Springer Series on Evidence-Based Crime Policy
Series Editors: Lawrence W. Sherman, University of Pennsylvania Heather Strang, Australian National University Crime prevention and criminal justice policies are domains of great and growing importance around the world. Despite the rigorous research done in this field, policy decisions are often based more on ideology or speculation than on science. One reason for this may be a lack of comprehensive presentations of the key research affecting policy deliberations. While scientific studies of crime prevention and criminal policy have become more numerous in recent years, they remain widely scattered across a wide range of journals and countries. The Springer Series on Evidence-Based Crime Policy aims to pull this evidence together while presenting new research results. This combination in each book should provide, between two covers (or in electronic searches), the best evidence on each topic of crime policy. The series will publish primary research on crime policies and criminal justice practices, raising critical questions or providing guidance to policy change. The series will try to make it easier for research findings to become key components in decisions about crime and justice policy. The editors welcome proposals for both monographs and edited volumes. There will be a special emphasis on studies using rigorous methods (especially field experiments) to assess crime prevention interventions in areas such as policing, corrections, juvenile justice and crime prevention. Published in Cooperation with the Campbell Crime and Justice Group
For further volumes: http://www.springer.com/series/8396
Bruce J. Doran · Melissa B. Burgess
Putting Fear of Crime on the Map Investigating Perceptions of Crime Using Geographic Information Systems
123
Bruce J. Doran Fenner School of Environment & Society The Australian National University Canberra, ACT 2601, Australia
[email protected] Melissa B. Burgess Fenner School of Environment & Society The Australian National University Canberra, ACT 2601, Australia
ISBN 978-1-4419-5646-0 e-ISBN 978-1-4419-5647-7 DOI 10.1007/978-1-4419-5647-7 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011934029 © Springer Science+Business Media, LLC 2012 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Dedication
We would like to dedicate this book to the many survey participants who gave willingly their time and experience – without their contribution, the research in Wollongong and Kings Cross would not have been possible. The two-stage interview process used in the Wollongong study provided a means for informal discussions in addition to the survey itself. Very often people chose to share their thoughts on policing or crime in the area and to describe personal experiences, or those of work colleagues, friends or family. The following accounts are the stories of some of the participants and in many ways these personal reflections provide powerful insights into the impact of fear of crime at an individual level.
A Real Estate Agent Leaves At the time of the survey, Amelia1 worked in the Piccadilly area of Wollongong. She was a community-minded person who took great pride in the fact that she had raised a number of adopted children and was a key person in the local business community. She worked for a real estate agency and was based in the Piccadilly shopping mall, the key feature of the precinct and a focus of crime, disorder and fear in the CBD area. The mall, despite being next to the main railway station that commuters used to access the CBD, was poorly utilized. The area had long proven to be a serious challenge for the police, the Wollongong City Council (WCC) and business residents of the local community. Amelia firmly believed that her job provided her with the potential to make positive changes in the area. As a senior real estate agent who primarily dealt in commercial property, she was able to encourage buyers who she felt were likely to have a beneficial presence in the area. An example of this was the ongoing negotiations she was handling with a university who were considering the purchase of a motel above the mall. It was well known that the motel functioned as an informal brothel and centre for drug dealing. Amelia felt that a university-run research 1 The names of respondents have been changed to protect privacy, but the content of their stories have not been altered.
v
vi
Dedication
facility would dramatically change the dynamics. One afternoon while locking up the shopfront for the agency, Amelia was approached from behind and doused with petrol. She was then confronted by a drug addict she knew well – someone she was not normally bothered by but who, on this occasion, was high and did not seem to recognize her. Amelia frantically pleaded with the addict as he waved a lighter and threatened to ignite her. After several terrifying moments, her attacker seemed to lose interest and walked away. Amelia was someone conditioned to minor disorder. She knew by name many of the addicts, including her attacker, who used the methadone clinics. She was also understanding of the weekend alcohol-related problems, as well as the youths involved in tagging graffiti. However, the very direct and personal attack she experienced outside her shopfront that afternoon was too much for her. The attack took place between the first and second stages of the interview. Several weeks later, she had moved to another area and a community-minded person, who genuinely believed Piccadilly could change for the better, had left the area for good.
A Cobbler Who Wouldn’t Eat Outside Tony was a huge man standing well over 6.5 feet tall. He ran a small shoe repair shop that opened directly out onto the Crown Street Mall. The area was not in the core of crime hotspot for the CBD but was a focus of social disorder on the weekends. Tony told me that he had recently retired at the age of 35 from a specialist military unit in the Australia Defence Force and had located in Wollongong for family reasons. He had described how he had been a victim of several serious crimes in the 12 months preceding the survey. One crime was particularly fear inspiring – a group of youths had attacked him with an iron bar while he was getting from his shop to his car at the back of the building after work. However, when asked about the incident, Tony explained that this did not bother him because of his self-defence training and that he was easily able to disarm the attackers. It was all the more striking then, to hear him talk about how he would never eat his lunch or take breaks in the mall area directly outside his shop. His fear in this case related to the fact that, in his judgement, there was a chance of being attacked with a syringe and this was not a risk he was prepared to take. He explained that his first priority was his family and that if he was a victim of a syringe attack, he may no longer be able to act as a provider. It is hard to imagine a more capable guardian than Tony, yet his avoidance behaviour meant that he was effectively removed from the mall area only metres away from his shopfront.
A Night on the Town Goes Wrong John ran a small shop below the Crown Street Mall that sold specialist figurines for dungeons and dragons-type board games. He was a soft spoken small man who was 20 years old. He maintained a calm demeanour during the interview but passionately
Dedication
vii
related a story at the conclusion of the exercise. He undid the top three buttons of his shirt and revealed some massive scarring around the base of his neck. I could see that he had grown a beard to hide some of them. John went on to explain that two years back, he had been walking through Crown Street Mall late at night one weekend. The mall often serves as a conduit between two night club strips at either end of the vehicle-free area. He had left friends and was going to ‘kick on’ at some of the clubs alone on Keira Street. He found himself suddenly surrounded by three men with skateboards – without warning or provocation; they picked up the boards and attacked him. He was seriously injured but able to walk after the incident and attempted to get help from passers-by. When this proved to be unsuccessful, he attempted to catch a taxi from Crown Street to the hospital, only a kilometre away but up a steep incline. When no taxis would stop he was forced to walk to the hospital. After buttoning up his shirt, he stated strongly that he was determined not to let the experience ‘beat him’. Many months later, I was conducting a social disorder assessment in the mall at 4 am and was aghast to see John walking determinedly, and alone, through the paved walkway area. Here appeared a classic manifestation of the risk-victimization paradox – a young man who was relatively more likely to become a victim of crime displaying an apparently irrational lack of fear. I had the strong impression though that John was carrying something to protect himself.
A Husband Threatens to Take the Law into His Own Hands Probably one of the more horrific accounts related to me while conducting interviews was the experience of Michelle, a petite mid-30s dress-shop owner, who worked at the bottom end of Crown Street Mall. I could see that she was nervous as the survey moved through a section on victimization over the past 12 months. At the end of the interview, her husband came from the back of the shop to join the conversation. They were both very keen to know what the survey data would be used for – would it be used to police antisocial behaviour in the mall? Who would have access to the results? Was the study simply an academic exercise? It emerged that their concern stemmed from a serious attack that had taken place a number of months prior to the interview. Michelle had been accosted in her shop by a much larger woman who demanded cash from the register. When Michelle refused, the woman became violent and threw her against a display. Her attacker then went into a frenzy, kicking and punching her repeatedly, as well as bodily picking her up and throwing her around the shop, as the smaller woman desperately tried to fend off the blows. The assault continued for some minutes before the offender left the shop. Michelle was badly shaken and had sustained serious facial injuries that required surgery. She pressed charges, as the offender was generally known in the CBD. However, a week later, Michelle’s attacker was back in her shop to threaten her again. Michelle and her husband spoke of their frustration with the authorities – in their opinion, the system had completely failed them and had left them both feeling vulnerable. Michelle no longer felt secure in the shop by herself, so her husband, who was self-employed, had moved to the rear of the shop and established an office. He emphasized strongly
viii
Dedication
that he would not tolerate any future intimidation and that if it were to happen again, he would take the law into his own hands. The responses to the survey varied considerably. Some respondents told of how they carried their car keys when leaving work so that sharp ends protruded from between their fingers – if they needed to defend themselves they were ready. Other people spoke on their mobile phones when walking in public to avoid being addressed by strangers. Many people drew very detailed cognitive maps which outlined the areas they avoided because they were afraid of being robbed, beaten or attacked. In some cases, people were prisoners not in their homes but in their workplace, as is the case in the example below.
Some of the cognitive maps drawn by a survey respondent in Wollongong outlining areas they avoided around their workplace (the hollow arrow indicates the location of their workplace)
The point of these stories is not to overemphasize the shocking nature of some of the experiences but rather to acknowledge the individual stories and behavioural responses that are somewhat masked by, and lay behind, the collective spatial analyses presented in the Kings Cross and Wollongong studies. These accounts also serve to reinforce the fundamental assumption behind this book, namely that fear of crime is a significant problem for society because it prompts people to adopt protective and
Dedication
ix
avoidance behaviours. These behaviours have many consequences at the level of the individual and community. They are complex and can be hard to understand but they also provide a lens through which to examine interactions between members of a community, public space and relationships with crime and disorder. We use techniques from behavioural geography in conjunction with Geographic Information Systems, to develop an approach which we hope will contribute to the literature on fear of crime as well as the management of the problem. The approach is relatively simple, but strongly grounded in well-established principles of cognitive mapping. It is transferable to other contexts and situations – in the final chapter we outline many possible future applications and avenues for research. We would again like to thank the people who participated in the Wollongong and Kings Cross studies as it is their contribution that allows us to ‘put fear on the map’.
Series Foreword
All over the world, politicians and policy makers are increasingly inclined to claim that their proposals are ‘evidence-based’. Social scientists have even caught this spirit of evidence, which may show in their occasional use of the malapropism of ‘evidence-based research’ (thus implying the existence of some other legitimate category of research that is not based on evidence, perhaps including what Peter Reuter and others describe as ‘mythical numbers’i ). Even when policies can clearly cite a relevant body of research, however, scientists cannot agree on what makes a policy ‘evidence-based’.ii The present book series must therefore grapple with a series of challenges to its very name, let alone the ordinary hurdles of good research. One challenge is about the scope of evidence that is embraced by the concept of ‘evidence-based’ anything. In forensic evidence, courts usually offer a very broad invitation to facts and measures in support of a hypothesis that bears on the case. In the United States they even allow theories of causation to be presented to juries, a practice widely attacked as ‘junk science’ until the US Supreme Court barred the use of theories that had not been tested, at least in the federal courts (Daubert v. Merell Dow, 1993). While many court decisions may still turn on theories that most scientists would dismiss as not adequately evidence-based, the standard at least requires some evidence. A far narrower scope for what is ‘evidence-based’ has been implied by those who focus on ‘what works’, or the impact of programmes on outcomes.iii Readers might expect a series on evidence-based crime prevention to use that boundary. They will, perhaps, be pleasantly surprised that we do not. As any definition of good medical practice holds, an accurate diagnosis is a prerequisite to choosing an appropriate treatment. Similarly, it is just as important to know ‘what is’ as to know ‘what works’. Tools and evidence for classifying crimes and criminals, for analysing trends and patterns in criminal events, understanding how crimes are committed and may therefore be prevented – all these are essential forms of evidence for the broader enterprise of crime prevention. Even research that focuses on interventions is usually accompanied by descriptive and diagnostic data on the nature of the crime issue in question. An entire series of books can certainly afford to do the same. A further challenge is how rigorous a series should be in defining adequate evidence of cause and effect, or even descriptive estimates of crime patterns. Our aim
xi
xii
Series Foreword
is to publish the most rigorous evidence available on important crime problems. If, for example there are no randomized controlled trials on gun crime prevention, then the best possible quasi-experiments are a welcome addition to the policy debate. Despite the editors’ strong associations with experimental criminology, we do not insist that randomized trials are the only worthwhile source of evidence for policy. As Sherman has defined evidence-based policing,iv the best definition of rigor is that the evidence simply be ‘scientific’, with all the systematic care and precision required by the scientific method. The aim of this series is to help foster evidence-based crime prevention with a broader range of materials, and a more flexible medium, than is presently available. We invite readers to examine the series as a more rigorous, complete and independent source of evidence than may be available from government reports or programme delivery organizations. We invite submissions from authors who want their readers to have all the evidence produced by a particular project, and who have much more evidence to report than can fit in any one journal article. We invite subscriptions from libraries that require the most complete evidence available on crime and justice issues costing hundreds of billions of dollars for governments to address world-wide. We are grateful to both Springer and the Campbell Collaboration Crime and Justice Steering Group for their support in developing this series. And while the dedication of each book is the privilege of the authors, we would like to dedicate the series to the steadfast support of Jerry Lee, the greatest champion of evidence-based policy we know. Washington, DC April, 2011
Heather Strang Lawrence W. Sherman
Notes i. Reuter, P. (1987). “The (continued) vitality of mythical numbers”. Public Interest 75: 79–95. ii. The most elaborate attempt to do so can be found in a 2009 report of the National Research Council and Institute of Medicine, Preventing Mental, Emotional, and Behavioral Disorders Among Young People: Progress and Possibilities. Committee on Prevention of Mental Disorders and Substance Abuse Among Children, Youth and Young Adults: Research Advances and Promising Interventions. Mary Ellen O’Connell, Thomas Boat, and Kenneth E. Warner, Editors. Board on Children, Youth, and Families, Division of Behavioral and Social Sciences and Education. Washington, DC: The National Academies Press. iii. Sherman, L. W., D. P. Farrington, B. C. Welsh and D. L. Mackenzie (eds) (2002), EvidenceBased Crime Prevention. London, Routledge. iv. Sherman, L. W. Evidence Based Policing. Washington, DC, Police Foundation http://www. policefoundation.org/pdf/Sherman.pdf.
Acknowledgements
There are many people we would like to thank who have helped in producing this book. First, we owe a great deal to Professor Brian Lees from the University of New South Wales at the Australian Defence Force Academy, who was the principal PhD supervisor for both the Wollongong and Kings Cross projects. In general we are both grateful for the opportunities that have opened up through undertaking the research and for Brian’s guidance and feedback throughout. We are just two of many students who have benefited from his experience in GIS-based research and his ability to find topics that deal with relevant and interesting issues. It was his vision that identified a need for spatially explicit research into fear of crime. We would also like to extend our gratitude to Professor Peter Grabosky, who suggested that the research conducted in Wollongong and Kings Cross would make a valuable contribution to this series. Peter’s generosity, encouragement and willingness to promote our work have been of significant value. There are a number of specific acknowledgements to make regarding the Wollongong study, presented in Chapter 6 Dr Ron Horvarth provided important advice and introductions to personnel within the NSW Police Service during the initial stages of the project. Dr Chris Devery, NSW Police Force, provided feedback at various stages of the project and guidance on protocols for working alongside the NSW Police Service. Dr Jerry Ratcliffe shared his considerable expertise on crime mapping and policing issues and gave valuable comments on PhD thesis chapters and papers related to the study. Various members of the NSW Police Service, Wollongong Local Area Command, gave specific advice relating to the study site, helping to gain access to crime data for the region. A number of officers also attended seminars at the Wollongong City Council where they gave feedback on early results from the project. Bronwyn Richards, Sand Hall, Rada Jordan and Greg Doyle from the Wollongong City Council were all very supportive of the project from the fieldwork stage onwards. We are very grateful for the ideas they shared and for the opportunities they created for me to discuss and implement my research through workshops and seminars. I am also in their debt for guiding me to a number of secondary sources that were relevant to my project. Towards the end of the project, the Australian Institute of Criminology and the Local Government Association provided funding to travel to Brisbane and present a paper at a conference looking at graffiti and disorder. This was valuable for many reasons – most xiii
xiv
Acknowledgements
notably initial for the discussions held with NSW Police Superintendent Dave Darcy of the Kings Cross Local Area Command that ultimately led to the Kings Cross Study, presented in Chapter 7. Dave provided much insight regarding the implementation of the project. His early endorsement and continuing interest in fear-of-crime research is also valued. Inspector Gary Groves, NSW Police Force, was also instrumental during the interviewing stage of the project. Gary provided the materials necessary for interviewing, helped with interviewer training, a temporary office and assisted in distributing information fliers. All NSW Police officers stationed in Kings Cross and Woolloomooloo during 2004 are acknowledged for accommodating the research during this period. I thank Chris Devery, NSW Police Force, for liaising between the NSW Police Legal Services and the ANU regarding the exchange of crime data. Associate Professor Julie Stubbs, University of Sydney, also provided thorough and constructive feedback on my thesis chapter drafts. Julie gave particularly useful advice on the interviewing procedure and also liaised with her 2004 Masters of Criminology students to conduct the interviewing for this study. These students and Volunteers in Policing (VIP) are acknowledged for their time, professionalism and assistance in conducting the interviews. In particular, VIPs Warwick and Jim are acknowledged for their outstanding participation. Their assistance was central to the acquisition of the large dataset used in the research. Helen Steptoe, VIP, also provided immense support during the data entry phase of the project. Emeritus Professor Diana Howlett is acknowledged for funding the Howlett Honours Prize for Geography. Melissa was awarded this prize in 2004 and used the financial gift to purchase numerous fear-of-crime books that could not be sourced in Australian libraries. I specifically thank Douglas Grand, General Manager of the Kings Cross Licensing Accord, for his donation in 2004 to help with costs associated with the interviewing stage of the project. As with many research projects, special thanks should go to staff from the research department – the School of Resources, Environment and Society, now the Fenner School of Environment and Society at the Australian National University. Professor Peter Kanowski was head of department at the time and was always encouraging and willing to support the projects with conference and fieldwork funding. Karl Nissen and Steve Leahy have provided help with computer-related issues over many years. To the various members of the tea club over the years – Shawn Laffan, Kimberly Van Neil, Brian Lees, Clive Hilliker, Steve Leahy, Karl Nissen, Eugene Wallensky, Paul Carlile, Sanjeev Shrivastava, Sunil Sharma, Sandy Gilmore, Piers Bairstow and many others.
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . The Emergence of Fear of Crime As an Area of Research The Paradoxical Nature of the Fear of Crime . . . . . . . Current Trends in Fear of Crime Research . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
1 1 2 3 4
2 Why Is Fear of Crime a Serious Social Problem? . . . . . . . . . Individual Reactions . . . . . . . . . . . . . . . . . . . . . . . . . Hypothesized Links Between the Fear of Crime, Disorder and Crime Disorder and Decline Hypothesis . . . . . . . . . . . . . . . . . . . Economic Impact of Behavioural Responses to Fear of Crime . . . . Chapter Review: Potential Problems Not to Be Ignored and a Need for Spatially Explicit Research . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . .
. . . . .
9 9 11 13 16
. . . .
19 19
3 What Causes Fear of Crime? . . . . . . . . . . . . . . . . . Criminal Opportunity and Risk of Victimization Theories . . . Demographic Theories Explaining Fear of Crime . . . . . . . Victimization Hypothesis . . . . . . . . . . . . . . . . . . . Indirect Victimization Hypothesis . . . . . . . . . . . . . . Vulnerabilities Hypothesis . . . . . . . . . . . . . . . . . . Review: An Abundance of Contested Demographic Studies . . Social Theories Explaining Fear of Crime . . . . . . . . . . . Risk Society Hypothesis . . . . . . . . . . . . . . . . . . . Social Disorganization Hypothesis . . . . . . . . . . . . . . Review: Social Studies Emphasize the Inherent Complexity of ‘Fear’ of ‘Crime’ . . . . . . . . . . . . . . . . . . . . . . . . Environmental Theories Explaining Fear of Crime . . . . . . . The Disorder/Incivilities Hypothesis . . . . . . . . . . . . . Threatening and Safe Environments Theories . . . . . . . . Signal Crimes Perspective . . . . . . . . . . . . . . . . . . Review: Intuitive Environmental Studies into Cues Triggering Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
25 25 26 26 27 29 31 31 31 32
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
37 38 38 40 41
. . . . .
44
xv
xvi
Contents
Chapter Review: An Opening for Pertinent Environmental Studies . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Managing Fear of Crime . . . . . . . . . . . . . . . . . . . . . Policing Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . Case Study: The New York Police Department’s (NYPD) Policing Model . . . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Design and Fear of Crime . . . . . . . . . . . . . Chapter Review: Police, Community and Government Cooperation References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . .
51 51
. . . .
. . . .
55 56 61 61
. . . . . .
67 67 68
. . . . . .
. . . . . .
69 70 71 72 72 72
. . . .
75 76
. . . . . . . . .
. . . . . . . . .
79 79 81 82 84 85 86 88 88
. . . . . . . . .
. . . . . . . . .
95 95 96 98 99 101 103 115 133
. . . . . .
145
5 Investigating Fear of Crime . . . . . . . . . . . . . . . . . . . . . Defining Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . Fear Is an Emotion, Not Cognition . . . . . . . . . . . . . . . . . Fear in Relation to Other Emotional Reactions and Stimuli that Trigger Fear . . . . . . . . . . . . . . . . . . . . . . . . . . Crime Involves a Violation of Criminal Law . . . . . . . . . . . . Types of Fear of Crime: Personal and Altruistic Points of View . . Review: Key Issues to Consider When Defining Fear of Crime . . . Measuring Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . Problems with Cognitive Approaches to Measuring Fear of Crime Improvements Through Affective Approaches to Measuring Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . . Behavioural Approaches to Measuring Fear of Crime . . . . . . . Review: A Preference for Avoidance-Based Measures in Fear-of-Crime Studies . . . . . . . . . . . . . . . . . . . . . . . Analysing Fear-of-Crime Data . . . . . . . . . . . . . . . . . . . . Advantages of Spatial Analyses of Fear of Crime . . . . . . . . . Spatial Cognition and Cognitive Mapping . . . . . . . . . . . . . The Beginning of Fear Mapping . . . . . . . . . . . . . . . . . . Activity Diaries and Daily Routines . . . . . . . . . . . . . . . . Geographic Information Systems and Fear of Crime . . . . . . . Chapter Review: A New Direction with Avoidance Mapping . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 The Wollongong Study . . . . . . . . . . . . . . . . . . . The Goals of the Wollongong Study . . . . . . . . . . . . . Research Setting . . . . . . . . . . . . . . . . . . . . . . . . Logic Behind Study Site Selection . . . . . . . . . . . . . The Central Business District of Wollongong . . . . . . . Crime and Fear of Crime in Wollongong . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . Discussion of Spatial Outputs . . . . . . . . . . . . . . . Integrating the Key Spatiotemporal Findings with Police and Community Initiatives in Wollongong: The Degree of Institutional Involvement . . . . . . . . . . . . . . . . . .
44 45
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . .
Contents
xvii
Assessments of Techniques and Approaches Developed in Wollongong Study . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 The Kings Cross Study . . . . . . . . . . . . . . . . . . . . . . . Background to the Kings Cross Study . . . . . . . . . . . . . . . . Goals of the Kings Cross Study . . . . . . . . . . . . . . . . . . . . Research Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . Geographic Location . . . . . . . . . . . . . . . . . . . . . . . . Historical Background . . . . . . . . . . . . . . . . . . . . . . . Demographic Characteristics . . . . . . . . . . . . . . . . . . . . Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Interviewing Approach . . . . . . . . . . . . . . . . . . . . . . . Survey Design and Questions . . . . . . . . . . . . . . . . . . . Spatial Data Visualization . . . . . . . . . . . . . . . . . . . . . Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . Sample Characteristics . . . . . . . . . . . . . . . . . . . . . . . People Are Afraid of Crime in Kings Cross . . . . . . . . . . . . People Avoid Specific Areas of Kings Cross Due to Fear of Crime Exploring the Underlying Reasons for Fear of Crime . . . . . . . Integrating the Fear Mapping Results with Policy and Community Crime and Fear-of-Crime Prevention . . . . . . . . Addressing Crime . . . . . . . . . . . . . . . . . . . . . . . . . Targeting Pertinent Signs of Disorder and Incivility . . . . . . . . Assessments of Techniques and Approaches Developed in the Kings Cross Study . . . . . . . . . . . . . . . . . . . . . . . . . General Summary of the Kings Cross Study . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . .
155 155 156 156 156 159 162 162 164 168 168 168 172 173 173 177 181 188
. . . . . .
221 221 223
. . . . . .
240 242 243
. . . . .
251
. . . . .
251
. . . . .
254
. . . .
. . . .
256 261 262 263
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
269
8 Future Avenues for Fear Mapping: Potential Applications and Improvements . . . . . . . . . . . . . . . . . . . . . . . Has Collective Avoidance Behaviour Changed in Wollongong and Kings Cross? . . . . . . . . . . . . . . . . . . . . . . . . Investigating Behavioural Responses in Relation to Different Types of Crime . . . . . . . . . . . . . . . . . . . . . . . . . Further Avenues for Investigating Links Between Fear, Crime and Disorder . . . . . . . . . . . . . . . . . . . . . . . . . . . Broken Windows Theory in the Transit Context . . . . . . . . Fear Mapping and Advances in Spatial Technology . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . .
. . . .
. . . . . . . . . . . . . . . . . .
149 151
. . . .
Chapter 1
Introduction
The Emergence of Fear of Crime As an Area of Research The fear of crime first began to emerge as an issue of concern in the mid-1960s when national public opinion polls in the United States began to incorporate open-ended questions relating to the public perception of crime (Furstenberg, 1971; McIntyre, 1967; Poveda, 1972). Furstenberg (1971) notes that it is difficult to pinpoint exactly when the issue began to gain momentum but broadly links this to a general concern about crime, and racial and economic conflict in the 10 years prior to the 1970s. Before this, crime had only been given slight attention in public opinion polls (McIntyre, 1967), with the surveys conducted in 1966 by the President’s Commission on Crime providing virtually the only source of information on the public reaction to crime (Furstenberg, 1971). The findings from these surveys were published in a large volume entitled “The Challenge of Crime in a Free Society”, which involved the work of numerous commissioners, staff members of the President’s Commission on Crime and consultants from every part of America (PCLEAJ, 1967). The report was forthright in arguing that the fear of crime was eroding the basic quality of life of many Americans. Studies in two high-crime areas showed that fear of crime was causing 43% of respondents to stay off the streets at night, 35% to not speak to neighbours and 21% to use cars or cabs at night. In addition, 20% of respondents said they would like to move to another neighbourhood because of their fear of crime. The findings from the national survey were generally found to support the results from these local studies with one-third of a representative sample of Americans stating they felt unsafe to walk alone in their neighbourhoods at night. One-third of respondents also said they kept firearms or watchdogs for protection against criminals. The report also found that fear of crime varied according to race, income, sex and experience of victimization. Women, people of non-Caucasian origin and of lower income levels were found to have the highest average scores of fear. The report emphasized that a number of the findings were less intuitive than would be imagined. Fear of crime was found to be less closely associated with having been a victim of crime than might have been supposed. On a broader level, fear of crime was not always highest in areas that had high rates of crime, according to official
B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_1, C Springer Science+Business Media, LLC 2012
1
2
1
Introduction
crime data, or victimization surveys. People were also found to be most fearful of the types of crimes that occurred least frequently. The general conclusions of the report were alarming. The commission felt that it could not state that the public’s fear of crime was exaggerated and concluded that people’s fears must be respected. Further, fear of crime was seen as a complex response, not simply a fear of death or injury but, at bottom, a fear of strangers. This was seen as one of the most dangerous aspects of the fear of crime as it damaged social order and, by reducing the level of sociability and mutual trust, could indeed make streets and public places more dangerous. The results from the study provided the impetus for further investigation (e.g. Conklin, 1971; Furstenberg, 1971; McIntyre, 1967; Poveda, 1972; Brooks, 1974). The findings from other national level surveys such as Gallup and Harris polls supported the general results from the President’s Commission on Crime (PCLEAJ, 1967) report and also showed that fear of crime had risen steadily since 1965 (Erskine, 1974; McIntyre, 1967). The midto-late 1970s saw a plethora of studies looking specifically into the fear of crime (e.g. Brooks, 1974; Clemente, 1977; Balkin, 1979; Hartnagel, 1979; Thomas and Hyman, 1977).
The Paradoxical Nature of the Fear of Crime The focus of much early research into the fear of crime centred on the degree to which fear was seen to be rational or irrational in relation to the actual occurrence of crime (e.g. Poveda, 1972; Brooks, 1974; Balkin, 1979). While fear of crime is not always negative, provoking people to protect themselves when they are threatened, it becomes problematic when out of proportion with the objective risks of victimization (Clark, 2003; Warr, 2000). Results from public opinion polls frequently showed that high levels of fear were being recorded not only in areas characterized by high rates of recorded crime, but those with low rates as well (e.g. PCLEAJ, 1967; Furstenberg, 1971). Similarly, the public was generally found to fear most the crimes that occurred least frequently (PCLEAJ, 1967; McIntyre, 1967). At the time, recorded crime rates were seen as an objective measure and the observed inconsistencies between fear of crime and victimization rates were often attributed to irrational individual perceptions (Balkin, 1979). Since then, the mismatch between the fear of and the incidence of crime has been found in numerous broad level studies set in cities in the United Kingdom, Switzerland, New Zealand and Australia (Borooah and Carcach, 1997; Box et al., 1988; Doeksen, 1997; Killias and Clerici, 2000). Even those considering high levels of unreported incidents have found fear of crime to exceed the real risk of crime (Liska et al., 1988; Painter, 1996; Taylor and Hale, 1986). This discrepancy between fear and actual risk has become known as the “paradox of fear” (e.g. Hollway and Jefferson, 1997; Warr, 1984). The paradox is most evident among women and the elderly who, despite experiencing lower rates of victimization, are consistently found to have higher rates of fear (e.g. Smith and Tortensson, 1997; Warr, 1984).
Current Trends in Fear of Crime Research
3
Rather than dismissing the fear of crime as an unwarranted area of research, many researchers have seen the discrepancies between official crime data and fear of crime surveys as justification for the fear of crime to be treated as a serious social problem in its own right (e.g. Poveda, 1972; Brooks, 1974). Garofalo (1981) went so far as to suggest that discussions over the apparent irrationality of fear had become an unnecessary impediment to researchers looking into the phenomenon. These sentiments were echoed much later by Lupton and Tulloch (1999). However, one could argue that this was an extreme stance as many of the studies looking into the paradoxical nature of fear of crime have been helpful in furthering the understanding of the sources of fear of crime among particular groups of society (e.g. Clemente, 1977; Hanson et al., 2000; Smith and Tortensson, 1997; Warr, 1984). Studies such as these essentially seek to answer calls for a deeper knowledge of the determinants of fear, without which many authors have argued fear of crime would remain elusive to address (e.g. Brooks, 1974; Balkin, 1979).
Current Trends in Fear of Crime Research As a research area, the fear of crime is now one of the most researched topics in contemporary criminology (Farrall et al., 2000). It receives considerable attention in other disciplines such as social ecology (e.g. Taylor and Covington, 1993; Wilson Doenges, 2000), social psychology (e.g. Van der Wuff et al., 1989; Farrall et al., 2000) and geography (e.g. Smith, 1987; Valentine, 1989; Pain, 1991, 1997, 2000; Koskela, 1999; Koskela and Pain, 2000; Thomas and Bromley, 2000). Hale (1996) estimated that over 200 articles, monographs or books had been devoted to the fear of crime. Some 15 years later, a search using the Current Contents engine lists over 400 published journal articles of crime between 1993 and 2011 that include the term “fear of crime” in the title or abstract. Further, research into the fear of crime has increased in countries outside of the United States, most noticeably the United Kingdom (e.g. Smith, 1987; Mayhew and White, 1997; Mirrlees-Black and Allen, 1998; Pain, 1997) and Australia (e.g. Brown and Polk, 1996; NCAVAC, 1998; Grabosky, 1995; Tulloch, 2000). There have been far fewer studies of fear of crime in developing nations but this seems to be changing (e.g. Chadee and Ditton, 2003; Karakus et al., 2010; Zhang et al., 2009). The seriousness and extent of the phenomenon is often illustrated by statistics from national or international crime surveys (e.g. Koskela, 1999; Smith, 1987). The findings from such surveys continue to show that 20–30% of people indicate that they feel very unsafe or fairly unsafe while out alone after dark (e.g. Mayhew and White, 1997; Mirrlees-Black and Allen, 1998). For some sectors of society, up to 60% of people report a degree of fear in this situation (e.g. Joseph, 1997; Thomas and Bromley, 2000). When described in these terms, the fear of crime appears to be a problem of truly striking dimensions (Farrall et al., 1997, 2000) which plagues many, if not most, communities (Reid et al., 1998). Scarborough et al. (2010) notes that the consistently identified relationships between demographic characteristics
4
1
Introduction
(e.g. age, race, gender) and fear of crime provide an “enduring frustration” for policy makers as these factors cannot be altered by government policy. Some have suggested that the fear of crime is a problem as great or greater than crime itself (Clemente, 1977; Brown and Polk, 1996; Oc and Tiesdell, 1997). Such claims are based upon the assumption that, in terms of impact upon urban living, perceptions of crime are often more important than the actuality (Oc and Tiesdell, 1997). Unlike crime, fear of crime is not restricted in its distribution in space and time, giving it the potential to be more widespread (Perkins and Taylor, 1996; Smith, 1987). In essence, unlike crime, which requires the convergence of a victim and an offender in time and space (Cohen and Felson, 1979), fear of crime only requires a victim. Further elevating fear of crime is the fact that those not directly victimized are indirectly victimized when they hear of the experiences of others (Covington and Taylor, 1991). A number of authors have noted an increased interest in the fear of crime in policy arenas over more recent years (Smith, 1987; Fishman and Mesch, 1996; Farrall et al., 1997; Keane, 1998; Farrall et al., 2000). Walklate (1998) attributes much of the interest in media and policy circles to the results from broad-scale victimization surveys which give rise to disturbing statements like the oft-quoted assertion that fear of crime causes many people to become prisoners in their own homes (e.g. Joseph, 1997). Such comments are intrinsically disturbing (Box et al., 1988) and demand that efforts be made to alleviate the fear of crime (Clemente, 1977). It is not surprising, therefore, that fear of crime has been paid close attention in political campaigns over time (e.g. Brown and Polk, 1996; Kelling and Coles, 1997). A further factor influencing the relationship between researchers and policy makers is that the motivation for many studies into fear of crime will translate into practical policies for reducing fear (Box et al., 1988). The continued research and interest in the topic reinforces the assertion that fear of crime is an intractable and resistant phenomenon (Nair et al., 1993; Tulloch et al., 1998). Hollway and Jefferson (1997) argue that despite the voluminous literature on fear of crime, it is fair to say that the area remains conceptually undeveloped and that most work remains largely descriptive. To some extent this provides support for Brooks’ (1974) suggestion that, because of its irrational qualities, fear of crime may be more difficult to combat than criminality itself. Garofalo (1981) noted that every advance that is made in the field seems to generate more questions than answers. However, the author also suggested that this should be expected, as part of the nature of complex social phenomena is that their complexity becomes more apparent the more closely they are examined. In general, it seems likely that the fear of crime will continue to remain high on the agendas of researchers and policy makers alike.
References Balkin, S. (1979). “Victimization rates, safety and fear of crime”. Social Problems 26(3): 343–358. Borooah, V. and C. Carcach (1997). “Crime and fear. Evidence from Australia”. The British Journal of Criminology 37(4): 635–657.
References
5
Box, S., C. Hale, et al. (1988). “Explaining fear of crime”. The British Journal of Criminology 37(4): 340–356. Brooks, J. (1974). “The fear of crime in the United States.” Crime and Delinquency 20: 241–245. Brown, M. and K. Polk (1996). “Taking fear of crime seriously: the Tasmanian approach to community crime prevention”. Crime and Delinquency 42(3): 398–420. Chadee, D. and J. Ditton (2003). “Are older people most afraid of crime? Revisiting Ferraro and LaGrange in Trinidad”. The British Journal of Criminology 43(2): 417–433. Clark, J. (2003). “Fear in fear-of-crime”. Psychiatry, Psychology and Law 102(267–282). Clemente, F. (1977). “Fear of crime in the United States: a multivariate analysis”. Social Forces 56(2): 519. Cohen, L. E. and M. Felson (1979). “Social change and crime rate trends: a routine activity approach”. American Sociological Review 44: 588–608. Conklin, J. E. (1971). “Dimensions of community response to the crime problem”. Social Problems 18: 373–385. Covington, J. and R. B. Taylor (1991). “Fear of crime in urban residential neighbourhoods: implications between – and within – neighbourhood sources for current models”. The Sociological Quarterly 32(2): 231–249. Doeksen, H. (1997). “Reducing crime and the fear of crime by reclaiming New Zealand’s suburban street”. Landscape and Urban Planning 39(2–3): 243–252. Erskine, H. (1974). “The polls: fear of violence and crime”. Public Opinion Quarterly 38(1): 131–145. Farrall, S., J. Bannister, et al. (1997). “Questioning the measurement of the ‘fear of crime’: findings from a major methodological study”. The British Journal of Criminology 37(4): 658–679. Farrall, S., J. Bannister, et al. (2000). “Social psychology and the fear of crime”. The British Journal of Criminology 40(3): 399–413. Fishman, G. and G. S. Mesch (1996). “Fear of crime in Israel: a multidimensional approach”. Social Science Quarterly 77(1): 76–89. Furstenberg, F. F., Jr. (1971). Public reaction to crime in the streets “The American Scholar”. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 3–12. Garofalo, J. (1981). “The fear of crime: causes and consequences”. Journal of Criminal Law and Criminology 72(2): 839. Grabosky, P. N. (1995). “Fear of crime, and fear reduction strategies”. Current Issues in Criminal Justice 7(1): 7–19. Hale, C. (1996). “Fear of crime: a review of the literature”. International Review of Victimology 4: 79–150. Hanson, R. F., D. W. Smith, D. G. Kilpatrick and J. R. Freedy (2000). “Crime-related fears and demographic diversity in Los Angeles county after the 1992 civil disturbances”. Journal of Community Psychology 28(6): 607–623. Hartnagel, T. F. (1979). “The perception and fear of crime: implications for neighbourhood cohesion, social activity and community affect”. Social Forces 58(1): 176–193. Hollway, W. and T. Jefferson (1997). “The risk society in an age of anxiety: situating fear of crime”. The British Journal of Sociology 48(2): 255. Joseph, J. (1997). “Fear of crime among black elderly”. Journal of Black Studies 27(5): 698–717. Karakus, O., E. F. McGarrell, et al. (2010). “Fear of crime among citizens of Turkey.” Journal of Criminal Justice 38(2): 174–184. Keane, C. (1998). “Evaluating the influence of fear of crime as an environmental mobility restrictor on women’s routine activities”. Environment and Behavior 30(1): 60–74. Kelling, G. L. and C. M. Coles (1997). Fixing broken windows: restoring order and reducing crime in our communities. New York, NY, Touchstone. Killias, M. and C. Clerici (2000). “Different measures of vulnerability in their relation to different dimensions of fear of crime”. The British Journal of Criminology 40(3): 437–450. Koskela, H. (1999). “Gendered exclusions: women’s fear of violence and changing relations to space”. Geografiska Annaler 81B(2): 111–124.
6
1
Introduction
Koskela, H. and R. Pain (2000). “Revisiting fear and place: women’s fear of attack and the built environment”. Geoforum 31(2): 269–280. Liska, A. E., A. Sanchirico, et al. (1988). “Fear of crime and constrained behavior specifying and estimating a reciprocal effects model”. Social Forces 66(3): 827–838. Lupton, D. and J. Tulloch (1999). “Theorizing fear of crime: beyond the rational/irrational opposition”. The British Journal of Sociology 50(3): 507–523. Mayhew, P. and P. White. (1997). “Home Office Research and Statistics Directorate Findings No 57: The 1996 international crime victimisation survey.” From http://www.homeoffice.gov.uk/ rds//pdfs/r57.pdf. McIntyre, J. (1967). “Public attitudes toward crime and law enforcement”. The Annals of the American Academy (November): 34–46. Mirrlees-Black, C. and J. Allen (1998). Concern about crime: Findings from the 1998 British Crime Survey. Research Findings No 83. London, Home Office Research, Development and Statistics Directorate. Nair, G., J. Ditton, et al. (1993). “Environmental improvements and the fear of crime: the sad case of the ‘Pond’ area in Glasgow”. The British Journal of Criminology 33(4): 555–561. National Campaign Against Violence and Crime (1998). Fear of crime – summary volume. Canberra, NCAVAC. Oc, T. and S. Tiesdell (1997). Safer city centres: reviving the public realm. London, Chapman. Pain, R. (1991). “Space, sexual violence and social control: integrating geographical and feminist analyses of women’s fear of crime.” Progress in Human Geography 15(4): 415–431. Pain, R. (2000). “Place, social relations and the fear of crime: a review.” Progress in Human Geography 24(3): 365–387. Pain, R. H. (1997). “ ‘Old age’ and ageism in urban research: the case of fear of crime”. International Journal of Urban & Regional Research 21(1): 117–128. Painter, K. (1996). “The influence of street lighting improvements on crime, fear and pedestrian street use, after dark”. Landscape and Urban Planning 35(2–3): 193–201. Perkins, D. D. and R. B. Taylor (1996). “Ecological assessments of community disorder: their relationship to fear of crime and theoretical implications”. American Journal of Community Psychology 24(1): 63–107. Poveda, T. G. (1972). “The fear of crime in a small American town”. Crime and Delinquency 18(2): 147–153. President’s Commission on Law Enforcement and Administration of Justice (1967). The Challenge of Crime in Free Society. Washington United States Government Printing Office. Reid, L., J. T. Roberts, et al. (1998). “Fear of crime and collective action: an analysis of coping strategies”. Sociological Inquiry 68(3): 312–328. Scarborough, B. K., T. Z. Like-Haislip, et al. (2010). “Assessing the relationship between individual characteristics, neighborhood context, and fear of crime”. Journal of Criminal Justice 38(4): 819–826. Smith, S. J. (1987). “Fear of crime: beyond a geography of deviance”. Progress in Human Geography 11: 1–23. Smith, W. R. and M. Tortensson (1997). “Gender differences in risk perception and neutralising fear of crime”. The British Journal of Criminology 37(4): 603–634. Taylor, R. B. and J. Covington (1993). “Community structural change and fear of crime”. Social Problems 40(3): 374–395. Taylor, R. and M. Hale (1986). “Criminology: testing alternative models of fear of crime”. Journal of Criminal Law and Criminology 77: 151–189. Thomas, C. and R. Bromley (2000). “City-centre revitalisation: problems of fragmentation and fear in the evening and night-time city”. Urban Studies 37(8): 1403–1429. Thomas, C. W. and J. M. Hyman (1977). “Perceptions of crime, fear of victimization and public perceptions of police performance”. Journal of Police Science and Administration 5(3): 305–317. Tulloch, M. (2000). “The meaning of age differences in the fear of crime.” The British Journal of Criminology 40(3): 451–467.
