on the specific levels they might discuss within this cone. Keith Harries (1974) includes the following levels: regional, intermetropolitan, macro-intra urban, and micro-intra urban. Paul and Patricia Brantingham (1981) discuss State, tract, census block group, and census block levels, in accord with easily available census data. 3. These have come to be known as Compstat (computer comparison statistics) meetings. But an exclusive focus on the crime maps overlooks the broader command and control processes, as well as the strategizing, that occur in the meetings. 4. This study did not examine the possibility that crime was displaced to a remote location. For example, in the fall of 1997, residents of Kensington, a neighborhood considerably north and east of downtown Philadelphia, expressed concern about the dramatic rise in prostitution in their locale. Many of those working the street had reportedly been displaced from the downtown area as a result of Business Improvement District (BID) activities. 5. To be selected, sites were required to have at least 20 hard- and 20 soft- crime calls in the selection year. Hard- and soft-crime calls in the experimental and treatment sites ranged from 56 to 628 (with a mean of 183). 6. The time period used for compiling the burglary history record is not clear. 7. Questions of what makes a guardian capable (and in whose eyes) remain to be explored. References Anderson, D., and Pease, K. 1997. "Biting Back: From Kirkholt to Huddersfield--Preventing Repeat Victimization." In Situational Crime Prevention: Successful Case Studies, ed. R.V. Clarke. Albany, New York: Harrow and Heston. Appleyard, D. 1981. Livable Streets. Berkeley, California: University of California Press. Atlas, R., and LeBlanc, W. 1994. "Environmental Barriers to Crime." Ergonomics in Design (October): 9-16. Barker, R.G. 1968. Ecological Psychology. Stanford, California: Stanford University Press. Barker, R.G., Dembo, T., and Lewin, K. 1943. "Frustration and Regression." In Child Behavior and Development, ed. R.G. Barker, J.S. Kounin, and H.F. Wright. New York, New York: McGraw-Hill. Barnes, G.S. 1995. "Defining and Optimizing Displacement." In Crime and Place, ed. J.E. Eck and D. Weisburd, pp. 95-114. Monsey, New York: Criminal Justice Press. Baum, A., Davis, A.G., and Aiello, J.R. 1978. "Crowding and Neighborhood Mediation of Urban Density." Journal of Population 1: 266- 279. Bennett, T.H. 1995. "Identifying, Explaining, and Targeting Burglary 'Hot Spots.' " European Journal of Criminal Policy and Research 3: 113-123. Block, C.R. 1993. "STAC Hot Spot Areas: A Statistical Tool for Law Enforcement Decisions." Paper presented at the annual meeting on Situational Crime Prevention, June, Miami, Florida. Block, C.R. 1996. The GeoArchive: An Information Foundation for Community Policing. Statistical Analysis Center. Chicago: Illinois Criminal Justice Information Authority. Unpublished paper. Block, R.L., and Block, C.R. 1995. "Space, Place, and Crime: Hot Spot Areas and Hot Places of Liquor-related Crime." In Crime and Place, ed. J.E. Eck and D. Weisburd, pp. 145-183. Monsey, New York: Criminal Justice Press. Bourgois, P. 1989. "In Search of Horatio Alger: Culture and Ideology in the Crack Economy." Contemporary Drug Problems 16: 619-650. Bourgois, P. 1996. In Search of Respect. Cambridge, England: Cambridge University Press. Braga, A. 1997. Telephone conversation with author, November 5. Braga, A., Weisburd, D.L., Waring, E.L., Mazerolle, L.G., Spelman, W., Gajewski, F. 1997. "Problem Solving in Violent Crime Places: A Randomized Controlled Experiment." Paper presented at the annual meeting of the American Society of Criminology, November, San Diego, California. Brantingham, P., and Brantingham, P. 1981. "Notes on the Geometry of Crime." In Environmental Criminology, ed. P. Brantingham and P. Brantingham. Beverly Hills, California: Sage Publications. Brantingham, P., Dyreson, D.A., and Brantingham, P. 1976. "Crime Seen Through a Cone of Resolution." American Behavioral Scientist 20: 261- 274. Buckley, J., and Kane, R. 1997. "Using Ecological Theories of Crime to Guide Parole Release Decisions: A Proposed Model." Paper presented at the annual meeting of the American Society of Criminology, November, San Diego, California. Buerger, M.E., Cohn, E.G., and Petrosino, A.J. 1995. "Defining the 'Hot Spots of Crime': Operationalizing Theoretical Concepts for Field Research." In Crime and Place, ed. J.E. Eck and D. Weisburd, pp. 237- 257. Monsey, New York: Criminal Justice Press. Bursik, R.J. 1988. "Social Disorganization and Theories of Crime and Delinquency." Criminology 26: 519-551. Bursik, R.J., Jr. and Grasmick, H.G. 1993. Neighborhoods and Crime: The Dimensions of Effective Social Control. New York, New York: Lexington Books. Buzawa, E.S., and Buzawa, C.G. 1990. Domestic Violence. Newbury Park, California: Sage Publications. Chicago Community Policing Evaluation Consortium. 1996. Community Policing in Chicago, Year 3. Chicago, Illinois: Illinois Criminal Justice Information Authority. Clarke, R.V. 1995. "Situational Crime Prevention." In Building a Safer Society: Strategic Approaches to Crime Prevention, ed. M. Tonry and D. Farrington, pp. 91-150. Chicago, Illinois: University of Chicago Press. Clarke, R.V., and Weisburd, D. 1994. "Diffusion of Crime Control Benefits: Observations on the Reverse of Displacement." In Crime Prevention Studies, vol. 2, ed. R.V. Clarke, pp. 165-184. Monsey, New York: Criminal Justice Press. Cohen, D., and Nisbett, R.E. 1994. "Self-protection and the Culture of Honor: Explaining Southern Violence." Personality and Social Psychology Bulletin 20: 551-567. Craik, K.H., and Appleyard, D. 1980. "Streets of San Francisco: Brunswik's Lens Model Applied to Urban Inference and Assessment." Journal of Social Issues 36: 72-85. Eck, J. 1995. "Review Essay: Examining Routine Activity Theory." Justice Quarterly 12: 783-797. Eck, J.E., and Wartell, J. 1998. "Improving the Management of Rental Properties with Drug Problems: A Randomized Experiment." In Civil Remedies, ed. L. Mazerolle and J. Roehl. Monsey, New York: Criminal Justice Press. Eck, J.E., and Weisburd, D. 1995. "Crime Places in Crime Theory." In Crime and Place (Crime Prevention Studies, Volume 4), ed. J.E. Eck and D. Weisburd, pp. 1-33. Monsey, New York: Criminal Justice Press. Ellingworth, D., Farrell, G., and Pease, K. 1995. "A Victim Is a Victim Is a Victim?" British Journal of Criminology 35: 327-399. Fagan, J., and Davies, G. 1997. "Crime in Public Housing: Two-way Diffusion Effects in Surrounding Neighborhoods." Paper presented at Workshop on Identification and Evaluation of Methods for Measuring and Analyzing Crime Patterns and Trends with GIS, City University of New York. February. Faris, R. 1948. Social Disorganization. New