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
bruce.doran@anu.edu.au
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 nego-
tiations 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 infor-
mal 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 disor-
der. 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 prob-
lems, 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 per-
son, 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 hos-
pital. 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 inter-
views 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 conver-
sation. 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 sys-
tem 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 analy-
ses 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 tech-
niques 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 cat-
egory 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 mea-
sures 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 usu-
ally 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 evi-
dence 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 avail-
able. 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 sub-
scriptions 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 Heather Strang

April, 2011 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), Evidence-
Based 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 abil-
ity 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 con-
ference 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 val-
ued. 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 thor-
ough 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, professional-
ism and assistance in conducting the interviews. In particular, VIPs Warwick and
Jim are acknowledged for their outstanding participation. Their assistance was cen-
tral 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 fund-
ing. 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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
The Emergence of Fear of Crime As an Area of Research . . . . . . . . 1
The Paradoxical Nature of the Fear of Crime . . . . . . . . . . . . . . . 2
Current Trends in Fear of Crime Research . . . . . . . . . . . . . . . . 3
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2 Why Is Fear of Crime a Serious Social Problem? . . . . . . . . . . . 9
Individual Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Hypothesized Links Between the Fear of Crime, Disorder and Crime . . 11
Disorder and Decline Hypothesis . . . . . . . . . . . . . . . . . . . . . 13
Economic Impact of Behavioural Responses to Fear of Crime . . . . . . 16
Chapter Review: Potential Problems Not to Be Ignored
and a Need for Spatially Explicit Research . . . . . . . . . . . . . . . . 19
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3 What Causes Fear of Crime? . . . . . . . . . . . . . . . . . . . . . . 25
Criminal Opportunity and Risk of Victimization Theories . . . . . . . . 25
Demographic Theories Explaining Fear of Crime . . . . . . . . . . . . 26
Victimization Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . 26
Indirect Victimization Hypothesis . . . . . . . . . . . . . . . . . . . 27
Vulnerabilities Hypothesis . . . . . . . . . . . . . . . . . . . . . . . 29
Review: An Abundance of Contested Demographic Studies . . . . . . . 31
Social Theories Explaining Fear of Crime . . . . . . . . . . . . . . . . 31
Risk Society Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . 31
Social Disorganization Hypothesis . . . . . . . . . . . . . . . . . . . 32
Review: Social Studies Emphasize the Inherent Complexity of
‘Fear’ of ‘Crime’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Environmental Theories Explaining Fear of Crime . . . . . . . . . . . . 38
The Disorder/Incivilities Hypothesis . . . . . . . . . . . . . . . . . . 38
Threatening and Safe Environments Theories . . . . . . . . . . . . . 40
Signal Crimes Perspective . . . . . . . . . . . . . . . . . . . . . . . 41
Review: Intuitive Environmental Studies into Cues Triggering
Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

xv
xvi Contents

Chapter Review: An Opening for Pertinent Environmental Studies . . . 44

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4 Managing Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . 51
Policing Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Case Study: The New York Police Department’s (NYPD)
Policing Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Environmental Design and Fear of Crime . . . . . . . . . . . . . . . . 56
Chapter Review: Police, Community and Government Cooperation . . . 61
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5 Investigating Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . 67
Defining Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Fear Is an Emotion, Not Cognition . . . . . . . . . . . . . . . . . . . 68
Fear in Relation to Other Emotional Reactions and Stimuli
that Trigger Fear . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Crime Involves a Violation of Criminal Law . . . . . . . . . . . . . . 70
Types of Fear of Crime: Personal and Altruistic Points of View . . . . 71
Review: Key Issues to Consider When Defining Fear of Crime . . . . . 72
Measuring Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . 72
Problems with Cognitive Approaches to Measuring Fear of Crime . . 72
Improvements Through Affective Approaches to Measuring
Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Behavioural Approaches to Measuring Fear of Crime . . . . . . . . . 76
Review: A Preference for Avoidance-Based Measures
in Fear-of-Crime Studies . . . . . . . . . . . . . . . . . . . . . . . . . 79
Analysing Fear-of-Crime Data . . . . . . . . . . . . . . . . . . . . . . 79
Advantages of Spatial Analyses of Fear of Crime . . . . . . . . . . . 81
Spatial Cognition and Cognitive Mapping . . . . . . . . . . . . . . . 82
The Beginning of Fear Mapping . . . . . . . . . . . . . . . . . . . . 84
Activity Diaries and Daily Routines . . . . . . . . . . . . . . . . . . 85
Geographic Information Systems and Fear of Crime . . . . . . . . . 86
Chapter Review: A New Direction with Avoidance Mapping . . . . . . 88
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
6 The Wollongong Study . . . . . . . . . . . . . . . . . . . . . . . . . 95
The Goals of the Wollongong Study . . . . . . . . . . . . . . . . . . . 95
Research Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Logic Behind Study Site Selection . . . . . . . . . . . . . . . . . . . 98
The Central Business District of Wollongong . . . . . . . . . . . . . 99
Crime and Fear of Crime in Wollongong . . . . . . . . . . . . . . . 101
Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Discussion of Spatial Outputs . . . . . . . . . . . . . . . . . . . . . 133
Integrating the Key Spatiotemporal Findings with Police
and Community Initiatives in Wollongong: The Degree of
Institutional Involvement . . . . . . . . . . . . . . . . . . . . . . . . 145
Contents xvii