References
7
Tulloch, J., D. Lupton, et al. (1998). Fear of crime volume one. Canberra, Centre for Cross Cultural Research. Valentine, G. (1989). “The geography of women’s fear”. Area 21(4): 385–390. van der Wuff, A., L. van Staaldiunen, et al. (1989). Fear of crime in residential environments: testing a social psychological model. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 395–414. Walklate, S. (1998). “Crime and community: fear or trust?”. The British Journal of Sociology 49(4): 550–569. Warr, M. (1984). “Fear of victimization: why are women and the elderly more afraid”. Social Science Quarterly 65(3): 681–702. Warr, M. (2000). “Fear of crime in the United States: avenues for research and policy.” Criminal Justice 4: 452–489. Warr, M. and C. G. Ellison (2000). “Rethinking social reactions to crime: personal and altruistic fear in family households”. American Journal of Sociology 106(3): 551–578. Wilson-Doenges, G. (2000). “An exploration of sense of community and fear of crime in gated communities”. Environment and Behavior 32(5): 597–611. Zhang, L. N., S. F. Messner, et al. (2009). “Guanxi and fear of crime in contemporary urban China”. The British Journal of Criminology 49(4): 472–490.
Chapter 2
Why Is Fear of Crime a Serious Social Problem?
Individual Reactions There is a general consensus in the literature that the most significant effect of fear of crime is the reduced quality of life it imposes on those affected by it (Bannister and Fyfe, 2001; Box et al., 1988; Brown and Polk, 1996; Fisher and Nasar, 1992; Grabosky, 1995; Green et al., 2002; Fishman and Mesch, 1996; Mirrlees-Black and Allen, 1998; Nasar et al., 1993; Wilson-Doenges, 2000; Oc and Tiesdell, 1997; Tiesdell and Oc, 1998). The impact of fear of crime ranges from detrimental physiological changes to psychological reactions and behavioural adaptations. In terms of physiological changes, fear of crime is associated with increased heart rate, rapid breathing, decreased salivation and increased galvanic skin response (Warr, 2000). Endocrinic changes, such as the release of adrenaline into the bloodstream, may also occur to prepare us for a ‘fight or flight’ response (Skogan and Maxfield, 1981). Additonally, according to Kovecses (1990), fear is more generally associated with physical agitation; increased heart rate; lapses in heart beat; blood leaving face; shrinking of skin; straightening of hair; drop in body temperature; inability to move, breathe or speak; involuntary releases of bowels or bladder; sweating; nervousness; and dryness in the mouth. From a psychological perspective, fear of crime can produce negative feelings of anger, outrage, frustration, violation and helplessness (Ferraro and LaGrange, 2000; Warr, 2000). These feelings can extend to those of anxiety, distrust of others, alienation and dissatisfaction with life (Miceli et al., 2004; Morrall et al., 2010). Fear of crime is also strongly correlated with mental health and sometimes triggers mental illness (Green et al., 2002; Miceli et al., 2004), which in more acute or chronic cases can lead to advanced states of depression and long-term trauma (Ferraro and LaGrange, 2000; Spelman, 2004). Alongside these wide-ranging physiological and psychological effects, fear of crime can prompt people to change their behaviour. At the level of the individual, people generally respond to the fear of crime by adopting protective or avoidance behaviours (Box et al., 1988; Keane, 1998; Liska et al., 1988; Reid et al., 1998; Riger et al., 1982; Warr, 1985). The structural constraints and role obligations dictated by lifestyles and routine daily activities may circumscribe people’s ability to use precautionary tactics such as avoidance behaviours (Riger et al., 1982). Under
B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_2, C Springer Science+Business Media, LLC 2012
9
10
2
Why Is Fear of Crime a Serious Social Problem?
these conditions, it appears that people are more likely to adopt protective measures, such as carrying a weapon, learning self-defence techniques, installing anti-burglary equipment or acquiring watch dogs (Cubbage and Smith, 2009; Krahn and Kennedy, 1985; Liska et al., 1988). Nasar et al. (1993) and Nasar and Jones (1997) conducted a series of investigations into the fear of crime at the Ohio State University campus which had a focus on protective and avoidance behaviours. The studies revealed that the campus was characterized by a climate of fear (Nasar and Jones, 1997), as 50% of survey respondents expressed safety concerns about routes they used on campus, while 73% indicated that they avoided areas they deemed unsafe (Nasar et al., 1993). When asked if they would carry some form of protection if they had to walk a particular route at night, 91% of the sample said they would (Nasar and Jones, 1997). On a broader scale, Teske and Arnold (1991) discuss results from a comparative victimization study in the United States and the Federal Republic of Germany which further indicate that people in a climate of fear are more likely to adopt protective measures. The authors found that survey respondents from Texas were 12 times more likely to have a gun in their houses for security purposes and were generally more likely to have installed security devices than respondents from Baden-Württemberg. The authors emphasize that Texas respondents were much more likely to have been the victims of a burglary, to know victims of a burglary and to feel that they may be victims of a burglary in the next year. In contrast to protective measures, avoidance behaviour primarily aims to reduce the risk of individuals being exposed to victimization, rather than reduce the risk of being victimized when exposed to threat (Skogan and Maxfield, 1981). Avoidance strategies often cause people to restrict their behaviour to places or times perceived to be safe or avoid certain activities they may perceive as dangerous, such as travelling by public transport, walking on certain streets or attending social activities (Box et al., 1988; Liska et al., 1988; Pantazis, 2000). Such behaviour, despite being a rational human reaction (Oc and Tiesdell, 1997), leads people to remove themselves from social activities and increases levels of distrust for others (Smith, 1987; Ross and Mirowsky, 2000; Wilson-Doenges, 2000). Keane (1998) investigated the influence of fear of crime as an environmental mobility restrictor on women’s routine movements. He found that a significant number of women were worried about walking alone in their area after dark and walking alone to their cars in a parking area. Of these women, a considerable number reported that they would change their behaviour and walk alone in their neighbourhoods and use parking areas more often if they felt safer. Keane (1998) concluded that increasing feelings of safety would increase women’s lifestyle choices and freedom of movement. Similar evidence for avoidance behaviours having a negative impact on the quality of people’s lives has been found by Liska et al. (1988). The authors found that constrained or avoidance behaviour increased, rather than decreased, fear. They suggest that avoidance behaviours may serve to decrease emotion-based fear in a dangerous situation, but may accentuate risk-based fear associated with anticipating a dangerous situation. Pantazis (2000) has likened the patterns associated with avoidance behaviours to current debates on poverty and social exclusion, which focus on people’s ability
Hypothesized Links Between the Fear of Crime, Disorder and Crime
11
to participate in activities that others take for granted. A further parallel between social exclusion and the fear of crime relates to the unequal impact these problems have upon different elements of society. In general, levels of crime and poverty are higher among groups in society that experience a greater degree of social exclusion (Brennan et al., 2000; Hirschfield and Bowers, 1997). In a similar vein, the fear of crime has been consistently found to be higher in the poorest and most deprived neighbourhoods (Smith, 1987) and among women, the elderly and those with less education (e.g. Ferraro, 1995; Garofalo, 1979; Smith and Hill, 1991; Thomas and Bromley, 2000; Warr, 1984). Indeed, there is a common assertion that older people are prone to becoming “prisoners of fear” (Joseph, 1997; Pain, 2000; Stephens, 1999). Thus, the avoidance behaviours that individuals adopt in relation to their fear of crime have the potential to exert a substantial effect on the autonomy of many social groups and are a worthy area for ongoing research. However, the influence of such responses is not contained to the level of the individual, as fear of crime and the behavioural adaptations it prompts can have wide-ranging impact at the community level.
Hypothesized Links Between the Fear of Crime, Disorder and Crime In their widely quoted1 paper titled ‘Broken Windows’, Wilson and Kelling (1982) put forth a theory outlining a negative feedback loop whereby unchecked incivilities and disorder not only lead to fear of crime, but also crime itself. Using the broken window as a symbol for all types of disorder, their account of this causal relationship between disorder, fear and crime is now commonly referred to as the broken windows hypothesis or thesis (e.g. Harcourt, 1998; Sampson and Raudenbush, 1999; Loukaitou-Sideris, 1999). Broken windows hypothesis has proven highly influential in subsequent research and policy developments (e.g. Bratton, 1995, 1996; Skogan, 1990; Taylor and Covington, 1993; Tiesdell and Oc, 1998; Sampson and Raudenbush, 1999). The underlying tenet of the broken windows hypothesis is based on the assumption that if a window is broken and left unrepaired (or disorder is left unchecked) then more windows will be broken (more disorder will occur) (Wilson and Kelling, 1982). The authors of the thesis draw on the incivilities/disorder hypothesis to suggest that an unrepaired broken window (untended disorderly behaviour) becomes a signal that no one cares and leads to a breakdown in community controls. This lack of response creates conditions under which social and physical disorder can flourish. Responding prudently and fearfully, both residents and passers-by perceive these areas as uncontrolled and unsafe. They accordingly change their activities to stay
1 For additional information and interpretations see Doran and Lees, 2005; Gibbons, 2004; Greene, 1999; Harcourt, 1998; Millie and Herrington, 2005.
12
2
Why Is Fear of Crime a Serious Social Problem?
off the streets and avoid areas perceived as unsafe. By doing so, the general public relinquish their roles of mutual support with fellow citizens and weaken forms of informal social control such as natural surveillance. Where the social fabric of a neighbourhood is undermined in this way, criminals, both opportunistic and professional, believe they have reduced chances of being caught or identified and will consequently operate more actively or invade the area (Wilson and Kelling, 1982). This leads to an influx of criminals, increased social and physical disorder and eventually the onset of serious crime. Various studies have supported the notion that social and physical incivilities and the presence of serious crime may act to increase the fear of crime (e.g. Borooah and Carcach, 1997; Covington and Taylor, 1991; Perkins and Taylor, 1996; Rountree and Land, 1996; Taylor and Covington, 1993). Thus, the fear of crime can be seen as one of the first steps in a positive feedback loop, because it results in residents adopting protective and avoidance behaviours which contribute to the breakdown of informal social control, more fear of crime and crime itself. This feedback cycle is illustrated in Fig. 2.1 below. There has been considerable debate over the validity of the broken windows hypothesis. Many researchers and practitioners readily accept the theory and it has therefore had considerable influence on research, policy and practice (see Doran and Lees, 2005; Harcourt, 1998; Stephens, 1999; Xu et al., 2005). The elements of broken windows hypothesis have also been used as a basis for the disorder and decline hypothesis (Skogan, 1986, 1990) which is described in more detail below. However, numerous critics also discount the fundamental assumptions of the broken windows
A ‘broken window’ is left unrepaired
This signals a breakdown in informal social controls
People physically and socially withdrawl from the community, avoiding uncontrollable/unsafe areas
More disorderly behaviour & broken windows (increased social and physical incivilities)
People become afraid of crime
There is an influx of criminals & more serious crime
Fig. 2.1 Flow chart illustrating the cycle of the broken windows hypothesis, highlighting the role of fear of crime
Disorder and Decline Hypothesis
13
hypothesis (e.g. Bowling, 1999; Greene, 1999; Harcourt, 1998; Taylor, 2001). Innes (2004) contends that there is a serious lack of empirical evidence supporting the thesis. Harcourt (1998) criticizes broken windows hypothesis and policing strategies based on it, highlighting the fact that they neglect numerous other complex factors that also contribute to crime. The proposition that people respond equally to both ‘broken windows’ and ‘broken people’ has also been challenged (Innes, 2004). One avenue that has not yet been explored thoroughly comprises the spatial and temporal components of the hypothesis – many of the links outlined in the cycle relate to the areas where social and physical disorder become concentrated, or the general public adopt behaviours which, over time, create conditions under which crime can flourish. The spatial and temporal scales at which these processes are likely to be operating are likely to vary considerably from short term (hours or days) to much longer term (years).
Disorder and Decline Hypothesis Skogan’s (1986, 1990) disorder and decline hypothesis expands upon the broken windows hypothesis (see Fig. 2.2 below). Like the broken windows hypothesis, the disorder and decline hypothesis begins with the justification that people gather information about the level of crime and safety in their neighbourhood through environmental cues (Skogan and Maxfield, 1981). Skogan (1990) maintains that signs of disorder are associated with high levels of risk and imply that neighbourhood systems of social control have broken down.2 When people encounter signs of disorder they physically withdraw from those areas, confining their activities to those times and routes perceived as the safest. This reduces the amount of informal social surveillance that occurs naturally with pedestrian activity (Skogan, 1986; Skogan and Maxfield, 1981). However unlike Wilson and Kelling, Skogan elaborates on the added psychological withdrawal of residents from the streets (Skogan, 1986). Skogan and Maxfield (1981) assert that crime and disorder, through fear of crime, generate suspicion and distrust. This, in turn, has an atomising effect upon individuals and households (Skogan and Maxfield, 1981).3 Skogan then argues that disorder restricts the neighbourhood potential for organizational life and mobilization (Skogan, 1986). In addition Skogan (1986) emphasizes spatial considerations and proposes that perceptions of disorder could cause a decrease in the geographic area that people feel responsible for. This further serves to weaken community mechanisms of
2
Skogan (1990) specifically defines disorder as ‘direct, behavioral evidence of disorganization’. Crime and disorder undermine people’s trust that their neighbours share common goals and norms (Skogan and Maxfield, 1981). This can lead to hostility and antipathy (Skogan, 1990). Disorder reduces resident confidence that their individual and collective actions can overcome disorder, (Skogan, 1990; Skogan and Maxfield, 1981). 3
14
2
Disorder / Incivilities
Interpreted as a breakdown in community controls
People become afraid of crime
Why Is Fear of Crime a Serious Social Problem?
Demographic collapse of the neighborhood
Deteriorating business conditions & local housing market
There is an influx criminals & crime
Physical & psychological withdrawal of the community
Delinquency and deviance among youth
Weakening of the processes of informal social control
Decline in the organizational life of the neighborhood
Fig. 2.2 Flow chart illustrating the disorder and decline hypothesis
informal social control and surveillance.4 With a decrease in social control and community-level capacity to combat disorder, Skogan mirrors Wilson and Kelling’s argument in stating the neighbourhood will invite ‘outside troublemakers’ who bring additional crime and disorder (Skogan, 1986). Skogan also elaborates on the economic impact of disorder on affected neighbourhoods. The first point he makes is in relation to a deterioration of local business conditions (Skogan, 1986). With fewer people on the streets, there will be fewer business customers resulting in shops being forced to close down. These empty shops are likely to remain abandoned
4 Skogan explains this using the concept of ‘territoriality’, which is a ‘set of attitudes and behaviours regarding the regulation of the boundary that surrounds people’s personal household space’ (Skogan, 1986). He claims that with healthy levels of territoriality residents will conduct surveillance over a wide area (Skogan, 1986). Surveillance is facilitated by personal recognition of one’s neighbours and a belief that local standards of appropriate public behaviour are widely shared (Skogan, 1990). These factors diminish, thereby negating the underlying necessities for social surveillance and the psychological defence of public space.
Disorder and Decline Hypothesis
15
or be converted to non-retail establishments. Economic forces favour those traditionally ‘unsavoury’ businesses, such as bars, transient hotels, x-rated outlets and massage parlours. The author argues that these businesses, and the ‘unsavoury’ people they attract, will further decrease the desirability of the area for people with a low tolerance for disorder (Skogan, 1986). Skogan’s second assumption is that, with an increasingly bad reputation, the local housing market becomes unstable (Skogan, 1990). Residents who are able to move relocate to other areas, and fewer people want to move into or invest in the area. Skogan states that this leads to a downward turn in the real estate market of affected areas and causes further deterioration and abandonment of buildings (Skogan, 1990).5 At this point, the disorder and decline hypothesis implies that disorder and these consequent social and economic problems continue to ‘feed on themselves, spiralling neighbourhoods deeper into decline’ (Skogan, 1986). Feedback processes ensure fear of crime increases until it is ‘incapacitating’ (Skogan, 1986; Skogan and Maxfield, 1981). The end of this cycle is characterized by a demographic collapse of the neighbourhood, when crime and disorder continue but there are few residents left to define it as a problem (Skogan, 1986). Schuerman and Kobrin (1986) argue that those areas characterized by at least three decades of high crime are ‘lost territory to the rest of society’ (in Skogan, 1986). Skogan (1990) cemented his theory on the links between disorder and serious crime with empirical research. Disorder was linked more strongly with higher crime levels than were other neighbourhood characteristics, such as poverty, instability in the housing market, and predominantly minority racial composition among residents. Further, the investigation found that disorder, both directly and as a precursor to crime, played an important role in neighbourhood decline. A number of researchers have supported Skogan’s (1990) findings. For example Borooah and Carcach (1997) investigated fear of personal and housing crime in relation to a common set of explanatory variables. The authors concluded that lack of neighbourhood cohesion, neighbourhood incivility and perception of relatively high neighbourhood crime levels contributed significantly to the probability of being afraid and to the risk of victimization. Similarly, in their own study, Ross and Mirowsky (2000) declare disorder and decay are highly correlated with crime and share many indicators. Kelling and Coles (1997) also stated that Skogan’s research supports the broken windows hypothesis. Thus some researchers have also concluded that fear of crime creates an environment where crime is more likely (Millie and Herrington, 2005). Others have also gone so far as to say that fear of crime is now a larger problem than crime itself (Bennett, 1991; Farrall et al., 2000; Warr, 1984; Hale, 1996). In contrast, Markowitz et al. (2001) point out that studies supporting the broken windows and disorder and decline theories are largely based on cross-sectional data. As the theory is longitudinal in nature more evidence is necessary to confirm
5 Nevertheless, Skogan does recognize that other factors play an important role in determining demand for property (Skogan, 1986).
16
2
Why Is Fear of Crime a Serious Social Problem?
the causal effect of disorder. However, Markowitz et al. (2001) do acknowledge that disorder may increase crime indirectly through its effect in increasing fear of crime and decreasing social involvement and collective efficacy. Harcourt (1998) also found that Skogan’s data did not support the claim that crime is related to disorder. While Harcourt confirmed that certain crimes like physical assault and robbery are at first significantly related to disorder, he argues that this relationship disappears when the variables of neighbourhood poverty, stability and race are held constant. Similarly, Sampson and Raudenbush (1999) did not find convincing evidence to support the strong versions of the broken windows or disorder and declines theories. Disorder was only a moderate correlate of predatory crime, and varied consistently with antecedent neighbourhood characteristics. Despite the lack of evidence for a direct association between disorder and crime, the authors suggest that if disorder operates in a cascading fashion by undermining residential stability and discouraging efforts of building collective responses, it would indirectly have an effect on crime. While emphasizing that it is not the disorder that causes the crime, but rather poor social control that causes both, this scenario is essentially the same as that outlined by Skogan (1990), where fear of crime plays an important role in determining the actions of residents within a community.
Economic Impact of Behavioural Responses to Fear of Crime The potential for the fear of crime to have a negative economic impact upon society has been recognized by a number of authors other than Skogan in his disorder and decline hypothesis (e.g. Brown and Polk, 1996; Grabosky, 1995; Hamermesh, 1999b; Liska et al., 1988; Oc and Tiesdell, 1997). Individuals who respond to the fear of crime by adopting avoidance behaviours incur a cost to both themselves and society (Oc and Tiesdell, 1997), as they keep away from the restaurants, shops, jobs and residences located in areas they perceive as dangerous (Liska et al., 1988). The opportunity costs associated with such behaviour, while difficult to quantify, are likely to be substantial (Oc and Tiesdell, 1997; Ayers and Levitt, 1998). Jackson and Gray (2010) note that there can be ‘hidden costs’ associated with such actions, through spending time or money on protective measures. A number of researchers have paralleled Skogan’s assertion that fear of crime has a negative impact on the housing market as a result of discouraging homebuyers and causing out migration (Katzman, 1980 in Smith, 1987; Gibbons, 2004; Oc and Tiesdell, 1997). Retail businesses suffer a shortage of customers as the most affluent people leave the neighbourhood and people generally avoid the streets (Conklin, 1971; Oc and Tiesdell, 1997). In turn businesses close down, relocate and new investment is suppressed, further reducing the activity and attraction of the area (Garofalo, 1981; Spelman, 2004; Oc and Tiesdell, 1997). The negative economic impact associated with the avoidance of retail areas has been linked to the attraction of youths to such environments. For example, Brown and Polk (1996) discuss what they term the ‘mall problem’ in Australia. By providing a day and night gathering and entertainment venue, shopping malls often prove an attractive environment for
Economic Impact of Behavioural Responses to Fear of Crime
17
unemployed and disengaged youths. This frequently results in malls becoming associated with problems, such as drinking, abusive language, fighting and drug use. The authors argue that such behaviours serve to work against the intended commercial function of malls by frightening away potential customers. A number of authors have identified similar trends in Britain (e.g. Oc and Tiesdell, 1997; Thomas and Bromley, 2000; Tiesdell and Oc, 1998). Thomas and Bromley (2000) observe that, despite the fact that many British cities have a thriving night-time economy, entertainment is largely centred around the ‘pub-and-club’ youth culture. The authors argue that the association of youth with threatening behaviour, such as heavy drinking, drugs and violent incidents has reduced the attraction of many city centres for a broader spectrum of the population. Oc and Tiesdell (1997) suggest that this denies large numbers of men and even greater numbers of women the use of city centres at night and has a significant economic and employment cost. On a broader scale, Warr and Ellison (2000) state that fear of crime and the consequent avoidance of dangerous places is so common and recognized in urban areas that it affects the ecology and economies of US cities. Avoidance behaviours resulting from safety concerns may lead to mass cancellations and financial problems in tourist destinations (Ferraro, 1995; Mawby et al., 2000). Brunt et al. (2000: 422) found in a survey of British holidaymakers that 42% of respondents said they had ruled out at least one country because of crime-related problems. Cothran and Cothran (1998) term this dependence of tourism demand upon perceptions of safety the ‘safety elasticity of demand’. The authors argue that tourism is a discretionary activity and, no matter how attractive a destination is, tourists will stay away if they feel their safety cannot be guaranteed. In the case of Mexico, they suggest that if American tourists began to act upon increasing levels of fear of crime by visiting alternative destinations the results for the Mexican tourist industry would be disastrous. Hamermesh (1999a) investigated the timing of work in the United States and found that work in the evenings and at night had declined sharply between the 1970s and 1990s. Using the assumption that fear of crime is most likely to have an effect during the evening and at night, Hamermesh (1999b) investigated the effect of crime and the fear of crime on the timing of work. The author found that higher homicide rates significantly deterred working in the evening and at night and argued that criminal activity imposes a negative externality on the labour market because crime, or the fear of crime, generates departures from optimal patterns of work timing. The author describes this behaviour in terms of a trade-off where higher crime rates reduce the incentive to labour to the point where it becomes insufficient for some of the workers to overcome their fear of crime. This impacts upon workers as they implicitly forego some earnings, and affects society because production shifts away from times when the marginal worker will be more productive. The author estimates that the impact of homicide rates on work timing costs the USA between $4 and $10 billion a year. Protective behaviours can also have direct economic impact on individuals and communities. Target hardening through the use of various security measures in fortifying their homes and places of work, such as outside lighting systems, watch dogs,
18
2
Why Is Fear of Crime a Serious Social Problem?
extra locks and weapons (Liska et al., 1988; Skogan and Maxfield, 1981; Teske and Arnold, 1991) incur a direct cost to the individual. Helsley and Strange (1999) suggest that fear of crime in the United States has led to increased spending on private security. Ayers and Levitt (1998) emphasize the fact that private expenditure on selfprotection potentially dwarfs the $100 billion spent on criminal justice each year in the United States. Not only does fear of crime affect the economies of the local neighbourhood and individuals, but also that of the wider government. Schemes designed by governments to reduce the fear of crime also involve significant cost. For example, investment in CCTV surveillance systems by central and local British government between 1994 and 1997 has been estimated to be in excess of £100 million (Norris and Armstrong, 1998 in Ditton 2000). There are significant time and monetary costs associated with increased public policing in affected communities (Murray et al., 2001). State or local council resources are also used in the upkeep of affected areas and the management of disorder. The firms providing security measures could be seen as deriving economic benefit from the fear of crime. Indeed Davis (1990) goes so far as to suggest that the market provision of security generates its own paranoid demand. Others express less extreme views but nonetheless attribute part of the rapid growth in the security industry to the fear of crime (e.g. Lymes, 1997; Helsley and Strange, 1999). The avoidance and protective behaviours that people adopt to cope with the fear of crime have the potential to generate negative, and in some cases, positive externalities. People who perceive that their neighbourhood is deteriorating often act on their fear of crime and choose to leave the city (Kelling and Coles, 1997). Where this takes place, the people and firms that reallocate their activities burden society with an indirect monetary cost (Hamermesh, 1999b). People remaining in areas where more prosperous citizens have left potentially lack the resources to protect themselves against crime. For example, Dililio (1996) argues that the relative lack of financial and political resources experienced by law-abiding people in inner-city black communities in the United States limits their ability to target-harden their homes, stores, parks and schools and may be partly responsible for the high rates of criminal victimization in these communities. Other studies have established strong links between the concentration of economic disadvantage and crime (Krivo and Peterson, 1996; Weatherburn et al., 1999). Freeman et al. (1996) suggest that the spatial concentration of crime in poor neighbourhoods is based on a positive externality that criminals create for each other. The externality exists because, if police resources are held constant, criminals stand a smaller chance of being caught if there are more of them in an area. Protective measures have also been linked to the redistribution of crime between communities. For example, Helsley and Strange (1999) argue that protective actions such as the building of gated communities or the adoption of target-hardening procedures have the sole objective of diverting or deterring criminals ex ante and have the potential to impose negative externalities which impact upon other sections of society. The authors investigate a number of aspects of gating on the level and spatial distribution of crime with the key result being that gating, by diverting crime to the business district, can reduce legitimate employment opportunities and increase
References
19
the number of active criminals and the aggregate level of crime. Ayers and Levitt (1998) investigated the effect of Lojack, a small, unobservable radio transmitter hidden within vehicles, and found that its use yields positive externalities through general deterrence. However, the authors note that as most forms of personal protective measures are highly visible they are more likely to redistribute, rather than reduce, the occurrence of crime. Hence protective measures that generate positive externalities are likely to be in the minority.
Chapter Review: Potential Problems Not to Be Ignored and a Need for Spatially Explicit Research It is commonly accepted that fear of crime is a major social problem (Liska et al., 1988). Studies have confirmed that fear of crime disrupts neighbourhood cohesion (Nasar et al., 1993); fractures the sense of community and neighbourhood (Box et al., 1988; Ross and Mirowsky, 2000); creates interpersonal distrust (Garofalo, 1981); breaks down social relations and attachment (Spelman, 2004); leads to social isolation (Doeksen, 1997; Ross and Mirowsky, 2000); adds to an erosion of social control and social order (Ross and Mirowsky, 2000); damages the public image of a community and causes avoidance behaviour in potential visitors (Doeksen, 1997; Nasar et al., 1993; Skogan, 1990; Warr, 2000); and causes a removal of ‘eyes on the street’ and informal natural surveillance (Jacobs, 1961; Painter, 1996; Samuels and Judd, 2002). A common thread running through these varied and serious impacts are the protective and avoidance behaviours that people adopt in relation to their fear of crime. The well-known broken windows hypothesis (Wilson and Kelling, 1982) and Skogan’s (1986, 1990) disorder and decline hypothesis have provided theoretical frameworks which outline potential interactions over space and time between crime, disorder and fear. Despite rigorous debate about the efficacy of such hypotheses, there is a consensus among much extant research that fear of crime and the associated protective and avoidance behaviours evident at the individual level have the potential to have a collective and detrimental impact at the community level. Given the heavy emphasis of temporal factors and potential impact in specific areas or neighbourhoods, it is also clear that there are avenues for explicitly spatial research into the hypothesized links between crime, disorder and fear.
References Ayers, I. and S. D. Levitt (1998). “Measuring positive externalities from unobservable victim precaution: an empirical analysis of Lojack”. The Quarterly Journal of Economics 113(1): 43–77. Bannister, J. and N. Fyfe (2001). “Introduction: fear and the city”. Urban Studies 38(5–6): 807–813. Bennett, T. (1991). “The effectiveness of a police-initiated fear-reducing strategy”. The British Journal of Criminology 31(1): 1–14.
20
2
Why Is Fear of Crime a Serious Social Problem?
Borooah, V. and C. Carcach (1997). “Crime and fear. Evidence from Australia”. The British Journal of Criminology 37(4): 635–657. Bowling, B. (1999). “The rise and fall of New York murder: zero tolerance or crack’s decline?”. The British Journal of Criminology 39(4): 531–554. Box, S., C. Hale, et al. (1988). “Explaining fear of crime”. The British Journal of Criminology 37(4): 340–356. Bratton, W. J. (1995). Great expectations: how higher expectations for police departments can lead to a decrease in crime. Paper presented to the National Institute of Justice Policing Research Institute “Measuring what matters conference”. Washington DC, 28 November. Bratton, W. J. (1996). Cutting crime and restoring order: what America can learn from New York’s finest. Heritage Lecture 573. Brennan, A., J. Rhodes and P. Tyler (2000). “The nature of social exclusion in England and the role of the labour market”. Oxford Review of Economic Policy 16(1): 129–146. Brown, M. and K. Polk (1996). “Taking fear of crime seriously: the Tasmanian approach to community crime prevention”. Crime and Delinquency 42(3): 398–420. Brunt, P., R. Mawby, et al. (2000). “Tourist victimisation and the fear of crime on holiday”. Tourism Management 21(4): 417–424. Conklin, J. E. (1971). “Dimensions of community response to the crime problem”. Social Problems 18: 373–385. Cothran, D. A. and C. C. Cothran (1998). “Promise or political risk for Mexican tourism”. Annals of Tourism Research 25(2): 477–497. Covington, J. and R. B. Taylor (1991). “Fear of crime in urban residential neighbourhoods: implications between – and within – neighbourhood sources for current models”. The Sociological Quarterly 32(2): 231–249. Cubbage, C. J. and C. L. Smith (2009). “The function of security in reducing women’s fear of crime in open public spaces: a case study of serial sex attacks at a Western Australian university”. Security Journal 22(1): 73–86. Davis, M. (1990). City of quartz: excavating the future in Los Angeles. New York, NY, Verso. Dililio, J. J. (1996). “Help wanted: economists, crime and public policy”. Journal of Economic Perspectives 10(1): 3–24. Doeksen, H. (1997). “Reducing crime and the fear of crime by reclaiming New Zealand’s suburban street”. Landscape and Urban Planning 39(2–3): 243–252. Doran, B. J. and B. G. Lees (2005). “Investigating the spatiotemporal links between disorder, crime, and the fear of crime”. Professional Geographer 57(1): 1–12. Farrall, S., J. Bannister, et al. (2000). “Social psychology and the fear of crime”. The British Journal of Criminology 40(3): 399–413. Ferraro, K. F. (1995). Fear of crime: interpreting victimisation risk. Albany, NY, State University of New York Press. Ferraro, K. F. and R. LaGrange (2000). The measurement of fear of crime. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 277–308. Fisher, B. S. and J. L. Nasar (1992). “Fear of crime in relation to three exterior site features prospect, refuge and escape”. Environment and Behavior 24(1): 214–239. Fishman, G. and G. S. Mesch (1996). “Fear of crime in Israel: a multidimensional approach”. Social Science Quarterly 77(1): 76–89. Freeman, S., J. Groger, et al. (1996). “The spatial concentration of crime”. Journal of Urban Economics 40(2): 216–231. Garofalo, J. (1979). “Victimisation and the fear of crime”. Journal of Research in Crime and Delinquency 16: 80–97. Garofalo, J. (1981). “The fear of crime: causes and consequences”. Journal of Criminal Law and Criminology 72(2): 839. Gibbons, S. (2004). “The costs of urban property crime”. Economic Journal 114(499): F441–F463. Grabosky, P. N. (1995). “Fear of crime, and fear reduction strategies”. Current Issues in Criminal Justice 7(1): 7–19.
References
21
Green, G., J. M. Gilbertson, et al. (2002). “Fear of crime and health in residential tower blocks – a case study in Liverpool, UK”. European Journal of Public Health 12(1): 10–15. Greene, J. (1999). “Zero tolerance: a case study of police policies and practices in New York city”. Crime and Delinquency 45(2): 171–187. Hale, C. (1996). “Fear of crime: a review of the literature”. International Review of Victimology 4: 79–150. Hamermesh, D. S. (1999a). “The timing of work over time”. The Economic Journal 109(37–66). Hamermesh, D. S. (1999b). “Crime and the timing of work”. Journal of Urban Economics 45: 311–330. Harcourt, B. E. (1998). “Reflecting on the subject: a critique of the social influence conception and deterrence, the broken windows theory, and order maintenance policing New York style”. Michigan Law Review 97(2): 291–389. Helsley, R. W. and W. C. Strange (1999). “Gated communities and the economic geography of crime”. Journal of Urban Economics 4: 80–105. Hirschfield, A. and K. J. Bowers (1997). “The effect of social cohesion on levels of recorded crime and disadvantaged areas”. Urban Studies 34(8): 1275–1295. Innes, M. (2004). “Signal crimes and signal disorders: notes on deviance as communicative action”. The British Journal of Sociology 55(3): 335–355. Jackson, J. and E. Gray (2010). “Functional fear and public insecurities about crime”. The British Journal of Criminology 50(1): 1–22. Jacobs, J. (1961). The death and life of great American cities. New York, NY, Vintage Books. Joseph, J. (1997). “Fear of crime among black elderly”. Journal of Black Studies 27(5): 698–717. Katzman, M. T. (1980). “The contribution of crime to urban decline.” Urban Studies 17(3): 277– 286. Keane, C. (1998). “Evaluating the influence of fear of crime as an environmental mobility restrictor on women’s routine activities”. Environment and Behavior 30(1): 60–74. Kelling, G. L. and C. M. Coles (1997). Fixing broken windows: restoring order and reducing crime in our communities. New York, NY, Touchstone. Kovecses, Z. (1990). Emotion concepts. New York, NY, Springer. Krahn, H. and L. W. Kennedy (1985). “Producing personal safety: the effects of crime rates, police force size, and fear of crime”. Criminology 23(4): 697–710. Krivo, L. J. and R. D. Peterson (1996). “Extremely disadvantaged neighbourhoods and urban crime”. Social Forces 75(2): 619–648. Liska, A. E., A. Sanchirico, et al. (1988). “Fear of crime and constrained behavior specifying and estimating a reciprocal effects model”. Social Forces 66(3): 827–838. Loukaitou-Sideris, A. (1999). “Hot spots of bus stop crime: the importance of environmental attributes”. Journal of the American Planning Association 65(4): 395–412. Lymes, D. (1997). “The fortification of suburbia: investigating the rise of enclave communities”. Landscape and Urban Planning 39: 187–203. Markowitz, F. E., P. E. Bellair, et al. (2001). “Extending social disorganization theory: modeling the relationships between cohesion, disorder, and fear”. Criminology 39(2): 293. Mawby, R. I., P. Brunt, et al. (2000). “Fear of crime among British holidaymakers”. The British Journal of Criminology 40(3): 468–479. Miceli, R., M. Roccato, et al. (2004). “Fear of crime in Italy – spread and determinants”. Environment and Behavior 36(6): 776–789. Millie, A. and V. Herrington (2005). “Bridging the gap: understanding reassurance policing”. The Howard Journal of Criminal Justice 44(1): 41. Mirrlees-Black, C. and J. Allen (1998). Concern about crime: Findings from the 1998 British Crime Survey. Research Findings No 83. London, Home Office Research, Development and Statistics Directorate. Morrall, P., P. Marshall, et al. (2010). “Crime and health: a preliminary study into the effects of crime on the mental health of UK university students”. Journal of Psychiatric and Mental Health Nursing 17(9): 821–828.
22
2
Why Is Fear of Crime a Serious Social Problem?
Murray, A. T., I. McGuffog, et al. (2001). “Exploratory spatial data analysis techniques for examining urban crime”. The British Journal of Criminology 41(2): 309–329. Nasar, J. L., B. Fisher, et al. (1993). “Proximate physical cues to fear of crime”. Landscape and Urban Planning 26: 161–178. Nasar, J. L. and K. M. Jones (1997). “Landscapes of fear and stress”. Environment and Behavior 29(3): 291–323. Norris, C. and G. Armstrong. (1998). CCTV and the rise of mass surveillance society. Crime unlimited? Questions for the 21st century. P. Carlen and R. Morgan (Eds.). Macmillan, London: 76–98. Oc, T. and S. Tiesdell (1997). Safer city centres: reviving the public realm. London, Chapman. Pain, R. (2000). “Place, social relations and the fear of crime: a review”. Progress in Human Geography 24(3): 365–387. Painter, K. (1996). “The influence of street lighting improvements on crime, fear and pedestrian street use, after dark”. Landscape and Urban Planning 35(2–3): 193–201. Pantazis, C. (2000). “‘Fear of crime’, vulnerability and poverty”. The British Journal of Criminology 40(3): 414–436. Perkins, D. D. and R. B. Taylor (1996). “Ecological assessments of community disorder: their relationship to fear of crime and theoretical implications”. American Journal of Community Psychology 24(1): 63–107. Reid, L., J. T. Roberts, et al. (1998). “Fear of crime and collective action: an analysis of coping strategies”. Sociological Inquiry 68(3): 312–328. Riger, S., M. T. Gordon and R. K. LeBailly (1982). “Coping with urban crime: women’s use of precautionary behaviors”. American Journal of Community Psychology 10(4): 369–386. Ross, C. E. and J. Mirowsky (2000). “Disorder and decay: the concept and measurement of perceived neighborhood disorder”. Urban Affairs Review 34(3): 412–433. Rountree, P. W. and K. C. Land (1996). “Perceived risk versus fear of crime: empirical evidence of conceptually distinct reactions in survey data”. Social Forces 74(4): 1353–1377. 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). Samuels, R. and B. Judd (2002). Public housing estate renewal: Interventions and the epidemiology of victimisation. Housing, Crime and Stronger Communities Conference, Melbourne, Australian Institute of Criminology & Australian Housing and Urban Research Institute. Schuerman, L. and S. Kobrin. (1986). Community careers in crime. Crime and justice: A review of research, communities and crime, Vol. 8. A. J. Reiss and M. Tonry (Eds.). The University of Chicago Press, Chicago and London: 67–100. Skogan, W. G. (1986). Fear of crime and neighbourhood change. Communities and crime. A. J. Reiss and M. Tonry (Eds.). University of Chicago press, Chicago, IL: 203–230. Skogan, W. G. (1990). Disorder and decline: crime and the spiral decay in American neighbourhoods. Los Angeles, CA, University of California Press. Skogan, W. G. and M. G. Maxfield (1981). Coping with crime: individual and neighborhood reactions. Beverly Hills, CA, Sage Publications. Smith, S. J. (1987). “Fear of crime: beyond a geography of deviance”. Progress in Human Geography 11: 1–23. Smith, L. N. and G. D. Hill (1991). “Victimisation and fear of crime”. Criminal Justice and Behaviour 18(2): 217–239. Spelman, W. (2004). “Optimal targeting of incivility-reduction strategies”. Journal of Quantitative Criminology 20(1): 63–88. Stephens, D. W. (1999). Measuring what matters. Measuring what matters: Proceedings from the Police Research Institute meetings. R. H. Langworthy, National Institute of Justice; Office of Community Oriented & Policing Services. Taylor, R. B. (2001). Breaking away from broken windows. Boulder, CO, Westview Press.
References
23
Taylor, R. B. and J. Covington (1993). “Community structural change and fear of crime”. Social Problems 40(3): 374–395. Teske, R. H. C. J. and H. R. Arnold (1991). A comparative victimization study in the United States and the federal republic of Germany: a description of the principal results. Victims and criminal justice. G. Kaiser, H. Kury and H. J. Albrecht (Eds.). Max Planck Institute, Freiburg: 3–44. Thomas, C. and R. Bromley (2000). “City-centre revitalisation: problems of fragmentation and fear in the evening and night-time city”. Urban Studies 37(8): 1403–1429. Tiesdell, S. and T. Oc (1998). “Beyond ‘fortress’ and ‘panoptic’ cities – towards a safer urban public realm”. Environment and Planning B: Planning and Design 25: 639–655. Warr, M. (1984). “Fear of victimization: why are women and the elderly more afraid”. Social Science Quarterly 65(3): 681–702. Warr, M. (1985). “Fear of rape among urban women”. Social Problems 32(3): 238–250. Warr, M. (2000). “Fear of crime in the United States: avenues for research and policy.” Criminal Justice 4: 452–489. Warr, M. and C. G. Ellison (2000). “Rethinking social reactions to crime: personal and altruistic fear in family households”. American Journal of Sociology 106(3): 551–578. Weatherburn, D., B. Lind, et al. (1999). ““Hotbeds of crime?” crime and public housing in urban Sydney”. Crime and Delinquency 45(2): 256–271. Wilson, J. Q. and G. L. Kelling (1982, March). “The police and neighbourhood safety: broken windows”. The Atlantic Monthly: 29–38. Wilson-Doenges, G. (2000). “An exploration of sense of community and fear of crime in gated communities”. Environment and Behavior 32(5): 597–611. Xu, Y., M. Fielder, et al. (2005). “Discovering the impact of community policing: the broken windows thesis, collective efficacy and citizen’s judgement”. Journal of Research in Crime and Delinquency 42(2): 147–186.
Chapter 3
What Causes Fear of Crime?
Criminal Opportunity and Risk of Victimization Theories While Cohen and Felson’s (1979) routine activities hypothesis, also known as the criminal opportunity or risk of victimization hypothesis, was developed to explain where and when people engage in crime, it has also been adapted to assist understanding of fear of crime (e.g. Ferraro, 1995). It proposes that rationally motivated offenders commit crime when opportunities, in space and time, provide a potential victim and an absence of capable guardians (Cohen and Felson, 1979). These opportunities are systematically related to the routine activities of the potential victims and their guardians1 (Cohen and Felson, 1979). Variation in routine activities differentially places people at risk of victimization by structuring their convergence in time and space with motivated offenders. This convergence increases their likelihood of victimization2 (Cohen and Felson, 1979). In a similar vein to offenders who assess environments in order to increase their opportunity for crime, potential victims may also make judgements when defining places and times as risky or threatening (Brantingham and Brantingham, 1993; Ferraro, 1995). When applied in conjunction with micro-scale perspectives, such as symbolic interactionism, criminal opportunity hypotheses facilitate analyses which seek to explain the spatial and temporal distribution of fear of crime3 (Ferraro, 1995). However, multiple studies concur that fear of crime, and people’s perception of risk of victimization, far exceeds the reality of actual crime rates and levels (e.g. see: Cozens, 2002; Liska et al., 1988; Miceli et al., 2004; Nelson et al., 2001; Smith, 1987; Taylor and Hale, 1986; Tulloch, 1998). This applies even when assuming a liberal amount of unreported crime (Liska et al., 1988; Painter,
1
See also: Bursik, 1988; Cochran et al., 2000; Vold et al., 2002; Walklate, 2003 Criminal opportunity theory branches into numerous related theories focusing on routine activities affecting people’s risk of victimization. For example Clarke (1980) and Cornish and Clarke (1986) propose the rational choice theory. Similarly, Miethe (1990) propose the structural choice theory. 3 For example, the micro-scale environmental cues and the wider macro-scale structural and geographic influences are taken into account (Ferraro, 1995). 2
B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_3, C Springer Science+Business Media, LLC 2012
25
26
3 What Causes Fear of Crime?
1996; Taylor and Hale, 1986). Fear of crime thus appears out of proportion to the objective risks of victimization (Warr, 2000). Therefore, it is paramount that researchers hoping to influence the design of fear-reduction strategies investigate other potential factors associated with fear of crime. The first of these factors relate to characteristics of demographic groups experiencing relatively high levels of fear of crime.
