sign in sign up
<<
Home , Collective efficacy, Crime

U.S. Department of Office of Justice Programs National Institute of Justice National Institute of Justice R e s e a r c h i n B r i e f Julie E. Samuels, Acting Director February 2001 Issues and Findings Disorder in Urban Neighborhoods— Discussed in this Brief: The link between disorder and ; Does It Lead to Crime? specifically, whether manifesta- tions of social and physical disor- By Robert J. Sampson and Stephen W. Raudenbush der, such as public drunkenness, graffiti, and broken windows, According to a now-familiar thesis, social Disorder is indeed related to crime. The lead directly to more serious and physical disorder in urban neighbor- broken windows metaphor is apt insofar as offenses. The study, part of the hoods can, if unchecked, lead to serious it asserts that physical signs of decay sig- long-range Project on Human crime. The reasoning is that even such nal neighbors’ unwillingness to confront Development in Chicago minor public incivilities as drinking in strangers, intervene when a crime is being Neighborhoods, assesses the the street, spray-painting graffiti, and committed, or ask the to respond. “broken windows” thesis and breaking windows can escalate into preda- Disorder may in be more useful than its implications for crime control tory crime because prospective offenders crime for understanding certain troubling policy and practice. assume from these manifestations of dis- urban processes, such as the Key issues: The assumption that that area residents are indifferent of many of the Nation’s urban cores. That social and physical disorder can to what happens in their neighborhood.1 is because disorder can be observed, escalate to serious crime has had The “broken windows” thesis has greatly while crime, by contrast, is largely unob- a major influence on enforce- influenced crime control policy, with served. But the contention that disorder ment in many urban areas, result- New York City best exemplifying the use is an essential cause in the pathway to ing in police crackdowns on even of aggressive police tactics to stem disor- predatory crime is open to question. Re- minor incivilities. The research, der. Many other cities have adopted simi- ported here are the results of research that conducted in 196 Chicago neigh- lar “zero tolerance” policies, cracking revisits the assumption of disorder as pro- borhoods, assesses this thesis, down on even the most minor offenses. viding cues that entice potential predators. proposing that crime stems from the same sources as disorder— There is no doubt that understanding Rethinking disorder structural characteristics of certain physical and social disorder in public neighborhoods, most notably The research was part of the Project on concentrated poverty. spaces is fundamental to understanding urban neighborhoods. Certainly, visual Human Development in Chicago Neigh- “,” defined as signs of decay silently but forcefully con- borhoods, a long-term study of the cohesion among neighborhood vey messages about affected neighbor- antecedents of antisocial and criminal residents combined with shared hoods. Disorder triggers attributions and behavior being conducted among a large expectations for informal social predictions in the minds of and group of people in a number of Chicago control of public space, is pro- outsiders alike, changing the calculus neighborhoods. (For a description, see posed as a major social process of prospective homebuyers, real “The Roots of the Study: The Project on inhibiting both crime and disor- Human Development in Chicago Neighbor- der. Disorder was measured by agents, agents, and investors. direct observation rather than The extent of disorder reflects the extent hoods.”) The major goal of this phase of through the subjective percep- of residents’ effectiveness in improving the study was to rethink the consequences tions of neighborhood residents. their neighborhoods and may affect their of disorder and examine its sources. The informal mech- willingness to sustain their activism. anism of collective efficacy (and the broken windows thesis as This publication summarizes the authors’ article, “Systematic Social Observation of Public Spaces: A New Look at Disorder well) focuses on what is visible in in Urban Neighborhoods,” which appeared in American Journal of 105 (3) (November 1999): 603–51. © 1999 by the University of Chicago Press. All reserved. The summary is published with permission of the University of public places. continued… Chicago Press, publisher of the American Journal of Sociology. R e s e a r c h i n B r i e f