York, New York: Ronald Press. Farrell, G. 1995. "Preventing Repeat Victimization." In Building a Safer Society: Strategic Approaches to Crime Prevention, ed. M. Tonry and D. Farrington, pp. 469-534. Chicago, Illinois: University of Chicago Press. Felson, M. 1986. "Linking Criminal Choices, Routine Activities, Informal Control, and Criminal Outcomes." In The Reasoning Criminal, ed. D. Cornish and R. Clarke, pp. 119-128. New York, New York: Springer- Verlag. Felson, M. 1994. Crime in Everyday Life. Thousand Oaks, California: Pine Forge Press. Felson, M. 1995. "Those Who Discourage Crime." In Crime and Place, vol. 4, ed. J.E. Eck and D. Weisburd, pp. 53-66. Monsey, New York: Criminal Justice Press. Fisher, B., and Nasar, J.L. 1994. "Defining the Concept of Hot Spots of Fear: A Multilevel Study." Journal of Environmental Psychology 13: 187- 206. Fisher, B., and Nasar, J. 1995. "Fear Spots in Relation to Microlevel Physical Cues: Exploring the Overlooked." Journal of Research in Crime and Delinquency 32: 214-239. Fowler, F.J., and Mangione, T. 1986. "A Three-pronged Effort to Reduce Crime and Fear of Crime: The Hartford Experiment." In Community Crime Prevention, ed. D. Rosenbaum, pp. 87-108. Newbury Park, California: Sage Publications. Fox, K.A. 1983. "The Eco-behavioural View of Human Societies and Its Implications for Systems Science." International Journal of Systems Science 14: 895-914. Fox, K.A. 1984a. "Behavior Settings and Eco-behavioral Science: A New Arena for Mathematical Social Science Permitting a Richer and More Coherent View of Human Activities in Social Systems, Part I: Concepts, Measurements, and Linkages to Economic Data Systems, Time-allocation Matrices, and Social System Accounts." Mathematical Social Sciences 7: 117-138. Fox, K.A. 1984b. "Behavioral Settings and Eco-behavioral Science: A New Arena for Mathematical Social Science Permitting a Richer and More Coherent View of Human Activities in Social Systems, Part II: Relationships to Established Disciplines and Needs for Mathematical Development." Mathematical Social Sciences 7: 139-165. Giacomazzi, A.L., McGarrell, E.F., and Thurman, Q.C. n.d. "Reducing Disorder, Fear and Crime in Public Housing: An Evaluation of a Drug Crime Elimination Program in Spokane, Washington." Draft final report. Forthcoming. Washington State University, Spokane, Washington. National Institute of Justice grant number 93-IJ-CX-0054. Giggs, J.A. 1990. "Drug Abuse and Urban Ecological Structure: The Nottingham Case." In Spatial Epidemiology, ed. R.W. Thomas, pp. 218- 239. London, England: Pion. Glade, J. 1856. "Localities of Crime in Suffolk." Journal of Statistical Society of London 19: 102-106. Green, L. 1995. "Cleaning Up Drug Hot Spots in Oakland, California: The Displacement and Diffusion Effects." Justice Quarterly 12: 737-754. Green, L. 1996. Policing Places with Drug Problems. Thousand Oaks, California: Sage Publications. Hannan, M.T. 1971. Aggregation and Disaggregation in Sociology. Lexington, Massachusetts: Lexington Books. Harries, K. 1974. The Geography of Crime and Justice. New York, New York: McGraw-Hill. Hindelang, M.J., Gottfredson, M.R., and Garofalo, J. 1978. Victims of Personal Crime: An Empirical Foundation for a Theory of Personal Victimization. Cambridge, Massachusetts: Ballinger. Hindelang, M.J., Gottfredson, M., and Garofalo, J. 1979. A Theory of Personal Victimization. Cambridge, Massachusetts: Ballinger. Hope, T. 1995. "Community Crime Prevention." In Building a Safer Society, ed. M. Tonry and D. Farrington, pp. 21-90. Chicago, Illinois: University of Chicago Press. Jacobs, J. 1961. The Death and Life of Great American Cities. New York, New York: Random House. Jacobs, J. 1968. "Community on the City Streets." In The Search for Community in Modern America, ed. E.D. Baltzell, pp. 74-93. New York, New York: Harper and Row. Kelling, G.L., and Coles, C.M. 1996. Fixing Broken Windows: Restoring Order and Reducing Crime in Our Communities. New York, New York: Free Press. Kennedy, L.W., and Forde, D.R. 1990a. "Risky Lifestyles and Dangerous Results: Routine Activities and Exposure to Crime." Sociology and Social Research 74: 208-211. Kennedy, L.W., and Forde, D.R. 1990b. "Routine Activities and Crime: An Analysis of Victimization in Canada." Criminology 28: 137-151. Knutsson, J. 1997. "Restoring Public Order in a City Park." In Crime Prevention Studies, ed. R. Homel, pp. 133-151. Monsey, New York: Criminal Justice Press. Koper, C.S. 1995. "Just Enough Police Presence: Reducing Crime and Disorderly Behavior by Optimizing Police Patrol in Crime Hot Spots." Justice Quarterly 12: 649-672. Lab, S.P. 1992. Crime Prevention: Approaches, Practices and Evaluations. 2d ed. Cincinnati, Ohio: Anderson. Lauritsen, J.L., and Quinet, K.F.D. 1995. "Repeat Victimization Among Adolescents and Young Adults," Journal of Quantitative Criminology 11: 143-166. Lauritsen, J.L., Sampson, R.J., and Laub, J.H. 1991. "The Link Between Offending and Victimization Among Adolescents." Criminology 29: 265- 291. La Vigne, N. 1996. "Safe Transport: Security by Design on the Washington Metro." In Preventing Mass Transit Crime (Crime Prevention Studies), vol. 6, ed. R.V. Clarke, pp. 163-197. Monsey, New York: Criminal Justice Press. La Vigne, N. 1997. "How Hot Is That Spot?: The Utility and Application of Place-based Theories of Crime." Paper presented at the International Situational Crime Prevention Conference, June, Oslo, Norway. Lurigio, A.J., Davis, R.C., Regulus, T.A., Popkin, S.A., Dantzken, M.L., Smith, B., and Ouellet, L. 1998. "More Effective Place Management: An Evaluation of Cook County's Narcotics Nuisance Abatement Unit." In Civil Remedies, ed. L. Mazerolle and J. Roehl. Monsey, New York: Criminal Justice Press. Maltz, M., Gordon, A.C., and Friedman, W. 1991. Mapping Crime in Its Community Setting. New York, New York: Springer-Verlag. Mann, P.H. 1954. "The Concept of Neighborliness." American Journal of Sociology 60: 163-168. Mazerolle, L.G. and Terrill, W. 1997. "Problem-oriented Policing in Public Housing: Identifying the Distribution of Problem Places." Policing 20: 235-255. Mazerolle, L.G., Kadleck, K., and Roehl, J. 1997. "Controlling Drug and Disorder Problems: The Role of Place Managers," Unpublished paper. Cincinnati, Ohio: University of Cincinnati. Mazerolle, L.G., Belluci, C., and Gajewski, F. in press. "Crime Mapping in Police Departments: The Challenges of Building a Mapping System." In Crime Mapping, ed. T. McEwen and D. Weisburd. Monsey, New York: Criminal Justice Press. McEwen, J.T., and Taxman, F.S. 1995. "Applications of Computer Mapping to Police Operations." In Crime and Place, vol. 4, ed. J.E. Eck and D. Weisburd, pp. 259-284. Monsey, New York: Criminal Justice Press. McGuire, P. 1997. Telephone conversations with