Assessments of Techniques and Approaches Developed

in Wollongong Study . . . . . . . . . . . . . . . . . . . . . . . . . . 149
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
7 The Kings Cross Study . . . . . . . . . . . . . . . . . . . . . . . . . 155
Background to the Kings Cross Study . . . . . . . . . . . . . . . . . . 155
Goals of the Kings Cross Study . . . . . . . . . . . . . . . . . . . . . . 156
Research Setting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Geographic Location . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . 159
Demographic Characteristics . . . . . . . . . . . . . . . . . . . . . . 162
Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Fear of Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
Interviewing Approach . . . . . . . . . . . . . . . . . . . . . . . . . 168
Survey Design and Questions . . . . . . . . . . . . . . . . . . . . . 168
Spatial Data Visualization . . . . . . . . . . . . . . . . . . . . . . . 172
Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Sample Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . 173
People Are Afraid of Crime in Kings Cross . . . . . . . . . . . . . . 177
People Avoid Specific Areas of Kings Cross Due to Fear of Crime . . 181
Exploring the Underlying Reasons for Fear of Crime . . . . . . . . . 188
Integrating the Fear Mapping Results with Policy
and Community Crime and Fear-of-Crime Prevention . . . . . . . . . . 221
Addressing Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Targeting Pertinent Signs of Disorder and Incivility . . . . . . . . . . 223
Assessments of Techniques and Approaches Developed in the
Kings Cross Study . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
General Summary of the Kings Cross Study . . . . . . . . . . . . . . . 242
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
8 Future Avenues for Fear Mapping: Potential Applications
and Improvements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
Has Collective Avoidance Behaviour Changed in Wollongong
and Kings Cross? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
Investigating Behavioural Responses in Relation to Different
Types of Crime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
Further Avenues for Investigating Links Between Fear, Crime
and Disorder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
Broken Windows Theory in the Transit Context . . . . . . . . . . . . . 261
Fear Mapping and Advances in Spatial Technology . . . . . . . . . . . 262
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
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 con-
cern 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 com-
missioners, staff members of the President’s Commission on Crime and consultants
from every part of America (PCLEAJ, 1967). The report was forthright in argu-
ing 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 stud-
ies 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 1
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_1,

C Springer Science+Business Media, LLC 2012
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 pro-
vided 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 mid-
to-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 victim-
ization (Clark, 2003; Warr, 2000). Results from public opinion polls frequently
showed that high levels of fear were being recorded not only in areas character-
ized 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 incon-
sistencies 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 lev-
els 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 sen-
timents 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 para-
doxical 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 pol-
icy 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 victimiza-
tion 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 politi-
cal 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: impli-
cations 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 indi-
vidual 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 pub-
lic 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 blood-
stream, may also occur to prepare us for a ‘fight or flight’ response (Skogan and
Maxfield, 1981). Additonally, according to Kovecses (1990), fear is more gener-
ally associated with physical agitation; increased heart rate; lapses in heart beat;
blood leaving face; shrinking of skin; straightening of hair; drop in body tempera-
ture; 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 dic-
tated 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 9
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_2,

