Broken Windows Policing

by

Jacinta M. Gau, Alesha Cameron

• LAST REVIEWED: 27 March 2019
• LAST MODIFIED: 27 March 2019
• DOI: 10.1093/obo/9780195396607-0265

Introduction

The 1980s saw sweeping changes occur in policing nationwide. Disorder (sometimes also called incivilities) rose to the top of the police agenda with the publication of studies showing that physical decay and socially undesirable behaviors inspire more fear than crime does. Crime is a relatively rare event, but physical disorder (graffiti, vandalism, and the like) and social disorder (such as aggressive panhandling or people being intoxicated in public) are far more prevalent. Broken windows theory drew from concepts embedded within criminological and social psychological theories. According to this perspective, the cause of crime is disorder that goes unchecked and is permitted to spread throughout a neighborhood or community. Disorder is theorized to scare people and makes them believe that their neighborhood is unsafe. These people subsequently withdraw from public spaces. The disorderly environment and empty streets invite crime and criminals. Offenders feel emboldened to prey on people and property because the environmental cues suggest a low likelihood that anyone will intervene or call the police. At this point, the neighborhood’s disorder problem becomes a crime problem. In the broken windows viewpoint, police are the front line of disorder reduction and control. Police are seen as needing to actively combat disorder in order to make neighborhood residents feel safe so that they will continue participating in the social fabric of their community. Broken windows theory does not contain a set of directions for precisely how police should go about preventing and eliminating disorder. Police leaders wanting to use the tenets of broken windows theory in their communities have to figure out how to put these concepts into practice. What evolved to be called order maintenance policing still varies from agency to agency. One of the most popular (and controversial) strategies is to aggressively enforce laws against nuisance and public-order offending (loitering, public drunkenness, and so on). The signature tactic of order maintenance policing is the street stop, which is a brief field detention and questioning. Police officers who have reasonable suspicion to believe an individual is engaged in criminal behavior can detain that person for questioning. Aggressive enforcement using street stops as a core tactic is not universal and there are other options (e.g., community policing, target hardening, utilizing government services, conveying reliable information) for police agencies to engage in disorder-reduction activities. The disorder-reduction strategy relying on street stops and arrests for low-level offenses goes by names such as “broken-windows policing” and “zero-tolerance policing.” The term “aggressive order-maintenance policing” is adopted for present purposes, as some broken-windows proponents have taken issue with terms like “zero tolerance.” This bibliography provides an overview of studies of broken windows theory and of some of the police efforts to employ the logic of this theory to reduce disorder, fear, and crime. Methodological rigor has been a recurrent topic in the discussion about the merits of broken windows theory and order maintenance policing, so this will be reflected in the bibliographies where relevant.

Tests of Broken Windows Theory

Broken windows theory should be understood as analytically separate from the policing strategy premised upon it. Broken windows theory predicts that unchecked disorder sparks fear and drives people indoors or causes them to move out of the neighborhood altogether. Social ties between neighbors break down and social control in public spaces declines. This physical and social decay creates an environment in which offenders thrive. The accuracy of the prediction that police can control crime by managing disorder rests upon the validity of the assumption that disorder causes crime. Several researchers have attempted to test this assumption. Data sources have included survey-based reports of people’s perceptions of disorder and crime, systematic social observation of public spaces, and official data from police departments. While some studies appear to support the claim that disorder causes crime, this support is limited and inconsistent. The evidence leans toward the conclusion that disorder and crime are interrelated problems but that disorder cannot be isolated as a singularly influential determinant from which a crime problem will eventually emerge. Skogan 1990 was the first empirical test of broken windows theory and concluded that the disorder-crime relationship appeared sound. Harcourt 2001, however, reanalyzes the data Skogan 1990 used, and finds that the disorder-crime relationship was significantly weaker than Skogan had claimed, even nonexistent in some cities. Taylor 2001 likewise finds that disorder is a weak, inconsistent predictor of later crime rates. Sampson and Raudenbush 1999 suggests that racial and social composition of a neighborhood predicted perceived disorder more than direct observations of disorder did. Sampson and Raudenbush 2004 uncovers evidence that neighborhood-level collective efficacy accounted for the presence of both disorder and crime. Gau and Pratt 2008 finds that people do not make distinctions between disorder and crime in their neighborhoods, though Gau and Pratt 2010 determines that this distinction does appear larger in disorderly areas. Harcourt and Ludwig 2006 finds that disorder had no effect on crime. Skogan 2015 concludes that disorder plays a role in encouraging fear of crime, eroding community stability, and weakening informal social control.

• Gau, J. M., N. Corsaro, and R. K. Brunson. 2014. Revisiting broken windows theory: A test of the mediation impact of social mechanisms on the disorder–fear relationship. Journal of Criminal Justice 42.6: 579–588.

  Although previous research on broken windows theory suggests that disorder initiates fear of crime, which then leads to crime, this article uses individual and census-tract level data to examine the mediating effects of both social cohesion and social control in relation to fear of crime. Findings from this article suggest that fear of crime outcomes are caused partially by disorder through the mediating effects of social cohesion and control.