Issues and Findings The research produced an alternative common standards that underlie any col- … continued interpretation of the link between disorder lective effort on their part to establish and crime, one that sees many elements social order and safety. These efforts Key findings: The study suggests of disorder as part and parcel of crime are initiated or otherwise pursued infor- that disorder does not directly itself. Typical activities categorized as mally through relatively noncoercive promote crime, although the two phenomena are related, and that social disorder, such as soliciting prosti- means, and are an expression of the self- collective efficacy is a significant tutes and loitering, and incivilities like regulating capacity of a social unit. Thus, factor in explaining levels of painting graffiti are of either neighborhood residents might use these crime and disorder. crime or ordinance violations. The forces means—informal social control mecha- producing these minor may be the nisms—to intervene in preventing truan- Disorder and crime alike were same as those that produce more serious cy, public drinking, , or other found to stem from certain neigh- crimes, with the difference only in the manifestations of disorder. borhood structural characteristics, degree of seriousness. Viewed this way, notably concentrated poverty. disorder and crime are manifestations of The degree of informal social control , arguably one of the the same phenomenon. is not the same in all neighborhoods. best measures of , was Where the rules of comportment are among the offenses for which What behind crime and unclear and people mistrust one another, there was no direct relationship they are unlikely to take action against disorder? with disorder. Disorder was directly disorder and crime. Where there is cohe- linked only to the level of . The study proposes that both crime sion and mutual among neighbors, and disorder stem from structural char- the likelihood is greater that they will In neighborhoods where col- acteristics specific to certain neighbor- lective efficacy was strong, rates share a willingness to intervene for the of violence were low, regardless hoods, most notably concentrated poverty common good. This link of cohesion and of sociodemographic composi- and the associated absence of social trust with shared expectations for inter- tion and the amount of disorder resources. The concentration of disad- vening in support of neighborhood social observed. Collective efficacy also vantage refers not only to low incomes control has been termed “collective effi- appears to deter disorder: Where but also to high unemployment, a high cacy,” a key social process proposed in it was strong, observed levels of ratio of financial dependence of one part this research as an inhibitor of both physical and social disorder were of the on another, and lack of crime and disorder.4 low, after controlling for sociode- investment potential.2 mographic characteristics and These two sets of forces—structural residents’ perceptions of how Structural constraints are not necessarily characteristics of neighborhoods and much crime and disorder there or solely economic. Residential stability, human intervention—are interrelated, was in the neighborhood. typically measured by levels of home working jointly and reciprocally to affect The findings imply that although and transience, has long been crime and disorder. Concentrated disad- reducing disorder may reduce considered a key of strong urban vantage and residential instability under- crime, this happens indirectly, by social organization3 and its absence a mine collective efficacy, in turn fostering stabilizing neighborhoods via lost opportunity for residents to build a increased crime and, by implication, pub- collective efficacy. stake in the . Still other social lic disorder. If the broken windows thesis constraints, among them inordinate pop- is correct, and disorder directly causes Target audience: Local officials and policy- ulation density (which can overwhelm crime, then disorder should mediate the makers, particularly those in urban public services) and mixed land use, are effects of neighborhood structural charac- areas; researchers, particularly also proposed as obstacles to overcoming teristics and collective efficacy on crime. those focused on violence public incivilities. By contrast, if disorder is a manifestation prevention. of the same forces that produce crime, At the same time these social constraints then collective efficacy and structural may promote crime and disorder, there characteristics should account for the are forces working to inhibit them. Com- relationship between disorder and crime. munity residents are assumed to want to In other words, the disorder-crime link live in safe environments free of preda- would be spurious. tory crime and disorder, and they share

2 R e s e a r c h i n B r i e f

Measuring disorder fear of crime. The study reported here In this type of measurement, named assessed visual cues independently “systematic social observation” (SSO), The method of measuring disorder of residents’ perceptions—the amount the means of observation are inde- was different from the one used by the of disorder in the neighborhoods stud- pendent of what is observed. The majority of studies that link signs of ied was measured by directly observ- method is systematic in the sense that disorder with fear of crime and crime ing what was happening on the streets observation and recording are done in victimization. Because these studies during the day. This method of obser- a way that permits replication. What is rely on surveys conducted among resi- vation is consistent with the informal observed are “natural social phenome- dents, they tap subjective perceptions social control mechanism of collective na”—events and their consequences, and they generally find that percep- efficacy, which also focuses on what is viewed more or less as they occur.5 tions of disorder are associated with visible in public places.

The Roots of the Study: The Project on Human Development in Chicago Neighborhoods