C Springer Science+Business Media, LLC 2012
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 cam-
pus 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 trav-
elling 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 them-
selves 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 rou-
tine 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 park-
ing 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 evi-
dence 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 sit-
uation, 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 win-
dows hypothesis or thesis (e.g. Harcourt, 1998; Sampson and Raudenbush, 1999;
Loukaitou-Sideris, 1999). Broken windows hypothesis has proven highly influen-
tial 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 assump-
tion 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 sug-
gest 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

1For 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 pub-
lic 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 pro-
fessional, 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 even-
tually 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 bro-
ken 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 People physically and socially

social controls withdrawl from the community,
avoiding uncontrollable/unsafe areas

More disorderly behaviour & broken

windows (increased social and People become afraid of crime
physical incivilities)

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 the-
sis. 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 tempo-
ral 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 neigh-
bourhood systems of social control have broken down.2 When people encounter
signs of disorder they physically withdraw from those areas, confining their activ-
ities 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 peo-
ple feel responsible for. This further serves to weaken community mechanisms of

2 Skogan (1990) specifically defines disorder as ‘direct, behavioral evidence of disorganization’.

3 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).
14 2 Why Is Fear of Crime a Serious Social Problem?

Disorder / Incivilities Demographic collapse of

the neighborhood

Interpreted as a
breakdown in community Deteriorating business
controls conditions & local housing
market

People become afraid of

crime There is an influx criminals
& crime

Physical & psychological

withdrawal of the Delinquency and deviance
community among youth

Weakening of the Decline in the

processes of informal organizational life of the
social control 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 eco-
nomic 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 tradi-

tionally ‘unsavoury’ businesses, such as bars, transient hotels, x-rated outlets and
massage parlours. The author argues that these businesses, and the ‘unsavoury’ peo-
ple 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 resi-
dents 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 resi-
dents. 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 exam-
ple 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 rela-
tively 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 con-
cluded 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 bro-
ken 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 disor-
der. 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 discourag-
ing 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 out-
lined 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 soci-
ety 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 afflu-
ent 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 asso-
ciated 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, enter-
tainment is largely centred around the ‘pub-and-club’ youth culture. The authors
argue that the association of youth with threatening behaviour, such as heavy drink-
ing, 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 cancel-
lations 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 crim-
inal 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 forti-
fying 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) sug-
gest 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 self-
protection 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 gov-
ernments 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 para-
noid 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 exter-
nalities. 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 soci-
ety 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 exter-
nality 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 pro-
cedures 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 spa-
tial 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 pro-
tective 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 asso-
ciated 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: impli-
cations 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 epidemi-
ology 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 neighbour-
hoods. 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 win-
dows 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 under-
standing 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 oppor-
tunities are systematically related to the routine activities of the potential victims and
their guardians1 (Cohen and Felson, 1979). Variation in routine activities differen-
tially 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 analy-
ses 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
2 Criminal opportunity theory branches into numerous related theories focusing on routine activ-
ities 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).

B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series 25
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_3,

C Springer Science+Business Media, LLC 2012
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 concep-
tion (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 demo-
graphic hypothesis seeks to explain why some socio-demographic groups are more
afraid of crime than others. This knowledge is important in providing an under-
standing 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 experi-

ence 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, pre-
vious 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 vic-
timization. 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 self-
image (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, reac-
tions 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 dif-
ferent 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 under-
lying factors, the surveying methods and fear of crime measures could also account
for some of the variation. Generally, victimization is assessed in surveys by ask-
ing 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 hypoth-
esis 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 perspec-
tive, 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 sup-
porting indirect victimization through the media have taken a number of different
approaches. These are known as the cultivation,5 substitution,6 resonance,7 social-
comparison8 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 fos-
tering 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 differ-
ences 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 con-
ditions 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 rela-
tionship between interpersonal communication, rather than the media, and fear of
crime. The interpersonal communication hypothesis assumes that knowledge of oth-
ers’ 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 cul-
tural 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 com-
pare 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 socio-
demographic 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 includ-
ing 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 sensitiv-
ity 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 suscep-
tibility 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 vulner-
abilities 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 vic-
timization (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, char-
acteristics, and social circumstances of the fearful . . .’ and how they ‘acquire a sense
of subset ability’ (Sacco and Glackman, 1987). Some researchers extend this fur-
ther 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, indi-