• Gau, J. M., and T. C. Pratt. 2008. Broken windows or window dressing? Citizens’ (in)ability to tell the difference between disorder and crime. Criminology & Public Policy 7.2: 163–194.

  This study uses survey data measuring people’s assessments of the extent of disorder and crime in their neighborhoods to determine whether these two constructs are empirically separable. Results from confirmatory factor analyses suggest that although a two-factor model fits the data, the strong between-factor correlation indicates an absence of discriminant validity. The authors conclude that people’s inability to discern the difference between disorder and crime might account for the consistently high relationship found between these two neighborhood problems.

• Gau, J. M., and T. C. Pratt. 2010. Revisiting broken windows theory: Examining the sources of the discriminant validity of perceived disorder and crime. Journal of Criminal Justice 38.4: 758–766.

  This article analyzes the relationship between concentrated disadvantage and perceptions of disorder and crime as separate problems. The authors of this article utilize survey data collected in 2003 and 2006. Findings suggest concentrated disadvantage affects respondents’ perceptions of fear and crime as separate social problems. Results from this study support the broken windows notion that orderly areas should be kept orderly. The occurrence of varying perceptions of disorder and crime as separate problems is also a key finding in this study.

• Harcourt, B. E. 2001. Illusion of order: The false promise of broken windows policing. Cambridge, MA: Harvard Univ. Press.

  Harcourt situates broken windows theory within the politically conservative movement underway in the 1980s. Harcourt maintains that the social meaning of disorder is more important than disorder itself, and that social meaning can change across time and be different across groups. In reanalyzing Skogan’s data, Harcourt finds that robbery is the only crime statistically related to disorder once neighborhood disadvantage (e.g., poverty) is controlled for. Additionally, even this relationship disappears when the five neighborhoods in Newark are excluded from the analysis.

• Harcourt, B. E., and J. Ludwig. 2006. Broken windows: New evidence from New York City and a five-city social experiment. The University of Chicago Law Review 73.1: 271–320.

  The Moving to Opportunity (MTO) project provides housing vouchers to relocate low-income families to less disadvantaged communities that have fewer signs of both physical and social disorder. This article presents evidence on the disorder and crime relationship proposed in broken windows theory by using MTO data from five cities in the United States. Since disorder was found to have no effect on crime, results from this study do not support the claim that reducing signs of disorder in turn reduces crime.

• Sampson, R. J., and S. W. Raudenbush. 1999. Systematic social observation of public spaces: A new look at disorder in urban neighborhoods. American Journal of Sociology 105.3: 603–651.

  Utilizing a combination of systematic social observation, US census data from 1990, household survey data, and Chicago Police Department data from 1995, the authors of this article analyze the origins and effects of public disorder. In testing the broken windows hypothesis, which states that disorder is directly related to serious crime, findings from this study did not support this claim. The authors conclude that the disorder-crime relationship varies according to neighborhood characteristics, such as collective efficacy.

• Sampson, R. J., and S. W. Raudenbush. 2004. Seeing disorder: Neighborhood stigma and the social construction of “broken windows.” Social Psychology Quarterly 67.4: 319–342.

  This article examines variations in individual perceptions of disorder within and between neighborhoods. The authors use neighborhood survey data, US census data, official crime data, and systematic social observation. Results suggest that racial and social composition of a neighborhood predicted perceived disorder more than direct observations of disorder. The findings from this study reveal that the broken windows hypothesis on reducing disorder may have limited effects on perceptions of neighborhoods with large minority populations.

• Skogan, W. G. 1990. Disorder and decline: Crime and the spiral of decay in American neighborhoods. Berkeley: Univ. of California Press.

  Using data collected between 1977 and 1983 from residents in high-crime neighborhoods in six major cities, Skogan estimates the empirical relationship between disorder and crime. Using survey-based measures of neighborhood-level disorder and robbery victimization, he finds a significant relationship between disorder and robbery, controlling for relevant covariates. He follows up this analysis with a multi-city comparison of different types of order-maintenance policing strategies, which reveal mixed results in terms of how police successfully reduced disorder, fear, and crime.

• Skogan, W. 2015. Disorder and decline: The state of research. Journal of Research in Crime and Delinquency 52.4: 464–485.

  This article summarizes the research available on social and physical disorder indicators that are used in community settings and specifies approaches for measuring disorder. In support of broken windows theory, the author concludes that disorder plays a role in encouraging fear of crime, eroding community stability, and weakening informal social control.

• Taylor, R. B. 2001. Breaking away from broken windows: Baltimore neighborhoods and the nationwide fight against crime, grime, fear, and decline. Boulder, CO: Westview Press.

  Taylor critiques both the tenets behind broken windows theory and the methods previous researchers used to empirically test the theory. Taylor notes that although disorder and crime frequently covary, it is hard to prove that disorder causes crime. Many other community-level factors and social processes are at play. Taylor’s analysis of survey-based and independently assessed disorder measures from Boston neighborhoods were weak, inconsistent predictors of later crime rates in those neighborhoods.