T he Project on Human Development Catherine T. MacArthur Foundation; the Adolescent Girls: The Role of Depres- in Chicago Neighborhoods is a long-range National Institute of Mental Health; the sion in the Development of Delinquency, study of the way influence Administration on Children, and by Dawn A. Obeidallah and Felton J. Earls people’s social development. As a study of the U.S. Department of Health (Research Preview, Washington, DC: U.S. of crime in the context of community, the and Human Services; and the U.S. Depart- Department of Justice, National Institute project examines not just the activities of ment of . of Justice, July 1999, NCJ FS 000244). people and their communities, but also the activities of people in their communi- The research team includes Felton J. Earls, Attitudes Toward Crime, Police, and ties. Social scientists from a range of fields investigator and director of the the Law: Individual and Neighborhood are conducting the study and NIJ has pub- project, Harvard Medical School; Stephen Differences, by Robert J. Sampson and lished several reports of the findings. L. Buka, coprincipal investigator, Harvard D.J. Bartusch (Research Preview, Washing- School of Public Health; Robert J. Samp- ton, DC: U.S. Department of Justice, The study topics. The social, economic, son, scientific director for community National Institute of Justice, June 1999, organizational, political, and cultural design, University of Chicago; Stephen FS 000240). structures of Chicago’s neighborhoods Raudenbush, scientific director for analy- are being examined, as are the changes sis, University of Michigan; Jeanne Linking Community Factors and Indi- taking place in them over time. The Brooks-Gunn, scientific director for lon- vidual Development, by Felton J. Earls other component of the project is a gitudinal design, Columbia University (Research Preview, Washington, DC: U.S. series of long-range assessments of the Teachers College; Maya Carlson, policy Department of Justice, National Institute personal characteristics and changing analyst, Harvard Medical School; and of Justice, September 1998, FS 000230). circumstances of children, adolescents, Daniel Kindlon, research associate, Har- Neighborhood Collective Efficacy— and their primary caregivers. Informa- vard Medical School. Does It Help Reduce Violence? by Robert tion has been collected from nearly J. Sampson, Stephen W. Raudenbush, 9,000 residents of 343 Chicago neigh- NIJ reports on the project. To date, and Felton Earls (Research Preview, borhoods, more than 2,800 key commu- researchers have amassed a wealth of Washington, DC: U.S. Department of nity leaders, and a sample of more than information that reveals significant ways Justice, National Institute of Justice, 6,000 children and adolescents (ranging in which the of a April 1998, FS 000203). in age from birth to 18 years). The study neighborhood shapes and determines behavior and identifies the developmen- aims to unravel processes working at the Project on Human Development in tal pathways that lead individuals toward levels of the individual, , and com- Chicago Neighborhoods: A Research or away from a variety of antisocial munity that determine what makes some Update, by Felton J. Earls and Christy behaviors. This knowledge should help neighborhoods safe and law-abiding and Visher (Research in Brief, Washington, practitioners and policymakers develop others dangerous. DC: U.S. Department of Justice, National effective strategies for prevention, inter- The National Institute of Justice is con- Institute of Justice, February 1997, NCJ vention, treatment, and rehabilitation, ducting the study in partnership with 163603). as well as . Among NIJ publica- the Harvard School of Public Health. The tions of project findings are the following: project is cofunded by the John D. and

3 R e s e a r c h i n B r i e f

To measure disorder, trained observers public—to be.) Police records were efficacy acts to inhibit disorder. This videotaped what was happening on examined for counts of three types of finding also held after controlling for the face blocks6 of more than 23,000 crime—homicide, robbery, and burgla- sociodemographic characteristics, type streets in 196 neighborhoods that var- ry. Neighborhood structural character- of land use, and residents’ perceptions ied by race/ethnicity and social class. istics believed to be key to explaining of the amount of crime and disorder. As the observers drove and filmed, the level of crime and disorder were they produced a permanent visual measured by examining the extent of Broken windows revisited record that would be accessible at poverty, the concentration of immi- Overall, the findings did not support any time. They also logged the obser- grants, and residential stability.7 the thesis that disorder directly causes vations they made on each face block. crime. First, although it is true that Counted as signs of physical disorder Residents’ ability to act as “guardians” where -reported violence was were such items as garbage on the who exercise informal social control high, levels of disorder detected by streets, litter, graffiti, abandoned cars, of their neighborhood may also be SSO tended to be high, the relation- and needles and syringes. Counted as affected by population density and ship was not strong. Second—and signs of social disorder were such activ- type of land use. Presumably, the more more important—is the finding, noted ities as loitering, public consumption people per unit of space, the greater above, that the level of disorder varied of , public intoxication, pre- the anonymity and the more difficult strongly with neighborhood structural sumed drug sales, and the presence for residents to identify wrongdoers. characteristics, poverty among them. of groups of young people manifesting Similarly, the commercial traffic den- Once these characteristics and collec- signs of membership. sity accompanying mixed-use land use (in which residential and commer- tive efficacy were taken into account, the connection between disorder and Obtaining information about cial development are combined) has been shown to be related to crime and crime vanished in most instances. neighborhoods disorder and may inhibit social inter- Homicide, arguably one of the best To find out the extent of neighborhood action and . Because measures of violence, was among the collective efficacy, some 3,800 resi- these two factors affect opportunities offenses for which there was no direct dents of these neighborhoods were for crime, the study took them into relationship with disorder. interviewed. From the interviews came account. information about how much informal The implication is that disorder and crime have similar roots: The forces social control was exercised to contain What explains disorder? disorder and crime and how much that generate disorder also generate cohesion residents saw in their neigh- The analysis revealed that a neighbor- crime. It is the structural characteris- borhoods. People were asked, for hood’s structural characteristics matter tics of neighborhoods, as well as neigh- example, about the likelihood their greatly in affecting levels of disorder. borhood cohesion and informal social neighbors would take action if they saw Poverty was the single most important control—not levels of disorder—that children misbehaving in public and factor found to influence the level of most affect crime. Where collective whether the neighbors were willing to disorder in the Chicago neighborhoods efficacy was strong, rates of violence help each other. The information about studied. Disorder tends to be high not were low regardless of sociodemo- social cohesion and informal social only where levels of poverty are high graphic composition and observed dis- control was then combined to produce but also where immigrant order. Levels of homicide and a measure of collective efficacy. are concentrated. And regardless of were affected by the amount of poverty sociodemographic characteristics, and the strength of collective efficacy. Five measures of the extent of crime neighborhoods where land use is Immigrant concentration also affected were used. Neighborhood residents mixed tend to have higher levels of crime: In neighborhoods where immi- were asked whether they or a member physical and social disorder. grants were concentrated, crime victim- of their household had recently been ization levels (as reported by the victimized either by or In neighborhoods where collective survey) tended to be higher. What is by a burglary or . (They were also efficacy was strong, the levels of phys- more, disorder did not act as a media- asked separately how much of a prob- ical and social disorder observed were tor between neighborhood structural lem they believed various social inci- correspondingly low. This finding is con- characteristics and predatory crime. vilities—for example, drinking in sistent with the idea that collective 4 R e s e a r c h i n B r i e f