Demographic Theories Explaining Fear of Crime The demographic theories have dominated fear of crime research since its conception (Farrall et al., 2000). They examine whether people’s fear of crime is associated with their experiences of crime or feelings of vulnerability. Ultimately, each demographic hypothesis seeks to explain why some socio-demographic groups are more afraid of crime than others. This knowledge is important in providing an understanding of the nature of public fear of crime, which is a valuable component of many fear-reduction initiatives. The group of demographic theories comprises the victimization hypothesis, indirect-victimization hypothesis and vulnerabilities hypothesis.
Victimization Hypothesis The victimization hypothesis posits a positive relationship between direct experience of victimization and fear of crime (Crank et al., 2003; Skogan and Maxfield, 1981). Direct victimization recognizes only those victims who have been directly affected by the actions of an offender or incur some immediate loss following a victimization (Clark, 2003; Mesch, 2000). Under the victimization theory, previous experiences of direct victimization increase one’s sensitivity to risk. Past victims therefore have an increased likelihood of defining situations as dangerous and perceiving the risks of victimization as greater (Mesch, 2000). Drawing on Janoff-Bulman’s (1985) three ‘theories of reality’, Clark (2003) discusses the stages of emotional loss that victims endure following criminal victimization. The first of these losses is the desecration of the belief in one’s personal invulnerability, that victimization ‘won’t happen to me’. Similarly, the belief in the ‘social law’ that ‘good people do not get hurt’ is also defeated. In turn, this translates into the third emotional loss, which involves a detrimental turn in one’s selfimage (in Clark, 2003). Notions of self-worth are affected as victims ‘. . . recognise their self limitations, powerlessness, helplessness and neediness . . .’ (Clark, 2003). Societal attributions of blame are also said to reinforce these views and lead the victims to have less trust in themselves and others (Janoff-Bulman, 1985 in Clark, 2003). It is hypothesized that these reactions following victimization represent a new sense of personal vulnerability, which could result in increased fear of crime. In addition to this, victimization can create reactions of confusion, shock, helplessness,
Demographic Theories Explaining Fear of Crime
27
fear and anxiety and can lead to depression and post-traumatic stress disorder, reactions that could further increase fear of crime (Clark, 2003). In this sense, the victimization theory is related to the vulnerabilities and the indirect-victimization theses, which are discussed later in this section. A multitude of studies have investigated the victimization hypothesis, with different studies obtaining different results (Borooah and Carcach, 1997). Numerous studies find a positive relationship between experience of victimization and fear of crime (Crank et al., 2003). Of these, many confirm a strong and direct relationship in support of the theory (e.g. Akers et al., 1987; Cates et al., 2003; Ferraro, 1995; Katz et al., 2003; Smith and Hill, 1991; Skogan, 1990). Others find only a positive but weak relationship exists (e.g. Akers et al., 1987; Cates et al., 2003; Evans and Fletcher, 2000; Garofalo, 1979; Katz et al., 2003; Liska et al., 1988). In contrast, there are studies that either fail to find an association (e.g. Borooah and Carcach, 1997; Rader, 2004), or indeed find a negative association, between victimization and fear (Evans and Fletcher, 2000). Overall, the mixed results have prompted some researchers to conclude ‘. . . there is little consistent evidence to suggest that personal (direct) victimization has an impact on fear of crime’ (Katz et al., 2003). The victimization theory thus remains unsubstantiated (Borooah and Carach, 1997). While the conflicting evidence may be a consequence that fear and experience of victimization depends on other underlying factors, the surveying methods and fear of crime measures could also account for some of the variation. Generally, victimization is assessed in surveys by asking respondents about their experiences in the 12–14 months prior to the survey (Gray and O’Conner, 1990; Akers et al., 1987; Evans and Fletcher, 2000). The given time period may also not be relevant to many respondents. For example, people may either still feel the impact of victimization beyond this timeframe (Evans and Fletcher, 2000) or have long been implementing fear neutralization techniques. Regardless, as the victimization thesis makes intuitive sense (Crank et al., 2003), few researchers have been able to elucidate why previous victims of crime may not be afraid of crime (Katz et al., 2003). Agnew (1985) suggested that previous victims may neutralize their fear of crime by employing techniques, such as denial of injury or damage, acceptance of responsibility or denial of future vulnerability (cited in Katz et al., 2003). Although it is a major coping task for victims to rebuild their views of the world and themselves following victimization, a victim’s sensitivity to fear of crime is reduced over time (Mukherjee and Carach, 1998 in Clark, 2003).
Indirect Victimization Hypothesis The indirect victimization hypothesis accounts for the host of studies which find that ‘non-victims’ also experience fear of crime. The indirect victimization hypothesis recognizes people can experience victimization vicariously and may experience the same emotions that result from a direct victimization when they hear of others’ crime encounters (Clark, 2003; Hanson et al., 2000). The signal crimes perspective, discussed later, even suggests that crime and disorder have the same effect
28
3 What Causes Fear of Crime?
regardless of whether they are encountered in person or indirectly (Innes et al., 2002).4 Indirect-victimization research focuses on how crime information is obtained. Findings point towards exposure to crime through media accounts and interpersonal communication (Rountree and Land, 1996). The Media and Fear of Crime Many studies suggest that fear of crime is a product of media exposure (Killias and Clerici, 2000; Romer et al., 2003; Weitzer and Kubrin, 2004). Researchers supporting indirect victimization through the media have taken a number of different approaches. These are known as the cultivation,5 substitution,6 resonance,7 socialcomparison8 and interpersonal-diffusion9 hypotheses. Overall, they argue the media exacerbates perceptions of risk of victimization, and therefore induces fear of crime (Lane and Meeker, 2003a). In contrast, some researchers discredit the link between media exposure and fear of crime (Lane and Meeker, 2003b; Romer et al., 2003). Other researchers find no relationship between fear of crime and the media when demographic characteristics or neighbourhood levels of crime are examined (Katz et al., 2003).10
4 The signal crimes theory names directly encountered crimes ‘situated signal crimes’ and indirectly encountered crimes ‘disembedded signal crimes’ (Innes et al., 2002). 5 Cultivation theorists argue the media portrays an unrealistic world rife with crime, thereby fostering perceptions of increased risk and provoking fear of crime. The cultivation thesis argues media crime coverage has a uniform effect regardless of the audience (see Heath and Gilbert, 1996; Jopson, 1995; Lupton and Tulloch, 1999; Romer et al., 2003; Skogan and Maxfield, 1981; Totaro, 1988; Tulloch, 2000; Weitzer and Kubrin, 2004; Williams and Dickinson, 1993). 6 In contrast, the substitution thesis suggests that audience characteristics and contextual differences affect the impact of media on fear of crime. It propounds media exposure has a greater influence on fear of crime experienced by non-victims than victims (see Chiricos et al., 1997; Heath and Gilbert, 1996; Lane and Meeker, 2003b; Weitzer and Kubrin, 2004). 7 The resonance thesis, while also acknowledging that media affects audiences differently, expects the opposite reaction to the cultivation thesis. It considers the media influences fear of crime only when the crime coverage resonates with crime experiences of the audience, acting to mutually reinforce present feelings of fear (Weitzer and Kubrin, 2004). 8 In line with the resonance thesis, the social comparison thesis focuses on crime coverage pertinent to the audience’s locality. It proposes that crime reported in one’s neighbourhood fosters fear of crime, whereas crime reported in remote areas does not (Romer et al., 2003). 9 The interpersonal diffusion thesis also reflects the regional relevance of crime reports. It argues fear of crime is amplified when crime accounts resonate with the audience’s direct or indirect experiences of victimization. Only when media reports are compounded with other local sources of information about crime does fear of crime increase (Romer et al., 2003). 10 Supporters of the real-world thesis argue that fear of crime is more a result of objective conditions including personal victimization, actual crime rates and neighbourhood characteristics, than sensationalist stories reported by the media (Chiricos et al., 2000; Lupton and Tulloch, 1999; Weitzer and Kubrin, 2004). Additionally the operationalization and measurement of fear of crime can alter the relationship between media exposure and fear of crime (Heath and Gilbert, 1996).
Demographic Theories Explaining Fear of Crime
29
Interpersonal Communication and Fear of Crime The second element of the indirect victimization hypothesis focuses on the relationship between interpersonal communication, rather than the media, and fear of crime. The interpersonal communication hypothesis assumes that knowledge of others’ experience of criminal victimization spreads throughout the social networks of a community (Mawby et al., 2000; Taylor and Hale, 1986). Using the same explanation as cultivation theorists, it is presumed that knowledge about crime attained through interpersonal communication adds a crime multiplier and therefore increases the perceived risk of victimization (Taylor and Hale, 1986). It is thought that this effect will be maximized for people who are well entrenched in social networks (Lewis and Salem, 1986; Skogan and Maxfield, 1981; Crank et al., 2003). Generally, researchers find that there is a stronger relationship between fear of crime and indirect victimization than direct victimization (Katz et al., 2003; Mawby et al., 2000). For instance, using the same dataset many researchers have found that vicarious experience of victimization significantly increases fear of crime, while direct experience of victimization was not significantly related to fear of crime (Lewis and Salem, 1986; Skogan and Maxfield, 1981; Katz et al., 2003). Skogan and Maxfield (1981) concluded that indirect victimization is more common and widespread than direct victimization and should logically have a stronger effect on fear of crime. Later, Hale (1996) stated that the fear-of-crime response could be greater via indirect victimization because hearing about crime ‘allows one’s imagination full scope without perhaps the same urgency to find some coping strategy . . .’ (quoted in Katz et al., 2003). It is also likely these stories will be about local events and local victims, and hold the potential for greater personal impact for those hearing about them (Skogan and Maxfield, 1981). Once indirect knowledge about victimization is obtained, fear of crime is also unlikely to dissipate rapidly (Taylor and Hale, 1986). Adding to this, many cultural geographers have gone on to state that certain areas of a neighbourhood are feared because of their reputation, which can be considered a consequence of interpersonal communication (Koskela and Pain, 2000). Ferraro (1995) argues that indirect victimization has a strong effect on constrained behaviour in such areas (Ewald, 2000). When non-victims hear about a criminal victimization they will compare themselves to the victim. They may distinguish themselves from, or relate to, the victim by either what they did or the kind of people they are (Clark, 2003). Thus, the indirect victimization theory is influenced by notions of vulnerability and socio-demographic background (Skogan and Maxfield, 1981 in Taylor and Hale, 1986).
Vulnerabilities Hypothesis The vulnerabilities hypothesis is based on the assumption that different sociodemographic groups experience different levels of fear of crime and exhibit this fear differently (Warr, 2000; Liska et al., 1988). The vulnerabilities hypothesis also
30
3 What Causes Fear of Crime?
explains two other trends. After taking the risk of victimization into account, many studies conclude that typically fearful socio-demographic groups, like women and the elderly, are the least likely to be victimized (Katz et al., 2003; Painter, 1996; Pantazis, 2000; Taylor and Hale, 1986). Vulnerabilities are used to account for this discrepancy and its converse, namely an apparently lower-than-warranted fear of crime in some groups, such as young men, who have greater actual risks of victimization (Katz et al., 2003; Lane and Meeker, 2003b). Stinchcombe (1978) first introduced the concept of vulnerability. Perloff (1983: 43) later defines it as ‘. . . a belief that one is susceptible to future negative outcomes and unprotected from danger or misfortune’. Vulnerability is determined by three major factors, namely exposure to risk, loss of control and seriousness of consequences (Killias, 1990). Essentially it is not based on objective threat, yet occurs if one perceives himself or herself as vulnerable to criminal victimization (Katz et al., 2003). The concept of vulnerability highlights the importance of including anticipatory fear, or anxiety, in fear-of-crime research (Sacco and Glackman, 1987). It also explains that fear of crime, in contrast to perceived risk, depends on one’s perception of the seriousness of a particular offence and one’s risk sensitivity to it (Cates et al., 2003; Mesch, 2000). This is mirrored by other researchers such as Wurff et al. (1988) who argue that fear is ‘. . . the perception of a threat to some aspect of well-being, concurrent with the feeling of inability to meet the challenge . . .’. Skogan and Maxfield (1981) distinguish physical vulnerabilities from social vulnerabilities. Physical vulnerability refers to one’s perception of his/her susceptibility to attack, ability to resist an attack and ability to recover health following an attack (McCoy et al., 1996; Skogan and Maxfield, 1981). Such physical vulnerabilities include health, body size, self-defence capabilities and disabilities. Social vulnerability reflects how a person’s position in society differentially affects his/her exposure to victimization and his/her capacity to cope with the consequences of victimization (McCoy et al., 1996; Ortega, 1987; Skogan and Maxfield, 1981). Social vulnerabilities are a function of an individual’s position in society. They include income, residential status, educational level, ethnic background, living alone and experiences of victimization (Skogan and Maxfield, 1981). Purist vulnerabilities theorists do not see objective conditions in the external world as the source of the public’s fear of crime. Instead they encourage research on those who experience fear of crime to be more sensitive to the ‘. . . biographies, characteristics, and social circumstances of the fearful . . .’ and how they ‘acquire a sense of subset ability’ (Sacco and Glackman, 1987). Some researchers extend this further by pointing out that using general socio-demographic predictors to account for fear of crime masks potentially significant individual psychological factors, which should be considered (Farrall et al., 2000).
Social Theories Explaining Fear of Crime
31
Review: An Abundance of Contested Demographic Studies The demographic theories look at people’s experiences of victimization, indirect victimization and their vulnerabilities to explain fear of crime. The demographic theories largely account for the seemingly different levels of fear exhibited by different socio-demographic groups. Despite an abundance of research, the demographic theories remain contested. This adds credence to the notion that additional factors and complexities may be associated with fear of crime.
Social Theories Explaining Fear of Crime The social theories discussed in this section argue that fear of crime reflects a general state of anxiety caused by a change or breakdown of a range of different societal factors. This section starts with the two most prominent hypotheses, the risk society and social disorganization hypotheses. The social disorganization hypothesis branches into a framework of various independent, but inter-related models (Covington and Taylor, 1991), which are also discussed below. These models include the subcultural diversity, social integration, community concern and social change hypotheses. The purpose of this chapter is to review the core components of different theories, rather than test them rigorously and as such the chapter does not fully engage with all angles of debate evident in the literature on this issue. Rather, social theories are discussed because they attempt to explain fear of crime and frequently contribute towards fear-reduction strategies
Risk Society Hypothesis Drawing upon notions of the ‘risk society’, fear of crime is conceptualized as an expression of people’s wider feelings of insecurity or uncertainty about life. Risk society theorists commonly propose that fear of crime provides an outlet to express general feelings of anxiety that predominate in everyday lives. While the literature on risk society is extensive, a few pertinent points are emphasized here. According to Beck (1992), the founder of the hypothesis, processes of industrialization produce numerous new, unforeseen, unpredictable and uncontrollable risks (Dean, 1999; Ewald, 2000; Lupton, 1999). The risks are extensive, irreversible and affect all individuals regardless of their social position or class (Beck, 1992; Ewald, 2000; Girling et al., 2000). Furthermore the risks are incalculable and unsatisfactorily insurable, making them additionally threatening and anticipatory (Beck, 1992; Dean, 1999; Ewald, 2000). According to Beck (1992), a risk society, defined by the statement ‘I am afraid’, emerges with these risks.
32
3 What Causes Fear of Crime?
Lianos and Douglas (2000) come to a similar conclusion. They contend present societies are in a state of ‘dangerization’11 which is portrayed by a continuous detection of potential threats, which ultimately results in fear and anxiety.12 When in a state of dangerization, the unknown ‘other’ is perceived as dangerous (Lianos and Douglas, 2000). This person usually operates beyond one’s managed territory and possesses differences in his/her appearance or behaviours. As a result deviance is associated with unknown individuals or groups, who consequently trigger fear and avoidance behaviours (Lianos and Douglas, 2000).13 In turn, the signs and behaviours associated with these groups become automatic indicators of dangerousness (Lianos and Douglas, 2000; Rose, 2000). Beck (1992) similarly claims that it is not the risks themselves that cause fear and unease but those people who represent the risks. The underlying state of anxiety14 is projected onto other individuals or social groups. Numerous other theorists agree that crime becomes a scapegoat for intangible insecurities and anxieties (Ewald, 2000; Hollway and Jefferson, 2000; Lupton and Tulloch, 1999).15 In a risk society not only is anxiety a part of everyday life, but so too is crime and the threat of crime (Stanko, 2000).16 Researchers should be aware of this possibility, as it affects fear-of-crime measurement approaches. Survey questions should therefore be as specific and precise as possible in targeting fear of actual ‘crime’. Similarly, survey questions should be specific in targeting ‘fear’ of, not concern about, crime. This is pertinent to the social disorganization group of hypotheses, discussed below.
Social Disorganization Hypothesis The social disorganization hypothesis implies that fear of crime is linked to concern about the destruction of social organization. Since its origins in the 1920s and 11
Like Beck’s thesis, dangerization is brought about by a change in institutional control over collective social interaction (Lianos and Douglas, 2000). 12 Stanko (2000) argues that we live in an age fraught with uncertainty. Hope and Sparks (2000) echo similar sentiments and state that ‘. . . fear reaches down into the unilluminated corners of the inner life . . .’ 13 These ‘others’ are generally depicted as dangerous in adherence with existing biases and discriminations (Lianos and Douglas, 2000). 14 Sparks also argues that fear is never caused by a specific target (Sparks, 1992). See also Dammert and Malone, 2003; Hope and Sparks, 2000; Gottfredson, 1984; Lupton and Tulloch, 1999; Mawby et al., 2000; Pain, 2000; and Stanko, 2000. 15 Hollway and Jefferson (2000) argue ‘unconscious’ anxieties are displaced onto more tangent external threats (Hollway and Jefferson, 2000). They report criminals are easily identifiable targets and thus provide ‘. . . a repository for anxieties about other fears that are more intractable and are diffuse for the individual . . .’ (Lupton and Tulloch, 1999). Ewald (2000) also asserts that the psychological experiences associated with victimization, such as feelings of loss of control, are similar to those anxieties of the risk society and therefore crime becomes a suitable scapegoat. 16 With fear of crime at the forefront of the risk society, the control and prevention of risk becomes a preoccupation of everyday living (Vaughn, 2002; Walklate, 2000).
Social Theories Explaining Fear of Crime
33
formal naming in 1942 by Shaw and McKay, the social disorganization hypothesis has dominated criminological perspectives attempting to explain neighbourhood crime (Cochran et al., 2000; Sun et al., 2004; Taylor and Covington, 1993). While originally focusing on how the destruction of community social organization can ultimately lead to crime and delinquency, it now encompasses fear of crime (Bursik, 1988; Taylor and Covington, 1993). Bursik (1988) defines social disorganization as ‘the inability of local communities to realise the common values of their residents or solve community experienced problems’. Sampson and Groves (1989) amend this description to include the concept of social control,17 defining social disorganization as ‘the inability of a neighbourhood to achieve the common goals of its residents and maintain effective social controls’. Social disorganization hypothesis is therefore dependent upon a community having common values and social norms. The inherent proposition underlying these definitions is that community structures affect a community’s ability to maintain public order, constrain residents from violating social norms and therefore fend off crime and fear (Markowitz et al., 2001; Taylor and Covington, 1993). Social disorganization theory proposes that the destruction of community social organizations ultimately leads to crime and delinquency (Bursik, 1988; Taylor and Covington, 1993). Early work focused on processes of urbanization that led to the erosion of the informal social controls that had governed traditional rural communities of the United States (Taylor and Covington, 1993). Heterogeneity and rapid population turnover seemingly undermined the social ties between neighbours, ‘limiting their ability to agree on common sets of values or to solve commonly experienced problems’ (Bursik, 1988). In turn this prevented residents from organizing collectively against those groups migrating into their neighbourhoods and prevented them from adequately controlling public antisocial behaviour, particularly that of new immigrants (Bursik, 1988; Taylor and Covington, 1993). A breakdown in familial controls and the presence of unsupervised youth groups within a neighbourhood were also central to the social disorganization theory (Taylor and Covington, 1993). The urban settings for social disorganization research were subject to rapid urbanization following an influx of immigrants from rural United States and Europe (Taylor and Covington, 1993). These immigrants were believed to have been ‘ill-equipped to supervise children assimilating the values of urban United States’. Due to the high population turnover and concern about the values of others within the neighbourhood, local adults were reportedly hesitant to reprimand youths participating in deviant activities (Taylor and Covington, 1993). Social mistrust also caused local adults to withdraw from nonconforming families, anticipating opposition and anxious that retaliation may result should they attempt to reform and informally control delinquent behaviour (Maccoby et al., 1958 in Taylor and Covington, 1993). In the event that residents did not exercise order, it was feared
17 Social control refers to the ‘capacity of the society to regulate itself according to the desired principles and values’ or the ‘ability of social groups or institutions to make norms all rules effective’ (Janowitz, 1975 and Reiss, 1951 cited respectively in Sampson, 1986).
34
3 What Causes Fear of Crime?
that the youth would continue their delinquencies, which would eventually escalate in severity (Wilson and Kelling, 1982; Taylor and Covington, 1993). Thus, ineffective parental supervision of youths, a traditional means of informal control, is an essential tenet of social disorganization theory (Taylor and Covington, 1993). Similar processes of social disorganization have also been put forth as mechanisms that lead to fear of crime. In 1978, Hunter proposed that social disorganization produces signs of physical and social incivility (Taylor and Covington, 1993). These incivilities, such as the presence of unsupervised youth in the streets, are negatively interpreted by residents as alluding to a lack of social control in the neighbourhood (Taylor and Covington, 1993). The idea of social disorganization has been supported in various longitudinal studies (Bursik and Webb, 1982; Markowitz et al., 2001; Sun et al., 2004; Taylor and Covington, 1993). These, and other cross-sectional studies, generally suggest that changes in racial composition are most strongly associated with disorder (Taylor and Covington, 1993). However, Sampson and Groves (1989) argue this research has failed to measure any mediating variables and therefore cannot be used to support the hypothesis. They proposed a model that has been hailed as ‘the most complete examination of the systemic social disorganisation model’ (Bursik and Grasmick, 1993). Sampson and Grove’s (1989) model predicted that neighbourhoods with low socio-economic status, high residential mobility, racial heterogeneity and family disruption would have disrupted local social organizations (Sun et al., 2004). Social disorganization was characterized by weak local friendship networks, low organizational participation and unsupervised youth groups. Sampson and Groves then predicted that these characteristics limit the capacity residents have to control behaviour, which in turn leads to increased neighbourhood crime and delinquency. In testing their model, Sampson and Groves confirmed crime rates were higher in areas of social disorganization, and that the aforementioned characteristics mediated the effect of ethnic heterogeneity, population turnover and social class on crime rates (Markowitz et al., 2001; Sun et al., 2004). However, social disorganization theory encountered some inevitable criticism – the theoretical concept has been rebuked as being poorly defined, and failing to distinguish between the causes and consequences of social disorganization (Markowitz et al., 2001; Sun et al., 2004; Taylor and Covington, 1993). This combined with the longitudinal component of the theory and problems of empirical testing saw a decline in its prevalence among the literature until the mid-1980s (Markowitz et al., 2001; Sun et al., 2004). In an attempt to counter criticisms of this nature, Bursik (1988) more succinctly defined social disorganization as ‘the inability of local communities to realise the common values of their residents or solve community experienced problems’ (Lane and Meeker, 2003b). Sampson and Groves (1989) later amended this description slightly to include the concept of social control, defining social disorganization as ‘the inability of a neighbourhood to achieve the common goals of its residents and maintain effective social controls’ (Markowitz et al., 2001; Sun et al., 2004; Taylor and Covington, 1993). Social disorganization theory is therefore dependent upon a community having common or dominant values and social norms. The inherent proposition underlying both of these definitions is that it is community structure that affects the ability of a neighbourhood to
Social Theories Explaining Fear of Crime
35
maintain public order, constrain its residents from violating such social norms and therefore fend off crime and fear (Markowitz et al., 2001; Taylor and Covington, 1993). Despite criticisms, the prominence of the social disorganization hypothesis means that it should be acknowledged. Furthermore, the presence of the hypothesis indicates fear-of-crime survey questions should be developed to target fear, so results are not confused with concern about crime or social disorganization. This would save confusion when interpreting research findings. The same conclusion can be made from the following discussion of the related subcultural diversity, social integration, community concern and social change hypotheses. Subcultural Diversity Hypothesis The subcultural diversity hypothesis proposes that fear of crime develops when residents live in close proximity to individuals of differing cultural backgrounds. This was presented by Merry (1981) who theorises that the behaviours of those who are racially, ethnically and culturally different are difficult to interpret (cited in Lane and Meeker, 2003b). When residents cannot understand different behaviours, they become uncertain about and mistrust these ‘others’. The residents believe the ‘others’ have different social values, attitudes and community commitment (Covington and Taylor, 1991; Lane and Meeker, 2003b). In the longer term, they are consequently perceived as being dangerous and fear of crime results (Lane and Meeker, 2003b).18 Numerous studies support the hypothesis, finding racial diversity is related to increased fear of crime (Chiricos et al., 1997; Covington and Taylor, 1991; Lane and Meeker, 2003b; Taylor and Covington, 1993).19 In opposition to the subcultural diversity hypothesis,20 Chiricos et al. (1997) found that racial composition has no consequence on fear of crime when other relevant factors are controlled. With the subcultural diversity hypothesis, fear of crime can be thought of as an ‘anxiety endangered through the confrontation of difference’ (Ditton et al., 2000).21 This further emphasizes the need for fear-of-crime survey questions to focus on fear of a specific crime, in order to minimize the potential for confusion with anxiety related to diversity.
18 This is considered particularly pertinent in communities with poor social networks (Lane and Meeker, 2003b). 19 Katz et al. (2003) note that the majority of research supporting the subcultural diversity theory use ethnicity or racial heterogeneity to measure cultural background (Katz et al., 2003). They argue these measures are less relevant to subcultural diversity than to conflict theory. While similar, conflict theory proposes ‘the greater the presence of minority populations, the more threatened majority group members will feel’ (Blalock, 1967; Katz et al., 2003). 20 As with any of the explanatory theories, the effect of subcultural diversity may also be dependent on the measure of fear used. For example, Thompson et al. (1992) found that perceived safety was related to racial composition, while fear of being the victim of specific crimes was not. 21 A variation of the subcultural diversity theory posits that flux in subcultural diversity, as opposed to static subcultural diversity, causes residents to perceive their neighbourhood as in a state of disorder and decline (Lane and Meeker, 2003b).
36
3 What Causes Fear of Crime?
Social Integration/Neighbourhood Cohesion Hypotheses The social integration hypothesis, also known as the neighbourhood cohesion hypothesis, proposes that poor social integration in a community leads to increased fear of crime (Crank et al., 2003). Social integration can be considered as ‘the capacity of the community to exert social control over its members and passersby, thereby enforcing local versions of right and seemly conduct’ (Janowitz, 1978 cited in Skogan and Maxfield, 1981). The social integration hypothesis depends upon additional concepts of social support, social capital and collective efficacy. Like many sociological terms these concepts are multifaceted and arguably ill-defined. The main descriptions are covered here. Social support refers to the frequency of contact residents have with one another, the amount of help they provide to one another and how satisfied they are with that support (Thompson and Krause, 1998). Social capital relates to social contact through community networks or associations and resident feelings of trust in one another, while collective efficacy concerns the level of cohesion between residents and their willingness to intervene on behalf of the common good (Lindstrom et al., 2003). A number of researchers find that low levels of social integration, social support, social capital and collective efficacy lead to fear of crime (Bellair, 1997; Markowitz et al., 2001). In contrast, Gibson et al. (2002) state it is ‘challenging to derive any definitive conclusions of the effects of social integration on fear of crime’.22 Community Concern Hypothesis The community concern hypothesis draws upon the disorder/incivilities and disorder and decline hypotheses, discussed shortly. The community concern hypothesis implies that fear of crime represents the opinion that one’s community is in a state of decline (Lane and Meeker, 2003b). People become concerned about the vitality, viability and quality of their neighbourhood when they encounter signs of physical and social decay (Taylor and Hale, 1986). They consequently worry that the problems present in their community may worsen and that their community, as a whole, is in a state of decline (Taylor and Hale, 1986). Residents become concerned that their neighbourhood is less safe than it was in the past and consequently feel afraid of crime (Covington and Taylor, 1991; Lane and Meeker, 2003b). The community concern theory also concludes that fear of crime is intensified when local social ties are weak (Conklin, 1971; Covington and Taylor, 1991; Garofalo and Laub, 1978 in Lane and Meeker, 2003b). Thus the theory is also related to notions of social integration. This temporal component of the community concern hypothesis lends the hypothesis its secondary title, known as the decline model (Lane and Meeker, 2003b). Researchers such as Taylor and Hale (1986) support the community concern hypothesis, finding that concern predicts fear of crime.
22 However, in comparing such studies it is important to consider the varying operationlizations of the concepts inherent in the theory and how they are measured (Crank et al., 2003)
Social Theories Explaining Fear of Crime
37
Social Change Hypothesis Furstenberg (1971) first put forward the social change model, and suggested that people most disturbed by social change are more concerned about crime (Furstenberg, 1971). According to the hypothesis, fear of crime eventuates when people resent the processes and features of social change, particularly those that denote adjustments to the prevailing status quo (Furstenberg, 1971). These social changes include diversifying racial heterogeneity, a declining economic base, increasing unemployment and high population turnover (Clark, 2003; Furstenberg, 1971). This could accompany shifts in the environment that may disrupt the identification of people and places that are perceived to be risky, which generates more anxiety (Skogan and Maxfield, 1981). Fear of crime therefore becomes a metaphor for resentment and anxiety following social change (Clark, 2003; Pantazis, 2000).23 Possibly due to the longitudinal nature of this hypothesis, few studies have tested the social change model. While Hunter et al. (2002) have lent some support for the model,24 Clark (2003) disputes that such research has only maintained the concept of fear of crime as an anxiety response to rapid change. Instead drawing upon Lotz’s (1979) study, Clark (2003) proposes that it is concern about crime, rather than fear, that correlates with rapid change.25
Review: Social Studies Emphasize the Inherent Complexity of ‘Fear’ of ‘Crime’ The social theories draw attention to how the social fabric of the environment can lead to fear of crime. According to the social theories, fear of crime can reflect • feelings of insecurity or uncertainty; • concerns about the destruction of community social organization; • fear of the unknown and the different; • concerns about poor social integration;
23 Taylor further proposes that fear of crime is provoked by ‘different types of modern risk’, a conclusion very similar to those made by risk society theorists (Hollway and Jefferson, 1997). This supports the concept that fear is more akin to a general sense of anxiety (Clark, 2003). 24 Hunter et al. (2002) found that fear of crime increased during immigrant boom periods. Smith et al. (2001) found that during a period of population growth residents are more likely to view the social context as ‘unpredictable and potentially risky in regard to their perceptions about personal safety from criminal victimisation’ (cited in Hunter et al., 2002). 25 Similarly, Lemert (1951) and others have suggested that changes in conditions, rather than the current level of neighbourhood problems, are the most significant bellwether of fear (Skogan and Maxfield, 1981).
38
3 What Causes Fear of Crime?
• concerns about communities being in a state of decline; • concerns arising from rapid social changes. By suggesting that ‘fear of crime’ is not always a ‘fear’ of ‘crime’, the social theories emphasize the complexity underlying fear of crime and the importance of specifically targeting ‘fear’ of ‘crime’ in survey questions.
Environmental Theories Explaining Fear of Crime Environmental theories focus on cues in the external environment that trigger fear of crime. This set of theories is particularly relevant to strategies targeting fear of crime, as they seek to identify factors in the environment that can be altered to potentially reduce fear. The first of the environmental theories is the disorder/incivilities hypothesis, which pioneered such research. The subsequent theories include the threatening and safe environments theories and the signal crimes perspective.
The Disorder/Incivilities Hypothesis The disorder or incivilities hypothesis advises that there is a positive relationship between fear of crime and people’s perceptions of the social and physical characteristics of an environment (Crank et al., 2003; Millie and Herrington, 2005; Nasar et al., 1993; Tulloch, 2000). In particular it is signs of disorder or visible cues in an environment that signify a lack of order and trigger fear of crime (Ross and Mirowsky, 1999). According to Wilson (1968), disorder and incivilities are violations of ‘standards of right and seemly conduct’. Originally, fear of crime studies were primarily concerned with criminal acts and actual infractions of law (Phillips and Smith, 2003). However, the disorder/incivilities hypothesis draws attention to activities and objects that violate norms without violating the law (Ross and Mirowsky, 1999). Numerous studies reveal that the signs of disorder at the forefront of the public’s mind are those that are not legally criminal acts (Phillips and Smith, 2003; Stephens, 1999). More often they include lower-level breaches of community standards or ‘soft’ crimes that are frequently encountered in everyday life (Carvalho and Lewis, 2003; Millie and Herrington, 2005; Phillips and Smith, 2003; Skogan, 1990). Incivilities/disorder theorists (e.g. Nasar et al., 1993) argue that incivilities generate fear of crime in areas where there is no real criminal activity. Incivilities generate fear because they are perceived to be warning signs of underlying crime and criminal threat (Mirrlees-Black and Allen, 1998; Tulloch, 2000). They indicate a breakdown in social norms of behaviour and community relinquishment of both formal and informal social controls and support systems (Perkins and Taylor, 1996; Nasar and Jones, 1997; Rountree and Land, 1996; Tulloch, 2000).
Environmental Theories Explaining Fear of Crime
39
Disorder highlights the inability of community members to mobilize resources and deal with problems such as crime (Skogan, 1990; Taylor, 1999). This also includes the inability, or neglect, among officers of the state and external agencies to cope with crime (Perkins and Taylor, 1996). An impression of neighbourhood level vulnerability to crime is generated, which translates into fear (Painter, 1996; Nasar and Jones, 1997; Rountree and Land, 1996; Taylor and Hale, 1986). Further, incivilities act as warning signals of impending danger because they are associated with things people fear (Skogan and Maxfield, 1981). Thus, the presence of disorder creates increased perceptions of criminogenic risk (Crank et al., 2003).26 The disorder/incivilities hypothesis assumes that these incivilities are interpreted similarly regardless of the particular situation or local context (Taylor and Gottfredson, 1986). An encounter with disorder can either be ‘direct’ or ‘less targeted’ (Phillips and Smith, 2003). A ‘direct’ encounter refers to those situations whereby an individual is the direct target of an intentional act of deviance. A ‘less targeted’ encounter occurs when an individual observes or hears about an intentional action directed at another person or group of people (Phillips and Smith, 2003). Signs of disorder can also be encountered after the act. This is more often the case with signs of physical disorder. Hunter (1978) and Lewis and Maxfield (1980) identified disorder as being both ‘social’ and ‘physical’ in nature (Robinson et al., 2003). ‘Incivilities’ is an allencompassing label, which characterizes these disorders (Mirrlees-Black and Allen, 1998; Ross and Mirowsky, 1999). Social incivilities encompass disorder that involves people and their behaviours (Ross and Mirowsky, 1999; Skogan, 1999). Social disorder denotes people violating social norms or official laws, or acting in an unpredictable and threatening manner (Painter, 1996; Perkins and Taylor, 1996; Ross and Mirowsky, 1999; Skogan, 1999), including drug users, sex workers, beggars, gangs and people perceived to be behaving violently (Ferraro, 1995; Painter, 1996; Perkins and Taylor, 1996; Ross and Mirowsky, 1999; Skogan and Maxfield, 1981; Tulloch, 2000). Physical disorder refers to a neighbourhood’s overall physical appearance and signs of negligence or unchecked decay (Ross and Mirowsky, 1999; Skogan, 1999). They can also be the by-products of social disorder that has not been managed or taken care of by the community over time. Physical disorder includes abandoned buildings, graffiti, damaged property and broken streetlights (Doeksen, 1997; Painter, 1996; Ross and Mirowsky, 1999; Skogan, 1999). While not legally defined crimes, both social and physical signs of disorder trigger fear of crime. Likewise, so do the threatening environments.
26 However, some researchers state it is not merely the presence of incivilities that triggers fear of crime, but rather a change in the presence of incivilities, which is accompanied by changing community satisfaction and changing perceptions of relative risk, that triggers fear of crime (Robinson et al., 2003; Taylor and Gottfredson, 1986).
40
3 What Causes Fear of Crime?
Threatening and Safe Environments Theories Although similar to the disorder/incivilities hypothesis and signal crimes hypothesis, which link fear of crime with certain cues in the environment, the threatening environments hypothesis does not reflect a breakdown in social control. Threatening environments instead provide an all-encompassing label for those objects and acts that generate fear of crime, but these cues are not limited solely to disorder. However, like disorder, threatening environments can be either physical or social in nature. Threatening physical environments are a manifestation of urban planning or lack thereof. While signs of disorder are not necessarily present in threatening environments, they may generate fear because they are perceived as attractive sites for criminal victimization. Threatening physical environments commonly have characteristics that prohibit natural surveillance. Some researchers refer to this as ‘a lack of prospect’, ‘blocked prospect’ or ‘concealment’ (e.g. Fisher and Nasar, 1995; Nasar et al., 1993). For example they may have poor street lighting and barriers that prevent visibility to others, thereby creating hiding spots for offenders (DTUPA, 2002; Painter, 1996). These barriers include the presence of alcoves, too many bushes and overgrown vegetation (Cozens, 2002; Kuo and Sullivan, 2001; Newman, 1972; Fisher and Nasar, 1995). Similarly, threatening physical environments may have entrapment spots, which block the escape avenues of victims (DTUPA, 2002; Fisher and Nasar, 1995). There is a second characteristic, independent of urban planning, that can affect whether an environment is considered threatening. The literature indicates that fear of crime is influenced by time of day (Nasar and Jones, 1997). Researchers agree that people have increased fear after dark (Brantingham and Brantingham, 1993; Cubbage and Smith, 2009; Doran and Lees, 2005; Fisher and Nasar, 1995; Painter, 1996; Samuels and Judd, 2002). The reduction in visibility and recognition abilities and the creation of blind spots, shadows and potential places of entrapment play a role in the physical environment (Painter, 1996). The change in the social character of environments during the night is also likely to be an influencing factor (Koskela, 1999). Threatening social environments may also generate fear while not representing disorder. For example, an absence of pedestrian activity and the notion of a lack of natural surveillance or ‘eyes on the street’ induce fear (Jacobs, 1961; Samuels and Judd, 2002). This is partly based on Jacobs’s (1961) premise that criminals do not want to be observed, as it increases their risk of being reported and apprehended. Social surveillance increases the perceived risk of detection for offenders, prompting them to participate in criminal activity in less populated areas (Jacobs, 1961). In line with this, there is the perception that unaccompanied individuals are more attractive targets for victimization (e.g. Painter, 1996). A lack of social surveillance could also increase a potential victim’s fear of crime for two more reasons. First, there is a lack of potential witnesses who could seek help from police or other authorities (Jacobs, 1961; Samuels and Judd, 2002) and second, there is a lack of capable guardians who could help resist an attack (Painter, 1996). Conversely,
Environmental Theories Explaining Fear of Crime
41
social surveillance arguably reduces fear of crime, as can the other environmental factors discussed below (Doeksen, 1997; Loukaitou-Sideris, 1999). Safe environments, the opposite of threatening environments, could potentially help mitigate fear of crime. Safe environments theoretically lack those environmental cues that trigger the public’s fear of crime, for example areas to hide and signs of social or physical disorder. They would also contain other environmental cues that reinforce perceptions of safety. Very little information in the fear of crime literature has been gathered on such ‘safe cues’ or ‘control signals’. Nasar (1998) discusses cues he labels as ‘likeable features’, which could trigger people to feel safe. These include signs of ‘naturalness’ (for example vegetation and mountains), ‘upkeep/civilities’ (well-maintained areas), ‘openness’ (open spaces and scenery), ‘historical significance’ (features with a historical feel) and ‘order’ (organization and compatibility of features) (Nasar, 1998). Cozens (2002) additionally suggested that ‘upkeep/civilities’ and ‘order’ can decrease fear of crime. Vegetation, despite potentially being a source of fear when causing concealment and areas to hide, has also been found to reduce fear of crime in some studies (Kuo and Sullivan, 2001). Appleton (1975) proposes that the public is more inclined to feel safe in environments that have adequate prospect to create opportunities for surveillance (Yokohari et al., 2006). In similar vein the signal crimes perspective, discussed below, emphasizes the presence of ‘control signals’ in an environment (Innes et al., 2002; Millie and Herrington, 2005). Control signals are defined as ‘acts of social control that communicate a message to the public’ (Innes, 2004a). Police and town planners generally put such signals in place in an attempt to reassure the community and they have a positive effect by reducing perceptions of criminogenic risk (Innes et al., 2002; Millie and Herrington, 2005). While logical, there is the potential for control signals to inadvertently have a negative impact upon public perceptions of security (Innes, 2004a). For example, the presence of closed-circuit television (CCTV) cameras, which may in part be erected to reduce fear of crime, could simultaneously denote the presence of an unsafe element to some sectors of society.
Signal Crimes Perspective The signal crimes perspective, put forward by Innes et al. (2002), refines some of the generalizations inherent in the disorder/incivilities hypothesis. It draws on social semiotics and symbolic interactionist sociology to illustrate how the wider social character of the environment shapes the way crime and disorder are interpreted and rendered meaningful. The signal crimes perspective argues that different crimes and disorders have a disproportionate impact on how people interpret them, and the extent to which they connote criminogenic risk. It also recognizes that although community members may share common values, different individuals and groups vary in the way they interpret crime and disorder (Innes, 2004a; Innes et al., 2002).
42
3 What Causes Fear of Crime?
A brief theoretical background in semiotics and signs is necessary for the understanding of signal crimes. Semiotics theory advises that signs are objects27 or acts that mean something to someone in a context (Innes, 2004b). Social semiotics examines signs in light of how their meaning in different cultural and situational contexts will vary. Signs are composed of two components, the first being the ‘expression’ and denotative description (Innes, 2004a). The second component is the ‘content’ and connotative description. According to Eco (1976), signals are defined as signs that have an effect. The effect of a signal can be ‘affective’ (changing how the onlooker feels), ‘cognitive’ (changing how the onlooker thinks), ‘behavioural’ (changing how the onlooker acts) or a mixture of each (Innes, 2004a). All signals therefore have an expression, content and effect, which in combination, act to establish meaning and differentiate signals from other signs (Innes, 2004a). The signal crimes perspective differentiates ‘signal crimes’ and ‘signal disorders’. With regard to expression, ‘signal crimes’ encompass those signals that denote criminal incidents, for example a mugging. The content is that they indicate the presence of criminogenic risk. In this example it is the risk of mugging (Innes, 2004a). ‘Signal disorders’ follow on from the disorder/incivilities hypothesis. In semiotics terms, while not directly denoting a legally criminal incident, signal disorders28 also connote criminogenic risk (Innes, 2004a). Instead of supposing that all crimes and disorders generically lead to fear of crime, as with some disorder/incivilities theorists and the positivist view of crime, the signal crimes perspective focuses on how and why different signal crimes have a different effect, despite having the same content (Innes et al., 2002). Innes et al. (2002) refer to Slovic’s (1992) hypothesis that proposed different risks have different ‘signal values’. The signal value refers to the extent, strong or weak, a signal crime shapes one’s perception of risk. Strong signal crimes are those acts or objects that are serious enough to generate a ‘significant degree of public awareness’ (Innes et al., 2002). Weak signal crimes do not generate such perceptions of criminogenic risk, when encountered in isolation. However, an accumulative impact occurs when numerous weak signals are encountered in succession or combination (either temporally or spatially). They are then interpreted as a strong signal (Innes, 2004a; Innes et al., 2002). Another addition to the disorder/incivilities hypothesis is the situational relevance of signal crimes. The signal crimes perspective contends that identical objects and acts may be signal crimes in one environment and not another (Innes, 2004a). The content and effect of a signal crime is highly contextualized and situational (Innes et al., 2002). Therefore one’s interpretation of a signal crime is sensitive to characteristics of the social and physical environment in which it is located
27 An object is anything that can be indicated, everything that is pointed to or referred to (Blumer, 1969). 28 As discussed in the previous section, disorders can either be social or physical in their denotative expression.