author, September. Document faxed, September 25. Merry, S.E. 1981. "Defensible Space Undefended: Social Factors in Crime Control Through Environmental Design." Urban Affairs Quarterly 16: 397-422. Messner, S.F. 1983. "Regional and Racial Effects on the Urban Homicide Rate: The Subculture of Violence Revisited." American Journal of Sociology 88: 997-1007. Morris, T. 1957. The Criminal Area: A Study in Social Ecology. London, England: Routledge and Kegan Paul. Newman, O., and Franck, K. 1982. "The Effects of Building Size on Personal Crime and Fear of Crime." Population and Environment 5: 203- 220. Nisbett, R.E. 1993. "Violence and U.S. Regional Culture." American Psychologist 48: 441-449. Pease, K. 1991. "The Kirkholt Project: Preventing Burglary on a British Public Housing Estate." Security Journal 2: 73-77. Perkins, D.D., Florin, P., Rich, R.C., Wandersman, A., and Chavis, D.M. 1990. "Participation and the Social and Physical Environment of Residential Blocks: Crime and Community Context." American Journal of Community Psychology 18: 83-115. Perkins, D.D., Meeks, J.W., and Taylor, R.B. 1992. "The Physical Environment of Street Blocks and Resident Perceptions of Crime and Disorder: Implications for Theory and Measurement." Journal of Environmental Psychology 12: 21-34. Polvi, N., Looman, T., Humphries, C., and Pease, K. 1991. "The Time- course of Repeat Burglary Victimization." British Journal of Criminology 31: 411-414. Popkin, S., Gwiasda, V.E., Amendiola, J.M., Anderson, A.A., Hanson, G., Johnson, W.A., Martel, E., Olson, L.M., and Rosenbaum, D.P. 1997. "The Hidden War: The Battle to Control Crime in Chicago's Public Housing." Draft final report: 93-IJ-CX-0037; 95-IJ-CX-0011). Chicago, Illinois: Abt Associates Inc. Reiss, A.J., Jr. 1986. "Why Are Communities Important in Understanding Crime?" In Communities and Crime, vol. 8, ed. A.J. Reiss, Jr., and M. Tonry, pp. 1-34. Chicago, Illinois: University of Chicago Press. Rengert, G. 1996. The Geography of Illegal Drugs. Boulder, Colorado: Westview. Rengert, G., and Wasilchick, J. 1985. Suburban Burglary. Springfield, Illinois: Charles C. Thomas. Repetto, T.A. 1976. "Crime Prevention and the Displacement Phenomenon." Crime and Delinquency 22: 166-177. Roncek, D.W. 1981. "Dangerous Places: Crime and Residential Environment." Social Forces 60: 74-96. Roncek, D.W., and Bell, R. 1981. "Bars, Blocks and Crime." Journal of Environmental Systems 11: 35-47. Roncek, D.W., and Faggiani, D. 1985. "High Schools and Crime: A Replication." The Sociological Quarterly 26: 491-505. Roncek, D., and Maier, P.A. 1991. "Bars, Blocks, and Crimes Revisited: Linking the Theory of Routine Activities to the Empiricism of 'Hot Spots.' " Criminology 29: 725-753. Roncek, D.W., and Pravatiner, M.A. 1989. "Additional Evidence That Taverns Enhance Nearby Crime." Sociology and Social Research 73: 185- 188. Roncek, D.W., Bell, R., and Francik, J.M.A. 1981. "Housing Projects and Crime: Testing a Proximity Hypothesis." Social Problems 29: 151-166. Rosenbaum, D.P., and Lavrakas, P.J. 1995. "Self-reports About Place: The Application of Survey and Interview Methods to the Study of Small Areas." In Crime and Place, ed. J.E. Eck and D. Weisburd, pp. 285-314. Monsey, New York: Criminal Justice Press. Rossmo, D.K. 1995. "Place, Space, and Police Investigations: Hunting Serial Violent Criminals." In Crime and Place, ed. J.E. Eck and D. Weisburd, pp. 217-235. Monsey, New York: Criminal Justice Press. Sampson, R.J., and Grove, W.B. 1989. "Community Structure and Crime: Testing Social Disorganization Theory." American Journal of Sociology 94: 774-802. Sampson, R.J., and Lauritsen, J.L. 1994. "Violent Victimization and Offending: Individual, Situational, and Community-level Risk Factors." In Understanding and Preventing Violence (Volume 3: Social Influences), ed. A.J.J. Reiss and J.A. Roth, pp. 1-114. Washington, D.C.: National Academy Press. Sampson, R.J., Raudenbush, S.W., and Earls, F. 1997 "Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy." Science 277: 918-925. Shaw, C.R., and McKay, H.D. 1969. Juvenile Delinquency and Urban Areas, 2d ed. Chicago, Illinois: University of Chicago Press. Sherman, L.W. 1992. Policing Domestic Violence. New York, New York: Free Press. Sherman, L.W. 1995. "Hot Spots of Crime and Criminal Careers of Places." In Crime and Place, eds. J.E. Eck and D. Weisburd, pp. 35-52. Monsey, New York: Criminal Justice Press. Sherman, L.W., and Rogan, D.P. 1995. "Deterrent Effects of Police Raids on Crack Houses: A Randomized, Controlled Experiment." Justice Quarterly 12: 755-781. Sherman, L.W., and Weisburd, D. 1995. "General Deterrent Effects of Police Patrol in Crime 'Hot Spots': A Randomized, Controlled Trial." Justice Quarterly 12: 625-649. Sherman, L.W., Gartin, P.R., and Buerger, M.E. 1989. "Hot Spots of Predatory Crime: Routine Activities and the Criminology of Place." Criminology 27: 27-56. Snyder, B. 1995. "Crime and Grime Patrols May Not Start Until March." The Baltimore Sun January 6, 1995, B2. Spelman, W. 1995a. "Criminal Careers of Public Places." In Crime and Place, ed. J.E. Eck and D. Weisburd, pp. 115-144. Monsey, New York: Criminal Justice Press. Spelman, W. 1995b. "Once Bitten, Then What: Cross-sectional and Time- course Explanations of Repeat Victimization." British Journal of Criminology 35: 366-383. Sullivan, M. 1989. Getting Paid: Youth Crime and Work in the Inner City. Ithaca, New York: Cornell University Press. Sullivan, M.L. 1993. "Puerto Ricans in Sunset Park, Brooklyn: Poverty Amidst Ethnic and Economic Diversity." in In the Barrios: Latinos and the Underclass Debate, ed. J. Moore and R. Pinderhughes, pp. 1-25. New York, New York: Russell Sage Foundation. Taylor, R.B. 1988. Human Territorial Functioning. Cambridge, England: Cambridge University Press. Taylor, R.B. 1996. "Impacts of Crime and Grime on Reactions to Crime, Neighborhood Crime Changes, and Changes in Neighborhood Structure." Draft final report (93-IJ-CX-0022, National Institute of Justice). Philadelphia, Pennsylvania: Department of Criminal Justice, Temple University. Taylor, R.B. 1997. "Social Order and Disorder of Streetblocks and Neighborhoods: Ecology, Microecology and the Systemic Model of Social Disorganization." Journal of Research in Crime and Delinquency 33: 113- 155. Taylor, R.B., Gottfredson, S.D., and Brower, S.N. 1980. "The Defensibility of Defensible Space." In Understanding Crime, ed. T. Hirschi and M. Gottfredson, pp. 53-71. Beverly Hills, California: Sage Publications. Taylor, R.B., Gottfredson, S.D., and Brower, S. 1981. "Territorial Cognitions and Social Climate in Urban Neighborhoods." Basic and Applied Social Psychology 2: 289-303. Taylor, R.B., Gottfredson, S.D., and Brower, S. 1984. "Understanding Block Crime and Fear." Journal of Research in Crime and Delinquency 21: 303-331. Taylor, R.B., Koons, B., Kurtz, E., Greene, J., and Perkins, D. 1995. "Streetblocks with More Nonresidential Landuse Have More Physical Deterioration: Evidence from Baltimore and Philadelphia." Urban Affairs Review. 