The main exception was the link In particular, disorder may operate in tory crimes. Neighborhoods with high between disorder and robbery. Of all a cascading by motivating res- levels of disorder did not have higher five measures of crime—residents’ idents to move out of their neighbor- crime rates than neighborhoods with reports of violent victimization and hood, thereby increasing residential low levels of disorder once collective burglary or theft and police counts of instability. And because people move efficacy and neighborhood character- robbery, burglary, and homicide— a only if they have the financial means istics were taken into account. Visible direct association with disorder was to do so, outmigration would increase street-level disorder does not neces- found only for robbery. Evidently, rob- the concentration of poverty among sarily translate into high rates of vio- bers respond to visual cues of social those left behind. Since residential lence; hence, public disorder may not and physical disorder in a neighbor- instability and concentrated poverty be so “criminogenic” after all in cer- hood. These cues may entice them to are associated with lower collective tain neighborhood and social contexts. act, and this in turn undermines col- efficacy and higher crime and disor- The active ingredients of crime seem lective efficacy, producing a cycle der, over the course of time this process to be structural disadvantages and of yet more disorder and ultimately would lead to more crime and disorder. low levels of collective efficacy more more . than disorder. More important, the findings strongly Implications for crime control suggest that policies intended to reduce Tackling public disorder as a means crime by eradicating disorder solely of reducing crime leaves the common Although a basic tenet of the broken through tough law enforcement tactics origins of both, but especially the lat- windows thesis was not sustained by are misdirected. (For a list of recent ter, untouched. Perhaps more effective the study, the findings do not signify NIJ publications on policing disorder would be an approach that focuses on that disorder is irrelevant to under- and related topics, see “Crime, Disor- how residents’ efforts to stem disorder standing crime. Signs of physical and der, and Public Safety: Selected NIJ may reap unanticipated benefits in social disorder are highly visible cues Publications.”) Eradicating disorder greater collective efficacy, which in to which neighborhood residents may reduce crime indirectly by stabi- turn would lower crime in the long run. respond, and they potentially influ- lizing neighborhoods. This is implied Informally mobilizing a neighborhood ence migration, investment, and the in the finding that there is no direct cleanup, for example, would reduce overall viability of a neighborhood. link between disorder and most preda- physical disorder while building