Environmental Theories Explaining Fear of Crime
43
(Innes et al., 2002). Innes et al. (2002) use the example that graffiti in a neighbourhood with good social control might act as a signal crime because of its high ‘dissonance’ value, whereas it might go unnoticed in a neighbourhood with the presence of more serious crime and disorder. The signal crimes perspective essentially acknowledges that the disorder/incivilities hypothesis has merit in that that certain signal crimes and signal disorders are thought to be common throughout a community. Innes (2004a) draws on symbolic interactionist sociology to highlight the role of social reactions in defining deviant acts (Innes, 2004a).29 Slovic (1992) reasons that people do not define risk purely on the basis of the signal crime itself, but according to its nature and one’s personal context (Innes et al., 2002). Risk is dependent upon surrounding belief systems, such as those governing acceptable social norms (Innes et al., 2002). If community members share common social norms, then signal crimes may be commonly interpreted. However, the signal crimes perspective recognizes that there is not necessarily a consensus between community members on which acts or objects are considered signal crimes (Innes et al., 2002). Nor is it assumed that common signal crimes are interpreted in the same manner, to the same extent or have the same effect (Innes et al., 2002; Innes, 2004a).30 Signal crimes are interpreted in light of an individual’s past experiences with similar objects, personal values and concerns (Innes, 2004a). An assessment of the situation and prediction about the likelihood of future risks then takes place (Innes, 2004a). Consequently, a particular personal reaction to the signal crime occurs (Innes et al., 2002). Thus, the signal crimes perspective recognizes that individuals vary in the way they interpret and render meaningful signs of disorder. Similarly, different signal crimes vary in their effect on people. As mentioned above, there are a variety of cognitive, affective and behavioural reactions people can exercise after encountering a signal crime. By their definition, signal crimes always induce a cognitive and affective reaction, adversely altering criminogenic risk perceptions and causing feelings of heightened fear and anxiety (Innes et al., 2002). Subsequently the affected people may also adopt a behavioural change in order to protect themselves from victimization (Innes et al., 2002).
29 Symbolic interactionism is a label for an ‘approach to the study of human group life and human conduct’ (Blumer, 1969). Symbolic interactionism contends the meaning of objects and things is derived from the social interaction one has with one’s fellows (Blumer, 1969). 30 This is relevant to different individuals and socio-demographic groups. For example factors such as age, gender and experience of previous victimization may shape how certain signal crimes are interpreted and rendered meaningful (Innes, 2004a).
44
3 What Causes Fear of Crime?
Review: Intuitive Environmental Studies into Cues Triggering Fear of Crime Environmental theories propose that signs of disorder (also known as incivilities or signal crimes/disorders) and other stimuli in threatening environments can trigger fear of crime. While environmental theories are well established, different components of the theories have not been fully examined. New research could specifically determine what environmental cues trigger fear of crime in different environments. These studies could, for example, pay attention to potential differences in the content, effect or signal value of different environmental cues in different situational contexts.
Chapter Review: An Opening for Pertinent Environmental Studies Criminal opportunity and risk of victimization theories argue that crime is the primary cause of fear of crime. Drawing on the literature, it is evident that while crime certainly does lead to fear of crime, there is also evidence that fear of crime can occur in areas characterized by low crime rates. Therefore, research into the other factors associated with fear of crime is justified. An extensive body of research has tested demographic theories by examining the potential associations between fear of crime and victimization, indirect victimization and personal feelings of vulnerability. The findings from such research are frequently contested and it is unlikely further studies into these associations will provide new information or substantially progress the fear of crime research field. Similarly, numerous studies have examined the various social theories that propose fear of crime is caused by, and actually represents, risk society feelings or concern about social disorganization. While there may be a set of relationships that can be explored, general feelings of uncertainty or concern cannot substitute fear of crime. Consequently fear of crime studies should use survey questions that minimize the likelihood of producing results that could represent fear of crime as something other than ‘fear’ of ‘crime’. There is clear evidence that environmental cues, for example signs of disorder and other stimuli in threatening environments, can trigger fear of crime. Despite the fact that several studies have investigated the link between fear of crime and environmental cues, it appears there is room for more research into environmental theories and the associated behavioural responses that individuals adopt in relation to perceptions of risk.
References
45
References Agnew, R. S. (1985). “Neutralizing the impact of crime.” Criminal Justice and Behaviour 12(2): 221–239. Akers, R. L., A. J. Greca, et al. (1987). “Fear of crime and victimization among the elderly in different types of communities”. Criminology 25(3): 487. Beck, U. (1992). Risk society: towards a new modernity. London, Sage. Bellair, P. E. (1997). “Social interaction and community crime: examining the importance of neighbor networks.” Criminology 35(4): 677–703. Blalock, H. (1967). Toward a theory of minority group relations. New York, NY, Wiley. Blumer, H. (1969). Symbolic interactionism. Berkley, CA, University of California Press. Borooah, V. and C. Carcach (1997). “Crime and fear. evidence from Australia”. The British Journal of Criminology 37(4): 635–657. Brantingham, P. L. and P. J. Brantingham (1993). “Nodes, paths and edges – considerations on the complexity of crime and the physical-environment”. Journal of Environmental Psychology 13(1): 3–28. Bursik, R. J. (1988). “Social disorganisation and theories of crime and delinquency”. Criminology 26: 519–551. Bursik, R. J., Jr. and H. G. Grasmick (1993). Neighborhoods and crime: the dimensions of effective community control. New York, NY, Lexington Books. Bursik, R. J. and J. Webb (1982). “Communities change and patterns of delinquency”. American Journal of sociology 88: 24–42. Carvalho, I. and D. A. Lewis (2003). “Beyond community: reactions to crime and disorder among inner-city residents”. Criminology 41(3): 779–812. Cates, J. A., D. A. Dian, et al. (2003). “Use of protection motivation theory to assess fear of crime in rural areas”. Psychology Crime & Law 9(3): 225–236. Chiricos, T., M. Hogan, et al. (1997). “Racial composition of neighborhood and fear of crime”. Criminology 35(1): 107. Chiricos, T., K. Padgett, et al. (2000). “Fear, TV news, and the reality of crime.” Criminology 38(3): 755. Clark, J. (2003). “Fear in fear-of-crime”. Psychiatry, Psychology and Law 102(267–282). Clarke, R. V. G. (1980). “Situational crime-prevention – theory and practice.” British Journal of Criminology 20(2): 136–147. Cochran, J. K., M. L. Bromley, et al. (2000). “Victimization and fear of crime in an entertainment district crime “hot spot:” a test of structural-choice theory”. American Journal of Criminal Justice 24(2): 189. Cohen, L. E. and M. Felson (1979). “Social change and crime rate trends: a routine activity approach”. American Sociological Review 44: 588–608. Conklin, J. E. (1971). “Dimensions of community response to the crime problem”. Social Problems 18: 373–385. Cornish, D. and R. V. Clark. (1986). The reasoning criminal: rational choice perspectives on offending. New York, NY, Springer. Covington, J. and R. B. Taylor (1991). “Fear of crime in urban residential neighbourhoods: implications between – and within – neighbourhood sources for current models”. The Sociological Quarterly 32(2): 231–249. Cozens, P. M. (2002). “Sustainable urban development and crime prevention through environmental design for the British city. Towards an effective urban environmentalism for the 21st century”. Cities 19(2): 129–137. Crank, J. P., A. Giacomazzi, et al. (2003). “Fear of crime in a nonurban setting”. Journal of Criminal Justice 31(3): 249–263. Cubbage, C. J. and C. L. Smith (2009). “The function of security in reducing women’s fear of crime in open public spaces: a case study of serial sex attacks at a Western Australian university”. Security Journal 22(1): 73–86.
46
3 What Causes Fear of Crime?
Dammert, L. and M. F. T. Malone. (2003). “Fear of crime or fear of life? Public insecurities in Chile.” Bulletin of Latin American Research 22(1): 79–101. Dean, M. (1999). Risk, calculable and incalculable. Risk and sociocultural theory. D. Lupton (Ed.). Cambridge University Press, Cambridge: 131. Department for Transport Urban Planning and the Arts (DTUPA) (2002). Crime prevention through environmental design and urban design: Design Issues for safe neighbourhoods. Adelaide. Ditton, J., J. Bannister, et al. (2000). Afraid or angry? recalibrating the fear of crime. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 535–552. Doeksen, H. (1997). “Reducing crime and the fear of crime by reclaiming New Zealand’s suburban street”. Landscape and Urban Planning 39(2–3): 243–252. Doran, B. J. and B. G. Lees (2005). “Investigating the spatiotemporal links between disorder, crime, and the fear of crime”. Professional Geographer 57(1): 1–12. Eco, U. (1976). A theory of semiotics. Bloomington, IN, Indiana University Press. Evans, D. J. and M. Fletcher (2000). “Fear of crime: testing alternative hypotheses”. Applied Geography 20(4): 395–411. Ewald, U. (2000). Criminal victimisation and social adaptation in Germany: fear of crime and risk perception in the new Germany. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London: 166. Farrall, S., J. Bannister, et al. (2000). “Social psychology and the fear of crime”. The British Journal of Criminology 40(3): 399–413. Ferraro, K. F. (1995). Fear of crime: interpreting victimisation risk. Albany, NY, State University of New York Press. Fisher, B. S. and J. L. Nasar (1995). “Fear spots in relation to microlevel physical cues: exploring the overlooked”. Journal of Research in Crime and Delinquency 32(2): 214–239. Furstenberg, F. F., Jr. (1971). Public reaction to crime in the streets “The American Scholar”. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 3–12. Garofalo, J. (1979). “Victimisation and the fear of crime”. Journal of Research in Crime and Delinquency 16: 80–97. Garofalo, J. and J. Laub. (1978). “The fear of crime: broadening our perspective.” Victimology 3: 242–253. Gibson, C., J. Zhao, et al. (2002). “Sociological measurement confusion, paradigmatic imperfection, and etiological nirvana: striking a pragmatic balance in pursuing science”. Justice Quarterly: JQ 19(4): 793. Girling, E., I. Loader, et al. (2000). Crime and social change in Middle England: questions of order in an English town. London, Routledge. Gottfredson, M. R. (1984). Victims of crime: the dimensions of risk. London, H.M.S.O. Gray, D. E. and M. E. OÇonner (1990). “Concern about to fear of crime in an Australian rural community”. Australia and New Zealand Journal of Criminology 23: 284–298. Hale, C. (1996). “Fear of crime: a review of the literature.” International Review of Victimology 4: 79–150. Hanson, R. F., D. W. Smith, et al. (2000). “Crime related fears and demographic diversity in Los Angeles county after 1992 civil disturbances”. Journal of Community Psychology 28(6): 607–623. Heath, L. and K. Gilbert. (1996). “Mass media and fear of crime.” American Behavioral Scientist 39(4): 379–386. Hollway, W. and T. Jefferson. (1997). “The risk society in an age of anxiety: situating fear of crime.” The British Journal of Sociology 48(2): 255. Hollway, W. and T. Jefferson (2000). The role of anxiety in fear of crime. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London. Hope, T. and R. Sparks. (2000). Introduction: risk, insecurity and the politics of law and order. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London.
References
47
Hunter, A. (1978). “Symbols of incivility: social disorder and fear of crime in urban neighborhoods”, Paper presented at the Annual Meeting of the American Society of Criminology, Dallas, TX (1978). Symbols of incivility: social disorder and fear of crime in urban neighborhoods. Annual Meeting of the American Society of Criminology. Dallas, TX. Hunter, L. M., R. S. Krannich, et al. (2002). “Rural migration, rapid growth, and fear of crime.” Rural Sociology 67(1): 71–89. Innes, M. (2004a). “Reinventing tradition? Reassurance, neighbourhood security and policing.” Criminal Justice 4(2): 151–171. Innes, M. (2004b). “Signal crimes and signal disorders: notes on deviance as communicative action.” The British Journal of Sociology 55(3): 335–355. Innes, M., N. Fielding, et al. (2002). Signal crime and control signals: towards an evidence based framework for reassurance policing. Guilford, University of Surrey. Jacobs, J. (1961). The death and life of great American cities. New York, NY, Vintage Books. Jopson, D. (1995). Fear that feeds on violent crime. The Sydney Morning Herald. Sydney. Katz, C. M., V. J. Webb, et al. (2003). “Fear of gangs: a test of alternative theoretical models”. Justice Quarterly 20(1): 95–130. Killias, M. (1990). New methodological perspectives for victimisation surveys: the potentials of computer assisted telephone surveys and some related innovations. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 315–330. Killias, M. and C. Clerici (2000). “Different measures of vulnerability in their relation to different dimensions of fear of crime”. The British Journal of Criminology 40(3): 437–450. Koskela, H. (1999). “Gendered exclusions: women’s fear of violence and changing relations to space”. Geografiska Annaler 81B(2): 111–124. Koskela, H. and R. Pain (2000). “Revisiting fear and place: women’s fear of attack and the built environment”. Geoforum 31(2): 269–280. Kuo, F. E. and W. C. Sullivan (2001). “Environment and crime in the inner city – does vegetation reduce crime?”. Environment and Behavior 33(3): 343–367. Lane, J. and J. W. Meeker (2003a). “Ethnicity, information sources, and fear of crime”. Deviant Behavior 24(1): 1–26. Lane, J. and J. W. Meeker (2003b). “Fear of gang the crime: a look at three theoretical models”. Law and Society Review 37(2): 425–456. Lewis, D. A. and M. G. Maxfield. (1980). “Fear in the neighborhoods – an investigation of the impact of crime.” Journal of Research in Crime and Delinquency 17(2): 160–189. Lewis, D. A. and G. Salem (1986). Fear of crime: incivility and the production of a social problem. New Brunswick, NJ, Transaction Inc. Lianos, M. and M. Douglas (2000). Dangerization and the end of deviance: the institutional environment. Criminology and social theory. D. Garland and R. Sparks (Eds.). Oxford University Press, Oxford. Lindstrom, M., J. Merlo, et al. (2003). “Social capital and sense of insecurity in the neighbourhood: a population-based multilevel analysis in Malmo, Sweden”. Social Science and Medicine 56(5): 1111–1120. Liska, A. E., A. Sanchirico, et al. (1988). “Fear of crime and constrained behavior specifying and estimating a reciprocal effects model”. Social Forces 66(3): 827–838. Loukaitou-Sideris, A. (1999). “Hot spots of bus stop crime: The importance of environmental attributes”. Journal of the American Planning Association 65(4): 395–412. Lupton, D. (1999). “Dangerous places and the unpredictable stranger: Constructions of fear of crime”. Australian and New Zealand Journal of Criminology 32(1): 1–15. Lupton, D. and J. Tulloch (1999). “Theorizing fear of crime: beyond the rational/irrational opposition”. The British Journal of Sociology 50(3): 507–523. Markowitz, F. E., P. E. Bellair, et al. (2001). “Extending social disorganization theory: modeling the relationships between cohesion, disorder, and fear”. Criminology 39(2): 293. Mawby, R. I., P. Brunt, et al. (2000). “Fear of crime among British holidaymakers”. The British Journal of Criminology 40(3): 468–479.
48
3 What Causes Fear of Crime?
McCoy, V. H., J. D. Wooldredge, et al. (1996). “Lifestyles of the old and not so fearful: life situation and older persons fear of crime”. Journal of criminal justice 24(3): 191–205. Mesch, G. S. (2000). “Perceptions of risk, lifestyle activities, and fear of crime”. Deviant Behavior 21(1): 47–62. Miceli, R., M. Roccato, et al. (2004). “Fear of crime in Italy – spread and determinants”. Environment and Behavior 36(6): 776–789. Miethe, T. D. (1990). “Opportunity, choice, and criminal victimization – a test of a theoreticalmodel.” Journal of Research in Crime and Delinquency 27(3): 243–266. Millie, A. and V. Herrington (2005). “Bridging the gap: understanding reassurance policing”. The Howard Journal of Criminal Justice 44(1): 41. Mirrlees-Black, C. and J. Allen (1998). Concern about crime: Findings from the 1998 British Crime Survey. Research Findings No 83. London, Home Office Research, Development and Statistics Directorate. Nasar, J. L. (1998). The evaluative image of the city. California, Sage. Nasar, J. L., B. Fisher, et al. (1993). “Proximate physical cues to fear of crime”. Landscape and Urban Planning 26: 161–178. Nasar, J. L. and K. M. Jones. (1997). “Landscapes of fear and stress.” Environment and Behavior 29(3): 291–323. Nelson, A., R. Bromley, et al. (2001). “Identifying micro-spatial and temporal patterns of violent crime and disorder in a British city centre”. Applied Geography 21: 249–274. Newman, O. (1972). Defensible space: crime prevention through urban design. New York, NY, Macmillan. Ortega, S. T. (1987). “Race and gender effects on fear of crime: an interactive model with age”. Criminology 25(1): 133. Pain, R. (2000). “Place, social relations and the fear of crime: a review.” Progress in Human Geography 24(3): 365–387. Painter, K. (1996). “The influence of street lighting improvements on crime, fear and pedestrian street use, after dark”. Landscape and Urban Planning 35(2–3): 193–201. Pantazis, C. (2000). “‘Fear of Crime’, vulnerability and poverty”. The British Journal of Criminology 40(3): 414–436. Perkins, D. D. and R. B. Taylor (1996). “Ecological assessments of community disorder: their relationship to fear of crime and theoretical implications”. American Journal of Community Psychology 24(1): 63–107. Perloff, L. S. (1983). “Perceptions of vulnerability to victimization”. Journal of Social Issues 39(2): 41–61. Phillips, T. and P. Smith (2003). “Everyday incivility: towards a benchmark”. Sociological Review 51(1): 85–108. Rader, N. E. (2004). “The threat of victimization: a theoretical reconceptualization of fear of crime”. Sociological Spectrum 24(6): 689–704. Robinson, J. B., B. A. Lawton, et al. (2003). “Multilevel longitudinal impacts of incivilities: fear of crime, expected safety, and block satisfaction”. Journal of Quantitative Criminology 19(3): 237–274. Romer, D., K. H. Jamieson, et al. (2003). “Television news and the cultivation of fear of crime”. Journal of Communication 53(1): 88–104. Rose, N. (2000). Government and control. Criminology and social theory. D. Garland and R. Sparks (Eds.). Oxford University Press, Oxford. Ross, C. E. and J. Mirowsky (1999). “Disorder and decay: the concept and measurement of perceived neighborhood disorder”. Urban Affairs Review 34(3): 412–433. Rountree, P. W. and K. C. Land (1996). “Perceived risk versus fear of crime: empirical evidence of conceptually distinct reactions in survey data”. Social Forces 74(4): 1353–1377. Sacco, V. F. and W. Glackman (1987). Vulnerability, locus of control, and worry about crime. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 415–428.
References
49
Sampson, R. J. (1986). Crime in cities: the effects of formal and informal social control. Communities and crime. A. J. Reiss and M. Tonry (Eds.). University of Chicago press, Chicago: 271–312. Sampson, R. and B. Groves (1989). “Community structure and crime: testing social disorganisation theory”. American Journal of sociology 94: 774–802. Samuels, R. and B. Judd (2002). Public housing estate renewal: Interventions and the epidemiology of victimisation. Housing, Crime and Stronger Communities Conference, Melbourne, Australian Institute of Criminology & Australian Housing and Urban Research Institute. Skogan, W. G. (1990). Disorder and decline: crime and the spiral decay in American neighbourhoods. Los Angeles, CA, University of California Press. Skogan, W. G. (1999). Measuring what matters: Crime, disorder and fear. Measuring what matters: proceedings from the police research institute meetings. R. H. Langworthy (Ed.). National Institute of Justice, Washington, DC. Skogan, W. G. and M. G. Maxfield (1981). Coping with crime : individual and neighborhood reactions. Beverly Hills, CA, Sage Publications. Smith, S. J. (1987). “Fear of crime: beyond a geography of deviance”. Progress in Human Geography 11: 1–23. Smith, L. N. and G. D. Hill (1991).“Victimisation and fear of crime.” Criminal Justice and Behaviour 18(2): 217–239. Sparks, R. (1992). Television and the drama of crime: moral tales and the place of crime in public life. Buckinghamshire, Open University Press. Stanko, E. A. (2000). Victims R us: the life history of fear of crime and the politicisation of violence. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London. Stephens, D. W. (1999). Measuring what matters. Measuring what matters: Proceedings from the Police Research Institute meetings. R. H. Langworthy, National Institute of Justice; Office of Community Oriented & Policing Services. Stinchcombe, A. L. (1978). Theoretical models in social history. New York, NY, Academic Press. Sun, I. Y., R. Triplett, et al. (2004). “Neighbourhood characteristics and crime: a test of Sampson and Groves’ model of social disorganisation”. Western Criminology Review 5(1). Taylor, R. B. (1999). The incivilities thesis: Theory, measurement, and policy. Measuring what matters: Proceedings from the Police Research Institute meetings. R. H. Langworthy, National Institute of Justice; Office of Community Oriented & Policing Services. Taylor, R. B. and J. Covington (1993). “Community structural change and fear of crime”. Social Problems 40(3): 374–395. Taylor, R. B. and S. D. Gottfredson (1986). Environmental design, crime and prevention: an examination of community dynamics. Communities and crime. A. J. Reiss and M. Tonry (Eds.). University of Chicago Press, Chicago, IL: 387–416. Taylor, R. B. and M. Hale (1986). “Testing alternative models of fear of crime”. The Journal of Criminal Law and Criminology 77(1): 151–189. Thompson, E. E. and N. Krause (1998). “Living alone and neighborhood characteristics as predictors of social support in late life”. Journals of Gerontology Series B-Psychological Sciences and Social Sciences 53(6): S354–S364. Thompson, C. Y., W. B. Bankston, et al. (1992). “Parity and disparity among three measures of fear of crime: a research note.” Deviant Behavior 13: 373–389. Totaro, P. (1988). Sydney plays it safe by staying at home. The Sydney Morning Herald. Sydney. Tulloch, J. (1998). Quantitative review. Fear of crime. J. Tulloch, D. Lupton, W. Blood et al. (Eds.). National Campaign Against Violence and Crime (NCAVAC), Canberra. Tulloch, M. (2000). “The meaning of age differences in the fear of crime”. The British Journal of Criminology 40(3): 451–467. Vaughn, D. (2002). Signals and interpretive work: the role of culture in a theory of practical action. Culture in mind: toward a sociology of culture and cognition. K. Cerulo (Ed.). Routledge, New York, NY.
50
3 What Causes Fear of Crime?
Vold, G. B., T. J. Bernard, et al. (2002). Theoretical criminology. New York, NY, Oxford University Press. Walklate, S. (2000). Trust and the problem of community in the inner-city. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London. Walklate, S. (2003). Understanding criminology: current theoretical debates, second edition. Buckingham, Open University press. Warr, M. (2000). “Fear of crime in the United States: avenues for research and policy.” Criminal Justice 4: 452–489. Warr, M. and C. G. Ellison (2000). “Rethinking social reactions to crime: personal and altruistic fear in family households”. American Journal of Sociology 106(3): 551–578. Weitzer, R. and C. E. Kubrin (2004). “Breaking news: how local TV news and real-world conditions affect fear of crime”. Justice Quarterly 21(3): 497–520. Williams, P. and J. Dickinson. (1993). “Fear of crime: read all about it. The realtionship between newspaper crime reporting and fear of crime.” The British Journal of Criminology 33(1): 33– 56. Wilson, J. (1968). “The urban unease: community versus the city”. The Public Interest 12: 25–39. Wilson, J. Q. and G. L. Kelling (1982, March). “The police and neighbourhood safety: broken windows”. The Atlantic Monthly: 29–38. Wurff, A. V. D., P. Stringer, and F. Timmer. (1988). Feelings of unsafety in residential surroundings. Environmental social psychology. D. Canter, C. Jesuino, L. Soczka, and G. Stephenson (Eds.). Kluwer, The Hague: 135–148. Yokohari, M., M. Amemiya, et al. (2006). “The history and future directions of greenways in Japanese new towns”. Landscape and Urban Planning 76(1–2): 210–222.
Chapter 4
Managing Fear of Crime
Policing Fear of Crime Fear of crime and other non-criminal community problems are not typically considered in conventional policing models. Instead, policing is traditionally reactive and oriented towards crime incidents, requiring an offence before police act (Xu et al., 2005). Even so, the police often deal with disorder and fear of crime more than actual crime (Glensor and Peak, 1996). Hence, many policing models are increasingly focusing on a more in-depth understanding of non-criminal problems, including fear of crime (Ashby and Longley, 2005). Addressing fear of crime therefore features in many ‘problem-oriented’, ‘zero-tolerance’ and ‘community-oriented’ policing models. Problem-oriented policing was initially developed by Goldstein (1979) and employed in the early 1980s by policing practitioners, such as Wilson and Kelling (1982), the initiators of the broken windows hypothesis. Under this model, the police aim to proactively prevent crime, rather than react to incidents. They deal with non-criminal problems that concern or cause harm to the community, for example disorder and fear of crime (CPOP, 2003; Sims et al., 2002), and identify public concerns in order to carry out thoroughly planned responses to those concerns (CPOP, 2003; Lawton et al., 2005). This process is based on the SARA model (Scanning, Analysis, Response and Assessment) and often involves other public agencies and the private sector, with the community being identified as a potentially important policing partner in dealing with problems like fear of crime (Sims et al., 2002; CPOP, 2003). Problem-oriented policing incorporates a framework for situational crime prevention when acting on identified problems which, in turn, draws on the criminal opportunity and risk of victimization theories by aiming to increase the risks to potential offenders and reduce the rewards or benefits from criminal activity (CPOP, 2003). Therefore unlike standard policing models, problem-oriented policing is geographically focused and allows localized intervention (Lawton et al., 2005). Zero-tolerance policing, also known as order-maintenance policing or disorder policing, is widely discussed in the fear of crime literature (Harcourt, 1998). While grounded in problem-oriented policing, zero-tolerance policing does not focus on
B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_4, C Springer Science+Business Media, LLC 2012
51
52
4 Managing Fear of Crime
police-community interaction. Zero-tolerance policing strategies draw on Wilson and Kelling’s (1982) broken windows hypothesis and Skogan’s (1990) disorder and decline hypothesis and attempts to combat disorder before it can lead to fear of crime and crime (Katz et al., 2003). It is based on the principle that police intervention in reducing disorder can also reverse those processes of neighbourhood decay and criminal activity (Crank et al., 2003; Katz et al., 2003; Novak et al., 1999). The New York Police Department’s (NYPD) zero-tolerance program of the 1990s is frequently cited as a successful example of this, however some critiques suggest the decline in New York’s crime levels were the result of other factors (e.g. Greene, 1999; Harcourt, 1998; Katz et al., 2003; Kelling and Coles, 1997). A case study on the well-known zero-tolerance policing strategy adopted by the NYPD is provided below to examine some of the issues surrounding the role of police–community partnerships in reducing the fear of crime. Stemming from problem-oriented policing, community-oriented policing or neighbourhood policing specifically promotes ‘community police partnerships, proactive problem-solving, and community engagement to address the causes of crime, fear of crime, and other community issues’ (Dietz, 1997). A police understanding of, and response to, public perceptions of crime and disorder is fostered (Baker and Wolfer, 2003; Dietz, 1997; Sims et al., 2002).1 Police empower and work with city agencies, businesses, service providers and the community at large to identify, prioritize and resolve citizen concerns (Adams et al., 2005; Glensor and Peak, 1996; Sims et al., 2002; Walklate, 2000). Surveillance activities like neighbourhood watch programmes are most common, whereby residents report any suspicious activity to the police. Such programmes are identified as helping reduce public fear of crime (Baker and Wolfer, 2003; Skogan and Maxfield, 1981; Tulloch, 1998). More intensive programmes include mobile citizen patrols, whereby community groups patrol the neighbourhood with the aim of interrupting criminal activities, apprehending offenders and making citizens arrests on behalf of the police (Baker and Wolfer, 2003; Kenney, 1987; Skogan and Maxfield, 1981). The dissemination of crime prevention information through newsletters and public meetings, which often involve the police, are also conventional (Kenney, 1987). Garofalo (1979) proposed that information about crime decreases fear of crime. According to this model, increased knowledge of local crime leads to an alteration of risk assessment, which then changes fear of crime levels. Skogan and Maxfield (1981) state that community-based initiatives which aim to reduce and prevent crime play a large role in independently helping to reduce fear of crime. For example, involvement in crime reduction initiatives potentially decreases fear of crime by reversing feelings of vulnerability, community concern and perceptions of social disorganization. Skogan (1986, 1990) further suggests
1 However despite this benefit, community-oriented policing is criticized as being a spatially generalist model that does not reflect local conditions (Bennett, 1991; Spelman, 2004).
Policing Fear of Crime
53
that participation also generates feelings of helpfulness, responsibility, territoriality and optimism, which may also reduce fear. Given this, the presence of disorder and crime may actually increase the well-being of a neighbourhood by encouraging preventative action and collective efficacy (Innes et al., 2002). Another aim of community-oriented policing is that if police and community appear to work cohesively then potential criminals could be deterred from behaving in an antisocial manner (Baker and Wolfer, 2003).2 Public cooperation with police and increased police visibility reduce public fear of crime (Adams et al., 2005; Dietz, 1997; Salmi et al., 2004).3 It is argued that such inter-agency approaches to reducing crime and fear of crime are most successful (e.g. Brown and Polk, 1996; Smith, 1987). Dixon (1995) goes so far as to argue that a safer community cannot be created by criminal justice agencies alone. Thus despite their limitations, proactive crime prevention components of these popular policing models do help to fight crime and fear of crime and have been adopted in many countries (Xu et al., 2005). Indeed, this is frequently reflected in the mission statements of police departments that have fear reduction as one of their primary objectives (e.g. Cordner, 1986; also see Table 4.1 below). In a review of the fear of crime in Australia, Grabosky (1995) argues that the general public view police services as the main government agency with the responsibility of managing the fear of crime. This statement seems to be generally applicable as others have expressed similar views regarding police services and appropriate public policy in relation to the fear of crime (Cordner, 1986; Bennett, 1991; Borooah and Carcach, 1997; Grabosky, 1995).The former NSW Police Commissioner, Ken Moroney, has even formally stated that fear of crime ‘is as debilitating as the crime itself’ (Cameron, 2002). As indicated above, fear of crime features in the primary mission statement of the New South Wales Police, being to provide ‘a safe NSW with a respected police force working with the community to reduce violence, crime and fear’ (NSW Police Force, 2011).
2 Reassurance policing emphasizes this notion even further by focusing on police visibility, familiarity and accessibility in an effort to thwart declining public confidence in the police (Povey, 2001 in Millie and Herrington, 2005). Reassurance policing places a strong emphasis on the reduction of disorder and fear of crime by focusing scarce police resources on the root causes of these issues (Millie and Herrington, 2005). 3 For example in terms of avoidance, Skogan and Hartnett (1997) found that residents in jurisdictions governed by community-oriented policing avoided fewer areas due to worrying about victimization than residents in non-COP neighbourhoods (Sims et al., 2002). Then again, Weisburd and Eck (2004) found that community-oriented policing only reduced fear of crime when implemented with models of problem-oriented policing.
Australia
Australia
New South Wales Police Force, Wollongong local area command Police Tasmania, Launceston
San Diego Police Department
United Kingdom United States
‘To serve the people of Tasmania by protecting life and property, enhancing community safety and reducing the incidence and fear of crime’ ‘Working with our communities to reduce crime, disorder and fear as the leading, caring and professional police service’ ‘We are committed to working together, within the department, in a problem solving partnership with communities, government agencies, private groups and individuals to fight crime and improve the quality of life for the people of San Diego’
United States
Los Angeles Police Department
Thames valley Police
‘The mission of the New York city police department is to enhance the quality of life in our city by working in partnership with the community and in accordance with constitutional rights to enforce the laws, preserve the peace, reduce fear, and provide for a safe environment’ ‘It is the mission of the Los Angeles police department to safeguard the lives and property of the people we serve, to reduce the incidence and fear of crime, and to enhance public safety while working with the diverse communities to improve their quality of life. Our mandate is to do so with honor and integrity, while at all times conducting ourselves with the highest ethical standards to maintain public confidence’ ‘To have police and the community working together to establish a safer environment by reducing violence, crime and fear’
United States
New York City Police Department
Mission statement
Country
Police department
City of San Diego (2011)
Milton Keynes Police (2011)
Glenorchy City Council (2011)
NSW Police Force (2011)
LAPD (2011)
NYPD (2011)
Source
Table 4.1 Examples of police services in western democracies having the reduction of the fear of crime as a primary objective
54 4 Managing Fear of Crime
Policing Fear of Crime
Case Study: The New York Police Department’s (NYPD) Policing Model Bratton (1995, 1996), who was NYPD Police Commissioner from 1994 to 1996, and Kelling and Coles (1997) describe the sequence of events that led to the NYPD’s adoption of a zero-tolerance policing model. During the 1970s, before the implementation of zero-tolerance policing, New York City mirrored the spiral of decay described in Wilson and Kelling’s (1982) broken windows thesis. Unchecked disorder was seen to be leading to more significant crime, disorder and widespread fear. The public began attempting to restore order themselves, which placed direct pressure on the NYPD and other official organizations to address order-related quality-of-life issues. NYPD responded with concerted efforts targeting fear of crime, graffiti, panhandling and homeless people in areas such as Bryant Park and Times Square and on the New York City subway. Order restoration continued to remain a priority throughout the 1980s and early 1990s. These zero-tolerance policing activities have been widely hailed for reducing the city’s crime levels (e.g. Bowling, 1999; Bratton, 1996; Greene, 1999). For example, the New York City Mayor’s Management Report (1998) lists reductions in felony crimes, increases in narcotics arrests and the continued policing of minor disorder as improvements in the quality of life for citizens and neighbourhoods. However, despite the emphasis on fear of crime prior to, and during, the implementation of zero-tolerance policing, debate over the success of the strategy has focused almost exclusively on crime (e.g. Bowling, 1999; Bratton, 1996; Greene, 1999). Bratton (1996) discusses the success of zero-tolerance policing in terms of the number of people arrested for qualityof-life crimes and states that fear was reduced following the aggressive control of disorder. Yet at no point does Bratton (1996) attempt to verify this claim with comparable data used for the section of his argument pertaining to crime. No research is cited which attempts to ascertain levels of fear and whether they have changed in relation to zero-tolerance policing. This suggests that fear of crime was used more for political leverage and not specifically as an issue to be dealt with, monitored and analyzed in the same focused manner as crime. It also means that an assessment of the effects of zero-tolerance policing on fear reduction has to be based on a wider discussion of police–community relations. Nevertheless, Bratton argued that police activities and police departments should expect to have an impact on crime, disorder and fear and that this should result in proactive tactics that focus on the problems that generate crime. This key assumption formed the basis of the four principles that continue to guide the patrol and investigative work of the NYPD: timely accurate
55
56
4 Managing Fear of Crime
intelligence, rapid deployment; effective tactics; and relentless follow-up and assessment. One of the arguments that Bratton (1996) uses to justify the zerotolerance approach is that it is based on community-oriented policing (COP). Bratton lists three elements he considers critical to the success of COP: partnership, problem solving and prevention. Partnership refers to the greater effectiveness of the police when they work with the community, not apart from it. Problem solving centres on the focus of police and the community to deal with crime and the signs of crime, while prevention is the logical outcome of the first two elements. Bratton claims that the dramatic drop in index crime in New York City in the 1990s is the result of COP that focuses on partnership, problem solving and prevention. However, others have criticized the style of policing adopted at the community and neighbourhood level. For example, Greene (1999) highlights that civil rights claims against the police for abusive conduct increased by 75% in the four years prior to 1999. Amnesty International (1996a, b) also raised concerns over the use of excessive force within the NYPD. They state that allegations of police brutality continued to rise between 1994 and 1996 while deaths in custody also rose substantially between 1993 and 1994. Harcourt (1998) criticizes New York style policing on the basis that it aims to watch, control, relocate and, ideally, exclude members of the community categorized as disorderly. In addition, those arrested for quality-of-life offences are burdened with a criminal record that may haunt them on future job and school applications. Such outcomes are at odds with the principles of partnership, problem solving and prevention that Bratton (1996) argues are central to COP. Hence, by using traditional policing methods and excessive force at the neighbourhood level, it seems likely that the NYPD will undermine the potential for the police to work effectively with some of the communities they aim to serve. However, Amnesty International (1996b) suggests that the police brutality within the NYPD is the result of police, in many cases, ignoring the NYPD’s own guidelines and point towards a significant gap between police policy and practice. It is thus unclear whether the NYPD example means that the disorder-removal and quality-of-life approach cannot significantly reduce fear, or whether well-founded policing policies simply broke down at the community–police level.
Environmental Design and Fear of Crime Strategies of Crime Prevention Through Environmental Design (CPTED) are ‘based on the theory that proper design and effective use of the built environment can reduce the incidence and fear of crime and make an improvement in the quality
Environmental Design and Fear of Crime
57
of life’ (Crowe, 1991). The primary goal of CPTED is to modify the physical environment so that it deters criminal activity, thereby making it safer for pedestrian activity, thus reducing fear of crime (JHSA, 1999).4 Equally important is the aim of encouraging people to use previously avoided public spaces (Oc and Tiesdell, 1997). These aims reflect the fundamental assumption in CPTED that environmental characteristics can be manipulated to effect human social behaviour, which subsequently reduces both the incidence of and the fear of crime (Crowe, 1991; Oc and Tiesdell, 1997; Steventon, 1996). While Jacobs (1961)5 is acknowledged as a forerunner in CPTED, Jeffery (1971) is seen to have initiated CPTED in his book. The author argues that urban design can play a role in crime prevention when security is considered in street and building plans (Jeffery, 1971). Despite these seminal works, modern CPTED strategies are based predominantly on Newman’s (1972) concept of ‘defensible space’ (Cozens et al., 2001). Newman (1972) drew on Jacobs’s insights to devise his theory of defensible space and proposed that altering the physical environment reduces opportunities for crime in urban areas (Newman, 1972).6 Defensible spaces primarily communicate residential control, have high prospects for natural surveillance and are difficult to escape from (Oc and Tiesdell, 1997; Schweitzer et al., 1999). Newman’s CPTED model therefore involves residents promoting surveillance opportunities, defining territorial boundaries, limiting access, eliminating conflicting uses, providing amenities and improving area aesthetics (Oc and Tiesdell, 1997; Pollack, 1980). Brantingham and Brantingham (1993) built on this logic by commenting that city planners can shape nodes, edges and paths in environments to affect broad patterns of crime through CPTED techniques. They drew on theories of situational crime prevention and the notion that criminal events require a convergence of victims, offenders and opportunity in space and time. Since these major CPTED theories, planners and policy makers have readily adopted the suggested principles. For example, the Department for Transport Urban Planning and the Arts in Australia encourages access controls that are designed to
4 The arrangement of urban form and activity, later dubbed CPTED, was identified by Pollack (1980) as one of three environmental-modification approaches to crime control. The other two approaches are the management of the environment (for example through police activity) and the use of protective devices (for example locks). 5 Jacobs proposed that feelings of safety in inner city areas are dependent on those areas being in continuous public use. Jacobs identified three main qualities of a safe city: territoriality, surveillance and social controls. To promote these there must be a clear demonstration between public and private space, buildings must be oriented to promote surveillance, and a diversity of street activities present to promote use and vitality (Jacobs, 1961; Oc and Tiesdell, 1997; Taylor and Gottfredson, 1986). 6 After studying crime in public housing, Newman observed that crime was discouraged from ‘zones of territorial influence’ that residents maintained surveillance over and defended (Newman, 1972; Pollack, 1980). Newman termed these areas defensible spaces, which he defined as a ‘range of mechanisms – real and symbolic barriers, strongly defined areas of influence, and improved opportunities for surveillance – that combine to bring an environment under the control of its residents’ (Newman, 1972).