30: 120-136. Thomas, R.W., ed. 1990a. Spatial Epidemiology. London, England: Pion. Thomas, R.W. 1990b. "Introduction: Issues in Spatial Epidemiology." In Spatial Epidemiology, ed. R.W. Thomas, pp. 1-14. London, England: Pion. Thomas, R.W. 1992. Geomedical Systems. New York, New York: Routledge. Titus, R.M. 1995. "Activity Theory and the Victim." European Journal of Criminal Policy and Research 3 (Autumn): 41-55. Van Dine, S. 1997. "It's a Beautiful Day in the Neighborhood." Remarks delivered at National Institute of Justice Crime Mapping Research Center Planning Conference. Washington, D.C. February. Wartenberg, D., and Greenberg, M. 1990. "Space-time Models for the Detection of Clusters of Disease." In Spatial Epidemiology, ed. R.W. Thomas, pp. 17-34. London, England: Pion. Weisburd, D. 1996. "What To Do About Crime." 1996. Paper presented at the National Institute of Justice Conference on Criminal Justice Research and Evaluation. Washington, D.C. Weisburd, D. 1997. Reorienting Crime Prevention Research and Policy: From the Causes of Criminality to the Context of Crime, Research Report. Washington, D.C.: U.S. Department of Justice, National Institute of Justice, NCJ 165041. Weisburd, D., and Green, L. 1995. "Policing Drug Hot Spots: The Jersey City Market Analysis Experiment." Justice Quarterly 12: 711-736. Weisburd, D., and Green, L. 1996. "Defining the Street-level Drug Market." In Drugs and Crime: Evaluating Policy Alternatives, ed. D.L. McKenzie and C.D. Uchida, pp. 61-76. Thousand Oaks, California: Sage Publications. White, R.C. 1932. "The Relation of Felonies to Environmental Factors in Indianapolis." Social Forces 46: 525-541. Wicker, A. 1972. "Processes Which Mediate Behavior-environment Congruence." Behavioral Science 17: 265-277. Wicker, A.W. 1979. Introduction to Ecological Psychology. Monterey, California: Brooks/Cole. Wicker, A.W. 1987. "Behavior Settings Reconsidered: Temporal Stages, Resources, Internal Dynamics and Context." In Handbook of Environmental Psychology, ed. D. Stokols and I. Altman, New York, New York: Wiley. Williams, T. 1989. The Cocaine Kids. Reading, Massachusetts: Addison- Wesley. Wilson, W.J. 1987. The Truly Disadvantaged. Chicago, Illinois: University of Chicago Press. Wilson, W.J. 1996. When Work Disappears: The World of the New Urban Poor. New York, New York: Knopf. Wolfgang, M., and Ferracuti, F. 1967. The Subculture of Violence. London, England: Tavistock. ------------------------------- Evaluating Community Youth Sanctioning Models: Neighborhood Dimensions and Beyond Gordon Bazemore, Ph.D., Florida Atlantic University "Community justice" is taking hold throughout North America and elsewhere in the world. Although the application of community sanctioning models with adult offenders is increasing, nowhere has the interest in these new approaches been greater than in the response to youth crime.[1] (See exhibit 1 for descriptions of four community youth sanctioning models.) While it is difficult to track each local application, at least 10 States are experimenting with, or in some cases reviving, community sanctioning programs for young offenders. Viewed as a whole, the movement toward community-based youth sanctioning and decisionmaking encompasses a wide array of processes, goals, and practical and philosophical rationales. The enthusiasm for community sanctioning as a new response to juvenile crime appears to be a result of both crisis and opportunity in juvenile justice. Advocates of these approaches view them as an opportunity to build support for a juvenile justice system frequent-ly criticized as insular and out of touch with community needs and expectations (Bazemore and Day, 1996). Stated and implicit objectives for youth community sanctioning initiatives include (but are not limited to) giving citizens a greater role in sanctioning, increasing citizen involvement in juvenile justice, providing a more meaningful and immediate response to delinquency, encouraging alternative dispute resolution, reducing fear of youth crime, improving monitoring of young offenders, and diverting more young offenders from the court. These more positive rationales are being joined with a practical and realistic sense of urgency driven by the fear that the survival of the juvenile justice system may be in jeopardy. In at least one State, it appears that one political motivation behind neighborhood sanctioning is to create a community response to misdemeanor and less serious offenses while policymakers pursue abolition of the juvenile court and transfer of jurisdiction over most felony offenses to criminal courts (Torbet, et al., 1996). In other jurisdictions such approaches are viewed as a means of strengthening community commitment and participation in what has become a closed and one-dimensional response to youth crime. While underlying philosophies in community youth sanctioning also run the gamut from traditional prevention and diversion paradigms to those emphasizing reparation and offender accountability (Umbreit, 1995; Bazemore, 1997a), to various adaptations of shaming (Kahan, 1996; Karp, 1997) (including reintegrative shaming) (Braithwaite and Mugford, 1994; Retzinger and Scheff, 1996), the four models described here (and presented in exhibit 1) share several common characteristics. As part of a growing but still emergent and loosely connected national and international movement for restorative justice (Zehr, 1990; Mesmer and Otto, 1992; Van Ness, 1993; Van Ness and Heetderks Strong, 1997; Braithwaite and Mugford, 1994), or community restorative justice (Young, 1995; Bazemore and Schiff, 1996), these approaches stand as case studies in an effort to give citizens, victims, and community groups an explicit decisionmaking role in an informal youth sanctioning process. Although an indepth description of important differences among the four models is beyond the scope of this paper, these models will be used here for the specific purpose of illustrating different trends and themes within this larger movement that have important implications for defining key dimensions of this emerging community response to youth crime. Yet, like many other initiatives in neighborhood or community justice (National Institute of Justice, 1996; Barajas, 1995), perhaps the most important shared characteristic of the new sanctioning models is the fact that relatively little is known about their impact, their objectives, or the nature of the process itself. Research in community sanctioning. Although there is a long tradition of evaluation research on community-based programs, research on community sanctioning is in its infancy. Relevant recent studies for the most part offer findings on specific applications of selected restorative justice sanctions and processes. Primarily, this research has focused on one model/process: victim-offender mediation (Umbreit and Coates, 1993; Umbreit, 1994) and one sanction: restitution (Bazemore and Schneider, 1985; Schneider, 1986; Schneider, 1990; Butts and Snyder, 1991).