Crime, Disorder, and Public Safety: Selected NIJ Publications

“Broken Windows” and Police Crime, Grime, Fear, and Decline: A “Measuring What Matters: Crime, , by George L. Kelling Longitudinal Look, by Ralph B. Taylor Disorder, and Fear,” by Wesley G. (Research Report, Washington, DC: (Research in Brief, Washington, DC: Skogan, in Measuring What Matters, U.S. Department of Justice, National U.S. Department of Justice, National ed. Langworthy: 37–53. Institute of Justice, October 1999, Institute of Justice, July 1999, NCJ NCJ 178259). 177603). “Measuring What Matters: A New Way of Thinking About Crime and “Crime Control, the Police, and “The Incivilities Thesis: Theory, Public Order,” by George Kelling, in Wars: Broken Windows and Measurement, and Policy,” by Ralph Measuring What Matters, ed. Cultural Pluralism,” by George L. B. Taylor, in Measuring What Matters: Langworthy: 27–35. Kelling, in Perspectives on Crime and Proceedings From the Police Research Justice: 1997–1998 Lecture Series Institute Meetings, ed. Robert H. Public Involvement: Community (Research Forum, Washington, DC: Langworthy (Research Report, Policing in Chicago, by Wesley G. U.S. Department of Justice, National Washington, DC: U.S. Department of Skogan et al. (Research Report, Institute of Justice, November 1998, Justice, National Institute of Justice Washington, DC: U.S. Department of NCJ 172851): 1–28. and Office of Community Oriented Justice, National Institute of Justice, Policing Services, July 1999, NCJ September 2000, NCJ 179557). 170610): 65–88.

5 R e s e a r c h i n B r i e f

collective efficacy by creating and 2. Wilson, William Julius, The Truly Disad- Research Preview, Washington, DC: U.S. strengthening social ties and increas- vantaged: The Inner City, the Underclass, and Department of Justice, National Institute , Chicago: University of Chicago of Justice, April 1998, FS 000203. ing awareness of the residents’ com- Press, 1987; Land, Kenneth, Patricia McCall, mitment to their neighborhood. Such a and Lawrence Cohen, “Structural Covariates 5. See Reiss, Albert J., Jr., “Systematic Obser- mobilization might also demonstrate of Homicide Rates: Are There Any Invariances vations of Natural Social Phenomena,” Socio- to participants and observers alike across Time and Space?” American Journal of logical Methodology, vol. 3, ed. Herbert Costner, San Francisco: Jossey-Bass, 1971: 3–33. that neighborhood residents could be Sociology 95 (1990): 922–963; and Hagan, John, and Ruth Peterson, eds., Crime and Inequality, relied on to maintain public order. By Stanford, CA: Stanford University Press, 1995. 6. The face block is the block segment on one contrast, a police-led crackdown on side of a street only. Thus, the buildings across disorder would probably produce a 3. Kasarda, John, and Morris Janowitz, “Com- the street from one another on any block count, for observation purposes, as two units. very different response by residents. munity Attachment in Mass ,” American Sociological Review 39 (1974): 328–339. 7. For more details of the way collective effica- Notes 4. Sampson, Robert J., Stephen Raudenbush, cy was defined and measured, see Sampson, and Felton J. Earls, “Neighborhoods and Violent Raudenbush, and Earls, “Neighborhoods and 1. The thesis was first elaborated by James Q. Crime: A Multilevel Study of Collective Effica- Violent Crime.” Wilson and George Kelling in “The Police and cy,” Science 277 (1997): 918–924. This article Neighborhood Safety: Broken Windows,” Atlantic was summarized as Neighborhood Collective Monthly 249 (3) (March 1982): 29–36, 38. Efficacy—Does It Help Reduce Violence?

Findings and conclusions of the research Robert J. Sampson, Ph.D., is Profes- Research for the Project on Human reported here are those of the authors and do sor of Sociology at the University of Development in Chicago Neighborhoods not necessarily reflect the official position or Chicago and Senior Research Fellow is supported by the National Institute policies of the U.S. Department of Justice. with the American Foundation. of Justice (grant 93–IJ–CX–K005), the Stephen W. Raudenbush, Ed.D., is John D. and Catherine T. MacArthur The National Institute of Justice is a Professor of Education and , Foundation, the National Institute of component of the Office of Justice and Senior Research Scientist at the Mental Health, the Administration on Programs, which also includes the Bureau Institute for , Children, Youth and Families, and the of Justice Assistance, the Bureau of Justice University of Michigan. U.S. Department of Education. Statistics, the Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime.

This and other NIJ publications can be found at and downloaded from NCJ 186049 the NIJ Web site (http://www.ojp.usdoj.gov/nij).

U.S. Department of Justice PRESORTED STANDARD Office of Justice Programs POSTAGE & FEES PAID National Institute of Justice DOJ/NIJ PERMIT NO. GÐ91

Washington, DC 20531 Official Business Penalty for Private Use $300