58
4 Managing Fear of Crime
keep unauthorized persons out of particular spaces. Such controls include doors, shrubs, fences and even lit porch lights (DTUPA, 2002; Wagner, 1997). Similarly, physical barriers are also used to create clear boundaries between public and private areas. These generally signify ownership and include fences, hedges, pavement treatments, art, signs, good maintenance and landscaping (DTUPA, 2002; Schweitzer et al., 1999). Signs of increased surveillance are also popular in CPTED projects. These include the presence of neighbourhood watch signs and even porches and mailboxes, which increase opportunities for surveillance (Oc and Tiesdell, 1997; Schweitzer et al., 1999). Similarly, the British Crime and Disorder Act 1998 requires all local authorities to take crime and disorder into account in all aspects of decision making (Cozens et al., 2001). The British Department of Environment’s Secured by Design scheme further provides an accolade for housing schemes that meet specific CPTED design criteria (Kitchen, 2002). The criteria incorporate key principles such as aiming to create defensible space, territoriality and natural surveillance while minimizing escape routes, crime generators and fear generators (Kitchen, 2002). Another British approach, New Urbanism, draws on Jacobs’s (1961) works. New urbanism recognizes the importance of promoting human activity in the environment in order to achieve safety. A major feature is the encouragement of natural surveillance (Kitchen, 2002). A number of researchers have suggested that the fear of crime can be successfully reduced through such environmental or order-related improvements. Painter (1996) essentially identifies darkness and disorder as pivotal environmental cues that heighten fear in pedestrians and argues that good-quality street lighting can make a substantial contribution as a fear-reducing strategy. The author tested her assumptions in an experiment that looked at the impact of lighting improvements on crime, disorder and fear in three urban streets. The results showed a marked reduction in fear of physical attack and over 90% of pedestrians interviewed in all locations thought fear of crime in the surrounding area had gone down. In a similar study, Herbert and Davidson (1994) found that improved lighting in two British cities significantly reduced the fear of crime. Table 4.2 shows how the public perceived a number of problems central to the fear of crime such as fear of going out after dark and a range of physical and social incivilities to have decreased following the lighting improvements. However, the success of reducing fear of crime through environmental design has been questioned. Herbert and Davidson (1994) suggest that the astonishing influence of improved street lighting on the fear of crime in their study may be due to a halo effect. They describe this as the process where a single change appeared to stimulate other changes in aspects of local life, some of which had no obvious links to the actual environmental modification. Painter (1996) suggests that the effectiveness of the lighting strategy in her study was due to it altering the behaviour of the public, including potential offenders. The author suggests that improved sight lines, increased perceived risks of offending, increased pedestrian density and traffic flow and the associated enhanced natural surveillance all contribute to the potential for improved street lighting to reduce the fear of crime. However, the author
Environmental Design and Fear of Crime
59
Table 4.2 Problems in Herbert and Davidson’s (1994) study thought to be more common before and after relighting Hull
Cardiff
Problem
Before %
After %
Before %
After %
Afraid to go out after dark Burglary Rubbish/litter Theft of/from cars Noise nuisance Youths hanging around Vandalism Drunks Robbery Street lighting Being pestered
39 34 44 26 18 17 17 16 13 10 5
13 18 41 15 10 8 11 10 6 4 3
61 47 35 52 25 49 44 16 29 20 12
28 26 35 34 14 20 23 7 13 4 9
tempers her optimistic suggestions by emphasizing that if lighting is to be an effective strategy, planners need to be clear about the mechanisms they are expecting to induce in a specific environmental and social setting. In contrast to the successful projects described by Herbert and Davidson (1994) and Painter (1996), Nair et al. (1993) found that a wide range of environmental improvements, including improved street lighting, failed to reduce the fear of crime in an area of Glasgow, Scotland. Survey respondents later told the authors that the improvements were mostly made in areas where there was little public use and that they would have preferred the changes to have been made in more heavily used public spaces. Attempts to reduce fear through environmental improvements have been strongly criticized by others (e.g. Stanko, 1995; Koskela and Pain, 2000). Koskela and Pain (2000) list a number of projects, such as the work of Nair et al. (1993), which have failed to produce long-term benefits in terms of fear reduction. They argue that ‘designing out fear strategies’ only deal with one immediate and visible source of fear and leave few alternatives in the face of failure. Similarly, Stanko (1995) argues that environmental improvements, if not coupled with increased safety within private houses and relationships, will not significantly reduce women’s fear of crime. Koskela and Pain (2000) suggest that physical and social cues to fear are inextricably linked and that social connotations often explain why some places are regarded as particularly frightening. For example, the authors argue that women’s routine avoidance of certain areas is largely underpinned not by fear of concrete structures but by fear of unknown men. Some researchers have even gone so far as to say that CPTED actually exacerbate fear of crime. For example, Doeksen (1997) suggests that the growing concern for personal security in New Zealand and Australia has resulted in residents, private developers and engineers designing physical environments that emphasize separation over interaction. The author uses the example of how the important role of the colonial veranda as a means of promoting social surveillance within the streetscape has been reduced in many neighbourhoods because residents are literally fencing
60
4 Managing Fear of Crime
themselves off. Similarly, in the United States and the United Kingdom, protective responses to the fear of crime such as the rapid proliferation of gated communities and the implementation of closed-circuit television surveillance systems have been criticized for contributing to the atomization of communities and the breakdown of public life (e.g. Blakely and Snyder, 1997; Helsley and Strange, 1999; Lymes, 1997; Graham et al., 1998 in Ditton, 2000; Jackson and Gray, 2010). Where individuals become isolated, mistrusting and fearful in their homes, they are unlikely to form social ties with neighbours (Ross and Mirowsky, 1999). WilsonDoenges (2000) identified such patterns in gated communities in Orange County, California. Despite the fact that developers of gated communities aim to create a strong sense of community by providing access control and security walls, and that such measures increase levels of perceived safety (Lymes, 1997), residents from a high-income gated community reported significantly lower sense of community scores compared to those from a non-gated community (Wilson-Doenges, 2000). The negative effects of protective measures on the sense of community within residential areas are also evident within retail centres. Tiesdell and Oc (1998) describe the concept of the ‘fortress city’ which is based upon the separation of those who belong and ‘the other’. This entails the physical segregation, territorialization and defence of space with access controls. The authors argue that by isolating and defending particular territories, fortress cities protect only certain individuals or groups while undermining the public realm’s ideal qualities of social inclusivity, collectivity and universal accessibility. The cost of creating apparently safe, small environments through the use of target-hardening procedures may come at the cost of increasing the fear of wider public spaces (Brown and Polk, 1996) and may also displace the occurrence of crime onto areas or sectors of society that are unable to protect themselves to the same degree (Skogan and Maxfield, 1981; Herbert and Davidson, 1994; Davis, 1990; Tiesdell and Oc, 1998). For example, Davis (1998) outlines how the financial core of Los Angeles was protected by a barrage of security measures during the 1992 riots, while extensive damage was taking place in the old business district nearby. The implication of these criticisms is that any environmental improvements also need to impact upon social factors influencing the fear of crime (Koskela and Pain, 2000; Stanko, 1995). Logically, it should be possible, and may be more appropriate in some cases, to reduce fear through inducing social changes within the local community. Wikström (1995) describes a situation in Sweden where a particular street corner was identified as being a focus for disorder. The venue comprises of a number of restaurants and a bar that were frequented by upper-level secondary students and working class ‘rockers’. The author explains how the area was peaceful during the day and mostly at night. Only during the late evenings and especially at weekends did it become a ‘hotspot’ for stranger to stranger assault. Thus, the area provided a focus for crime and was also likely to inspire fear. Brown and Polk (1996) suggest a number of social measures to address the time-specific nature of disorder in Wikstrom’s (1995) example. The measures outlined by Brown and Polk (1996) were increasing police supervision of premises and public spaces at closing times, training bar staff in management techniques to lower confrontations
References
61
with patrons at closing times and encouraging management practices that would result in keeping more orderly premises. Oc and Tiesdell (1997) and Thomas and Bromley (2000) discuss similar examples to Wikstrom’s (1995) in British city centres but suggest fear-reduction strategies that are a combination of both physical and social measures. Thomas and Bromley (2000) advocate social initiatives such as encouraging a wider range of activities and repopulating city centres to increase natural surveillance in combination target-hardening measures to alleviate motorist anxieties.
Chapter Review: Police, Community and Government Cooperation Fear of crime can be managed through a diverse range of approaches adopted by police, communities and governments. While traditional policing models have failed to acknowledge fear of crime, many models now see fear of crime as fundamental to proactive policing and crime prevention. Nevertheless, with regard to fear of crime, these models are limited by poor knowledge, their generalized responses or their lack of community involvement. Community involvement in fear-reduction strategies can help reduce the fear of crime experienced by public participants. In addition, governments can potentially reduce fear of crime through policies and plans that improve social infrastructure and the design of the environment. The choice of primarily policing, social or environmental strategies to reduce the fear of crime is likely to be dependent upon the nature of the problem in different communities and settings. In some cases the causes of fear may be predominantly social, such as in Wikstrom’s (1995) example. In other situations fear of crime may result from a combination of physical and social cues (e.g. Oc and Tiesdell, 1997; Thomas and Bromley, 2000). Implicit in this argument and the call for strategies to be relevant to the specific environmental and social settings of local areas (e.g. Painter, 1996) is the need for flexibility on behalf of the communities and police involved and an understanding of where and when fear of crime is a problem. Strategies to reduce fear of crime may be based on either social or physical measures and need to be grounded in a solid understanding of the specific environmental and social settings of local areas. This requires flexibility on behalf of the police and local communities as well as an understanding of where and when fear of crime is a problem.
References Adams, R. E., W. M. Rohe, et al. (2005). “Awareness of community-oriented policing and neighborhood perceptions in five small to midsize cities”. Journal of Criminal Justice 33(1): 43–54. Amnesty International. (1996a). “USA: police brutality widespread problem in New York City”. Retrieved April 4th, 2011, from http://www.amnesty.org/en/library/info/AMR51/050/1996/en.
62
4 Managing Fear of Crime
Amnesty International. (1996b). “United States of America: police brutality and excessive force in the New York City Police Department”. Retrieved April 4th, 2011, from http://www.amnesty. org/en/library/info/AMR51/036/1996/en. Ashby, D. I. and P. A. Longley (2005). “Geocomputation, Geodemographics and Resource Allocation for Local Policing”. Transactions in GIS 9(1): 53. Baker, T. and L. Wolfer (2003). “The crime triangle: alcohol, drug use and vandalism”. Police Practice and Research 4(1): 47–61. Bennett, T. (1991). “The effectiveness of a police-initiated fear-reducing strategy.” British Journal of Criminology 31(1): 1–14. Blakely, E. J. and M. G. Snyder (1997). Fortress america: gated communities in the United States. Washington, DC, The Brookings Institution. Borooah, V. and C. Carcach (1997). “Crime and fear. Evidence from Australia”. The British Journal of Criminology 37(4): 635–657. Bowling, B. (1999). “The rise and fall of New York murder: zero tolerance or crack’s decline?”. The British Journal of Criminology 39(4): 531–554. Brantingham, P. L. and P. J. Brantingham (1993). “Nodes, Paths and Edges – Considerations on the Complexity of Crime and the Physical-Environment”. Journal of Environmental Psychology 13(1): 3–28. Bratton, W. J. (1995). Great expectations: how higher expectations for police departments can lead to a decrease in crime. Paper presented to the National Institute of Justice Policing Research Institute “Measuring what matters conference”. Washington DC, 28 November. Bratton, W. J. (1996). Cutting crime and restoring order: what America can learn from New York’s finest. Heritage Lecture 573. Brown, M. and K. Polk (1996). “Taking fear of crime seriously: the Tasmanian approach to community crime prevention”. Crime and Delinquency 42(3): 398–420. Cameron, D. (2002). “News and features – respect family – Moroney’s crime fighting motto” The Sydney Morning Herald, July 2, 2002. Centre for Problem-Oriented Policing (CPOP). (2003). “What is Problem-Oriented Policing?” Retrieved 19/04/2006, 2006, from http://www.popcenter.org/about-whatisPOP.htm; http:// www.popcenter.org/about-keyelements.htm; http://www.popcenter.org/about-SARA.htm; http://www.popcenter.org/about-situational.htm. City of San Diego. (2011). “Police Department: Mission Statement”. Retrieved April 4th, 2011, from http://www.sandiego.gov/police/about/mission.shtml. Cordner, G. W. (1986). “Fear of crime and the police: an evaluation of a fear reduction strategy”. Journal of Police Science and Administration 14(3): 223–233. Cozens, P., D. Hillier, et al. (2001). “Crime and the design of residential property – exploring the perceptions of planning professional, burglars and other users part 2”. Property Management 19(4): 222. Crank, J. P., A. Giacomazzi, et al. (2003). “Fear of crime in a nonurban setting”. Journal of Criminal Justice 31(3): 249–263. Crowe, T. D. (1991). “Safer Schools by Design”. Security Management 35(9): 81. Davis, M. (1990). City of quartz: excavating the future in Los Angeles. New York, NY, Verso. Davis, M. (1998). Ecology of fear: Los Angeles and the imagination of disaster. New York, NY, Metropolitan Books. Department for Transport Urban Planning and the Arts (DTUPA) (2002). Crime prevention through environmental design and urban design: Design Issues for safe neighbourhoods. Adelaide. Dietz, A. S. (1997). “Evaluating community policing: quality police service and fear of crime”. Policing 20(1): 83. Ditton, J. (2000). “Crime and the city.” The British Journal of Criminology 40(4): 692–709. Dixon, J. (1995). Creating safer communities. Australian violence: contemporary perspectives II. D. Chappell and S. Egger (Eds.). Australian Institute of Criminology, Canberra: 339–346. Doeksen, H. (1997). “Reducing crime and the fear of crime by reclaiming New Zealand’s suburban street”. Landscape and Urban Planning 39(2–3): 243–252.
References
63
Dubow, F., E. McCabe, et al. (1979). Reactions to crime: a critical review of the literature. US department of justice and LEA administration. Washington, DC, Government Printing Office. Garofalo, J. (1979). “Victimisation and the fear of crime”. Journal of Research in Crime and Delinquency 16: 80–97. Glenorchy City Council. (2011). “Police Tasmania – Launceston”. Retrieved April 4th, 2011, from http://www.infoline.tas.gov.au/content/Police_Tasmania___Launceston.GCC Glensor, R. W. and K. Peak (1996). “Implementing change: Community-oriented policing and problem solving”. FBI Law Enforcement Bulletin 65(7): 14. Goldstein, H. (1979). “Improving Policing: A Problem Oriented Approach”. Crime and Delinquency 25: 236–258. Grabosky, P. N. (1995). “Fear of crime, and fear reduction strategies”. Current Issues in Criminal Justice 7(1): 7–19. Greene, J. (1999). “Zero Tolerance: A case study of police policies and practices in New York City”. Crime and Delinquency 45(2): 171–187. Harcourt, B. E. (1998). “Reflecting on the subject: a critique of the social influence conception and deterence, the broken windows theory, and order maintenance policing New York style”. Michigan Law Review 97(2): 291–389. Helsley, R. W. and W. C. Strange (1999). “Gated communities and the economic geography of crime”. Journal of Urban Economics 4: 80–105. Herbert, D. and N. Davidson (1994). “Modifying the built environment: the impact of improved street lighting”. Geoform 25(3): 339–350. Innes, M., N. Fielding, et al. (2002). Signal crime and control signals: Towards an evidence based framework for reassurance policing. Guilford, University of Surrey. Jackson, J. and E. Gray (2010). “Functional fear and public insecurities about crime.” British Journal of Criminology 50(1): 1–22. Jacobs, J. (1961). The death and life of Great American cities. New York, NY, Vintage Books. Jeffery (1971). Crime prevention through environmental design. Beverly Hills, CA, Sage. John Howard Society of Alabama (JHSA) (1999). Fear of crime. Alabama, John Howard Society of Alabama. Katz, C. M., V. J. Webb, et al. (2003). “Fear of gangs: A test of alternative theoretical models”. Justice Quarterly 20(1): 95–130. Kelling, G. L. and C. M. Coles (1997). Fixing broken windows: restoring order and reducing crime in our communities. New York, NY, Touchstone. Kenney, D. J. (1987). Crime, fear and the New York subways: the role of citizen action. New York, NY, Prager. Kitchen, T. (2002). “Crime Prevention and the British Planning System: New Responsibilities and Older Challenges”. Planning Theory and Practice 3: 155–172. Koskela, H. and R. Pain (2000). “Revisiting fear and place: women’s fear of attack and the built environment”. Geoforum 31(2): 269–280. Lawton, B. A., R. Taylor, et al. (2005). “Police officers on drug corners in Philadelphia, drug crime, and violent crime: intended, data fusion, and displacement impacts”. Justice quarterly 22(4). Los Angeles Police Department (LAPD). (2011). “The Mission Statement of the LAPD”. Retrieved April 4th, 2011, from http://www.lapdonline.org/inside_the_lapd/content_basic_view/844 Lymes, D. (1997). “The fortification of suburbia: investigating the rise of enclave communities”. Landscape and Urban Planning 39: 187–203. Millie, A. and V. Herrington (2005). “Bridging the gap: understanding reassurance policing.” The Howard Journal of Criminal Justice 44(1): 41. Milton Keynes Police. (2011). “MK Police Intro – Police Aims”. Retrieved April 4th, 2011, from http://www.mkweb.co.uk/police_news/ Nair, G., J. Ditton, et al. (1993). “Environmental improvements and the fear of crime: the sad case of the ‘Pond’ area in Glasgow”. The British Journal of Criminology 33(4): 555–561. New York Police Department (NYPD). (2011). “About NYPD – Mission and Values”. Retrieved April 4th, 2011, from http://www.nyc.gov/html/nypd/html/home/mission.shtml.
64
4 Managing Fear of Crime
Newman, O. (1972). Defensible space: crime prevention through urban design. New York, NY, Macmillan. Novak, K. J., J. L. Hartman, et al. (1999). “The effects of aggressive policing of disorder on serious crime”. Policing 22(2): 171–190. NSW Police Force. (2011). “Profile Of Wollongong Local Area Command”. Retrieved April 4th, 2011, from http://www.policensw.com/region/southern/wollongong/lac/rw7.html Oc, T. and S. Tiesdell (1997). Safer city centres: reviving the public realm. London, Chapman. Painter, K. (1996). “The influence of street lighting improvements on crime, fear and pedestrian street use, after dark”. Landscape and Urban Planning 35(2–3): 193–201. Pollack, L. M. (1980). “Territoriality and fear of crime in elderly and nonelderly homeowners”. Journal of Social Psychology 111(1): 119. Ross, C. E. and J. Mirowsky (1999). “Disorder and decay: The concept and measurement of perceived neighborhood disorder”. Urban Affairs Review 34(3): 412–433. Salmi, S., M. Gronroos, et al. (2004). “The role of police visibility in fear of crime in Finland”. Policing-an International Journal of Police Strategies & Management 27(4): 573–591. Schweitzer, J. H., J. W. Kim, et al. (1999). “The impact of the built environment on crime and fear of crime in urban neighborhoods”. Journal of Urban Technology 6(3): 59–73. Sims, B., M. Hooper, et al. (2002). “Determinants of citizens’ attitudes toward police – Results of the Harrisburg Citizen Survey – 1999”. Policing-an International Journal of Police Strategies & Management 25(3): 457–471. Skogan, W. G. (1986). Fear of crime and neighbourhood change. Communities and crime. A. J. Reiss and M. Tonry (Eds.). University of Chicago Press, Chicago, IL: 203–230. Skogan, W. G. (1990). Disorder and decline: crime and the spiral decay in American neighbourhoods. Los Angeles, CA, University of California Press. Skogan, W. G. and S. M. Hartnett (1997). Community policing, Chicago style. New York and London, Oxford University Press. Skogan, W. G. and M. G. Maxfield (1981). Coping with crime: individual and neighborhood reactions. Beverly Hills, CA, Sage Publications. Smith, S. J. (1987). “Fear of crime: beyond a geography of deviance”. Progress in Human Geography 11: 1–23. Spelman, W. (2004). “Optimal targeting of incivility-reduction strategies.” Journal of Quantitative Criminology 20(1): 63–88. Stanko, E. A. (1995). “Women, crime and fear”. Annals of the American Academy of Political & Social Science 539: 46–58. Stanko, E. A. (2000). Victims R Us: the life history of fear of crime and the politicisation of violence. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London. Steventon, G. (1996). “Defensible space: a critical review of the theory and practice of a crime prevention strategy”. Urban Design 1(3): 235–245. Taylor, R. B. and S. D. Gottfredson (1986). Environmental design, crime and prevention: an examination of community dynamics. Communities and crime. A. J. Reiss and M. Tonry (Eds.). University of Chicago Press, Chicago: 387–416. Thomas, C. and R. Bromley (2000). “City-centre revitalisation: problems of fragmentation and fear in the evening and night-time city”. Urban Studies 37(8): 1403–1429. Tiesdell, S. and T. Oc (1998). “Beyond ‘fortress’ and ‘panoptic’ cities – towards a safer urban public realm”. Environment and Planning B: Planning and Design 25: 639–655. Tulloch, J. (1998). Quantitative Review. Fear of crime. J. Tulloch, D. Lupton, W. Blood, et al. (Eds.). National Campaign Against Violence and Crime (NCAVAC), Canberra. Wagner, A. E. (1997). “A study of traffic pattern modifications in an urban crime prevention program”. Journal of Criminal Justice 25(1): 19–30. Walklate, S. (2000). Trust and the problem of community in the inner-city. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London. Weisburd, D. and J. E. Eck (2004). “What can police do to reduce crime, disorder, and fear?” Annals of the American Academy of Political and Social Science 593: 42–65.
References
65
Wikström, P. H. (1995). Preventing city center street crimes. Building a safer society: strategic approaches to crime prevention. M. Tonry and D. P. Farrington (Eds.). University of Chicago Press, Chicago, IL: 429–468. Wilson, J. Q. and G. L. Kelling (1982, March). “The police and neighbourhood safety: broken windows”. The Atlantic Monthly: 29–38. Wilson-Doenges, G. (2000). “An exploration of sense of community and fear of crime in gated communities”. Environment and Behavior 32(5): 597–611. Xu, Y., M. Fielder, et al. (2005). “Discovering the impact of community policing: the broken windows thesis, collective efficacy and citizen’s judgement”. Journal of Research in Crime and Delinquency 42(2): 147–186.
Chapter 5
Investigating Fear of Crime
Defining Fear of Crime While fear of crime is easily interpreted during everyday discourse, it needs to be defined for research purposes (Skogan, 1999). The conceptual definition of fear of crime has clear consequences for its operationalization which in turn impacts how research results and findings are interpreted (Skogan, 1999). The definition of fear of crime and the research methods must therefore be clarified before investigations can be conducted and compared to other studies with any validity (Ferraro and LaGrange, 2000). Prior to 1980, researchers rarely explicitly defined fear of crime (Yin, 1980). In his comprehensive review of literature, Yin (1980) found that only Sundeen and Mathieu (1976) explicitly defined of fear of crime as ‘anxiety and concern that persons have of becoming a victim’. Five years later, Garofalo (1981) defined fear of crime as an emotional reaction characterized by a sense of danger and anxiety, produced by the threat of harm. Warr (1984) stated fear of crime had ‘acquired so many diverging meanings in the literature that it is in danger of losing any specificity whatsoever’. This problem was such that the concept of fear of crime and its research utility was considered ‘negligible’ (Ferraro and LaGrange, 1987). Comments such as this have continued well into the 1990s (e.g. Ewald, 2000; Stanko, 2000). Despite an abundance of studies on the topic, the literature still exhibits considerable confusion and ambiguity in relation to defining fear of crime (Pantazis, 2000; Warr, 2000). Fear of crime is equated with a diverging array of emotions, insecurities, concerns, perceptions or judgements, and attitudes or values (see Ditton et al., 2000; Ferraro and LaGrange, 2000; Furstenberg, 1971; Mawby et al., 2000; Warr, 2000). In order to define fear of crime, a strategy is needed to systematically unpack the concept (Ditton et al., 2000). It is useful to examine the individual terms ‘fear’ and ‘crime’ when defining the concept of ‘fear of crime’ as a whole. To do this, we draw upon and refine one of the most commonly used definitions, developed by Ferraro and LaGrange in 1987, which has proven influential in subsequent research (e.g. Ferraro, 1995; Rountree and Land, 1996; Tulloch, 1998). Ferraro and LaGrange (1987) define fear of crime as ‘the negative emotional reactions generated by crime or symbols associated with crime’.
B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_5, C Springer Science+Business Media, LLC 2012
67
68
5
Investigating Fear of Crime
Fear Is an Emotion, Not Cognition Much of the debate and confusion surrounding the concept of fear of crime arises from a failure to distinguish between emotion (i.e. what we feel) and cognition (i.e. what we think) (Ferraro, 1995; Warr, 2000). Defining fear as an emotion is therefore important because, although related, emotive and cognitive responses to crime are conceptually different. Thus studies confusing the two states could have markedly dissonant results that cannot validly be compared (Ferraro and LaGrange, 2000; Rountree and Land, 1996). According to Ferraro and LaGrange’s (1987) definition, fear of crime consists of ‘the negative emotional reactions’ [emphasis added]. Emotion is a distinctive mental state, a feeling state, which includes physical responses that prompt or restrain motivated behaviour (Carlson and Hatfield, 1992). In contrast, some researchers view fear of crime as a cognitive assessment. Cognitive assessments encompass people’s judgements about crime – their evaluation of personal risk (i.e. perceived risk) and their general concern about crime (Skogan, 1999). This differentiation was first described by Ferraro and LaGrange (1987) when arguing that fear of crime is strictly an emotional response. They designed the taxonomy shown in Table 5.1 to differentiate risk from fear, with perceptions of crime forming a continuum ranging from cognitive to affective. The cognitive perceptions relate to judgements of risk and the affective perceptions relate to fear reactions. The authors define the concept of fear of crime as being limited to the emotional reaction arising from crime, or the symbols that a person associates with crime (i.e. cells C and F of Table 5.1 below). Ferraro (1995) defines perceived risk as an acknowledgement of potential danger, real or imagined. This danger involves exposure to the chance of injury or loss (Ferraro, 1995). Assessments of risk or safety are people’s perceptions of the probability of someone being victimized (Ferraro and LaGrange, 2000; Skogan, 1999). The distinction between perceptions of risk and threat of victimization is cumbersome. According to Skogan (1999) they are distinct yet related. The author implies that perceptions of risk refer to actual rates of victimization and that threat refers to how at danger one personally is of being victimized, taking into account any strategies that have been adopted to reduce one’s vulnerability. However, Mesch (2000) Table 5.1 Taxonomy of crime perceptions developed by Ferraro and LaGrange (1987) Type of perception: cognitive and affective Level of reference
Judgements
Values
Emotions
General
A. Risks to others: crime or safety assessments D. Risk to self: safety of self
B. Concern about crime to others
C. Fear for others’ victimization
E. Concern about crime to self: personal intolerance
F. Fear for selfvictimization
Personal
Defining Fear of Crime
69
draws the distinction between perceived risk and fear of crime, finding that they are related to different predictors. Researchers also define fear of crime as a concern or worry about crime, which can be referred to as a value (Ferraro and LaGrange, 2000). However, concern is not linked to fear, but to a state of agitation regarding the level of crime in one’s environment and a belief that crime is a serious social problem (Furstenberg, 1971; Oc and Tiesdell, 1997). This is also regarded as being distinct from the perceptions of risk or threat. Skogan (1999) elaborates, stating that concern is ‘a judgment about the frequency or seriousness of events and conditions’ and is distinct from threat because people believe they are capable of dealing with such crime. Garofalo (1981) provides the example of people being more concerned than fearful when it comes to property crime because the threat of physical harm is low compared to personal crime (Garofalo, 1981). Similarly, worry about crime may be reduced by behavioural changes without impacting on fear (Tulloch et al., 1998). With these contrasting meanings, distinguishing between fear (an emotion) and either risk, concern or worry can help when attempting to validate or draw comparisons between different fear-of-crime studies (Lewis and Salem, 1986).
Fear in Relation to Other Emotional Reactions and Stimuli that Trigger Fear Ferraro and LaGrange’s (1987) definition of fear as an emotion fails to distinguish fear from other emotional reactions, like sadness, anger or despair (Warr, 2000). Some researchers argue that many surveys aimed at examining fear of crime are actually tapping into other emotions (e.g. Innes et al., 2002; Innes, 2004). Farrall and Ditton (1999) suggest that respondents are more likely to feel anger, outrage or annoyance rather than fear when thinking about crime. Thus distinguishing fear from these other emotions is important when comparing the potentially discordant results from fear-of-crime studies. This also reinforces the need to succinctly define and target fear when undertaking research in the area. Fear is one of the six primary human emotions essential for survival (Neill, 2001) and is considered to prompt one to protect oneself against loss when confronted by a risk (Clark, 2003). Essentially, fear is a negative emotion that describes feelings of alarm, dread or apprehension about tangible or perceived threats (Clark, 2003; Innes, 2004). Thus, fear is an emotion characterized by an expectation of danger that is produced by the threat of harm (Williams and Dickinson, 1993; Sluckin, 1979). Fear forewarns danger, promoting vigilance and a fight or flight response (Carlson and Hatfield, 1992; Oatley and Jenkins, 1996). In general, fear is determined by an object or stimulus that is expected to cause harm and is not qualitatively different from other forms of fear (Warr, 2000). However, it is important to clarify what makes fear of crime distinct from other forms of fear. Fear of crime is specifically the fear of being harmed during criminal victimization and it is generated by crime or symbols associated with crime (Warr, 2000). These symbols can be thought of as environmental cues that relate to some aspect of crime (Williams and Dickinson,
70
5
Investigating Fear of Crime
1993). Numerous problems, discussed later in this chapter, arise from studies tapping into people’s diffuse or ‘formless’1 feelings of fear, rather than specific or ‘concrete’2 fear of crime. Fear may be aroused by immediate danger, for example an armed attacker, but is often experienced as anticipating a potential threat (Carlson and Hatfield, 1992; Kaplan, 1973). This occurs when people react to environmental cues that imply danger because they are associated with crime (Garofalo, 1981; Warr, 2000). Psychologists identify the emotional reaction to potential threats as anxiety (Clark, 2003). Warr (1984) reasons that anxiety is much more common than fear associated with real encounters of crime (Warr, 2000). Garofalo (1981) also states that behavioural changes can result from such anticipatory fear. This perhaps prompted Ferraro’s (1995) amended definition of fear of crime as an ‘emotional response of dread or anxiety to crime or symbols that a person associated with crime’ [emphasis added]. This is the conceptual definition of fear of crime that is consequently used in this chapter. A further difficulty relating to defining the term arises from the fact that the ‘crime’ in ‘fear of crime’ is also subject to contention.
Crime Involves a Violation of Criminal Law The term ‘crime’ has escaped definition in much of the criminological literature, with many studies presuming crime is self-explanatory (Ewald, 2000). However, how people conceive crime influences their response to fear-of-crime survey questions. Defining crime is therefore a necessary component when defining ‘fear of crime’. Nevertheless, even when crime is defined, opposing theoretical approaches leads to contention (Sparks et al., 2001). The two mainstream legal and social definitions of crime are discussed here.3 Traditional jurisprudential definitions of crime describe it as an act in violation of criminal law. For example, Reiss (1986) defines crime as ‘an event or sequence of events in time and space that violates the criminal statute’. Criminal law, or statute, represents those norms of conduct within a society that are intended to influence, regulate and guide the behaviour of the public (Potas, 1996). However, these social norms are formalized and enforced by a political authority through legislature and the courts (Potas, 1996; Sutherland and Cressey, 1970).4 Therefore, as Stephen 1
Formless fear is a non-specific anxiety (Friedberg and R.a.F. Inc, 1983). Concrete fear is the fear of becoming the victim of a specific crime (Friedberg and R.a.F. Inc, 1983). 3 However, there are many more approaches to defining crime (see Vold et al., 2002; Walsh and Poole, 1983; White and Haines, 2004). 4 (Oc and Tiesdell, 1997) emphasize that the definition of crime reflects the social and political processes whereby certain actions are subjected to criminalization. As crime is dependent on those with the power to label, it can be used to censure certain groups of people. The legal definition of acceptable behaviour can be modified should public concern be acknowledged – for example the introduction of bylaws outlawing the consumption of alcohol in public spaces. 2
Defining Fear of Crime
71
(1983) states, crime also becomes an ‘act or omission in respect of which legal punishment may be inflicted’ (cited in Walsh and Poole, 1983). In contrast, social perspectives of crime propose that crimes are violations of any social code, whether defined by criminal law or not (Jeffery, 1971). These social codes or ‘laws of morality’ also guide public behaviour but are not traceable to a single universally recognized rule-making institution that can enforce them through sanctions for disobedience (Potas, 1996). As social norms of conduct characterize crime, this definition includes many acts not usually regarded as legally criminal, such as drug addiction and prostitution (Jeffery, 1971). The social concept of crime links most closely with the general public’s viewpoint. It is often acts of disorder, rather than legally defined crimes, that cause fear of crime (Oc and Tiesdell, 1997). Clarifying Ferraro’s (1995) fear of crime definition, ‘crime’ in the research presented in Chapters 6 and 7, is seen as a violation of criminal law, yet it is acknowledged that the threat of crime can be triggered by acts of disorder that infringe only social norms. Some researchers contend that fear of crime examined in numerous studies is not actually a true fear of ‘crime’. These social theorists conclude that fear of crime is actually an underlying formless fear caused by different societal problems (Lane and Meeker, 2003). Bearing this in mind, researchers must be vigilant to target fear of actual, legally defined ‘crime’ when devising survey questions. The relationship between fear of crime and a number of different variables proposed by social theories are discussed in the following section on ‘factors associated with fear of crime’. Before doing so, it is necessary to note that crime, in its everyday sense, can be delineated by type of crime, subject of victimization and fear.
Types of Fear of Crime: Personal and Altruistic Points of View Rountree and Land (1996) state that researchers have ‘generally ignored the potentially important distinctions between types or dimensions of fear of crime’. Two dimensions of fear of crime are identified, one concerning the type of victimization (personal or property) and the second concerning the subject of victimization (personal or altruistic). Fear of personal crime was distinguished from fear of property crime by Ferraro and LaGrange (1992). Levels of fear and reactions to fear will vary according to whether the threat of physical harm from victimization is targeted on the person or one’s property (Garofalo, 1981). Therefore it is essential that the type of victimization (personal or property) be specified in fear-of-crime studies. Warr (2000) has been a strong advocate for the study of altruistic fear of crime, which he argues is predominant in society. The author contends that individuals may not only fear for their own personal safety when in a dangerous environment, but also for the safety of other individuals whom they value. Altruistic fear is defined as ‘an emotional reaction to the perceived danger that a household member would be a crime victim’ by Beck et al. (2004). However, it is also likely that altruistic fear extends to those outside of the household to close family and friends, or even the public at large. Nevertheless, it is necessary that researchers distinguish personal
72
5
Investigating Fear of Crime
fear (fear for oneself) from altruistic fear (fear for others) in their investigations (Warr, 2000). Refining the various types of victimization in this way has led to some improvement and clarity in results (Lewis and Salem, 1986).
Review: Key Issues to Consider When Defining Fear of Crime Historically, researchers have failed to succinctly define ‘fear’ and ‘crime’. Conceptual confusion has arisen when researchers have not defined fear of crime and assume it is commonly comprehended. This section has outlined the problems of defining ‘fear’ and ‘crime’. Drawing from the literature, it is recommended that the following points should be taken into account when undertaking research into fear of crime: • • • • •
defining fear as an emotion, not a cognition; recognizing fear is distinct from other emotions; distinguishing fear triggered by the threat of crime from formless fear; focusing on fear of crime that involves a violation of criminal law; acknowledging fear of crime can be triggered by violations of social norms, known as acts of disorder; and • being mindful of the different types of fear of crime. In doing this, researchers can better-define and operationalize fear of crime, tailoring their research design appropriately to maximize the potential for useful results.
Measuring Fear of Crime In order to scientifically investigate fear of crime, the variables in question must be accurately measured (Ferraro and LaGrange, 2000). Researchers consistently dispute the method by which fear of crime should be measured. Thus there are significant contradictions in research findings, even when examining a single dataset (Rountree and Land, 1996; Stafford and Gall, 1984; Mesch, 2000). The extent of these measurement inconsistencies seriously impedes the ability of researchers to make useful generalizations which could be used in initiatives aimed at combating fear of crime (Ferraro and LaGrange, 2000). There are three major approaches used in fear-of-crime research – namely, cognitive, affective and behavioural measures.
Problems with Cognitive Approaches to Measuring Fear of Crime The research utility of traditional cognitive approaches to measuring fear of crime is highly criticized (Rountree and Land, 1996). Despite this, they are continually used in Australian and international crime and safety surveys. Cognitive approaches include global and value- or concern-based measures.
Measuring Fear of Crime
73
Global Measures The most widespread approach to measuring fear of crime is based on perceptions of risk. Survey respondents are typically asked to assess how safe their neighbourhood is or how likely they are to be victimized (Rountree and Land, 1996). The most popular question is ‘How safe do you feel, or would you feel, out alone in your neighbourhood at night’ or something similar (e.g. Ditton and Farrall, 2000; Borooah and Carcach, 1997; Killias and Clerici, 2000; Mawby et al., 2000; Mirrlees-Black and Allen, 1998; Pantazis, 2000; Walker, 1994). Respondents answer by choosing from a list of options such as I feel ‘very safe’, ‘reasonably safe’, or ‘somewhat safe’ (ABS, 2006; Liska et al., 1988; Pantazis, 2000). As these questions do not refer to a particular crime, they are often referred to as global measures (Pantazis, 2000). There are a number of problems associated with global measures. First, they are a cognitive approach, targeting what respondents think (Ferraro and LaGrange, 1988). By asking respondents ‘How safe do you feel. . .’, global measures confuse fear of crime with perceived risk, invoking a general assessment of safety in one’s neighbourhood (LaGrange and Ferraro, 1987). Ferraro and LaGrange state that while perceived risk may be an important predictor of actual fear, (Rountree and Land, 1996), peoples’ perceptions of risk of victimization are ‘vastly different’ from their feelings of fear of victimization (Ferraro and LaGrange, 2000). Thus, perceived risk is distinct from, and cannot be used to measure, people’s fear of crime (Pantazis, 2000; Rountree and Land, 1996). Furthermore, it is uncertain whether respondents’ answers to global measurement questions actually reflect their perceptions of risk in the area, knowledge of real risks of victimization or genuine emotional fear (Garofalo and Laub, 1978; Pantazis, 2000; Rountree and Land, 1996; Wilson and Kelling, 1982). Due to this ambiguity inherent in the respondents’ answers, such global measures are criticized as being vague and problematic (Rountree and Land, 1996). A similar global question asks ‘Is there any area right around here – that is, within a mile – where you would be afraid to walk alone at night?’ (LaGrange and Ferraro, 1987). This question is more likely to tap into the emotional aspect of fear because the word ‘afraid’ is used, however it is still ambiguous and seems excessively foreboding (LaGrange and Ferraro, 1987). The word ‘crime’, or a specific act or acts that constitute crime, is not mentioned in global measurement questions (LaGrange and Ferraro, 1987). Respondents may not be sure what they are meant to feel safe or unsafe from, and therefore could confuse their fear of crime with fear in general (Garofalo, 1979; LaGrange and Ferraro, 1987). This creates a conceptual issue for people with specific phobias that cause them to feel unsafe in certain areas. It also opens the door to the various social theories that argue, for example, that fear of crime actually reflects perceptions of subcultural diversity (Covington and Taylor, 1991; Hanson et al., 2000; Katz et al., 2003; Merry, 1981; Taylor and Hale, 1986). With global questions it is important not to assume that people stay home at night because they are afraid of crime, but rather for a diverging array of other reasons (LaGrange and Ferraro, 1987). There are ambiguities even when ‘crime’ is mentioned (Ferraro and LaGrange, 1987). Fear varies with the type of crime under consideration, for example it
74
5
Investigating Fear of Crime
depends on whether the crime involves a threat to one’s personal well-being or damage to one’s property (Skogan and Maxfield, 1981). In terms of personal crime, experiences of fear differ if, for instance, rape or robbery is considered. Global measurement questions conceal any differences in the level of fear of these different crimes (Ferraro and LaGrange, 2000). A lack of crime specificity in survey questions forces respondents to select their own conceptual references. This choice differs between people and, therefore, responses may not be comparable (Ferraro and LaGrange, 2000). Ferraro and LaGrange (1987) argue that the lack of crime specificity in global measurement questions overrides any of their usefulness as fear-of-crime measures. Another criticism of global measures also relating to a lack of specificity is the often-vague geographic reference to the area in which people live. The spatial frame of reference of global measurement questions, the ‘neighbourhood’, is not sufficiently defined and can be envisaged differently by different people (Ferraro and LaGrange, 2000; LaGrange and Ferraro, 1987). For instance, those respondents completing the same fear-of-crime survey may reside in completely different neighbourhoods and thus be referring to a separate environment in their response. For those respondents actually even living in the same neighbourhood their ideas of the boundaries of that neighbourhood may be quite discordant. This inhibits comparison of respondents’ answers. Furthermore, assuming that each respondent reflects on the same neighbourhood when answering the global measurement question, it is still unclear whether they are fearful in the entire neighbourhood or only certain parts of it. This is particularly relevant considering that crime levels and rates fluctuate dramatically within urban neighbourhoods (LaGrange and Ferraro, 1987). Fisher and Nasar (1995) argue strongly that much extant research is limited because studies using global measures cannot reveal the location of specific fear spots or what types of cues stimulate the fear-generating process in individuals or across groups. Finally, survey items asking respondents ‘do you feel, or ‘would you feel’ merge reality with the hypothetical, thereby creating a double-barrelled question (LaGrange and Ferraro, 1987). Tulloch (1998) suggests that global measures are hypothetical for many women and older people because they rarely, if ever, walk alone at night. LaGrange and Ferraro (1989) argue that it is methodologically inappropriate to use hypothetical scenarios since it is difficult for respondents to evaluate how they would feel (Ferraro and LaGrange, 2000). In addition, the use of hypothetical scenarios may exaggerate fear-of-crime levels because it could seem excessively threatening (LaGrange and Ferraro, 1989). Tulloch (1998) further argues that such measures potentially fail to assess how people engage with fear of crime in their daily routines, often producing poor model results. An example of this can be found in the research conducted by Nair et al. (1993) into the effect of environmental improvements on fear of crime. The authors found that significant lighting changes made in a park in Glasgow, Scotland, did not result in the expected substantial fall in fear of crime. The survey respondents later indicated that the improvements to the park were not relevant to their daily routines and that the net result was to turn a poorly lit bad area into a well-lit bad area. Such limitations have, in part, led to the call for researchers to contextualize their studies such that
Measuring Fear of Crime
75
the fear-of-crime issues investigated are relevant to the daily routines of the survey respondents (e.g. Nair et al., 1993; Smith and Tortensson, 1997; Tulloch, 1998). Value- or Concern-Based Measures Closely aligned with global questions are value- or concern-based measures. The terms ‘worry’ and ‘concern’ are often interchanged with fear in social surveys (Skogan and Maxfield, 1981 in LaGrange and Ferraro, 1987). However, instead of targeting emotional levels of fear, these questions evaluate people’s opinions of the seriousness of the level of crime in their neighbourhood (Furstenberg, 1971). Furstenberg (1971) provides the example of asking respondents to, ‘choose the single most serious domestic problem (from a list of 10) that you would like to see government do something about’. Another simpler version involves asking respondents if they are personally concerned about becoming a victim of crime (Jaehnig et al., 1981 in Ferraro and LaGrange, 1988). As discussed in Chapter 3, people’s concern or worry about crime is distinctly different from their fear of crime. People who are troubled by the problem of crime are not necessarily afraid of being personally victimized (Furstenberg, 1971).
Improvements Through Affective Approaches to Measuring Fear of Crime While cognitive approaches to measuring fear of crime involve people making judgements about how safe they feel, affective approaches aim to elicit more of an emotional response and aim to measure ‘fear of crime’ in a more literal sense. ‘Emotion-based measures’ is the most common term given to these approaches in the literature. Emotion-Based Measures In contrast to global measures and other types of cognitive approaches to measuring fear of crime, emotion-based measures make explicit reference to a specific crime (Ferraro and LaGrange, 2000). In doing this, they target ‘concrete’ fear by eliciting a personal, emotional reaction from the respondent (Ferraro and LaGrange, 1987; Rountree and Land, 1996; Scott, 2003). While this reaction may also depend on perceived risk, it is distinct from judgements or concerns about crime (Ferraro and LaGrange, 2000). Emotion-based questions include ‘how afraid are you of becoming a victim of . . .’ (Mawby et al., 2000; Rountree and Land, 1996). Respondents answer by choosing from a list of options such as I feel ‘very afraid’, ‘fairly afraid’ or ‘a bit afraid’ (Skogan, 1999). These questions allow respondents to visualize themselves as victims of the crime (Reid et al., 1998). The extent of the fear elicited by the specific crime mentioned in the survey question will depend on a number of factors. Fear of crime is initially based on the nature
76
5
Investigating Fear of Crime
and perceived seriousness of the offence in question which will vary according to community context, social group and the individual (Clark, 2003). Fear is also influenced by an individual’s risk sensitivity to the crime in question (Clark, 2003; Warr, 1984; Rountree and Land, 1996). Each of these factors is subconsciously assessed when a person thinks about a crime and affects the extent of the fear response, pointing out the importance of crime specificity (Clark, 2003). By referring to a specific crime and eliciting these personal considerations, emotion-based measures effectively overcome many problems with global measurement questions. However, they also result in highly subjective responses. People have differing perceptions about concepts like ‘a bit afraid’. Two respondents who state they feel ‘somewhat afraid’ may react completely differently, and therefore comparative analysis of cognitive and affective comments can be problematic. This problem and the hypothetical nature of the questions used restrict the utility of emotion-based measures to certain contexts. Few studies have gathered crime-specific data on fear and those that have generally rank crimes according to the level of fear that they produce (Warr, 2000).