[4] Research describing and evaluating the new community sanctioning approaches is critical for several reasons. On the positive side, these new models represent some of the most promising approaches for changing the nature of the sanctioning function in juvenile justice (Bazemore and Umbreit, 1995). Although community sanctioning is but one component of a wide range of restorative community justice interventions that specifically addresses prevention, rehabilitation, victim services, and public safety (Young, 1995), some have suggested that citizen involvement in decisions about the disposition of juvenile offenders is an important gateway to broader and deeper participation in all aspects of the response to youth crime and greater support for juvenile justice (Braithwaite and Mugford, 1994; Bazemore and Day, 1996). On the negative side, the movement to devolve justice to the neighborhood level is fraught with dangers ranging from concerns about "net-widening" (broadening the reach of the system to take in more offenders) (Polk, 1994), to power imbalances for young offenders and adults in conferencing settings (Umbreit and Stacy, 1995), to insensitivity to victims (Braithwaite and Parker, 1998), to the "tyranny of community" in cases where community dynamics have resulted in a variety of abuses (Griffiths et al, 1995). Need for an evaluation protocol. At this stage the primary obstacle to meaningful evaluation of these approaches is neither technical nor methodological.[5] Rather, a first step is to become clear about what the new models are trying to accomplish and how they are trying to accomplish it. An evaluation protocol is needed to define the intended outcomes in community youth sanctioning, describe intervention inputs, and provide theoretical rationales that link inputs, sanctioning processes, and outputs. Toward this end, this paper is limited to the "independent variable" in community youth sanctioning; i.e., those inputs or interventions expected to lead to the desired outcomes. Currently, there is little knowledge about how the four models work in practice and the principles that guide the informal decisionmaking processes employed. Although it is important to also articulate the intended outcomes of these and other similar approaches, the primary purpose of this paper is to suggest theoretical and practical ways in which the sanctioning process may vary. These variations can help to define the research questions and propositions most helpful in assessing the integrity of community youth sanctioning interventions. This, however, departs somewhat from the primary theme of this symposium: the role of the neighborhood as specific locus or place for justice processes to take place. In keeping with the theme, the neighborhood concept is used to define one broader dimension of accessibility in community youth sanctioning. However, although the community youth sanctioning models share a commitment to making justice processes more accessible to those most affected by crime, even cursory observation of these processes will confirm that much more is going on than the fact that they are occurring in neighborhoods rather than in courts. Therefore, I will describe two additional categories that help to define theoretical dimensions in community youth sanctioning-- community justice and restorative justice.[6] Defining dimensions of variation Neighborhood sanctioning boards or community panels are nothing new and, in some ways, are not unique. Indeed, such programs have a long history in the United States and the rest of North America.[7] Clearly, the four models in the case examples presented here share with both current and earlier community-based criminal justice approaches an emphasis on bringing a more accessible and "user-friendly" justice process to the neighborhood (National Institute of Justice, 1996). What distinguishes this new crop of community sanctioning models from most current efforts to decentralize courts and court functions--as well as from traditional sanctioning and neighborhood dispute resolution initiatives (McGillis and Mullen, 1977; Garafalo and Connelly, 1980)--are several important themes and a primary unifying focus. This focus moves beyond the concern with neighborhood alone and suggests that this dimension of influence may, in fact, be far less relevant in some parts of the country than others. But while "neighborhood" may not be relevant to young offenders, crime victims, and other citizens participating in community sanctioning processes, "community" in one form or another and relationships will be. These concepts suggest a range of focal concerns related to restorative justice and community justice, including an effort to identify and engage those most affected by crime in problem-oriented and preventive solutions; changing the role of justice agencies to one of strengthening community capacity against crime; a commitment to repairing harm to victims and victimized communities; and actively involving victims, other citizens, and offenders in an informal, nonadversarial dispositional or decisionmaking process aimed in part at holding offenders personally accountable (Bazemore, 1997b; Stuart, 1995a). The integrity of interventions At this early stage in the development and evaluation of community youth sanctioning models, it is critical to establish theoretical and value-based criteria for answering the question, how do we know it when we see it? That is, how do evaluators know that a process referred to as a family group conference, for example, has been carried out in a way consistent with the principles of restorative justice or reintegrative shaming that inform it? Can evaluators know--when an offender recidivates 2 months after participating in such a conference or when the victim feels dissatisfied--that the theory of intervention underlying this approach was incorrect or inappropriate? Or was this apparent failure due to the fact that the conference was inadequately implemented? The first concern with determining whether an intervention has actually occurred is a construct validity issue of the utmost importance in evaluating any new initiative. The possibility of multiple interpretations of terms like "mediation," "shaming," and even "involvement" leaves new processes open to the phenomenon in which the name of very