Behavioural Approaches to Measuring Fear of Crime Ditton et al. (2000) criticize fear-of-crime research as being ‘trapped within an overly restrictive methodological and theoretical framework’. In a similar vein, Warr (2000) argues that ‘the study of fear seems to have stalled at a rudimentary phase of development, a situation that is in danger of turning into outright stagnation’. A major factor in this lack of progress is due to continual use of these problematic cognitive and affective questions in surveys (Ditton et al., 2000; Warr, 2000). One potential means of avoiding some of the limitations of global measures of fear of crime is to use behavioural measures. At a general level, behavioural measures would seem to be appropriate, given the common finding that people respond to fear by modifying their behaviour (Samuels and Judd, 2002; Tulloch, 2000; Warr, 2000). As Skogan (1999) indicates, fear is validated when it manifests through behaviour. Behavioural approaches eliminate much of the subjectivity associated with responses from cognitive or affective questions. By focusing on fear of crime through behavioural responses, researchers can measure and compare fear more reliably than other techniques. In fact, Hale (1996) argues that behaviour is a more accurate guide to fear levels than reported statements about fear level. This notion prompted Warr (2000) to state that behaviour may be the best indicator of fear. Behavioural approaches examine the protective actions and avoidance strategies adopted by people attempting to reduce fear and hold more potential to relate to the routines of the survey respondents (Gabriel and Greve, 2003; Samuels and Judd, 2002; Smith and Tortensson, 1997; Tulloch, 1998, 2000). Protection-Based Measures People who are afraid of crime, either in their home environment, or out in their neighbourhood, are likely to use self-protection (Ferraro, 1995; Tewksbury and
Measuring Fear of Crime
77
Mustaine, 2003). To determine the types of self-protection employed by survey respondents, they are usually asked questions such as ‘in general have you limited or changed your activities in the past year because of crime (yes or no)’ (Liska et al., 1988). A list of protective actions from which respondents can then choose is often provided (see DeFronzo, 1979; Gray and O’Connor, 1990; Sundeen and Mathieu, 1976). Protective actions are employed to either limit one’s exposure to risk or reduce one’s chances of being victimized when exposed to risk (Skogan and Maxfield, 1981). Many of these actions also therefore make people feel less afraid of crime (Vacha and McLaughlin, 2004). Protective actions generally include individual coping strategies or collective actions. Individual coping strategies are diverse and extensive. In terms of protection against property crime, people adopt ‘targethardening efforts’ (Skogan and Maxfield, 1981) through creating physical barriers for offenders to overcome by locking their doors when leaving home (Warr, 2000), installing extra security locks, bars and systems (Carvalho and Lewis, 2003) and keeping trained watch dogs (Williams et al., 1994). Psychological barriers to deter offenders are also employed such as installing car and home alarms (Reid et al., 1998) and leaving lights or timed appliances like radios and television sets on at home when they are out (Krahn and Kennedy, 1985; Warr and Ellison, 2000). Other coping strategies that protect against, or minimize, the negative consequences of property loss and damage include the engraving of valuables and the purchase of theft and vandalism insurance (Williams et al., 1994) and the use of police property identification systems (Toseland, 1982). The individual coping strategies that people employ to protect themselves against personal crime are also wide ranging. These commonly include the carrying of a weapon such as a handgun or mace to use when warding off or defending against an attacker (DeFronzo, 1979; Kenney, 1987; Reid et al., 1998). Personal alarms and whistles are also carried to drive away attackers and alert passers-by of the problem. For those who choose not to arm themselves in any way, they often simply increase their level of alertness (Reid et al., 1998) and walk faster during those moments of fear. They may also choose to drive a car or use other ‘safe’ methods of travel through feared areas rather than walk (Warr and Ellison, 2000). When at home, people may also refuse to open the door to a stranger (Warr, 1985). Collective actions that are used to protect against crime, and consequently fear of crime, often transcend the boundaries between personal and property crime. A widespread response is people walking in pairs or groups when in feared areas (Carvalho and Lewis, 2003; Nasar et al., 1993). Other collective actions include the organization of ‘neighbourhood watches’ (Reid et al., 1998). Williams et al. (1994) find that these collective actions are more common than personal coping strategies. While these protection-based measures overcome problems of subjectivity, they have generally been subject to the limitations of vague geographic references associated with the global measurement framework. In comparison to the multitude of cognitive and affective studies, relatively little information has been collected on the protective measures adopted by people in response to fear of crime, especially in response to fear of different specific crimes (Reid et al., 1998). Additionally, little is known about the different socio-demographic groups who employ such protective
78
5
Investigating Fear of Crime
measures and if the use of self-protection is related to an individual’s proximity to potential offenders (Tewksbury and Mustaine, 2003). Avoidance-Based Measures As discussed earlier, avoidance is documented as one of the most frequent behavioural responses to fear of crime (Garofalo, 1981). Avoidance refers to ‘those actions taken to decrease the chance exposure to crime by removing or distancing oneself from situations in which the risk of victimisation is perceived to be high’ (DuBow et al., 1979). Often people restrict their movements to safe places at safe times or refuse to leave their homes at all, particularly during the night (Pantazis, 2000; Samuels and Judd, 2002). Some residents even choose to avoid the neighbourhood altogether by moving (Carvalho and Lewis, 2003; Reid et al., 1998). Because collective avoidance is central to many of the negative consequences on affected communities, avoidance-based measures are particularly relevant to the study of fear of crime and any associated fear-reduction strategies. As mentioned, research into avoidance generally involves asking respondents if they avoid any areas because they feel unsafe (Ditton and Farrall, 2000) or something similar to ‘do you avoid certain places and areas of the city because of the possibility of crimes of violence’ (Gomme, 1986). The response to these avoidance-based items in fear-of-crime surveys predominantly features only a yes or no possibility. As such, these studies have only been useful for broad-level macro analyses of fear of crime and avoidance behaviour. Behavioural approaches that measure the actual behaviour of respondents (e.g. Fisher and Nasar, 1992; Nasar and Jones, 1997; Nasar et al., 1993) have the potential to overcome the limitations of global measures relating to the hypothetical nature of survey questions and vague geographic references. For example, avoidance-based questions more recently include a spatial element, with a request that those avoided areas be illustrated on a map (Doran and Lees, 2005; Nasar and Jones, 1997; Nasar et al., 1993). When assessing collective behavioural responses, it is appropriate that mapping restricts the scope of the question to a geographic reference that is defined and common to all respondents which enables a more accurate comparison across an area or ‘neighbourhood’. Despite these benefits, the utility of behavioural measures has often been limited because of a lack of crime specificity. In comparison to the multitude of cognitive and affective studies, relatively little information has been collected on the behavioural reactions adopted by people in response to fear of crime, especially in response to fear of different specific crimes (Reid et al., 1998). Additionally, little is known about the different socio-demographic groups who employ such measures and if the use of self-protection or avoidance is related to an individual’s proximity to potential offenders (Tewksbury and Mustaine, 2003). However not everyone will be able to adopt avoidance as a precautionary behaviour. Hindelang et al. (1978) discuss routine daily activity theory and call attention to a number of constraints that could affect whether people are able to adopt avoidance strategies. Skogan and Maxfield (1981) argue that social norms
Analysing Fear-of-Crime Data
79
and expectations of ‘role’ according to peoples’ socio-economic positions may preclude them from being able to avoid different areas because they are subject to ‘a host of formal and informal mechanisms which channel [their] activity in expected ways’. Other limitations may ‘derive from the operation of institutions’, namely where people are required to live and work. Skogan and Maxfield (1981) also provide some other examples of constraints on avoidance behaviour: people may lack the resources necessary to avoid feared areas, for instance they must use public transport or walk through feared areas if they do not have a car or a driving license. The authors argue that behaviourally based approaches to measuring fear of crime must consider such constraints (Skogan and Maxfield, 1981).
Review: A Preference for Avoidance-Based Measures in Fear-of-Crime Studies The range of different techniques used to measure fear of crime may be in part due to the complex and multifaceted nature of the issue. Researchers have generally measured fear of crime using three main approaches. The cognitive approach to measuring fear of crime is relatively easy to carry out. However, both the global and value- or concern-based measures are limited in their utility because they do not target actual ‘fear’ of crime. The wording of survey questions also results in responses that are difficult to interpret. While the affective approach to measuring fear of crime does target people’s emotional fear of crime, it too results in ambiguous and subjective findings due to lack of geographic specificity. Despite these restrictions, cognitive and affective approaches to measuring fear of crime have been useful for broadlevel analyses. In contrast, behavioural approaches to measuring fear of crime hold the potential to overcome much of the subjectivity and ambiguity inherent in cognitive and affective survey responses. As Smith (1987) noted in an earlier review, the observed effects of fear of crime on lifestyle are too marked to ignore. This still holds true some 20-plus years later, and it is in this area that future research needs to be conducted. Behavioural approaches, particularly avoidance-based measures, can also produce site-specific, or spatially explicit, results. This means they can be used to perform analyses using Geographic Information Systems (GIS). The potential advantages of spatial analyses of fear of crime are discussed in the concluding sections of this chapter.
Analysing Fear-of-Crime Data Fear of crime is typically analysed using traditional statistical techniques. Bivariate analyses dominate the field, with researchers using Pearson’s correlation co-efficient (r), Spearman’s rank (r) and Chi-Square analyses (e.g. Mirrlees-Black and Allen,
80
5
Investigating Fear of Crime
1998; Reid et al., 1998; Wilson-Doenges, 2000), often as a basis for more complex analyses (e.g. Karakus et al., 2010). Many studies also acknowledge the multidimensional nature of fear of crime and concentrate on the interactions between a multitude of dependent variables and fear of crime (Box et al., 1988; Carcach and Mukherjee, 1999; Ferraro and LaGrange, 1987). Given the contrasting and numerous approaches used to measure fear of crime, it is not surprising that different analytical techniques can give rise to conflicting or dissonant results, even when examining the same dataset (see LaGrange and Ferraro, 1989). One overarching disadvantage of many statistics-based studies is that they are spatially implicit in nature, in part due to the use of cognitive or affective measurement approaches and associated vague geographic references that are ingrained in survey questions. When findings are presented as the percentage of fearful people or fearful subgroups within a study area or region (e.g. Joseph, 1997; Mayhew and White, 1997; Mirrlees-Black and Allen, 1998; Thomas and Bromley, 2000), they are subject to the ecological fallacy and modifiable areal unit problem (MAUP). These issues are well known to urban geographers and arise when a researcher makes inferences about an individual based on area-level aggregations or when data are represented according to different administrative boundaries (e.g. Cromley and McLafferty, 2002; Longley et al., 2001; O’Sullivan and Unwin, 2010). Figures 5.1 and 5.2 below illustrate the MAUP and ecological fallacy with two hypothetical examples. In Fig. 5.1, it can be seen that different counts are derived for the same point dataset when different sets of boundaries are used. In Fig. 5.2, a hypothetical administrative boundary (e.g. a suburb or census district) envelopes a pocket of relative socio-economic disadvantage. In this instance, if the area were labelled as having 40% high-income houses, this would not be an accurate aggregation and could mask the subarea of disadvantage. From the perspective of the institutions responsible for addressing fear of crime, such as police and council services, the outputs from traditional statistical analyses make limited contribution to the ‘where’ and ‘when’ aspects of fear of crime, which are emphasized as fundamental components of fear management strategies (e.g. NCAVAC, 1998). While issues such as the MAUP and ecological fallacy are difficult to avoid entirely (Cromley and McLafferty, 2002; Monmonier, 1996), the use of geocoded data provides a more sensitive means of handling fine-scaled relationships (e.g. Doran et al., 2007). As such, the collection and analysis of spatially explicit fear-of-crime data can potentially contribute much to a ‘stagnant’ field (Warr, 2000) through the provision of information that is not constrained to area-level aggregation alone as a means of summarizing the geography of fear. As Goodchild (2004) notes, Only a fraction of 1 percent of the literature published in the social sciences takes a spatial perspective, so the potential for growth is still enormous.
It would seem that the use of GIS in fear of crime would be one such area where there remain many useful avenues open to investigation.
Analysing Fear-of-Crime Data Fig. 5.1 The effect of the MAUP
81
A)
B) 6 2 6 7
6
3
Fig. 5.2 The effect of the ecological fallacy
POCKET OF RELATIVE DISADVANTAGE
Advantages of Spatial Analyses of Fear of Crime Many researchers acknowledge a distinct spatiotemporal element to crime and fear of crime, which researchers should be sensitive to (Gold and Revill, 2003; Lemanski, 2004; Moran et al., 2003; Warr, 2000). Lupton and Tulloch (1999) call for research that explores the ‘dynamic situated and micro-contextual contexts in which fear of crime is generated and experienced’ (Lupton and Tulloch, 1999). By doing this through spatial analysis, fear-of-crime findings can be integrated with an understanding of the social and physical environment (Pain, 2000). As Samuels and Judd (2002) elaborate in the following statement, Mapping provides a spatially focused base for the interpretation of social indicators in their epidemiological context. Maps are setting specific, temporary sensitive, visual-diagnostic tools . . . allowing situational experience to be interpreted in light of the theory and practice of environmental design and community empowerment criminology.
Ashby and Longley (2005) state that these ‘geodemographic’ analyses lead to significantly improved police intelligence. For example, a spatial knowledge of fear of crime and avoidance patterns could allow for the targeting of limited resources to specific hotspots. Such locally tailored responses are also more likely to be effective than generalized strategies (Kitchen, 2002; Nelson et al., 2001; Skogan, 2004). In light of this, Fisher and Nasar (1995) believes that fear-of-crime studies missing a spatial element are vague and less informative than those that do.
82
5
Investigating Fear of Crime
Despite these benefits, few researchers have ventured into the world of spatial analyses of fear of crime. Doran and Lees (2005) used GIS to investigate the spatiotemporal links between crime, fear of crime and disorder – a study that is described in detail in Chapter 6. The study was conducted in a central business district (CBD) area and focused on sampling the working population. The fear mapping outputs helped councils, local agencies and the police to determine priority areas for fear reduction and appropriate measures in different locations earmarked for change in town planning initiatives. The spatial outputs also enabled a comparison of collective avoidance hotspots and concentrations of crime and disorder. It was evident that there existed many avenues for future research. In particular, the underlying motivations for avoidance behaviour in relation to specific fear of crime hotspots were not examined in detail. The techniques developed were suited to the comparison of different geographic areas and held the potential to objectively determine which socio-demographic groups are more afraid. Toseland (1982) states that such outputs could assist special efforts targeting these vulnerable groups. This focus on underlying motivations for avoidance behaviour and different responses to specific cues in fear-of-crime hotspots were investigated in the Kings Cross study – described in Chapter 7. At a macro scale, the spatial visualization of fear hotspots also allows for an investigation into the proposed idea that fear of crime is predominantly an urban, rather than rural, problem (Cates et al., 2003; Miceli et al., 2004; Yarwood, 2001). Therefore, fear mapping has the potential to provide an additional layer of understanding, as well as more localized and geographically relevant information than traditional statistical approaches. The foundation of fear mapping has its roots in behavioural geography and the associated use of cognitive mapping techniques in a GIS-based framework, which are discussed next.
Spatial Cognition and Cognitive Mapping Cognitive mapping, a technique that has been used extensively to gather geographic information in the broader area of behavioural geography (Kitchin, 1996), is likely to be an appropriate means of gathering spatially explicit information on fear of crime and avoidance behaviour. Golledge and Stimson (1997) argue that the surge of interest in behavioural research in human geography in the 1960s and 1970s stemmed from a desire to increase the geographer’s level of understanding of particular types of problems. This aligns well with understanding of the spatiotemporal nature of fear of crime. An understanding of how people develop cognitive maps, and how spatial cognition influences spatial choices and behaviour, is highly relevant to environmental criminology (Brantingham and Brantingham, 1993). In essence, cognitive mapping assists people in making spatial choices, such as determining which areas in which to commit crime or to avoid due to fear of crime (Brantingham and Brantingham, 1993; Downs and Stea, 1973). A cognitive map is a mental copy of one’s environment, featuring information about the relative spatial location, arrangements and properties of ‘phenomenon’ (Block, 1998; Downs and Stea, 1973; Sholl, 1996). Such phenomena include behaviourally relevant ‘landmarks’ that are visible reference points, like buildings, parks or street junctions
Analysing Fear-of-Crime Data
83
(Nasar, 1998). Spatial cognition involves the attribution of denotative meaning, or object recognition, to these geographic phenomena (Nasar, 1998). By definition, cognitive mapping refers to the process by which people comprehend and respond to the world around them (Downs, 1977). The end product of the cognitive mapping process is a cognitive map (Golledge and Stimson, 1997) which is a cross-section representing the world at one instant in time and reflects the world as a person believes it to be (Downs, 1977). Space is not considered only in terms of the physical environment (Koskela and Pain, 2000). Activities, specific events and processes become associated with the environmental context in which they take place (Koskela and Pain, 2000; Valentine, 1989). For example, a laneway may be associated with drug dealers. Therefore, when shaping and recalling information stored in one’s cognitive map, a person is aware of the environment as having distinct social and physical attributes (Burnett, 1976; Downs and Stea, 1973). Thus, character plays a vital role in social cognition and functions as an effective cue in retrieving spatial information (Tversky and Taylor, 1998). Space and events in space are intimately connected with the perception of time (Block, 1998). Therefore, landmarks and objects often have associated temporal properties and relationships (Block, 1998). As part of spatial cognition, or spatiotemporal reasoning, the ‘appearance, change, and disappearance of things in space and over time’ is also considered (Couclelis, 1998). Taking this into account, the presence of darkness in a particular environment (represented by darkness rather than a measurement of time) can trigger new attributes to be associated with that environment. Using the above example, the drug dealers in the laneway during the day may move to another location at night. Cognitive mapping is not only shaped by the physical, social and temporal properties of space, but also by one’s mental state (Orleans, 1973). The mind is the home of a person’s emotions, attitudes, needs and desires. The process of evaluating an environment is a function of these factors (Burnett, 1976; Orleans, 1973). This evaluation involves judgement and the assigning of a connotative meaning to the different phenomena and social activities within that environment (Husserl, 1973; Nasar, 1998). Continuing the previous example, onlookers could perceive the drug dealers as threatening, thereby connoting risk, or as harmless. They would take appropriate behavioural action, such as adopting avoidance or protective measures, depending on their judgement. While assessing the possible courses of action and making a spatial choice,5 cognitive information will also be influenced by one’s past experiences, present beliefs and especially the future expectations concerning the outcome of such a decision (Burnett, 1976; Downs and Stea, 1973; Kitchin, 1996; Kaplan, 1973; Jeffery, 1971; Mennis, 2003). In circumstances where onlookers perceive the drug dealers to be threatening, the concept of risk becomes attached to that specific laneway and the person may consequently avoid it (Nelson et al., 2001). The laneway then signals the need for avoidance and becomes an anchor point, which
5 Spatial choice is a function of knowledge of one’s location, what is likely to occur, whether it will be good or bad and possible courses of action (Nasar and Jones, 1997).
84
5
Investigating Fear of Crime
is similar to a landmark, only more personal and salient, in one’s cognitive map (Block, 1998; Couclelis et al., 1987). This avoidance behaviour therefore continues even in the absence of the original drug dealers. Thus, avoidance behaviour is part of an individual’s cognitive mapping process because it involves a response based on his/her perceived threat of crime. In conclusion, spatial behaviour is the result of the complex processes of spatial choice. Spatial behaviour6 and spatial choice are dependent on one’s cognitive map of the spatial environment (Burnett, 1976; Downs and Stea, 1973; Freundschuh, 1998) and is therefore a response to both the real and subjective worlds (Kitchin, 1996). However, despite the rational calculation involved in behaviour, inferences and spatial choices can be made without conscious thought (Nasar, 1998). Kitchin (1996) reviews the variety of techniques that can be used to gain information on the cognitive mapping process. These include asking respondents to draw a sketch map of an area (e.g. Walsh et al., 1981), estimate the distance and direction to locations (e.g. Day, 1976; Kirasic et al., 1984) or verbally describe a route or area (e.g. Vanetti and Allen, 1988).
The Beginning of Fear Mapping As mentioned above, cognitive mapping techniques have been successfully adapted to investigate fear of crime and develop fear mapping methodologies. The study of cognitive mapping originally involved evaluating environmental cognition by asking individuals to illustrate their mental maps of geographic regions, with landmarks, on paper. In line with this, Steinitz (1968) mapped ‘denotative meanings’, or people’s knowledge of a city (cited in Nasar, 1998). Later, environmental assessment became of interest where connotative meanings were mapped, or people’s feelings regarding places and activities in different areas of a city (Nasar, 1998). Newman (1972) created one of the first fear maps showing a site plan of designs that residents designated as dangerous. A year later, Gould produced a crude fear map of Philadelphia (cited in Nasar, 1998).7 In their various papers written approximately 15–20 years later, Fisher and Nasar made considerable contributions to the growth of fear mapping, the linking of certain environmental cues to fear of crime and the use of activity diaries (Fisher and Nasar, 1992, 1995; Nasar and Jones, 1997; Nasar et al., 1993). They conducted a number of micro-level behavioural studies investigating emotional levels of fear in university campus settings (e.g. Fisher and Nasar, 1992, 1995; Nasar and Jones, 1997). These studies assessed emotional levels of fear in area and
6 Spatial behaviour is ‘any form of human behaviour that involves or exhibits an interaction between the individual and one or more points in space’ (Louviere, 1976). 7 In 1976, Milgram and Jodelet also mapped perceived areas of danger in Paris (Nasar, 1998). Also, in 1977 Duncan (1997) mapped New York’s feared neighbourhoods (Oc and Tiesdell, 1997).
Analysing Fear-of-Crime Data
85
time-specific situations (e.g. Fisher and Nasar, 1992, 1995) while walking a particular route at night (Nasar and Jones, 1997). They provide the best examples of behavioural approaches that have sought to understand the actual protective and avoidance behaviours that survey respondents adopted in relation to fear of crime. In their 1990 ‘observations of behaviour’ study, Fisher and Nasar (1992) observed pedestrian activity to determine if people avoid walking in or near areas they judged as unsafe. By examining the most heavily avoided sites, they concluded that people avoid low-prospect/high-refuge areas. In 1991, Nasar et al. (1993) extended this research. Respondents were asked to circle areas that they avoided on a map. These individual maps were then aggregated to produce a coarse hierarchical map of fear. This was then used in more site-specific analyses of the links between feelings of safety and concealment, prospect and escape. Fisher and Nasar (1995) slightly amended this method in their later study, wherein they asked respondents to rate their perceived level of safety in eight predesignated areas on the provided map. The results similarly showed definitively that hotspots of fear occurred at the micro level. Nasar and Jones (1997) again explored fear mapping by asking 26 female respondents to walk a specified route between 8:15 pm and 10:00 pm and to carry a hand-held tape recorder. Respondents were asked to record feelings of safety or unsafeness and any emotional reactions or feelings generated as a result of particular elements of the surrounding environment or situation in general. Sites where respondents felt unsafe were documented on a map and aggregated to show the spatial distribution of fear comments by percentage. One of the main advantages of these studies was the ability to assess levels of fear in relation to actual activities and overcome some of the limitations of broader, global measures of fear. One prospect of extending this approach lies in the use of activity diaries provided as a means to assess fear of crime on a larger scale and in relation to the actual activities of respondents.
Activity Diaries and Daily Routines Activity diaries are another behavioural geography technique and are one of four main methods used to collect time-budget data. Golledge and Stimson (1997) outline these methods. The first is a recall method where activities of some specified period in the past are recalled with as much precision as possible, regarding the location and time of activities. The second is a recall method where activities for some normal period are recalled. The third is the diary method where subjects are asked to keep a diary for a specified period of time. The fourth is a game-based method basically used to investigate changes in the contingencies of the decision-making environment and often used in the ‘post diary’ or post-interview situations. In the context of fear of crime, the diary method of data collection is likely to be the most appropriate, as it records the actual activities of people. The other methods, by incorporating hypothetical procedures increase the likelihood that fear of crime will be measured in situations not relevant to the actual behaviour of survey respondents.
86
5
Investigating Fear of Crime
Farrall et al. (1997) describe a survey respondent who was listed as very worried about crime in a quantitative study but, in a qualitative study, he claimed only to be worried when out on the street and he came across a group of people (i.e. according to the authors’ measures he was a 1 at home, 4 or 5 when out on the streets). An activity-diary investigation of fear of crime can potentially overcome such limitations, as respondents can be asked about their emotional level of fear in relation to specific activities at specific times. A further potential advantage of using activity diaries to investigate fear of crime is that it will assess fear in situations that are part of the respondents’ daily routines. One could argue that the study that Nasar and Jones (1997) conducted, while overcoming some of the hypothetical issues associated with global measures of fear, still involved placing respondents in a situation that may not have been part of their daily routines. Time-space budgets can be used for a wide variety of purposes, following the processing of the data. The first step in processing time-space budgets generally involves the application of a classification scheme (Golledge and Stimson, 1997). Once classified, the data can be analysed to investigate specific issues in relation to the routines of respondents (e.g. Kwan, 1999, 2000a). Chadee and Ditton (2003) emphasize that the interaction between factors known to influence fear of crime is an important consideration of investigations of fear of crime. The authors use the example of the elderly only appearing to be more fearful if they live in large cities, are unmarried, live alone or are low-income earners and black. Activity-diary analyses have been shown to be sensitive to such interactions (e.g. Golledge and Stimson, 1997).
Geographic Information Systems and Fear of Crime Nasar (1998) proposed the use of GIS in fear mapping in order to increase accuracy. Given that recognition of the spatiotemporal nature of fear of crime is generally regarded as fundamental to any analysis of the phenomena (e.g. Nasar and Jones, 1997; Pain, 1997; NCAVAC, 1998; Thomas and Bromley, 2000), it is somewhat surprising that few studies have sought to use GIS in this area. By definition, a GIS contains a powerful set of tools which allows the collection, storage, retrieval, transformation and display of spatial data (Burrough and McDonnell, 1998). This data, or geographic information, is referenced to locations on the earth’s surface and not only includes the location of spatial objects, but also their attributes (Ding and Fotheringham, 1992; Martin, 1991). Mapping through GIS is therefore particularly useful when studying large and complex data with multiple attributes, where conventional inferential statistics and pattern recognition algorithms may fail (Kwan, 2000a). The use of GIS to model dynamic spatiotemporal phenomena is also well recognized (e.g. Egenhofer and Golledge, 1998; Maury and Gascuel, 1999, Kwan, 2000a, 2000b; Olsen and Doran, 2002). GIS is already widely used by police services to investigate patterns of criminal activity (e.g. Ashby and Longley, 2005; Baker and Wolfer, 2003; Bowers
Analysing Fear-of-Crime Data
87
et al., 2004; Harries, 1999; Murray et al., 2001; Nelson et al., 2001; Ratcliffe and McCullagh, 2001; Russo, 2001; Weisburd et al., 2004; Yarwood, 2001). Murray et al. (2001) suggest that GIS and crime mapping software have been the most influential computer-based tools for developing techniques to explain the occurrence of criminal activity. However, the authors also note that there exist many potential applications of GIS in this area. Similarly, Harries (1999) predicts that GIS adoption by police departments will increase rapidly in future because the technology is simultaneously becoming cheaper and more powerful. One of the major advantages of GIS is that it has the potential to increase the efficiency with which police resources are allocated. Ratcliffe and McCullagh (2001) outline how crime mapping is often used to identify the extent of a crime problem and to target resources to deal with the problem. This approach, first developed through NYPD’s CompStat procedure (Bratton, 1995, 1996), has now become popular in other law enforcement agencies (Ratcliffe and McCullagh, 2001). Using GIS in a similar manner to identify concentrations of fear may enable police services and local communities to address fear of crime in the same focused manner. An area that has received little attention in the literature is the potential spatiotemporal relationship between fear of crime and actual victimization. Considering that many studies have established or suggested links between fear of crime, social disorder and serious crime (e.g. Kelling and Coles, 1997; Perkins and Taylor, 1996; Skogan, 1990; Taylor and Covington, 1993; Wilson and Kelling, 1982), a spatiotemporal analysis of such potential links would be a useful addition to a fear mapping study. Further, the so-called risk-victimization paradox arises from the frequent observation that people with the least fear are at greatest risk and those with the greatest fear are at least risk (e.g. Oc and Tiesdell, 1997; Reid et al., 1998; Warr, 1984). At the level of the individual, it is quite possible that people frequenting areas or times where their fear of crime is low but crime rates are high may be more susceptible to victimization. On a broader level, many of the suggested links between fear of crime and the actual occurrence of crime are largely based on the collective nature of avoidance behaviour. Areas of low natural surveillance resulting from avoidance behaviours adopted by members within a community are said to provide opportunities not only for disorder, but also for crime itself to become established (e.g. Kelling and Coles, 1997; Skogan, 1990). However, to date there are no tools or analytical techniques available for identifying and collating the actual areas that the public avoid due to their fear of crime. If this can be done successfully, it would provide a framework to compare the collective nature of avoidance behaviour to concentrations of crime. As with investigations into the spatiotemporal distribution of crime, GIS-based analyses may prove useful and potentially provide new insights into these areas. The focus of the two studies described in Chapters 6 and 7 is to investigate these issues in Wollongong and Kings Cross – two different settings, one a regional town and another a densely populated inner city area, within Australia.
88
5
Investigating Fear of Crime
Chapter Review: A New Direction with Avoidance Mapping A review of the literature reveals that there are numerous problems with the commonly used cognitive and affective approaches to measuring fear of crime. Some of these limitations can be overcome by using behavioural measures and it is possible to strongly justify research into the area that is behaviour-based. First, if fear of crime is to be addressed on the basis of its most concerning influence on society, protective and avoidance behaviours must be taken into consideration. Second, avoidance-based behavioural measures are particularly applicable because they can be used in spatial investigations into fear of crime. Finally, spatial investigations and the use of techniques from behavioural geography hold the potential to provide new and useful information that cannot be gained through traditional statistical analyses. Such outputs are likely to be particularly useful for policy, planning and localized implementation of fear-reduction strategies. Despite these benefits, few researchers have conducted spatially explicit research into fear of crime. Those that have (e.g. Fisher and Nasar, 1992, 1995; Nasar, 1998) have strongly demonstrated utility of mapping fear at the micro level and have recommended the use of GIS in future applications. As such, there exists a genuine need and opening to ‘put fear on the map’ through the use of GIS and appropriate measurement techniques.
References Ashby, D. I. and P. A. Longley (2005). “Geocomputation, geodemographics and resource allocation for local policing”. Transactions in GIS 9(1): 53. Australian Bureau of Statistics (ABS) (2006). General social survey (cat no. 4159.0). Canberra, Australian Bureau of Statistics. Baker, T. and L. Wolfer (2003). “The crime triangle: alcohol, drug use and vandalism”. Police Practice and Research 4(1): 47–61. Beck, V. S., I. Travis and F. Lawrence (2004). “Sex offender notification and fear of victimization”. Journal of Criminal Justice 32(5): 455–463. Block, R. A. (1998). Psychological time and the processing of spatial information. Spatial and temporal reasoning in geographic information systems. M. Egenhofer and R. Golledge (Eds.). Oxford University Press, New York, NY: 119–130. Borooah, V. and C. Carcach (1997). “Crime and fear. Evidence from Australia”. The British Journal of Criminology 37(4): 635–657. Bowers, K. J., S. D. Johnon, et al. (2004). “Prospective hot-spotting: the future of crime mapping?”. The British Journal of Criminology 44: 641–658. Box, S., C. Hale, et al. (1988). “Explaining fear of crime”. The British Journal of Criminology 37(4): 340–356. Brantingham, P. L. and P. J. Brantingham (1993). “Nodes, paths and edges – considerations on the complexity of crime and the physical-environment”. Journal of Environmental Psychology 13(1): 3–28. Bratton, W. J. (1995). Great expectations: how higher expectations for police departments can lead to a decrease in crime. Paper presented to the National Institute of Justice Policing Research Institute “Measuring what matters conference”. Washington DC, 28 November. Bratton, W. J. (1996). Cutting crime and restoring order: what America can learn from New York’s finest. Heritage Lecture 573.
References
89
Burnett, P. (1976). Behavioural geography and the philosophy of the mind. Spatial choice and spatial behaviour. R. Golledge and G. Rushton (Eds.). Ohio State University Press, Ohio: 23–50. Burrough, P. A. and R. A. McDonnell (1998). Principles of geographical information systems. New York, NY, Oxford University Press. Carcach, C. and S. Mukherjee (1999, November). “Women’s fear of violence in the community” Trends and Issues in Crime and Criminal Justice: 1–6. Carlson, J. and E. Hatfield (1992). Psychology of emotion. Fort Worth, TX, Harcourt Brace Jovanovich College Publishers. Carvalho, I. and D. A. Lewis (2003). “Beyond community: reactions to crime and disorder among inner-city residents”. Criminology 41(3): 779–812. Cates, J. A., D. A. Dian, et al. (2003). “Use of protection motivation theory to assess fear of crime in rural areas”. Psychology Crime & Law 9(3): 225–236. Chadee, D. and J. Ditton (2003). “Are older people most afraid of crime? revisiting Ferraro and LaGrange in Trinidad”. The British Journal of Criminology 43(2): 417–433. Clark, J. (2003). “Fear in fear-of-crime”. Psychiatry, Psychology and Law 102(267–282). Couclelis, H. (1998). Aristotelian spatial dynamics in the age of GIS. Spatial and temporal reasoning in geographic information systems. M. Egenhofer and R. G. Golledge (Eds.). Oxford University Press, New York, NY: 109–118. Couclelis, H., R. G. Golledge, et al. (1987). “Exploring the anchor-point hypothesis of spatial cognition”. Journal of Environmental Psychology 7(2): 99–122. Covington, J. and R. B. Taylor (1991). “Fear of crime in urban residential neighbourhoods: implications between – and within – neighbourhood sources for current models”. The Sociological Quarterly 32(2): 231–249. Cromley, E. K. and S. L. McLafferty (2002). GIS and public health. New York, NY, Guilford Press. Day, R. A. (1976). “Urban distance cognition: review and contribution”. Australian geographer 13: 193–200. DeFronzo, J. (1979). “Fear of crime and handgun ownership”. Criminology 17(3): 331–339. Ding, Y. and S. Fotheringham (1992). “The integration of spatial analysis and GIS”. Computing, environment and urban Systems 16. Ditton, J., J. Bannister, et al. (2000). Afraid or angry? recalibrating the fear of crime. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 535–552. Ditton, J. and S. Farrall (2000). The fear of crime. Aldershot, Ashgate. Doran, B. J. and B. G. Lees (2005). “Investigating the spatiotemporal links between disorder, crime, and the fear of crime”. Professional Geographer 57(1): 1–12. Doran, B., J. McMillen, et al. (2007). “A GIS-based investigation of gaming venue catchments”. Transactions in GIS 11(4): 245–265. Downs, R. M. (1977). Maps in minds: reflections on cognitive mapping. New York, NY, Harper and Row. Downs, R. M. and D. Stea (1973). Cognitive maps and spatial behaviour: processes and products. Image and environment. R. M. Drowns and D. Stea (Eds.). Edward Arnold, Great Britain: 1–8. Dubow, F., E. McCabe, et al. (1979). Reactions to crime: a critical review of the literature. US Department of justice and LEA administration. Washington, DC, Government Printing Office. Egenhofer, M. J. and R. G. Golledge (1998). Spatial and temporal reasoning in geographic information systems. New York, NY, Oxford University Press. Ewald, U. (2000). Criminal victimisation and social adaptation in Germany: fear of crime and risk perception in the new Germany. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London: 166. Farrall, S., J. Bannister, et al. (1997). “Questioning the measurement of the ‘fear of crime’: Findings from a major methodological study”. The British Journal of Criminology 37(4): 658–679. Farrall, S. and J. Ditton (1999). “Improving the measurement of attitudinal responses: an example from a crime survey”. International Journal of Social Research Methodology 2(1): 55–68.
90
5
Investigating Fear of Crime
Ferraro, K. F. (1995). Fear of crime: interpreting victimisation risk. Albany, NY, State University of New York Press. Ferraro, K. F. and R. LaGrange (1987). “The measurement of fear of crime”. Sociological Review 57: 70–101. Ferraro, K. F. and R. L. LaGrange (1988). “Are older people afraid of crime?”. Journal of Aging Studies 2(3): 277–287. Ferraro, K. F. and R. L. LaGrange (1992). “Are older people most afraid of crime? reconsidering age differences in fear of victimization.”. Journal of Gerontology 47(5): S233. Ferraro, K. F. and R. LaGrange (2000). The measurement of fear of crime. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 277–308. Fisher, B. S. and J. L. Nasar (1992). “Fear of crime in relation to three exterior site features prospect, refuge and escape”. Environment and Behavior 24(1): 214–239. Fisher, B. S. and J. L. Nasar (1995). “Fear spots in relation to microlevel physical cues: exploring the overlooked”. Journal of Research in Crime and Delinquency 32(2): 214–239. Freundschuh, S. (1998). The relationship between geographic scale, distance, and time as expressed in natural discourse. Spatial and temporal reasoning in geographic information systems. M. J. Egenhofer and R. Golledge (Eds.). Oxford University Press, New York, NY: 155–166. Friedberg, A. and R. a. F. Inc (1983). America afraid: how fear of crime changes the way we live. New York, NY, New Americal Library Inc. Furstenberg, F. F., Jr. (1971). Public reaction to crime in the streets “The American Scholar”. The fear of crime. J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 3–12. Gabriel, U. and W. Greve (2003). “The psychology of fear of crime – conceptual and methodological perspectives”. The British Journal of Criminology 43(3): 600–614. Garofalo, J. (1979). “Victimisation and the fear of crime”. Journal of Research in Crime and Delinquency 16: 80–97. Garofalo, J. (1981). “The fear of crime: causes and consequences”. Journal of Criminal Law and Criminology 72(2): 839. Garofalo, J. and J. Laub (1978). “The fear of crime: broadening our perspective”. Victimology 3: 242–253. Gold, J. R. and G. Revill (2003). “Exploring landscapes of fear: marginality, spectacle and surveillance”. Capital & Class 80: 27. Golledge, R. D. and R. J. Stimson (1997). Spatial behavior: a geographic behavior. New York, NY, Guilford Press. Gomme, I. M. (1986). “Fear of crime among Canadians: a multi-variate analysis”. Journal of Criminal Justice 14(3): 249. Goodchild, M. F. (2004). “Social sciences: interest in GIS grows”. ArcNews 26: 1–4. Gray, D. E. and M. E. OÇonner (1990). “Concern about to fear of crime in an Australian rural community”. Australia and New Zealand Journal of Criminology 23: 284–298. Hale, C. (1996). “Fear of crime: a review of the literature”. International Review of Victimology 4: 79–150. Hanson, R. F., D. W. Smith, et al. (2000). “Crime related fears and demographic diversity in Los Angeles county after 1992 civil disturbances”. Journal of Community Psychology 28(6): 607–623. Harries, K. (1999). “Mapping Crime: Principle and Practice”. Retrieved April 4th, 2011, from http://www.ncjrs.org/html/nij/mapping/pdf.html Hindelang, M. J., M. R. Gottfredson, et al. (1978). Victims of personal crime: an empirical foundation for a theory of personal victimization. Cambridge, MA, Ballinger. Husserl, E. (1973). The possibility of cognition. Rules and meaning. M. Douglas (Ed.). Penguin Education, Baltimore, MD: 197. Innes, M. (2004). “Reinventing tradition? Reassurance, neighbourhood security and policing.” Criminal Justice 4(2): 151–171. Innes, M., N. Fielding, et al. (2002). Signal crime and control signals: towards an evidence based framework for reassurance policing. Guilford, University of Surrey.
References
91
Jeffery, C. R. (1971). Crime prevention through environmental design. Beverly Hills, CA, Sage. Joseph, J. (1997). “Fear of crime among black elderly”. Journal of Black Studies 27(5): 698–717. Kaplan, S. (1973). Cognitive maps in perception and thought. Image and environment. R. M. Downs and D. Stea (Eds.). Aldine, Chicago, IL. Karakus, O., E. F. McGarrell, et al. (2010). “Fear of crime among citizens of Turkey”. Journal of Criminal Justice 38(2): 174–184. Katz, C. M., V. J. Webb, et al. (2003). “Fear of gangs: a test of alternative theoretical models”. Justice Quarterly 20(1): 95–130. Kelling, G. L. and C. M. Coles (1997). Fixing broken windows: restoring order and reducing crime in our communities. New York, NY, Touchstone. Kenney, D. J. (1987). Crime, fear and the New York subways: the role of citizen action. New York, NY, Prager. Killias, M. and C. Clerici (2000). “Different measures of vulnerability in their relation to different dimensions of fear of crime”. The British Journal of Criminology 40(3): 437–450. Kirasic, K. C., G. L. Allen, et al. (1984). “Expression of configuration knowledge of large-scale environments: students’ performance of cognitive tasks”. Environment and Behavior 16(6): 687–712. Kitchen, T. (2002). “Crime prevention and the British planning system: new responsibilities and older challenges”. Planning Theory and Practice 3: 155–172. Kitchin, R. M. (1996). “Increasing the integrity of cognitive mapping research: appraising conceptual schematar of environment-behaviour interaction”. Progress in human geography 20(1): 56–84. Koskela, H. and R. Pain (2000). “Revisiting fear and place: women’s fear of attack and the built environment”. Geoforum 31(2): 269–280. Krahn, H. and L. W. Kennedy (1985). “Producing personal safety: the effects of crime rates, police force size, and fear of crime”. Criminology 23(4): 697–710. Kwan, M. (1999). “Gender and individual access to urban opportunities: a study using space-time measures.” The Professional Geographer 51(2): 210–227. Kwan, M. -P. (2000a). “Interactive geovisualization of activity-travel patterns using threedimensional geographical information systems: a methodological exploration with a large data set”. Transportation Research Part C 8: 185–203. Kwan, M. -P. (2000b). “Gender differences in space-time constraints”. Area 32(2): 145–156. LaGrange, R. L. and K. F. Ferraro (1987). The elderly’s fear of crime: a critical examination of the research. The fear of crime (2000). J. Ditton and S. Farrall (Eds.). Ashgate, Aldershot: 77–96. LaGrange, R. L. and K. F. Ferraro (1989). “Assessing age and gender differences in perceived risk and fear of crime”. Criminology 27(4): 697. Lane, J. and J. W. Meeker (2003). “Ethnicity, information sources, and fear of crime”. Deviant Behavior 24(1): 1–26. Lemanski, C. (2004). “A new apartheid? the spatial implications of fear of crime in Cape Town, South Africa”. Environment and Urbanization 16(2): 101–111. Lewis, D. A. and G. Salem (1986). Fear of crime: incivility and the production of a social problem. New Brunswick, NJ, Transaction Inc. Liska, A. E., A. Sanchirico, et al. (1988). “Fear of crime and constrained behavior specifying and estimating a reciprocal effects model”. Social Forces 66(3): 827–838. Longley, P. A., M. F. Goodchild, et al. (2001). Geographic information systems and science. Chichester, Wiley. Louviere, J. J. (1976). Information processing theory and functional form in spatial behaviour. Spatial choice and spatial behaviour. R. G. Golledge and G. Rushton (Eds.). Ohio State University Press, Columbus, OH: 211–246. Lupton, D. and J. Tulloch (1999). “Theorizing fear of crime: beyond the rational/irrational opposition”. The British Journal of Sociology 50(3): 507–523. Martin, D. (1991). Geographic information systems and their socioeconomic applications. London, Routledge.
92
5
Investigating Fear of Crime
Maury, O. and D. Gascuel (1999). “SHADYS (‘simulateur halieutique de dynamiques spatiales’), a GIS based numerical model of fisheries. Example application: the study of a marine protected area.” Aquatic Living Resources 12(2): 77–88. Mawby, R. I., P. Brunt, et al. (2000). “Fear of crime among British holidaymakers”. The British Journal of Criminology 40(3): 468–479. Mayhew, P. and P. White. (1997). “Home Office Research and Statistics Directorate Findings No 57: The 1996 international crime victimisation survey”. from http://www.homeoffice.gov.uk/ rds//pdfs/r57.pdf. Mennis, J. L. (2003). “Deivation and implementation of a semantic GIS data model informed by principles of cognition”. Computers, Environment and Urban Systems 27(5): 455–479. Merry, E. E. (1981). Urban danger: life in a neighbourhood of strangers. Philadelphia, PA, Temple University press. Mesch, G. S. (2000). “Perceptions of risk, lifestyle activities, and fear of crime”. Deviant Behavior 21(1): 47–62. Miceli, R., M. Roccato, et al. (2004). “Fear of crime in Italy – spread and determinants”. Environment and Behavior 36(6): 776–789. Mirrlees-Black, C. and J. Allen (1998). Concern about crime: Findings from the 1998 British Crime Survey. Research Findings No 83. London, Home Office Research, Development and Statistics Directorate. Monmonier, M. (1996). How to lie with maps. Chicago, IL, University of Chicago Press. Moran, L. J., B. Skeggs, et al. (2003). “The formation of fear in gay space: the ‘straights’ story” Capital & Class (80): 173. Murray, A. T., I. McGuffog, et al. (2001). “Exploratory spatial data analysis techniques for examining urban crime”. The British Journal of Criminology 41(2): 309–329. Nair, G., J. Ditton, et al. (1993). “Environmental improvements and the fear of crime: the sad case of the ‘Pond’ area in Glasgow”. The British Journal of Criminology 33(4): 555–561. Nasar, J. L. (1998). The evaluative image of the city. California, Sage. Nasar, J. L., B. Fisher, et al. (1993). “Proximate physical cues to fear of crime”. Landscape and Urban Planning 26: 161–178. Nasar, J. L. and K. M. Jones (1997). “Landscapes of fear and stress”. Environment and Behavior 29(3): 291–323. National Campaign Against Violence and Crime (1998). Fear of crime – summary volume. Canberra, NCAVAC. Neill, W. J. V. (2001). “Marketing the urban experience: reflections on the place of fear in the promotional strategies of Belfast, Detroit, and Berlin” Urban Studies 38(5/6): 815. Nelson, A., R. Bromley, et al. (2001). “Identifying micro-spatial and temporal patterns of violent crime and disorder in a British city centre”. Applied Geography 21: 249–274. Newman, O. (1972). Defensible space: crime prevention through urban design. New York, NY, Macmillan. Oatley, K. and J. Jenkins (1996). Understanding emotions. Cambridge, Blakwell Publishers. Oc, T. and S. Tiesdell (1997). Safer city centres : reviving the public realm. London, Chapman. Olsen, P. and B. Doran (2002). “Climatic modelling of the Australian distribution of the grass owl (Tyto capensis): is there an inland population?” Wildlife Research 29(2): 117–125. Orleans, P. (1973). Differential cognition of urban residents: Effects of social scale on mappings. Image and environment. R. M. Downs and D. Stea (Eds.). Aldine, Chicago, IL: 115–130. O’Sullivan, D. and D. J. Unwin (2010). Geographic information systems. Hoboken, NJ, Wiley. Pain, R. H. (1997). “ ‘Old age’ and ageism in urban research: the case of fear of crime”. International Journal of Urban & Regional Res 21(1): 117–128. Pain, R. (2000). “Place, social relations and the fear of crime: a review”. Progress in Human Geography 24(3): 365–387. Pantazis, C. (2000). “ ‘Fear of Crime’, vulnerability and poverty”. The British Journal of Criminology 40(3): 414–436.
References
93
Perkins, D. D. and R. B. Taylor (1996). “Ecological assessments of community disorder: their relationship to fear of crime and theoretical implications”. American Journal of Community Psychology 24(1): 63–107. Potas, I. (1996). Courts and sentencing. Crime and justice: an Australian textbook in criminology. K. M. Hazelhurst (Ed.). LBC Information Services, North Ryde, NSW. Ratcliffe, J. H. and M. J. McCullagh (2001). “Chasing ghosts?: police perception of high crime areas”. The British Journal of Criminology 41(2): 330–341. Reid, L., J. T. Roberts, et al. (1998). “Fear of crime and collective action: an analysis of coping strategies”. Sociological Inquiry 68(3): 312–328. Reiss, A. J. (1986). Why are communities important in understanding crime. Communities and crime. A. J. Reiss and M. Tonry (Eds.). University of Chicago press, Chicago, IL: 1–34. Rountree, P. W. and K. C. Land (1996). “Perceived risk versus fear of crime: empirical evidence of conceptually distinct reactions in survey data”. Social Forces 74(4): 1353–1377. Russo, J. (2001, Summer). From description to prediction: Crime mapping. TechBeat: 14. Samuels, R. and B. Judd (2002). Public housing estate renewal: Interventions and the epidemiology of victimisation. Housing, Crime and Stronger Communities Conference, Melbourne, Australian Institute of Criminology & Australian Housing and Urban Research Institute. Scott, H. (2003). “Stranger danger: explaining women’s fear of crime”. Western Criminology Review 4(3): 203–214. Sholl, M. J. (1996). From visual information to cognitive maps. The construction of cognitive maps. J. Portugali (Ed.). Kluwer, Dordrecht: 157–186. Skogan, W. G. (1990). Disorder and decline: crime and the spiral decay in American neighbourhoods. Los Angeles, CA, University of California Press. Skogan, W. G. (1999). Measuring what matters: crime, disorder and fear. Measuring what matters: proceedings from the police research institute meetings. R. H. Langworthy (Ed.). National Institute of Justice, Washington, DC. Skogan, W. G. (2004, May). “Preface” Annals of the American Academy: 1–14. Skogan, W. G. and M. G. Maxfield (1981). Coping with crime : individual and neighborhood reactions. Beverly Hills, CA, Sage Publications. Sluckin, W. (1979). Fear in animals and man. New York, NY, Van Nostrand Reinhold. Smith, S. J. (1987). “Fear of crime: beyond a geography of deviance” Progress in Human Geography 11: 1–23. Smith, W. R. and M. Tortensson (1997). “Gender differences in risk perception and neutralising fear of crime”. The British Journal of Criminology 37(4): 603–634. Sparks, R., E. Girling, et al. (2001). “Fear and everyday urban lives”. Urban Studies 38(5–6): 885–898. Stafford, M. C. and O. R. Gall (1984). “Victimization rates, exposure to risk, and fear of crime”. Criminology 22(2): 173. Stanko, E. A. (2000). Victims R us: the life history of fear of crime and the politicisation of violence. Crime, risk and insecurity. T. Hope and R. Sparks (Eds.). Routledge, London. Sundeen, R. A. and J. T. Mathieu (1976). The urban elderly: environments of fear. Crime and the elderly: challenge and response. J. Goldsmith and S. S. Goldsmith (Eds.). Lexington Books, Lexington, MA. Sutherland, E. H. and D. R. Cressey (1970). Criminology. Philadelphia, PA, J.B Lippincott Company. Taylor, R. B. and J. Covington (1993). “Community structural change and fear of crime”. Social Problems 40(3): 374–395. Taylor, R. B. and M. Hale (1986). “Testing alternative models of fear of crime”. The Journal of Criminal Law and Criminology 77(1): 151–189. Tewksbury, R. and E. E. Mustaine (2003). “College students’ lifestyles and self-protective behaviors – further considerations of the guardianship concept in routine activity theory”. Criminal Justice and Behavior 30(3): 302–327.
94
5
Investigating Fear of Crime
Thomas, C. and R. Bromley (2000). “City-centre revitalisation: problems of fragmentation and fear in the evening and night-time city”. Urban Studies 37(8): 1403–1429. Toseland, R. W. (1982). “Fear of crime: who is most vulnerable?” Journal of Criminal Justice 10(3): 199. Tulloch, J. (1998). Quantitative review. Fear of crime. J. Tulloch, D. Lupton, W. Blood, et al. (Eds.). National Campaign Against Violence and Crime (NCAVAC), Canberra. Tulloch, M. (2000). “The meaning of age differences in the fear of crime”. The British Journal of Criminology 40(3): 451–467. Tulloch, J., D. Lupton, et al. (1998). Fear of crime volume two. Canberra, Centre for Cross Cultural Research. Tversky, B. and H. A. Taylor (1998). Acquiring spatial and temporal knowledge from language. Spatial and temporal reasoning in geographic information systems. M. J. Egenhofer and R. Golledge (Eds.). Oxford University Press, New York, NY: 155–166. Vacha, E. F. and T. F. McLaughlin (2004). “Risky firearms behavior in low-income families of elementary school children: the impact of poverty, fear of crime, and crime victimization on keeping and storing firearms”. Journal of Family Violence 19(3): 175. Valentine, G. (1989). “The geography of women’s fear”. Area 21(4): 385–390. Vanetti, E. J. and G. L. Allen (1988). “Communicating environmental knowledge – the impact of verbal and spatial abilities on the production and comprehension of route directions.” Environment and Behavior 20(6): 667–682. Vold, G. B., T. J. Berbard, et al. (2002). Theoretical criminology, fifth edition. Oxford, Oxford University press. Walker, M. A. (1994). “Measuring concern about crime”. The British Journal of Criminology 34(3): 366–702. Walsh, D. and A. Poole (1983). A dictionary of criminology. London, Routledge & Kegan Paul. Walsh, D. A., I. K. Krauss, et al. (1981). Spatial ability, and environmental knowledge, and environmental use: the elderly. Spatial representation and behaviour across the lifespan: theory and application. L. S. Liben, A. H. Patterson and N. Newcombe (Eds.). Academic Press, New York, NY: 321–357. Warr, M. (1984). “Fear of victimization: why are women and the elderly more afraid”. Social Science Quarterly 65(3): 681–702. Warr, M. (1985). “Fear of rape among urban women”. Social Problems 32(3): 238–250. Warr, M. (2000). “Fear of crime in the United States: avenues for research and policy” Criminal Justice 4: 452–489. Warr, M. and C. G. Ellison (2000). “Rethinking social reactions to crime: personal and altruistic fear in family households.” American Journal of Sociology 106(3): 551–578. Weisburd, D., L. A. Wycoff, et al. (2004). Does crime just move around the corner? US Department of Justice. White, R. and F. Haines (2004). Crime and criminology. Oxford, Oxford University Press. Williams, P. and J. Dickinson (1993). “Fear of crime: read all about it. The relationship between newspaper crime reporting and fear of crime”. The British Journal of Criminology 33(1): 33–56. Williams, J. S., B. K. Singh, et al. (1994). “Urban youth, fear of crime, and resulting defensive actions.” Adolescence 29(114): 323. Wilson, J. Q. and G. L. Kelling (1982, March). “The police and neighbourhood safety: broken windows” The Atlantic Monthly: 29–38. Wilson-Doenges, G. (2000). “An exploration of sense of community and fear of crime in gated communities.” Environment and Behavior 32(5): 597–611. Yarwood, R. (2001). “Crime and policing in the British countryside: some agendas for contemporary geographical research”. Sociologia Ruralis 41(2): 201–219. Yin, P. P. (1980). “Fear of crime among the elderly: some issues and suggestions”. Social Problems 27(27): 492.
Chapter 6
The Wollongong Study
The Goals of the Wollongong Study Links between social and physical disorder, crime and fear of crime have been areas of research interest for some time (e.g. Wilson and Kelling, 1982; Skogan, 1990; Kelling and Coles, 1997). One of the most influential studies into this area was the work of Wilson and Kelling (1982) who put forth a theory outlining a causal relationship between disorder, fear and crime. The theory has since been referred to as the ‘broken windows’ thesis or theory (e.g. Harcourt, 1998; Sampson and Raudenbush, 1999; Loukaitou-Sideris, 1999) and has had a significant bearing upon subsequent research and policy developments (e.g. Taylor and Covington, 1993; Tiesdell and Oc, 1998; Skogan, 1990; Bratton, 1995, 1996; Sampson and Raudenbush, 1999). Despite this, there have been few studies that have used a spatially explicit approach to investigate potential spatiotemporal links between crime, disorder and fear. Thus, ‘mapping out fear of crime’ holds the potential to deliver baseline data and a localized means of looking into questions such as 1. Can techniques from behavioural geography be successfully combined with GIS to investigate the avoidance and protective behaviours that people adopt in relation to their fear? 2. When and where are people afraid of crime? 3. Do hotspots of crime, fear and disorder overlap? If so, what implication does this have for reducing fear of crime among the CBD working population? 4. How do localized, behavioural measures of fear compare with global approaches? This chapter first presents some background information on Wollongong, including the evolution and structure of the CBD area, crime rates over the years prior to the study and the survey technique adopted for the project. It then moves onto describe the cognitive mapping technique that was used to look at collective patterns of avoidance behaviour across the CBD at different times of the working day. These patterns are subsequently compared to distributions of social and physical disorder and used as a framework to discuss the spatiotemporal implications of the B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_6, C Springer Science+Business Media, LLC 2012
95
96
6
The Wollongong Study
broken windows thesis. The next section presents the findings from the activity diary analysis of protective behaviours and emotional levels of fear in relation to the daily routines of people working in the CBD. The study finished at the end of 2004 which proved to be fortuitous timing, as the Wollongong City Council was in the early stages of implementing a 5-year City Centre Revitalization Strategy (WCC, 2003) and developing a strategic vision, the ‘2020 plan’, which outlined goals that the city and broader community wanted to achieve in the short-medium term. This enabled key research findings from the Wollongong study to be used as a means of assessing the impact of proposed land use planning and design changes (Irwin et al., 2003). The results of the study were also integrated with a crime prevention and community safety plan (WCC, 2007).
Research Setting The study site for the project was the city of Wollongong, Australia, located approximately 80 km south of Sydney on the east coast of Australia (see Fig. 6.1).
( Cronulla (
(
Campbelltown
Picton
° (
Legend
! .
Study Site
(
Regional Centers
Bulli
! . Wollongong CBD
LGA Boundary Coastline Wollongong Roadnetwork
Tasman Sea 0
5
10
20
( Kilometers
Fig. 6.1 The location of Wollongong
Shell Harbour
Research Setting
97
Wollongong is the ninth largest city in Australia and the third largest in New South Wales (WCC, 2003), with a population of 181,000 people and key industry sectors based around manufacturing, mining, technology research and education (IRIS, 2004; ABS, 2003). It is recognized as being the major city in the Illawarra region (WCC, 2003) with the smaller centres of Bulli to the north and Shell Harbour to the south. Wollongong developed in the second half of the nineteenth century as an important industrial city with an economy based on coal, steel, engineering and clothing (Gupy et al., 2000). The past two decades have seen a shift, with activity in these sectors declining and growth taking place in health, education, hospitality, retailing, property, information technology and business services (WCC, 2003). The shift was marked by massive job losses in the early 1980s in steel, coal, engineering and clothing industries which have resulted in Wollongong struggling to establish a new identity and economic development path (Gupy et al., 2000). Future economic growth and diversification is anticipated in areas such as retail and wholesale trade, transport, telecommunications, business services, metal industries, community services, entertainment, accommodation and personal services (WCC, 2003). This is in line with the broad aim of the Wollongong City Council to further develop the city centre as a regional hub for higher-order services and facilities (Olsen, 2003). An important part of Wollongong’s regional identity relates to its rugby league heritage. It has long been a nursery for high-calibre players in the National Rugby League (NRL) competition. In 1998 the Illawarra Steelers merged with the powerful Sydney-based side, the Saint George Dragons, to become the Saint George-Illawarra Dragons (Fagan, 2009). The city draws immense pride from the performance of the merged entity, in particular their recent grand final triumph in 2010. Visually, the city centre of Wollongong is dominated by three elements of the physical setting: the steel works at Port Kembla, the escarpment and the ocean (Irwin et al., 2003, Figs. 6.2 and 6.3). The setting of Wollongong, lying between the coastline and the escarpment is regarded as an important part of the city’s identity (WCC, 2003).
Fig. 6.2 View looking south from the city centre of Wollongong towards the steel works at Port Kembla
98
6
The Wollongong Study
Fig. 6.3 View looking west from the city centre of Wollongong towards the escarpment
Logic Behind Study Site Selection The selection of the CBD as a study site was based on a number of considerations. First, the logic used was similar to Bennett (1991), who chose a study site with relatively high rates of crime in order to make comparisons between crime and fear of crime. One of the primary aims of this study is to investigate potential spatiotemporal links between fear of crime and the actual occurrence of crime. Hence, the selection of a study site that had a significant crime problem was logical. Wollongong has a significant crime problem, which is outlined in more detail below. According to a number of authors (e.g. Sampson and Groves, 1989; Markowitz et al., 2001), communities that experience rapid ecological change are more likely to show increases in crime and fear of crime. Given the economic and social shifts that have occurred in the Wollongong region in recent times, an increase in crimerelated problems is not entirely unexpected. Further, such trends make Wollongong an interesting and appropriate study site for a spatiotemporal investigation of fear of crime, crime itself and social and physical disorder. Second, Wollongong is spatially confined by the Tasman Sea to the east and an escarpment to the west, meaning the population of Wollongong is relatively confined. As such, it is well suited to the development of a method to analyse the spatiotemporal nature of the fear of crime. Larger potential study sites such as the CBD of Sydney were harder to define in terms of spatial extent. Other issues such as the likely commuting distance of respondents were a consideration. In terms of collecting activity diary data, larger study sites are likely to be more complex, with commuters travelling over greater distances (e.g. Kwan, 1999; 2000a, 2000b). Correspondingly, a large activity diary dataset would be needed in order to adequately analyse a study site of large proportions such as the CBD of Sydney. This was beyond the scope of this study. Thus, the smaller CBD of Wollongong was more suited to this project.
Research Setting
99
The Central Business District of Wollongong The Central Business District (CBD) of Wollongong is well defined and compact, comprising office blocks, several shopping mall complexes and public facilities such as parks within an area of approximately 1.5 × 0.8 kilometres (see Fig. 6.4 below). Historically, the city centre of Wollongong has always been structured around Crown Street (Lee, 1997). In 1948 it stretched from approximately 100 metres west of the rail line to Corrimal Street. By 1986, the city centre had spread considerably with growth taking place along Auburn Street and east towards Keira Street. There had also been expansion north of Crown Street. By 2003, the broader city centre was still concentrated around Crown Street between the rail line and Corrimal Street. However, it had expanded south to link up with some of the areas of the city centre that were isolated in 1986. Some growth was also evident along the northern and eastern edges of the city centre area (WCC, 2003). Commuters to the CBD show a strong dependency on motor vehicle transport, with a high proportion of people travelling to work by car (67%) with only 5.8% travelling by public transport (ABS,
Campbel
Victoria S
t
St
Loftus St
Rawson
St
e St
West St
t
Bank St
Church
Ellen St
Burelli S Stewart S McCabe treet t Park Georg Kembla S
t Kenny S
t
Atchison
St
Ave
Auburn S
Coastline Main Roads Minor Roads Rail Line Retail Core Government Offices West of Railway South of Burelli Street East of Corrimal Street Public Parks
t
Globe Ln
Gladstone
Legend
St
Market S Court Ln
Crown Street
Osborne
Smith St
Harbour St
St
St
Ro Hercules
on
l Street
s bin
Corrima
Rd
Denison
w
Ne
pto
Da
Keira S
ig
l Meters B
St
t hS
treet
l St
250
Glebe St
Beach S
t
Fig. 6.4 The CBD area of Wollongong (based on descriptions and maps in WCC, 2003)
100
6
The Wollongong Study
2003). A structure plan for the current city centre by the Wollongong City Council (WCC, 2003) describes six general areas of the CBD: – – – – – –
Retail core Commercial offices The area west of the railway The area south of Burelli Street The area east of Corrimal Street Public parks
The current city core or CBD is centred around Crown Street and occupies a smaller area. Figure 6.4 above shows the six main areas of the city centre as well as the extent of the current CBD. The retail core is centred around the Crown Street Mall area which is located along Crown Street between Keira Street and Kembla Street. The Crown Street Mall provides a wide range of shopping facilities with several large department stores and over 250 specialist stores (WCC, 2003). There is no vehicle access along Crown Street between Keira and Kembla Streets or along Church Street between Burelli and Market Streets. These areas are dominated by paved street zones and public seating, and entertainment facilities (see Figs. 6.5 and 6.6). Some parts of the retail core lie outside the Crown Street Mall area. Further west along Crown Street are a range of cafés, large department stores and some shops. An older mall complex, the Piccadilly Shopping area, is situated on the other side of the rail line. The retail core areas along Keira Street and along Crown Street east of Kembla Street are dominated by a wide range of restaurant and entertainment facilities, including a successful café culture (WCC, 2003; Irwin et al., 2003). Woolworths and Aldi supermarkets are located on the corners of Burelli and Kembla and Stewart and Corrimal Streets. The commercial office area comprises primarily the City Council, Commonwealth and NSW Government offices. The majority of office-based activity is centred around the commercial office area but some smaller commercial offices are spread south towards Swan Street and on Regent Street near
Fig. 6.5 Public seating area in the Crown Street Mall at the junction of Crown and Church streets
Research Setting
101
Fig. 6.6 Paved walkway and seating area in the Crown Street Mall at the junction of Crown and Keira Streets
the railway. Outside the commercial office area, few office buildings exceed a height of two or three floors (WCC, 2003). The area to the west of the rail line is dominated by hospital- and medical-related uses and medium-density housing. This area is also a major entry point into the CBD through Crown Street (WCC, 2003). The area south of Burelli Street contains a mix of residential, commercial and light industrial uses. The light industry and commercial offices are primarily in medium sized, two-story buildings of mixed quality (WCC, 2003). The area east of Corrimal Street is primarily residential. Exceptions to this are several large car dealerships located around the intersection of Corrimal and Crown Streets and a sports stadium and entertainment area located along Harbour Street. The main public parks in the CBD area are McCabe Park and Pioneer Park. McCabe Park is a large urban park with street edges to the north and east. The street edges have public parking facilities. The park itself contains grassed areas, playgrounds, ornamental gardens and memorial structures. Pioneer Park was formed over a former rest park and is essentially a public garden with a neighbourhood function. Market Square is a park on the site of an old market place. It is regarded as a site of heritage significance (WCC, 2003).
Crime and Fear of Crime in Wollongong Previous research has found crime and fear of crime to be major problems in the Wollongong area. A project looking specifically at fear of crime among women found that up to 75% of women surveyed experienced moderate to great fear (WCC, 1999). Similarly, a more recent community values survey found that crime was one of the issues residents were most concerned about (IRIS, 2002). Crime Trends in the Illawarra Region, 2001 An indication of the type of crime problems experienced in the broader Wollongong region can be gained by examining analyses of recorded crime data in the Illawarra
102
6
The Wollongong Study
Statistical Division (SD). An analysis of recorded crime trends between January 2000 and December 2001 in NSW by the Bureau of Crime Statistics and Research (Allen, 2002) showed the following trends. For the offence categories break and enter – non-dwelling, the Illawarra SD showed a significant upward monthly trend (up by 14%). The rate of break and enter-non-dwelling in the Illawarra SD is above the state average. Similarly, for the offence category motor vehicle theft, the Illawarra SD showed a significant upward trend (up by 27%). Only one SD outside Sydney recorded a rate of motor vehicle theft in 2001 that was higher than the state average in 2001 and this was the Illawarra SD. Further, Allen (2002) highlights that the Illawarra SD along with the Central Western Sydney SD have two of the highest rates of motor vehicle theft in the state. One other offence category, steal from person, showed a significant upward trend in the Illawarra SD. Steal-from-person offences were up by 41% during this period in the Illawarra SD but this was lower than the state rate in 2001. The only offence category showing decrease was for robbery with a firearm. This category was down by 25% during this period in the Illawarra SD but this did not represent a significant downward trend. Crime Hotspots at the LGA Level in NSW, 2002 The NSW Bureau of Crime Statistics and Research (BOSCAR, 2004) provide an analysis at the local government area (LGA) for crime hotspots in 2002. The analysis is based on ranking the top 25 LGAs for nine selected offences (assault, assault – DV related, sexual assault, robbery, break and enter – dwelling, break and enter – non-dwelling, motor vehicle theft, steal from motor vehicle and steal from person). The Wollongong LGA is listed as being a crime hotspot in 2002 for the offence categories of break and enter – dwelling, break and enter – non-dwelling, motor vehicle theft and steal from person. Table 6.1 below shows the four offences for which the Wollongong LGA is listed as being a crime hotspot in 2002. For break and enter – dwelling, Wollongong LGA is ranked 11th in the state with 3041 offences at a rate of 1615 per 100,000 of the population. For break and enter – nondwelling, the Wollongong LGA is ranked 23rd in the state with 1786 offences at a rate of 948.6 per 100,000 of the population. For motor vehicle theft, the Wollongong LGA is ranked 22nd in the state with 1676 offences with a rate of 890.2 per 100,000 Table 6.1 Offences for which the Wollongong LGA was recorded as being in the top 25 LGAs in NSW for 2002 (based on BOSCAR, 2004) Offence
Rank
Number of offences in 2002
Rate per 100,000 population
Break and enter – dwelling Break and enter – non-dwelling Motor vehicle theft Steal from person
11
3, 041
1,615.2
23
1, 786
948.6
22 21
1, 676 490
890.2 260.3
Research Setting
103
of the population. For the offence of steal from person, the Wollongong LGA is ranked 21st in the state with 490 offences at 260 per 100,000 of the population.
Methods There were four key components of the research methods which are listed below and described in detail in the sections following. (a) Fear of crime survey and analysis. This component was based on a survey of 260 people working in the CBD area, conducted in May–June 2002. (b) Disorder assessment. The disorder assessment included physical and social components: – 1 Physical disorder assessment conducted in June 2002. – 8 Social disorder assessments conducted in June–July 2002. (c) Spatial analysis of crime data. This component was based on crime data sourced through the NSW Police Service for Wollongong LGA. (d) Combinatory spatial analysis. A framework was developed to examine potential spatiotemporal links between crime, disorder and fear. Fear-of-Crime Survey and Analysis A five-part survey was designed to investigate the spatiotemporal nature of fear of crime in the CBD area of Wollongong. The survey was based on a voluntary sample. This involved the surveyor approaching numerous businesses in the CBD area and, where consent was given, conducting a face-to-face interview with respondents. Due to ethical requirements of human research, only people older than 18 years were interviewed and respondents were informed that the survey was in relation to fear of crime. Further, it was made clear to respondents that the study would not involve the dissemination of any personal information and that the analytical procedure would result in generalized results. The five sections of the survey were as follows: (1) (2) (3) (4) (5)
General factors known to influence fear of crime Questions on emotional levels of fear in relation to activity diaries Questions on protective behaviour in relation to activity diaries Vignettes assessing emotional levels of fear in hypothetical situations Cognitive mapping of avoidance behaviour
A pilot study was conducted to pretest the survey procedure. The results from the first pilot study indicated that more questions could be incorporated into the survey procedure, as the 15 minutes allocated to the interview process were not being fully utilized.
104
6
The Wollongong Study
General Factors Known to Influence Fear of Crime In general, a number of factors are known to influence fear of crime. These include age, sex, income status, housing type, previous victimization, ethnicity, perceptions of safety, media influence, length of time in neighbourhood, integration with neighbourhood and confidence in the police (e.g. Covington and Taylor, 1991; Box et al., 1988; Borooah and Carcach, 1997; Farrall et al., 2000). Age and sex were recorded as categorical variables. In the case of age, survey regulations dictated that only people over the age of 18 years could be interviewed. Age was recorded in 5-year blocks, starting at 18. Income status was assessed using the measure suggested by Farrall et al. (2000). This was based on the question, ‘How easily would you be able to find $600 suddenly, without resource to a bank loan?’ The responses to this question are based on a five-point Likert scale ranging from 1 equating to very easy to 5 equating to impossible. The measure by Farrall et al. (2000) used a figure of 200 British pounds. At the time of the survey, this figure was equivalent to approximately 600 Australian dollars. The type of housing was recorded after Borooah and Carcach (1997) according to whether respondents were renting from a government housing commission, were owner-occupiers or non-owner-occupiers. Ethnicity was assessed according to whether respondents considered themselves as coming from an English-speaking background. People who identified themselves as coming from a non-Englishspeaking background were assumed to be from an ethnic minority. Perception of media-related issues has been found to contribute to fear of crime (Box et al., 1988). This was recorded using the question, ‘How do you rate media coverage of crimerelated issues in the Wollongong region?’ Responses were based on a five-point Likert scale ranging from very understated to very overstated. Previous victimization was recorded according to Borooah and Carcach (1997). This involved asking respondents if they had been victims of certain crimes in the past 12 months. The crimes recorded were deliberate use of a weapon, attack or assault, threats of force or violence, theft and attempted theft and deliberate damage to property or tampering by vandals or thieves. The length of time that respondents had been living in their current neighbourhood was recorded according to whether they had been there less than 1 year, 1–2 years, 3–5 years and more than 5 years. Community integration was measured after Covington and Taylor (1991) who asked respondents, ‘Suppose some kids were spray painting a building near where you work. Do you think you or any of your neighbours would call the police?’ The response to this question was ‘Yes’ or ‘No’. For the purposes of this study, the phrasing of the question related to the work environment, as opposed to the neighbourhood environment in Covington and Taylor (1991). The purpose of changing the phrasing was related to the focus of this study on the working population of the CBD of Wollongong. Finally, the standard global measure of fear was also recorded (i.e. how safe do you feel when walking alone in the area around your home after dark?). The phrasing of this measure and the responses associated with it were based on Borooah and Carcach (1997).
Research Setting
105
Questions on Emotional Levels of Fear and Protective Behaviour in Relation to Activity Diaries Activity diaries were seen as a means of building on Fisher and Nasar’s (1992, 1995) and Nasar and Jones’s (1997) micro-level behavioural investigations of fear in specific spatial and temporal contexts. The activity diary approach is particularly similar to the procedure used by Nasar and Jones (1997) in their investigation of emotional levels of fear among 26 female respondents on the Ohio State University campus. The researchers asked the respondents to walk a specific pedestrian route on the campus which passed through a variety of landscapes. Respondents were given a hand-held tape recorder and asked to record emotional feelings of safety while walking the prescribed route. By using this approach, the authors were able to investigate site- and context-specific levels of emotion-based fear. In this study, the use of activity diaries provided a similar but slightly different framework for investigating emotional levels of fear. First, survey respondents were asked to record their activities using the diary method (Golledge and Stimson, 1997) and subsequently interviewed. During this follow-up interview respondents were asked about their emotional levels of fear in relation to the activities and times recorded in their diaries. The phrasing of the question used to assess emotion-based fear was based on Farrall et al. (2000). Respondents were also asked if they were adopting any protective behaviours in relation to the activities recorded in their diaries. The protective measures of having a dog, carrying something for defence, relying on self-defence training and ‘other’, as described by Krahn and Kennedy (1985) were recorded. Also included were categories relating to respondents who were making sure they were accompanied by a friend, carrying a mobile phone to call someone if they felt in danger and adopting no protective measures. The diary method involves the subject keeping a dairy for a specified period of time (Golledge and Stimson, 1997). In this study, the diary was for a period of one day during the week on which the respondent was working. The time of the diary was from 02:00 to 24:00. Prescribed time intervals of half-hour periods were used (e.g. Kwan, 2000a). Appendix 2 shows the template for the diary and the accompanying instruction sheet given to survey participants. Vignettes Assessing Emotional Levels of Fear in Hypothetical Situations Farrall et al. (2000) used a series of vignettes to assess emotional levels of fear in a range of hypothetical situations. The vignettes, originally based on the study by Van der Wurff et al. (1989), were used by Farrall et al. (2000) to describe their sample in terms of sensitivity to fear when exposed to the same set of hypothetical situations. This approach is used in this study, for the same purposes. Cognitive Mapping of Avoidance Behaviour Respondents were provided with a map of the CBD area and asked if they avoided any areas because they were afraid of being robbed, beaten or attacked during and
106
6
The Wollongong Study
after work hours. The map used to define the area of the CBD was based upon previous surveys conducted by the Wollongong City Council. The phrasing of the question, as with the activity diary analysis of emotion-based fear was designed to specifically tap the fear of personal crime (e.g. Farrall et al., 2000). Where respondents indicated that they avoided areas after work hours, they were asked to clarify the time that they started to avoid particular areas. Respondents that indicated that they avoided any areas because of their fear of crime were also asked to specify how hard they tried to avoid those areas on a scale of 1–5 (1 indicating very hard, 5 indicating not hard at all). Assessing the degree to which people avoided any area they indicated on the maps was designed to provide a weighting measure to be used when collating all of the maps. The same technique was used to assess avoidance behaviour in the neighbourhoods of the respondents. Respondents were asked to mark where they lived on a map of the area. A circle of 1.6 kilometres in diameter was then drawn around their home. This figure was taken from the general phrasing of the global measure, which typically relates to areas within one mile of respondents’ homes. GIS-Based Technique for Collating the Cognitive Maps of Avoidance The maps showing the areas avoided by the individual respondents were digitized and overlaid using ArcView GIS, to create maps of collective avoidance for different times of the day (see Fig. 6.7 below). The digitizing process was based upon tracing the outline of the areas that respondents indicated they avoided due to their fear of crime. Using a similar approach to create a simple index model (e.g. Chang, 2010), the degree of avoidance was used to weight the areas avoided by individuals when combined to create collective maps. Collective avoidance maps were made for the hours between 9 am and 5:30 pm, 5:30 pm and 7 pm and after 7 pm. This
Different degrees of avoidance and different times
Avoidance grids for individual respondents, weighted by how hard people tried to avoid specific areas, were combined for different time groupings to form collective avoidance maps
Fig. 6.7 Combinatory process used to collate individual avoidance grids
Research Setting
107
temporal segmentation was designed to investigate collective avoidance in relation to the general daily routines of people working in the CBD of Wollongong. Many of the cognitive maps that survey respondents drew to describe the areas they avoided because they were afraid of being robbed, beaten or attacked outlined areas which showed fine-scale detail. In some cases, areas indicated were street corners or seating areas no more than 10 metres in length or width. In order to minimize the loss of this fine-scale information on avoidance behaviour, an output cell size of 10 metres was selected when converting the input vector files to rasterbased coverages for the combinatory procedure. Disorder Assessment Physical Disorder A physical disorder assessment was conducted using a method similar to that of Sampson and Raudenbush (1999) who assessed disorder on a block-by-block level in Chicago. The disorder assessment in this study, however, was conducted on foot as opposed to by vehicle in the Sampson and Raudenbush study (1999). Table 6.2 below shows the different types of disorder recorded in the assessment. A weighting system was designed to gain a more accurate impression of how disorder was likely to impact upon the public. The weighting system was based upon recording the level of the different types of disorder as well as a weighting factor. At each of the blocks, the level of different types of disorder was assessed on a scale of 1–5 based on how extensive they were. Figures 6.8 and 6.9 show examples of the values given to different levels of graffiti. The weighting factor was based upon how Table 6.2 Types of physical disorder recorded in the physical disorder assessment (after Sampson and Raudenbush, 1999) and the weighting factor given to each type of disorder Disorder number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Type of physical disorder
Weighting factor
Cigarettes or cigars in street gutter Garbage or litter in street or sidewalk Empty beer bottles visible in street Tagging graffiti Gang graffiti Political message graffiti Graffiti painted over Abandoned cars/glass from smashed windscreens Abandoned/boarded-up houses Lack of exterior maintenance Vandalism to buildings Vandalism to public structures Condoms on sidewalk Needles/syringes/methadone capsules on sidewalk Evidence of homeless people
1 (not very visible) 2 (moderately visible) 3 (visible) 4 (quite visible) 4 (quite visible) 4 (quite visible) 2 (moderately visible) 5 (very visible) 5 (very visible) 3 (visible) 5 (very visible) 5 (very visible) 3 (visible) 5 (very visible) 5 (very visible)
108
6
The Wollongong Study
Fig. 6.8 An example of graffiti in Crown Lane, Wollongong, that was given a level value of 3
Fig. 6.9 An example of graffiti in McCabe Park, Wollongong, that was given a level value of 5 (i.e. very extensive)
visible the different types of disorders are. More visible types of disorder such as abandoned cars or buildings were given a higher factor weighting than cigarettes in the street gutter (see Table 6.2 for weighting factors). The latitude and longitude were taken down for each point where disorder data were recorded. Maps showing the distribution of disorder were created, one based upon simply the presence of
Research Setting
109
disorder and another based upon the weight multiplied by the level of disorder for each data point. Social Disorder As with the physical disorder assessment, the social disorder assessment was based on techniques outlined by Sampson and Raudenbush (1999). In their study, social disorder was assessed in a vehicle moving at approximately five miles per hour down every street in the study area. Trained observers recorded the presence or absence of adults loitering or congregating, drinking alcohol in public, peer groups with gang indicators present, public intoxication, adults fighting or arguing in a hostile manner, sale of drugs and prostitution on the street. Sampson and Raudenbush (1999) noted that temporal variation in social disorder is a particular problem in attempts to systematically observe it. The authors emphasize that the probability of finding adults loitering or drinking, of finding peer gangs hanging out, or of seeing prostitution or drug deals will depend on the time of day in which a face block is observed. In their study, the authors felt that they were able to account for temporal variation within their sample because face blocks were assessed between the hours of 07:00 and 19:00. However, it could be argued that this approach, by not recoding social disorder after 19:00, does not sufficiently account for potential temporal variation. For example, Thomas and Bromley (2000) describe how the decentralization of retail, office and leisure functions in British cities has been central to segmenting the use of facilities in these areas. The abrupt curtailment of functions related to the business day is followed by the ‘five o’clock flight’, where the working population leaves the city centre. Following this, the city is relatively abandoned for several hours until the onset of the ‘pub and club’ culture, which arrives later. The authors note that the ‘pub and club’ culture is frequently associated with types of social disorder such as heavy drinking, drug use and late-night violent incidents. In an attempt to capture this type of temporal variation, social disorder in this study was assessed for all hours of the day. Social disorder assessments were conducted between 06:00–12:00, 12:00–18:00, 18:00–24:00 and 00:00–06:00. Further, social disorder assessments were conducted on weekdays and the weekends. The purpose of this was to capture potential variation in social disorder related to the ‘pub and club’ culture. In a similar vein to the suggestion made Sampson and Raudenbush (1999), the probability of finding social disorder related to the ‘pub and club’ culture is likely to be greater on weekends, when these facilities receive their greatest use. As the focus of this project was to use GIS-based techniques, the precise location of incidents of social disorder was recorded at the time of observation. The social disorder assessments were conducted in a vehicle driven at approximately five miles per hour, except for the areas of Crown Street that had no vehicle access and the major pathway through the McCabe Park. In these areas, social disorder was assessed on foot. The combination of systematically assessing social disorder in a vehicle and on foot was suggested by Stephens (1999). The types of social disorder assessed in this project are shown below in Table 6.3.
110 Table 6.3 Types of social disorder recorded in the physical disorder assessment (after Sampson and Raudenbush, 1999)
6 Disorder number 1 2 3 4 5 6 7
The Wollongong Study
Type of social disorder Noise Truancy Loitering Public insults Prostitution Panhandling Adults fighting or arguing in a hostile manner Public drinking/public drunkenness Loud parties Street harassment of women Street harassment of elderly Drug dealing Homeless or mentally ill people Public urination
8 9 10 11 12 13 14
A number of authors describe certain venues that act as generators of social disorder, such as bars, clubs and pornographic theatres (e.g. Skogan, 1990; Wikstrom, 1995). The distribution of these types of venues was recorded for the CBD area of Wollongong and a hotspot map was also created using the same procedures as for social and physical disorder. The types of venues associated with social disorder that were recorded are shown below in Table 6.4. Spatial Analysis of Crime Data A map showing the distribution of general crime hotspots (i.e. based upon all types of recorded crime) was produced from geocoded crime data collected by the NSW Police Service for the Wollongong Local Area Command. The geocoding accuracy associated with the crime data was 55%, meaning recorded crime offences could be matched to an address and given a geographic coordinate for 55% of the data. Typically, the geocoding process delivers results where 25–75% of the target database records can be matched and given a geographic coordinate (Drummond, 1995). The database consisted of approximately 65,000 crimes which occurred
Table 6.4 Types of venues recorded that are associated with social disorder
Disorder number
Type of social disorder
1 2 3 4 5 6 7
Bars Bottle shops Night clubs Pornographic theatres Massage parlours Adult shops Methadone dispensaries
Research Setting
111
between March 1998 and August 2002. The hotspots of physical disorder and crime were mapped using the kernel density based function within the Crime Analysis Extension of ArcView 3.2 using search radii of 50 and 100 metres respectively. There are few guidelines regarding the selection of search radii for kernel density functions (Levine, 2002). In general, narrower search radii deliver results showing a finer mesh density estimate with well-defined ‘peaks’ and ‘valleys’. A larger bandwidth will lead to a smoother distribution and, therefore, less variability between areas (Levine, 2002). The selection of search radii of 50 metres for the creation of social and physical hotspot maps in this study was based on the likely impact of disorder on fear of crime and general visibility conditions in the CBD area of Wollongong. In most cases, visibility in the CBD of Wollongong is fairly restricted, with sight lines rarely extending beyond 50–80 metres (see Figs. 6.10 and 6.11 below). It was assumed that signs of disorder would be relatively difficult to detect beyond 50 meters in most areas. Hence, search radii of 50 meters were likely to capture a scale of detail relevant to people using public space in the CBD area. The selection of a broader search radius of 100 meters for the creation of the crime hotspot map was based on the desire to obtain a more generalized indication of the
Fig. 6.10 Typical view in the Crown Street Mall area, with sight lines of approximately 50 metres
112
6
The Wollongong Study
Fig. 6.11 Typical view looking west at the junction of Crown and Keira streets, with sight lines of approximately 80 metres
distribution of crime but to retain a scale of detail that was suitable for comparison with the maps outlining collective avoidance behaviour and social and physical disorder. Combinatory Spatial Analysis: Framework for a Spatiotemporal Comparison of Collective Avoidance Concentrations, Social and Physical Disorder and Crime In the context of disorder decline models such as the broken windows theory, the degree to which collective avoidance concentrations overlap crime or disorder hotspots is of interest. According to the theory, there is the potential for crime or disorder to expand into areas of poor natural surveillance over time (Kelling and Coles, 1997). However, there have been few, if any, spatiotemporal examinations of the links suggested between fear of crime, disorder and crime itself. The figures below provide an interpretive framework for the spatiotemporal links suggested by the broken windows theory. Figures 6.12a, b show the presence of a crime or disorder hotspot and the subsequent collective avoidance of the hotspot and the areas surrounding it. According to the broken windows theory, the low level of natural
Research Setting
113
a)
b)
c)
d)
Direction of expansion for crime or disorder hotspot
Direction of expansion for collective avoidance concentration Area representing crime or disorder hotspot
Fig. 6.12 Diagrammatic representation of the potential expansion of a crime or disorder hotspot in relation to a collective avoidance concentration
surveillance resulting from the collective avoidance in the areas adjacent to the crime or disorder hotspot creates the opportunity for the crime or disorder hotspot to expand (Fig. 6.12b) and occupy a larger area (Fig. 6.12c). The logical public response to this would be an expansion of the collective avoidance around the now larger crime or disorder hotspot (Fig. 6.12d). Figures 6.13a–d illustrate a situation where a collective concentration envelopes two crime or disorder hotspots. In this case, the direction of expansion of the crime or disorder hotspots into the areas of poor natural surveillance is towards each other (Fig. 6.13b). Over time, the crime or disorder hotspots could potentially join to create one continuous hotspot. As with the situation in Fig. 6.13d, the area of collective avoidance in this situation would expand in relation to a larger crime or disorder hotspot. Figures 6.14a–d illustrate another likely scenario where there is a partial overlap of a collective avoidance concentration and a crime or disorder hotspot. A situation such as this could exist where a social or physical barrier exists on one side of the crime or disorder hotspot. In this case, the expansion into the area of low natural surveillance creates a larger crime or disorder hotspot in one direction (Fig. 6.14b–c). In turn, this could result in a larger collective avoidance concentration in a direction away from the growth of the crime or disorder hotspot. The following process was used in order to examine the degree of overlap between collective avoidance concentrations and hotspots of disorder and crime. First, the maps for collective avoidance concentrations at different times (9 am– 5:30 pm, 5:30–7 pm, after 7 pm) were combined to create a generalized map of collective avoidance. In turn, this was then broken into two discrete classes, one
114
6 a)
b)
c)
d)
The Wollongong Study
Direction of expansion for crime or disorder hotspot
Direction of expansion for collective avoidance concentration Area representing crime or disorder hotspot
Fig. 6.13 Diagrammatic representation of the potential expansion and eventual linking of two crime or disorder hotspots in relation to a collective avoidance concentration
a)
c)
b)
d)
Direction of expansion for crime or disorder hotspot
Direction of expansion for collective avoidance concentration
Area representing crime or disorder hotspot
Fig. 6.14 Diagrammatic representation of the potential expansion of a crime or disorder hotspot in relation to a partially overlapping collective avoidance concentration
Research Setting
115
indicating the areas most strongly avoided, the other indicating areas that were not strongly avoided. This generalized output then formed the basis with which to compare generalized collective avoidance behaviour to the other elements of the broken windows theory, namely social and physical disorder and crime. The map showing the collective distribution of the strongly avoided areas was overlaid with separate maps showing the distribution of social and physical disorder and crime. In the case of social disorder, generalized maps were created in a similar manner used to combine the collective avoidance maps for various times. Social disorder maps were combined to create two general maps, one showing the distribution of social disorder on weekdays and another on weekends. This involved combining separate social disorder maps for the various times (6 am–12 noon, 12 noon–6 pm, 6 pm–12 midnight, 12 midnight–6 am) for weekdays and weekends. The display of the maps was designed to highlight the degree to which the generalized collective avoidance concentrations overlapped the separate maps showing the distribution of social and physical disorder and crime. This was achieved by making the map of the generalized collective avoidance concentrations semi-transparent.