traditional practices is simply changed to fit new trends (e.g., community policing, restorative justice), with little or no actual change in content. To avoid situations in which relabeled traditional practices are evaluated as restorative justice or community justice programs, it is important to establish definitions and criteria for determining whether and to what extent an intervention has occurred. More commonly, because consistency with underlying principles in processes such as mediation often varies along a continuum (Umbreit, n.d.), it is important in evaluation to have access to common criteria that allow for valid and reliable assessment of intervention integrity. Such criteria would reflect one or more dimensions of practical and theoretical importance to practitioners and participants in implementing a youth sanctioning model, and they could be used to compare different implementations of the same or contrasting models. Ideally, multiple dimensions could be grouped into theoretically meaningful categories based on underlying propositions about why each dimension is important. Such a classification also makes possible the development of independent variable categories associated with specific theories of intervention or logical propositions. These theories as propositions would be based on assumptions about the importance of such characteristics as where the intervention occurs, who is involved, participant roles, the nature of the process, and the importance attached to crime victim needs and reparation. Ultimately, such variable sets might allow for comparisons between, for example, the relative importance of a neighborhood location and active citizen involvement in sanctioning processes, or between an emphasis on victim input and a focus on participation of other stakeholders. For purposes of this discussion, what could be a wide range of dimensions for classifying the inputs of community justice decisionmaking is divided into three conceptual categories. First, several dimensions of variation that view the neighborhood as the locus of decisionmaking and reflect concerns related to place, proximity, and availability of justice services are classified here under the larger conceptual category of accessibility. Second, several characteristics referred to as community justice dimensions (Barajas, 1995) relate primarily to the role of the community as both the object of and coparticipant in intervention. Finally, the category of restorative justice refers to dimensions focused on reparation of harm, victim involvement and victim needs, and a process that defines an active role for victim, community, and offender.[8] Although overlap between these categories in practice makes clear division of specific community sanctioning interventions into these categories impossible, each category adds theoretically and practically important evaluation criteria that can be used in defining the inputs or independent variables in community sanctioning. In describing these dimensions, family group conferencing, community boards, circle sentencing, and victim-offender mediation are used as ideal types to illustrate the general range of variation in community sanctioning inputs. As the case examples (see exhibit 2) imply, there is substantial variation among the four youth sanctioning models in how the decisionmaking process is organized, the primary goals of participants, how and to what extent they prioritize efforts to involve citizens and build community ownership, and the role assigned to victims, community members, and offenders. Making sense of these differences as "inputs" into the sanctioning process can help evaluators assess the consistency and comprehensiveness of the implementation of various approaches. Beyond their relevance in these evolving models, such dimensions should prove useful in comparing other community sanctioning processes. Accessibility dimensions: Neighborhood location, flexibility, and informality Historically, the most common and predominant concept in the discussion and practice of community justice has been accessibility to justice services. In this regard, the new youth sanctioning models can be compared on the basis of three primary dimensions that address the extent to which justice processes and services are easily available and likely to be used by neighborhood residents who represent target users. These dimensions are labeled location/proximity, flexibility, and informality. Location/proximity. The assumption underlying concern with location-- distance as a primary barrier to participation in and satisfaction with justice services--has at various periods in recent history prompted a movement to physically decentralize justice services--often with a special focus on inner-city neighborhoods thought to need them most. Initially a primary motivator for community-based corrections, team and neighborhood policing, neighborhood dispute resolution, and foot patrol, this theme is also heard in the community sanctioning movement. Assessing variation in proximity as an independent variable involves measuring to what extent distance varies for individuals involved in a community sanctioning process and to what extent this affects overall satisfaction and other long- and short-term outcomes. Researchers should also examine whether residents are indeed more likely to participate in sanctioning processes in their own neighborhood and the extent to which lack of transportation may be an obstacle. At the program level, evaluators may wish to compare outcomes (such as client satisfaction) of programs held in neighborhoods with those held in more centralized settings. They may also wish to examine the extent to which neighborhood sanctioning programs vary in the average distance participants must travel to access justice services and processes. Turning to the four youth sanctioning models, it is clear that each takes place in neighborhood locations such as community centers, schools, churches, and other informal settings (see exhibit 3). Actual proximity to most users is unknown and has not been a topic of discussion or investigation in existing evaluations; it appears that in practice some models like victim-offender mediation (VOM) seem less concerned with neighborhood location than others. Mediation sessions may even be held in centralized locations (near courthouses, for example) to attract more clients or users, or in the victim's home if the offender agrees. While advocates of all models prefer holding youth sanctioning events close to participants, it is only in community boards that neighborhood setting appears to be a key concern. Flexibility. The flexibility dimension refers to efforts to adapt working hours, staffing, and services to the needs of neighborhood residents (Rottman, 1996; Stone, 1996). Decentralization may also be a component of this dimension (Clear, 1996), since an objective of such efforts may be (but is not necessarily) to allow and encourage managers to reorganize agencies to avoid multiple units, narrow specializations, and compartmentalization. Because these features add to citizens' difficulties in understanding (and thus accessing) the justice process and also limit the capacity of professionals to respond to the diverse needs of communities, this aspect frequently has been mentioned by advocates of community sanctioning (Stuart, 1995b; McElrae, 1993), as well as by more mainstream advocates of juvenile court reform (Edwards, 1996). One of the most theoretically interesting aspects of this dimension is the ability of justice professionals to adapt their professional roles to various and often diverse community needs outside their job description. Community policing initially was the prototype for this "role stretch" or "role blurring" as law enforcement officers became community police officers and found themselves with a new rationale for accepting an assignment that included service provision, prevention, capacity building, and problem solving (Stephens and Moore, 1988). The degree of flexibility in community youth sanctioning processes could be assessed by examining the roles of professional and paraprofessional staff and using client surveys and interviews. The latter would seek to determine if users perceive the process as rigid and narrow or if staff are viewed as willing to meet diverse local needs, rather than be bound by job descriptions and departmental protocols. Variation in flexibility among the four models seems again relatively minimal. But while practitioners of each model attempt to adapt the process somewhat to local needs, role stretch or role blurring is apparent primarily in circle sentencing, where police officers, judges, and prosecutors play the role of citizen participant in the circle process after fulfilling their legal responsibility (e.g., reading the charges) (Bazemore, 1997b). In two models, VOM and CB (community boards), justice system professionals are involved in only the most distant sense, usually as sources of referral. In family group conferencing (FGC) programs that use the Australian (or Wagga Wagga) model, a police officer is a facilitator of a sanctioning process. This role may vary depending on the extent to which the officer views his or her function as a coordinator and facilitator or as administering sanctions and leading a shaming process (Alder and Wundersitz, 1994; Umbreit and Stacy, 1995). The role of the coordinator as key staff person varies from the multifaceted one in circle sentencing to the very minor one on most community boards to the more focused role of the mediator in victim- offender mediation (see exhibit 3). Informality. Informality may also increase access by limiting formal legalistic barriers and procedures, which alienate many citizens from courts and other justice agencies. Getting away from rules and procedures, providing an array of nonlegal or paralegal services not typically provided by courts (e.g. counseling, victim support), and offering informal mediation and problem-solving options that allow citizens to speak and enhance the human and humanistic qualities of the process are all features of the informality dimension (National Institute of Justice, 1996; Zehr, 1990). The level of informality can be measured in part by the number and range of nonadversarial processes, but also by observation of how services are provided and the extent to which the process itself is rule- and procedure-driven or informal. Comparison of the formality of the four models (except most CBs, which tend to deliberate cases) reveals that decisionmaking is more or less achieved by consensus of participants (including victim and offender). In no case is there anything resembling an adversarial process. While all models are informal, process and protocol vary substantially, from ancient rituals involving passing the feather or talking stick, to the "script" of FGC, to the nondirective and facilitative approach now taught to many, if not most, mediators in VOM programs (Umbreit, 1994). The amount of time allowed for decisionmaking is also an indicator of the extent to which the process is driven by court case processing or caseload requirements (see exhibit 3). A central critique of the formal court process among VOM advocates, for example, has been the primary emphasis on speed and efficiency. Mediation practitioners insist on ensuring that time in VOM is based primarily on the needs of what are for them the primary clients of justice processes--victim and offender. Limits to improving access. The relative influence of having a justice decisionmaking or sanctioning process in one's neighborhood, the flexibility of process, and whether it is adaptable, user-friendly, and comfortable are empirical questions. For many, the various dimensions of accessibility are no doubt central, because citizens cannot easily experience justice from a distance and are unlikely to participate if the process is overly complex, unfriendly, or formal. All accessibility distinctions, however, appear to take as a given a constant flow of otherwise willing participants who are simply limited from involvement by logistical and "user-unfriendly" barriers. As the experience with community corrections indicates, however, simply locating courts and other agencies in neighborhoods, in the absence of involvement of community groups and residents, often results in isolated programs that may be said to be in but not of the community (Byrne, 1989; Clear, 1996). Breaking down formal barriers and increasing flexibility may increase citizen willingness to seek and receive assistance. But improving access does not change the primary role of citizens from service recipient to decisionmaker or partner in a justice process that provides a feeling of ownership over what services are provided and how they are delivered. It is in part this intentional effort to increase community involvement and ownership that distinguishes the accessibility dimensions from the next major category--community justice. Community justice dimensions: Role of the community Community justice has been defined as: . . . all variants of crime prevention and justice activities that explicitly include the community in their processes. Community justice is rooted in the actions that citizens, community organizations, and the criminal justice system can take to control crime and social disorder. Its central focus is community-level outcomes, shifting the emphasis from individual incidents to systemic patterns, from individual conscience to social mores, and from