Results Sample Characteristics Figure 6.15 shows the age distribution of survey respondents. It can be seen that the age categories for 18–23-year-olds and 30–35-year-olds are the best represented
40
Frequency
30
20
10
0 18−23 24−29 30−35 36−41 42−47 48−53 54−59 60−65 above66
Age Category
Fig. 6.15 Age distribution of respondents
116
6
The Wollongong Study
with 40 and 43 respondents or 17.1% and 18.4% of the sample. For the age categories between 30–35 years to 36–41 years, the number of respondents was 36 and 35, representing 30.3% of the sample. Approximately 23.9% of the sample fell into the age categories of 42–47 years and 48–53 years; 10.3% of the sample fell into the age categories for 54–59 years, 60–65 years and above 66 years. In terms of the sex distribution of survey respondents, the majority of the sample were females (n = 169, 72.2%) and a minority males (n = 65, 27.8% of the sample). This bias towards women is similar to other studies using voluntary samples. For example, Nasar and Jones (1997) conducted a voluntary sample to assess fear on a night-time walk at the Ohio State University campus. In their study, 26 females were surveyed. In a study by Nair et al. (1993) three-quarters of the respondents were women. The majority of the sample (n = 201, 85.9%) were from English-speaking backgrounds. A smaller number (n = 33, 14.1%) were from non-English-speaking backgrounds. Figure 6.16 shows the income distribution of respondents. It can be seen that the majority of the sample (n = 156, 66.6%) indicated that they would find it very easy or easy to access $600 suddenly without access to a bank loan; 17.9% of the sample indicated they may be able to access $600 without access to a bank loan, while 36 respondents (15.4% of the sample) indicated that this was not easy or impossible (Fig. 6.16). With respect to the housing type for the survey respondents, most of the sample (n = 149, 63.7%) were owner-occupiers. A smaller number (n = 79, 33.8%) were non-owner-occupiers while relatively few (n = 6, 2.6%) were from government housing commissions. The majority of the sample (n = 121, 51.7%) had been
100
Frequency
80
60
40
20
0 1 2 3 4 5 1 = Very Easy, 2 = Easy, 3 = Maybe, 4 = Not Easily, 5 = Impossible
Fig. 6.16 Income distribution of respondents
Research Setting
117
100
Frequency
80
60
40
20
0 1
2
3
4
5
1 = Very Understated, 2 = Quite Understated, 3 = Accurate, 4 = Quite Overstated, 5 = Very Overstated
Fig. 6.17 Media perception of crime-related issues in Wollongong among respondents
living in their current neighbourhood for over five years. Similar proportions (16.7% n = 41 and 17.5%, n = 39) of the sample had been living in their current neighbourhood for 3–5 years or 1–2 years respectively; 13.6% (n = 32) had been living in their current neighbourhood for under one year. Figure 6.17 shows the media perception of crime-related issues in Wollongong among survey respondents. Approximately 64.5% (n = 151) of the sample felt that crime-related issues in Wollongong were very understated or quite understated; 26.0% (n = 62) of the sample felt that crime-related issues in Wollongong were accurately represented, while 8.9% (n = 21) of the sample felt that crime-related issues were quite overstated. None of the respondents thought that crime-related issues were overstated. Figure 6.18 shows the experience of victimization among respondents in the 12 months prior to the time of interview: 40.2% (n = 94) of the sample had not experienced any type of crime. The proportions of the sample that had experienced the crimes of deliberate use of a weapon, attack or assault and threats of force or violence were 0.4% (n = 1), 1.7% (n = 4) and 4.7% (n = 11) respectively. Compared to the sample of Borooah and Carcach (1997), from which the measures of victimization used in this study were taken, the experience of crimes against the person is lower in this sample (8.8%). The proportion of respondents in the study by Borooah and Carcach (1997) that had been victims of a personal crime was 14%. Higher proportions of the sample had experienced theft or attempted theft 17.1% (n = 40) and vandalism 12.4% (n = 29). Approximately 23.5% (n = 55) of the sample had experienced more than one crime in the 12 months prior to the time of survey.
118
6
The Wollongong Study
100
Frequency
80
60
40
20
0 0 1 2 3 4 5 0 = none, 1 = Deliberate use of a weapon, 2 = attack or assault,
6
3 = Threats of force or violence, 4 = Theft or attempted theft, 5 = Vandalism, 6 = more than one crime
Fig. 6.18 Experience of victimization among respondents
Figure 6.19 shows the responses of the survey respondents to the global fear-ofcrime measure used in the survey (i.e. ‘How safe do you feel when walking along after dark in the area around your home?’). It can be seen that 54.7% (n = 128) of the sample indicated a degree of fear in this situation (not very safe or not safe at all); 45.3% (n = 106) of the sample indicated that they felt fairly safe or very safe. Table 6.5 below shows the responses of the sample to the vignettes of Van der Wurff et al. (1989). In parentheses are the responses to the same vignettes from the studies by Van der Wurff et al. (1989) and Farrall et al. (2000). Farrall et al. (2000) used the vignettes to make a general comparison between their sample and that of Van der Wurff et al. (1989) on the basis that differences in answers would reflect variations in sensitivities. Farrall et al. (2000) concluded that their sample was slightly more fearful, as it showed higher levels of fear to most vignettes was slightly more fearful. Using this logic, the sample in this study is slightly more fearful than the samples of Van der Wurff et al. (1989) and Farrall et al. (2000). In response to the question regarding the reporting of spray-painting offences in their neighbourhood (i.e. ‘Suppose some kids were spray painting a building near where you work. Do you think you or any of your neighbours would call the police?’), most of the sample (83.8%, n = 196) indicated that they would report spray painting in their neighbourhood to the police, while 16.2% (n = 38) said they wouldn’t. Fig. 6.20 shows the responses of the sample to the question regarding how well the police were thought to be performing their jobs. It can be seen that a relatively small proportion of the sample 2.6% (n = 6) felt that the police were performing their jobs very well; 42% (n = 98) of the sample felt that the police were
Research Setting
119
Frequency
90
70
50
30
10 1 2 3 4 1 = Very Safe, 2 = Fairly Safe, 4 = Not Very Safe, 5 = Not Safe at all
Fig. 6.19 Answers to global measure of fear amongst respondents
Table 6.5 Degree of safety in relation to the vignettes of Van der Wurff et al. (normal parentheses represent the sample of Farrall et al. (2000), square parentheses, the sample of Van der Wurff et al. (1989)) Vignette
Meana
Standard deviation
Modal answer
Doorbell Car Party Bus stop Telephone
3.00 (3.05) [2.35] 1.95 (2.53) [3.27] 1.38 (1.73) [3.77] 2.06 (2.53) [2.30] 3.06 (3.18) [1.96]
1.24 (1.28) [1.29] 0.95 (1.15) [1.23] 0.66 (0.86) [1.08] 1.00 (1.14) [1.14] 1.18 (1.19) [1.32]
4.00 not very afraid 2.00 quite afraid 1.00 very afraid 2.00 quite afraid 4.00 not very afraid
a Note:
Values are based on the Likert scale where 1 = very afraid to 5 = not afraid at all
performing their jobs quite well, 30.3% (n = 98) indicated that they did not know; 22.6% (n = 53) felt the police were not performing their jobs well and 2.6% (n = 6) the police were not performing their jobs well at all.
The Spatiotemporal Distribution of Collective Avoidance Concentrations Table 6.6 below shows the percentages of the sample that were adopting avoidance behaviour during the day and after dark for situations relating to their neighbourhood as well as the CBD area. It can be seen that the percentage of the sample adopting avoidance behaviour in the CBD area after dark is greater than for the neighbourhoods of respondents after dark (81.2% compared to 64.1%). It can also
120
6
The Wollongong Study
100
Frequency
80
60
40
20
0 1 2 3 4 5 1 = Very Well, 2 = Quite Well, 3 = Don't know, 4 = Not Very Well, 5 = Not Well at all
Fig. 6.20 Responses to the question regarding how well the police are thought to be performing their jobs Table 6.6 Percentages of respondents adopting avoidance behaviour in their neighbourhoods and the CBD area
Area
Percentage of respondents adopting avoidance behaviour during the day
Percentage of respondents adopting avoidance behaviour at night
CBD Neighbourhood
39.31 18.38
81.20 64.10
be seen that in both the CBD area and the neighbourhoods of respondents the percentage of the sample adopting avoidance behaviour increases substantially after dark. GIS-based analysis of avoidance behaviour in the neighbourhoods is not presented, as the sample was spatially too dispersed to adequately assess collective avoidance in the neighbourhood context. Figures 6.21, 6.22 and 6.23 show the collective avoidance hotspots for different times of the day for the sample of working people from the CBD of Wollongong. In general it can be seen that the collective avoidance hotspots are well defined in that the areas avoided are relatively specific. The distribution also changes noticeably over time. Between 09:00 and 17:30 (Fig. 6.21) there are two major hotspots, one centred around the McCabe Park area and another towards the western side of Crown Street. There are two smaller hotspots further down Crown Street, in what is a mall area. Between 17:30 and 19:00 the hotspots expand considerably. The most noticeable changes are that the hotspot in the west Crown Street area extends east to
Research Setting
121
250
Keira S
treet
Meters
Crown Street
McCabe Park
Burelli S
treet
Legend Minor Roads Main Roads
Degree of Avoidance Low Intensity
High Intensity
Fig. 6.21 Areas of the CBD avoided between 09:00 and 17:30 in relation to fear of crime
nearly join up with a large hotspot that has developed in the mall area. After 19:00, the hotspots recede to the areas around McCabe Park, west Crown Street and the mall area. The hotspots at this time are generally smaller, except for the hotspot in the mall area. Between 17:30 and 19:00 (Fig. 6.22) the collective avoidance concentration centred around the Piccadilly complex in west Crown Street extends eastwards to occupy most of Crown Street to effectively link up with a collective avoidance concentration that has formed in the Crown Mall area, which extends from the junction of Crown and Keira Streets to the junction of Kembla and Crown Streets. A smaller collective avoidance concentration has also formed around Globe Lane within the Crown Street Mall complex. The collective avoidance concentration centred around the McCabe Park area has grown in area to occupy most of the park and part of Burelli Street to the north. After 19:00 (Fig. 6.23), the collective avoidance concentrations have retreated in extent compared to the concentrations between 17:30 and 19:00. The concentrations centred around McCabe Park and Piccadilly areas occupy smaller areas than at the other times (Figs. 6.21 and 6.22). However, the collective avoidance concentration in the Crown Street Mall area has expanded, particularly around the junction of Market and Crown Streets. The extent of this concentration covers most of the paved areas in the Crown Mall that have no vehicle access. The collective avoidance concentration
122
6
The Wollongong Study
250
Keira S tr
eet
Meters
Crown Street
McCabe Park
Burelli S
treet
Legend Minor Roads Main Roads
Degree of Avoidance Low Intensity
High Intensity
Fig. 6.22 Areas of the CBD avoided between 17:30 and 19:00 in relation to fear of crime
centred around Globe Lane that is apparent between 17:30 and 19:00, is no longer present after 19:00. The Spatiotemporal Distribution of Physical and Social Disorder and Crime Figure 6.24 below shows the distribution of physical disorder hotspots within the CBD of Wollongong using the weighted locational data. It can be seen that the main hotspots are located along the west and middle Crown Street areas, the McCabe Park area and one hotspot on Keira Street. In general, the hotspots of physical disorder are smaller and more spatially confined than the collective avoidance hotspots shown in Figs. 6.21, 6.22 and 6.23. Figure 6.25 shows the ranking of different types of disorder after the weighting system was applied. It can be seen that tagging graffiti, garbage or litter and empty beer bottles in the street were the most highly ranked types of disorder. Graffiti was ranked the highest, showing it to be the most dominant type of physical disorder within the disorder hotspots shown in Fig. 6.24. Figures 6.26, 6.27, 6.28 and 6.29 show the spatial distribution of social disorder for different times on weekdays and weekends. Figure 6.26 shows concentrations of social disorder on weekdays during the day (i.e. between 06:00 and 18:00). It can be seen that there are far fewer hotspots of social disorder than there are for
Research Setting
123
250
Keira S
treet
Meters
Crown Street
McCabe Park
Burelli S
treet
Legend Minor Roads Main Roads
Degree of Avoidance Low Intensity
High Intensity
Fig. 6.23 Areas of the CBD avoided after 19:00 in relation to fear of crime
physical disorder (Fig. 6.24). On weekdays between these times, concentrations of social disorder are mainly around the northern end of McCabe Park and nearby on Burelli Street. An intense concentration is also evident along Denison Street near the junction of Crown Street. Other, less intense hotspots are distributed 100–200 metres to the north and south of Crown Street. Figure 6.27 shows the distribution of social disorder hotspots on weekdays at night (i.e. between 18:00 and 06:00). It can be seen that social disorder hotspots are dispersed widely across the CBD area between these times. Only one intense hotspot is evident along Keira Street, past the junction of Victoria Street. A cluster of less intense hotspots are concentrated along Crown Street between Kembla and Corrimal Streets. Also apparent is the lack of social disorder hotspots around the McCabe Park and Piccadilly areas between these times. Figures 6.28 and 6.29 show the spatial distribution of social disorder hotspots for weekends during the day (i.e. between 06:00 and 18:00) and at night (i.e. between 18:00 and 06:00). It can be seen that during the day that a cluster of social disorder hotspots are concentrated around the western end of the Crown Street Mall area between Market and Keira Streets. The cluster of hotspots extends along Globe Lane and east along Burelli Streets. A hotspot of medium intensity is located around the Piccadilly area between these times. Between these times, on the weekend, there are fewer hotspots in the minor streets surrounding Crown Street than there are
124
6
Meters
Corrima l Street
Keira S treet
250
The Wollongong Study
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Core avoidance hotspot Intensity of Physical Disorder Hotspot Low Intensity
High Intensity
1800 1500 1200 900 600
Abandoned cars/glass from … Evidence of homeless people
Graffiti painted over
0
Garbage or litter in street Empty beer bottles visible in street Cigarettes or cigars in street gutter Abandoned/boarded up houses Lack of exterior maintenance Vandalism to buildings Vandalism to public structures
300 Tagging graffiti
Weighting multiplied by level
Fig. 6.24 Physical disorder hotspots, based upon weighted data
Fig. 6.25 Ranking of different types of disorder recorded in the physical disorder assessment
125
Meters
Corrima
250
l Street
Keira S treet
Research Setting
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.26 Social disorder on weekdays during the day (i.e. between 6 am and 6 pm)
during the day on weekdays (Fig. 6.26). Figure 6.29 shows the distribution of social disorder hotspots on weekends at night (i.e. between 18:00 and 06:00). It can be seen that intense hotspots are located within the Crown Street Mall area and along Keira Street near the junction with Market Street. Smaller clusters of hotspots are centred around the Piccadilly area and the eastern end of Crown Street near the junction of Corrimal Street. A number of low-intensity hotspots are spread around the minor streets to the east and west of Keira Street and along Crown Street. Figures 6.30 and 6.31 show generalized (i.e. grouped) maps of social disorder on weekdays and weekends. These maps were a composite of social disorder for weekdays and weekends for day and night (i.e. all times). In general it is evident that social disorder hotspots on weekdays are more dispersed than on weekends. A noticeable difference is the lack of social disorder in the Crown Street Mall area during weekdays, whereas on weekends this area shows the greatest concentration of social disorder. On weekdays, hotspots of social disorder are evident in the Piccadilly area, the McCabe Park area, along Burelli Street towards Corrimal Street. Another cluster of hotspots stretches northwards along Keira Street. On weekends, social disorder hotspots are concentrated in the Crown Street Mall complex, around the junction of Keira and Market Streets and around the Piccadilly area. Figures 6.32 and 6.33 show the frequency of the different types of social disorder recorded during the assessments for weekdays and weekends respectively. It can be
Keira S treet
250 Meters
Crown Street
McCabe Park
The Wollongong Study
Street
6
Corrima l
126
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.27 Social disorder on weekdays at night (i.e. between 6 pm and 6 am)
Corrima
Keira S
Meters
l Street
treet
250
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.28 Social disorder on weekends during the day (i.e. between 6 am and 6 pm)
Research Setting
127
Corrima l
Meters
Street
Keira S treet
250
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.29 Social disorder on weekends at night (i.e. between 6 pm and 6 am)
Corrima
Keira S
Meters
l Street
treet
250
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.30 General hotspots of social disorder on weekdays (day and night grouped)
128
6
Corrima l
Meters
Street
Keira S treet
250
The Wollongong Study
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.31 General hotspots of social disorder on weekends (day and night grouped)
seen that on weekdays the most frequent type of social disorder recorded were loitering and noise (23 and 12 observations respectively). Public drinking, public insults and homeless people were recorded at low frequencies. On weekends (Fig. 6.33) it can be seen that, in general, more types of disorder were recorded and the frequencies were higher. Loitering, public drinking and noise were the most frequently recorded types of disorder (38, 22 and 14 observations respectively). On weekends, low frequencies (between 1 and 8 observations) were recorded for public insults, homelessness, adults fighting or arguing in public, street harassment of women and public urination. For some types of disorder, namely truancy, prostitution, panhandling, loud parties, street harassment of elderly and drug dealing, no observations were recorded on weekdays or weekends. Figure 6.34 shows the hotspots of clubs, bars, adult entertainment stores and other venues frequently associated with generating social disorder. It can be seen that the most evident concentration is along Crown Street between Keira Street and the rail line. Another concentration around the junction of Market and Keira Streets is also evident. A cluster of smaller hotspots is located near the junction of Corrimal and Crown Streets. The frequency of the different types of venues is shown in Fig. 6.35. Figure 6.36 below shows the general crime hotspot map for Wollongong between March 1998 and August 2002. It can be seen that there are several distinct hotspots within the CBD area, the largest of which is located in the eastern end of Crown
Research Setting
129
25
Frequency
20 15 10 5
Public urination
Drug dealing
Street harassment of elderly
Street harassment of women
Loud parties
Adults fighting or arguing in a hostile manner
Panhandling
Prostitution
Truancy
Homeless or mentally ill people
Public insults
Public drinking/ public drunkenness
Noise
Loitering
0
Fig. 6.32 Types of social disorder on weekdays
Street. Others are located on Keira Street between Burelli and Crown Streets, the northern part of Keira Street and in the western area of Crown Street. The Degree of Overlap Between Collective Avoidance Concentrations, Physical and Social Disorder and Crime The following figures show the overlap between general areas of collective avoidance and the primary elements of the broken windows theory, namely physical disorder, social disorder and crime itself. The general levels of avoidance were created by combining the avoidance grids for the various times (i.e. between 09:00 and 17:30, between 17:30 and 19:00 and after 19:00). The areas that were most heavily avoided were then selected and used to create a grid representing general avoidance. Figure 6.37 below shows the degree of overlap between general areas of avoidance and crime hotspots. It can be seen that there is a strong degree of overlap between the crime hotspot centred around the Piccadilly area and the general collective avoidance concentration. There is partial overlapping of collective avoidance concentrations and crime hotspots along the northern fringe of McCabe Park. In the Crown Street Mall area, there is partial overlapping with crime hotspots at the junctions of Crown and Keira streets and Crown and Kembla streets.
130
6
The Wollongong Study
40
Frequency
30
20
Drug dealing
Street harassment of elderly
Loud parties
Panhandling
Prostitution
Truancy
Public urination
Street harassment of women
Homeless or mentally ill people
Adults fighting or arguing in a hostile manner
Public insults
Noise
Public drinking/ public drunkenness
0
Loitering
10
Fig. 6.33 Types of social disorder on weekends
Figure 6.38 shows the degree of overlap between weighted physical disorder hotspots and general areas of collective avoidance. There is a strong degree of overlap between general areas of collective avoidance and the intense physical disorder hotspots around the Piccadilly area and east along Crown Street near the junction with Auburn Street. Also apparent near the Piccadilly area is the partial overlapping of the general collective avoidance area and two intense hotspots located along Gladstone Avenue and along Crown Street, approximately 100 m west of the junction with Gladstone Avenue. In the McCabe Park area there is a strong degree of overlap between the general area of collective avoidance and the cluster of physical disorder hotspots clustered along the western and central parts of the park. There is only slight overlapping of general collective avoidance of the Crown Mall area and physical disorder hotspots. Figure 6.39 shows the degree of overlap between general areas of collective avoidance and social disorder hotspots for weekdays. It can be seen that there is a strong degree of overlap between general areas of collective avoidance and the intense social disorder hotspot located near the Piccadilly area at the junction of Denison Street and Crown Street. There is also a strong degree of overlap between the general collective avoidance concentration and social disorder in the McCabe
Research Setting
131
Meters
Corrima l Street
Keira S treet
250
Crown Street
Burelli S
treet
McCabe Park
Legend Coastline Main Roads Minor Roads Rail Line Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.34 Social disorder hotspots for bars, clubs and adult entertainment stores
10 Number
8 6 4
Night clubs
Bottleshops
Pornographic theatres
Methadone dispensaries
Adult shops
Massage parlors
0
Bars
2
1
2
3
4
5
6
7
Fig. 6.35 The number of clubs, bars and nightclubs in the CBD area
132
6
Meters
Corrima l Street
Keira S treet
250
The Wollongong Study
CrownStreet
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Intensity of Crime Hotspot Low Intensity
High Intensity
Fig. 6.36 General crime hotspots within the CBD of Wollongong
Park area. There is relatively little overlap between the general collective avoidance concentration and social disorder on weekdays in the Crown Mall area. Figure 6.40 shows the degree of overlap between general areas of collective avoidance and hotspots of social disorder on weekends. As with the situation on weekdays, there is a strong degree of overlap between the general collective avoidance concentration and the intense social disorder hotspot in the Piccadilly area. In the Crown Street Mall area, there is a strong degree of overlap between general areas of collective avoidance and the cluster of intense social disorder hotspots. There is only a partial degree of overlap between the general collective avoidance concentration and social disorder around the McCabe Park area. Figures 6.39 and 6.40 show that there is no overlapping of the social disorder hotspots concentrated around the junction of Keira and Market Streets and the general collective avoidance concentration. Table 6.7 below summarizes the degree of overlap between the general collective avoidance concentration and hotspots of crime, social disorder on weekdays, social disorder on weekends and weighted physical disorder.
Research Setting
133
Meters
Corrima l Street
Keira S treet
250
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Core avoidance hotspot Intensity of Crime Hotspot Low Intensity
High Intensity
Fig. 6.37 The degree of overlap between general areas of avoidance and crime hotspots
Discussion of Spatial Outputs Potential Constraints on Social Interaction Resulting from Collective Avoidance Behaviour The use of cognitive mapping to investigate avoidance behaviour, and the subsequent GIS-based analysis, provides new insights into some of the issues that have been central to debates on fear of crime. One issue which has become an increasingly important aspect of such debates is the degree to which fear of crime impedes people’s freedom of movement (Pantazis, 2000). Liska et al. (1988) suggested that, to some extent, fear of crime is a social problem because it is assumed to constrain social interaction. The maps of collective avoidance for the CBD of Wollongong (Figs. 6.21, 6.22 and 6.23) show how fear of crime is likely to be constraining social interaction among the working population in the city. The implications of reduced social interaction in each of the key collective avoidance areas identified in Table 6.7 are discussed below. The Piccadilly Area In the Piccadilly area, the constant avoidance of the shopping complex and surrounding vicinity, at all times of the day, suggests that the working population in
134
6
Meters
Corrima l Street
Keira S treet
250
The Wollongong Study
Crown Street
McCabe Park
Burelli S
treetx
Legend Coastline Main Roads Minor Roads Rail Line Core avoidance hotspot Intensity of Physical Disorder Hotspot Low Intensity
High Intensity
Fig. 6.38 The degree of overlap between general areas of avoidance and weighted physical disorder hotspots
the CBD is effectively divided between people east and west of the rail line. The barriers to pedestrian movement along the rail line leave only one main pathway between west Crown Street and east Crown Street (Olsen, 2003). This pathway is the passage of Crown Street over the rail line. The strong avoidance of the Piccadilly area indicates that the majority of people in the CBD are reluctant to use this main pathway along Crown Street. This has a range of implications, the most apparent being the loss of potential social interaction between the working populations east and west of the rail line. Further, people west of the rail line, by being reluctant to utilize the main pathway along Crown Street are restricted in their ability to access the wider range of facilities and services available in the Crown Central Mall area. In this sense, the collective avoidance concentration in the Piccadilly area is a social barrier likely to reduce cohesion within the CBD community. This supports the findings of (Markowitz et al., 2001) who found fear to reduce cohesion on a broader scale. Gibson et al. (2002) suggest that social integration and community cohesion are important factors in terms of stabilizing or improving neighbourhood conditions. Thus, if the collective avoidance of the Piccadilly area continues, it is unlikely that conditions will improve and may in fact deteriorate over time. The actual centre of collective avoidance, the Piccadilly shopping complex and surrounding streets, is highly likely to be experiencing a substantial loss of potential customers for the businesses located in the area. This supports, and provides
Research Setting
135
Corrima
Keira S tr
Meters
l Street
eet
250
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Core avoidance hotspot Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.39 The degree of overlap between general areas of avoidance and social disorder hotspots for weekdays
concrete visual evidence for, arguments that the avoidance behaviours prompted by fear of crime must inevitably have an economic cost because people avoiding an area remove themselves as consumers (Oc and Tiesdell, 1997; Warr, 2000). The consistent avoidance of the area may also be creating greater opportunities for crime and disorder. Nodes of transport, such as rail stations, are often centres for certain types of disorder and crime (e.g. Loukaitou-Sideris, 1999). The close proximity of the Piccadilly centre to the rail station, in combination with poor natural surveillance associated with collective avoidance behaviour is likely to be providing favourable conditions for loitering, drug dealing and for prospective break-and-enter offenders to examine the area at their leisure during daylight hours. In terms of a policing response, the Piccadilly area is one where collective avoidance, social and physical disorder and crime itself show a strong degree of overlap. As such, a direct involvement of police in this area to control elements of disorder is appropriate. The McCabe Park Area The collective avoidance concentration in the McCabe Park area is similar to the one centred around the Piccadilly complex, in that it shows a constant avoidance of the area throughout the day by the public. Logically, the park should be used by
136
6
Meters
Corrima l Street
Keira S treet
250
The Wollongong Study
Crown Street
McCabe Park
Burelli S
treet
Legend Coastline Main Roads Minor Roads Rail Line Core avoidance hotspot Intensity of Social Disorder Hotspot Low Intensity
High Intensity
Fig. 6.40 The degree of overlap between general areas of avoidance and social disorder hotspots for weekends Table 6.7 Degree of overlap between different types of disorder, crime and collective avoidance concentrations at the three main centres of collective avoidance Degree of overlap with elements of disorder and crime Location
PDA
SDA1
SDA2
Crime
Piccadilly Mall McCabe Park
Strong Weak Strong
Strong Weak Strong
Strong Strong Weak
Strong Medium Medium
PDA physical disorder assessment, SDA1 social disorder assessment on weekdays, SDA2 social disorder assessment on weekends
people working in and around the Crown Street Mall area during lunch- and workbreak times, as it is the closest park to the mall area. The strong avoidance of the McCabe Park suggests that this is currently not happening. The amenities within the park designed for public use – seating areas, playgrounds and memorial sites – are likely to be underutilized. The social disorder assessments showed a consistent presence of public drinking and loitering in the park during the week. As with the Piccadilly area, the poor natural surveillance in the park may be creating favourable conditions for the types of disorder currently present, as well as serving to reinforce the avoidance of the area by the general public.
Research Setting
137
At the conclusion of the Wollongong study, the McCabe Park area was earmarked for redesign, in light of its current lack of use by the public (WCC, 2003; Irwin et al., 2003). The suggested changes to the park revolved around building on existing features in the park, improving lighting and the creation of seating areas that were intimate in nature. Changes to the park edge focus on strengthening the transition from street to park (Irwin et al., 2003: 74). In light of the collective avoidance of the park by the public, these suggested landscape design initiatives were inappropriate. The creation of seating areas of a more intimate nature would serve to reduce the potential for natural surveillance and as such, would be likely to discourage, rather than encourage, greater public use of the park. Further, the creation of secluded seating areas would be likely to create greater opportunities for the types of social disorder present in the park, namely loitering and public drinking. The suggested improvements to lighting may have results similar to the fear-reduction strategy described by Nair et al. (1993) in Glasgow, Scotland, where the authors concluded that the changes in lighting had simply turned a poorly lit bad area into a well-lit bad area. Some of the changes proposed to streets surrounding McCabe Park also seemed unlikely to prove beneficial in terms of reducing collective avoidance of the park area. For example, Irwin et al. (2003) suggested relocating current parking facilities in the median strip of Church Street where it abuts McCabe Park, to some of the residential streets joining Church Street such as Ellen and Bank Streets. Irwin et al. (2003) proposed that, following the relocation of parking, the median strip on Church Street be fully replanted with vegetation. While this was designed to build on the boulevard nature of Church Street where it abuts McCabe Park, it would most likely enhance the secluded nature of the park area and reduce the potential for natural surveillance from one side of the street to the other and into the park itself. The relocation of parking areas to Ellen and Bank Streets would also have had additional impact in terms of pedestrian activity. At the time of the Wollongong study, Church Street provided one of the main parking areas in the CBD. The daily movement of commuters who parked along Church Street between their work area and where they parked their cars gave rise to pedestrian movement, and hence natural surveillance around the park area before and immediately after work hours. The collective avoidance of the park by the public suggests that if an alternative route to the parking area were provided, it would be more heavily utilized. The relocation of parking facilities to Ellen and Bank Streets could result in commuters moving along Kembla Street when moving between their vehicles and their work areas, instead of along Church Street. This in turn would reduce pedestrian activity along Church Street and the resulting natural surveillance around the park. According to the broken windows theory and the logic presented in Figs. 6.12, 6.13 and 6.14, lower natural surveillance along the eastern edge of McCabe Park could create greater opportunities for social disorder currently persisting in the centre of the park and the eventual expansion of the collective avoidance concentration. The possible negative consequences of inappropriate landscape design initiatives in the McCabe Park area could have longer-term implications as well. The structure plan for the Wollongong city centre (WCC, 2003) advocates strategies that build on the current trend in Australian cities towards residential development on the fringe of core city areas. The structure plan specifically suggests a long-term increase in
138
6
The Wollongong Study
residential densities around and to the east of McCabe Park. If the shorter-term landscape design strategies in and around McCabe Park served to increase social disorder and collective avoidance as suggested above, they could jeopardize the success of the longer-term residential development proposals. Investors are less likely to involve themselves in an area characterized by social and physical disorder (Skogan, 1990). This places added emphasis on the selection of an appropriate short-term land use design strategy for the park. These observations, and others outlined for the Piccadilly and Crown Street Mall area, were presented at a number of workshops and strategic meetings that were part of the Wollongong City Council’s City Centre Revitalisation Plan (WCC, 2003). The concluding section of this chapter discusses how the core findings from the Wollongong study were integrated with a Crime Prevention and Community Safety Plan initiated in 2007 (WCC, 2007). The Crown Street Mall Area The collective avoidance of the Crown Street Mall area is different from that centred around the McCabe Park and Piccadilly areas. The collective avoidance concentration around the Crown Street Mall shows greater spatial and temporal variation. Evident in the collective avoidance concentration in this area between 17:30 and 19:00 is the fact that most people are avoiding the open walkways in the mall area, namely Globe Lane and the paved section of Crown Street between Keira and Kembla Streets. The strong increase in collective avoidance of this area following work hours appears to be a strong spatial representation of the five o’clock flight described by Thomas and Bromley (2000). There are some senses in which the rapid departure of people from the CBD area may act to reinforce the five o’clock flight. If people leaving their work areas are confronted with a near-vacant mall area, or people departing the area quickly, it is unlikely to create conditions under which they will want to stay after work hours in the CBD area. Perhaps adding to the five o’clock flight on weekdays is the presence of social disorder that occurs in the mall area on weekends. The overlaying of generalized collective avoidance areas for weekdays showed a high degree of overlap with social disorder hotspots on weekends in the Crown Street Mall. The cause for this disorder may relate to the mall area providing a gathering point for the ‘pub and club’ culture on weekends. The hotspot map of pubs and nightclubs showed concentrations on either side of the mall along Crown Street. It is possible that the mall, being well lit and located in between the concentrations of clubs, creates a convenient gathering point for people moving between the various clubs. According to the broken windows theory, the presence of disorder prompts avoidance of such areas (Wilson and Kelling, 1982; Kelling and Coles, 1997). It may be that people working in the CBD are aware of the social disorder that is present on weekends and this may contribute to their avoiding of the area during the weekdays after work hours. In this sense, the involvement of police to control or limit the presence of social disorder in the mall area on weekends may be of importance. As with the Piccadilly and McCabe Park areas, a number of landscape design changes were also proposed for the Crown Street Mall (Irwin et al., 2003). These
Research Setting
139
changes largely centred on physical changes to structures within the mall and aimed at improving pedestrian flow. In the main, these suggestions were likely to be potentially useful in terms of increasing natural surveillance. However, it was suggested that these initiatives should be combined with social measures that aim to address the five o’clock flight and collective avoidance of the mall area after work hours. Possible relevant social measures could be for the Wollongong City Council and mall management committee to encourage activities relevant to the working population of the CBD. The types of activities that are likely to be relevant to the CBD community include open-air coffee houses and staggered closing times for shopping venues. A number of such initiatives were established in 2007 as part of the Crime Prevention and Community Safety Plan (WCC, 2007). They are discussed in more detail in the final sections of this chapter regarding police–community partnerships and fear-reduction strategies. In order to gain further context and insights into the influence of fear of crime on the daily routines and behaviour of people working in the CBD of Wollongong, an activity diary analysis was also conducted as part of the study. The next section presents the techniques used and the key results from the activity diary analysis.
Activity Diary Analysis Data Preparation Typically, the first stage involved in the analysis of diary data is the classification of activities (Golledge and Stimson, 1997). The basis of classification stems largely from the focus of the research (e.g. Kwan, 2000b; Keuleers and Wets, 2001). In this study, the primary reason for using the activity diary approach was to examine protective behaviour and emotion-based fear in relation to specific situations in the daily routines of people working in the CBD of Wollongong. The classification of activities, therefore, focused on grouping the diary data according to the time of day and the commuting nature of the sample. As with other studies (e.g. Kwan, 2000b), the classification process involved a number of assumptions. It was assumed that the general activity pattern of respondents involved leaving their home, travelling to the CBD in a vehicle (bus, car or train), walking from the point of departure from the vehicle to their work area, work-based activities and the reverse sequence to return home. In addition to this, it was assumed that respondents also engaged in a number of recreational and mandatory activities. The recreational grouping included activities such as walks, bike rides, social outings and participation in sports. The mandatory grouping included activities such as dropping off or picking up children from school, shopping, appointments with a doctor and attending university or TAFE classes. Three relatively specific groupings were made, one for work breaks (e.g. lunch or scheduled breaks) and another for travelling to the bank and at the bank itself. The groupings for activities at home, work, travelling in a vehicle, travelling on foot and recreational activities were further split in relation to standard business hours (09:00–17:00). For the home grouping, it was assumed that most respondents
140
6
The Wollongong Study
Table 6.8 The 16 categories resulting from the classification procedure applied to the activity diary data Situation
Number of observations
Home before 08:30 Home between 08:30 and 17:30 Home after 17:30
231 106 229
Work before 09:00 Work between 09:00 and 17:00 Work after 17:00
119 232 135
Travelling in vehicle before 17:00 Travelling in vehicle after 17:00
221 199
Travelling by foot before 17:00 Travelling by foot after 17:00
164 102
Travelling to bank At bank Work break
17 30 86
Mandatory activities Recreational activities before 17:30 Recreational activities after 17:30
46 41 53
would have to leave their homes by 08:30 in order to get to work by 09:00 and would not be home before 17:30. For the recreational grouping, it was assumed that in situations where respondents were engaging in recreational activities after work hours, these would generally take place after 17:30. Following the classification procedure, 16 categories were created. The categories are shown below in Table 6.8. In order not to bias the results, only one observation per respondent was used for each of the categories. In cases where there was more than one observation per respondent in a particular category, an average level of emotion-based fear was taken. For example if a respondent was at work from 09:00 to 15:00 and showed an emotional level of fear of 4 at work before 12:00 and 3 after 12:00, for the situation ‘work between 09:00 and 17:00’, the respondent was given an average emotional level of fear of 3.5. For protective behaviours, the dominant type of protective behaviour was assigned to the observation. The process of assigning one observation per respondent in a category reduced the total number of observations from 10,211 to 2012 observations. Table 6.8 shows the number of observations in each of the categories and Table 6.9 shows the different categories of protective behaviour recorded. Protective behaviour was assessed by asking respondents if they were adopting any of the measures shown in Table 6.9 while engaging in different activities listed in their diaries. In a similar vein, emotional levels of fear were assessed by asking respondents how afraid they were of being robbed beaten or attacked while undertaking the different activities they had recorded.
Research Setting
141 Table 6.9 Coding system for protective behaviours
Protective behaviour 1 = Making sure you were accompanied by a friend 2 = Carrying something to defend yourself 3 = Relying on self-defence training 4 = Having a dog with you 5 = Carrying a mobile phone to call someone if you felt in danger 6 = Other, please specify: . . . . . . . . . . . . . . . . . . . . . . . . 7 = Not doing anything in particular to protect yourself 8 = More than one protective behaviour
Emotional Levels of Fear and Protective Behaviour in Relation to Daily Routines – General Results Table 6.10 shows the average level of emotion-based fear, percentage of respondents showing a degree of fear, percentage of respondents adopting a protective behaviour and percentage of respondents adopting more than one protective behaviour for each of the situations resulting from the classification of the activity diary data. In general the average levels of emotion-based fear do not indicate a definite degree of fear (i.e. less than 3). The only situation where the average level of emotion-based fear for all respondents was lower than 3 was while they were travelling to the bank (2.71). In all situations where the activity types were further segmented by time (i.e. home, work, travelling by vehicle, travelling on foot and recreational activities), average levels of emotion-based fear showed lower values (i.e. more fearful) with later times. For example, the average level of emotion-based fear at home between 08:30 was 4.71, between 08:39 and 17:30 it was 4.60 and after 17:00 it was 4.48. The percentage of respondents showing a definite degree of fear in the different situations varies considerably. The situations where the highest percentages were recorded were travelling to the bank (58.82%), travelling on foot after 17:00 (45.10%), at work after 17:00 (23.7%) and at the bank itself (20.0%). The situations where the lowest percentages were recorded were at home before 08:30 (3.90%), at home between 08:30 and 17:30 (4.71%), at home after 17:30 (6.55%), travelling by vehicle before 17:00 (3.17%) and recreational activities after 17:30 (3.78%). As with average levels of emotion-based fear, the percentage of respondents showing a degree of fear increases with time of day for the situations segmented by time. The exception to this pattern is for recreational activities after 17:30, where the percentage of respondents showing a degree of fear was lower than for recreational activities before 17:30 (i.e. 3.78% compared to 12.20%). The percentage of respondents adopting a protective behaviour in the different situations shows less variation than the percentage of respondents showing a degree of fear. In general, 40–50% of respondents adopted protective behaviours in the various situations. The highest percentages were recorded for situations where respondents were travelling to the bank (76.5%), travelling on foot after 17:00 (73.5%) and travelling in a vehicle after 17:00 (60.1%). The lowest percentages
86
46
41
53
Work break
Mandatory activities
Recreational activities before 17:30
Recreational activities after 17:30
164 102
Travelling by foot before 17:00 Travelling by foot after 17:00
30
221 199
Travelling in vehicle before 17:00 Travelling in vehicle after 17:00
At bank
119 232 135
Work before 09:00 Work between 09:00 and 17:00 Work after 17:00
17
231 106 229
Home before 08:30 Home between 08:30 and 17:30 Home after 17:30
Travelling to bank
Number of observations
Situation
4.56
4.30
4.33
3.78
12.20
13.04
16.30
20.0
58.82
14.20 45.10
3.17 9.55
12.61 16.40 23.7
3.90 4.71 6.55
41.5
41.5
34.7
52.9
53.3
76.5
48.2 73.5
49.3 60.1
40.4 44.8 48.1
46.3 50.0 46.7
Percentage of respondents adopting a protective behaviour
12.2
4.8
8.7
8.0
10.0
17.6
7.9 22.5
9.5 16.1
10.1 10.3 11.1
6.9 8.5 11.8
Percentage of respondents adopting more than one protective behaviour
6
3.84
3.47
2.71
4.00 3.10
4.48 4.22
4.00 3.90 3.61
4.71 4.60 4.48
Average level of emotion-based fear
Percentage of respondents showing a degree of fear (i.e.