individual good to the common good (Karp, 1997, p. 3). Like the neighborhood dimensions, the new practice and emerging theory of community justice suggest a preference for neighborhood-based, more accessible, and less formal justice services (National Institute of Justice, 1996) that to the greatest extent possible move the locus of the justice response closest to those affected by crime. Community justice uniquely changes the role of the community in the justice process. While accessibility may reinforce the strength of community justice efforts and may even make them possible, professionals and community groups may begin to move forward with this role change even in the absence of a neighborhood location. The change in the role of the community has four important dimensions: how the community is defined for purposes of intervention; the extent to which community change, or community-building, is a focus of intervention; the extent to which citizens and community groups are active participants; and the scope of discretion granted to citizens and community groups as collaborators in decisionmaking. 1. What is the "community" in community sanctioning? The first dimension, the definition of community, has been conceptually problematic in all neighborhood- or community-based interventions. Community is often an amorphous concept that more often confuses than clarifies issues of citizen involvement in government-sponsored processes. It is possible, however, to break the community into component parts for purposes of discussing citizen involvement in public life and government intervention with citizens and community groups. Community may be defined, for example, as a school, church, workplace, city block, extended family, tribe, or support group (Gardner, 1990). How community is operationalized for purposes of community justice sanctioning--who is at the table--will have tremendous importance for all other key dimensions of community justice (Bazemore, 1997b). Assessing the scope of the community definition in a community youth sanctioning process will involve contrasting both the intent of the process and the reality of participation. The latter can be measured only by observing who is present, noting the role of the participants, and interviewing citizen participants to determine their perceptions of the experience and their influence on the process. Intent is in part determined by the structure of various models and the importance attached to citizen involvement in each (see exhibit 4). While the community may be defined de facto as anyone who attends a community sanctioning meeting, the specific definition of community in each model runs the gamut from the most restricted focus on only victim and offender in traditional VOM, to the widest definition in circle sentencing, to the rather "hand-picked" community of appointed board members present at community board hearings. In comparing VOM and FGC, for example, one may examine outcomes to determine the impact of parties on mediation (e.g., family members, friends) who are essential participants in the FGC or CS process. Or one may compare mediation, FGC, or CS sessions in which the victim was not present. 2. Community as object of intervention. As James Byrne (1989, p. 10) has noted, "offender-based control strategies are incomplete, since they take a closed system view of correctional interventions: change the offender and not the community." The second dimension of community justice is best described as the extent to which the object of intervention changes from the offender to the community and community groups. In community policing, for example, the emphasis changed from arresting individuals to working with and through community members in problem solving and capacity building (Sparrow, Moore, and Kennedy, 1990). The latter were aimed at supporting citizens in attempting to change institutions and community groups rather than seeking to apprehend, or alter the behavior of, individual offenders. Such interventions, like those now becoming more common in corrections (Dooley, 1995) and other fields, seek to strengthen the capacity of the community, community groups, and socializing institutions to control and prevent crime (Barajas, 1995; Bazemore and Schiff, 1996). In contrast to both accessibility and restorative justice dimensions, a primary aspect of this work with the community as client is preventive. Efforts based on this perspective focus less on remedial and ameliorative services and more on proactive attempts to change conditions in neighborhoods and institutions believed to be criminogenic; less on targeting individual delinquents and at-risk youths and more on institutional change to promote youth development and personal growth (Polk and Kobrin, 1972; Lofquist, 1983; Pittman and Fleming, 1991). Much more information is needed on what interventions focusing on community rather than individual change actually look like in practice. However, in assessing variation on this dimension it is possible to distinguish commitment to the idea of community as the target of intervention in part by how community sanctioning staff view and operationalize their role. A primary measure might be the amount of time devoted to individual casework compared to organizing sanctioning panels, developing victim support groups, job development, community training, and other such tasks. In addition, researchers should examine the extent to which time is devoted to preventive efforts that seek to change educational or employment practices, increase citizen and community group skills in conflict resolution, provide services to victims, and support mentoring young offenders. The extent to which the community is targeted for intervention, at least in intent, differs substantially among the four youth sanctioning models (see exhibit 4). The content of the intervention directed at community change varies from simply attempting to increase the number of citizens participating in CBs to involving residents in a reintegrative shaming process in FGC (Braithwaite and Mugford, 1994). Intervention may also differ in scope. This is shown in the contrast between CS participants' holistic efforts to promote community healing, peacemaking, and capacity building for increased involvement (Stuart, 1995a; Bazemore, 1997b) on the one hand, and the micro-level and incidental efforts of VOM staff to involve the community by recruiting additional volunteer mediators on the other. What differences in outcomes for victim, offender, and community would be expected, for example, as a result of the CS focus compared with that of VOM, which is concerned first with victim outcomes, second with offender outcomes, and only very indirectly with the community? 3. Community as participant. While th