rect 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 fac-
tors. 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 indi-
viduals 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 dangerous-
ness (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 target-
ing 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 con-
cern 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 hypothe-
sis 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 disorga-
nization 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 vio-
lating 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 com-
munities 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 expe-
rienced 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 prin-
ciples 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, inef-
fective 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 pro-
duces 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 fam-
ily 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 dis-
tinguish 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 com-
munity experienced problems’ (Lane and Meeker, 2003b). Sampson and Groves
(1989) later amended this description slightly to include the concept of social con-
trol, 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 val-
ues 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 hypoth-
esis 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 diver-
sity 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 composi-
tion 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 ‘anx-
iety 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 disor-
der 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 prob-
lems 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 iden-
tification 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 poten-
tially 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 charac-
teristics 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 viola-
tions 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 under-
lying crime and criminal threat (Mirrlees-Black and Allen, 1998; Tulloch, 2000).
They indicate a breakdown in social norms of behaviour and community relinquish-
ment 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 vul-
nerability to crime is generated, which translates into fear (Painter, 1996; Nasar and
Jones, 1997; Rountree and Land, 1996; Taylor and Hale, 1986). Further, incivili-
ties 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 dis-
order/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 all-
encompassing 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 man-
ner (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 disor-
der 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 com-
munity 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 hypothe-
sis, 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 environ-
ments, they may generate fear because they are perceived as attractive sites for
criminal victimization. Threatening physical environments commonly have charac-
teristics 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 pre-
vent 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, prompt-
ing 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 environ-
ments, 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) addi-
tionally 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 inter-
preted 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 under-
standing 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 ‘affec-
tive’ (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 disor-
ders’. With regard to expression, ‘signal crimes’ encompass those signals that denote
criminal incidents, for example a mugging. The content is that they indicate the pres-
ence 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 differ-
ent ‘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 tem-
porally 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 rele-
vance 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 neigh-
bourhood 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 disor-
der/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 reac-
tions 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 per-
spective 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 disor-
der. 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 incivil-

ities 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 atten-
tion 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 pri-
mary 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 vulner-
ability. 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 exam-
ined 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 sev-
eral 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: impli-
cations 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 environ-
mental 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 imper-
fection, 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 neighbor-
hoods”, 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 neigh-
borhoods. 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 envi-
ronment. 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 theoretical-
model.” 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 neighbour-
hoods. Los Angeles, CA, University of California Press.
Skogan, W. G. (1999). Measuring what matters: Crime, disorder and fear. Measuring what mat-
ters: 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 exam-
ination 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 pre-
dictors 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 consid-
ered 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 increas-
ingly focusing on a more in-depth understanding of non-criminal problems, includ-
ing 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 con-
cerns 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 activ-
ity (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 51
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_4,

C Springer Science+Business Media, LLC 2012
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 under-
standing 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 iden-
tify, prioritize and resolve citizen concerns (Adams et al., 2005; Glensor and Peak,
1996; Sims et al., 2002; Walklate, 2000). Surveillance activities like neighbour-
hood 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, territorial-

ity 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 encourag-
ing preventative action and collective efficacy (Innes et al., 2002). Another aim of
community-oriented policing is that if police and community appear to work cohe-
sively 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 gener-
ally 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, famil-
iarity 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 juris-
dictions 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.
 54

Table 4.1 Examples of police services in western democracies having the reduction of the fear of crime as a primary objective

Police department Country Mission statement Source

New York City Police United States ‘The mission of the New York city police department is to enhance the NYPD (2011)
Department 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’
Los Angeles Police United States ‘It is the mission of the Los Angeles police department to safeguard LAPD (2011)
Department 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’
New South Wales Police Australia ‘To have police and the community working together to establish a NSW Police Force (2011)
Force, Wollongong safer environment by reducing violence, crime and fear’
local area command
Police Tasmania, Australia ‘To serve the people of Tasmania by protecting life and property, Glenorchy City Council (2011)
Launceston enhancing community safety and reducing the incidence and fear of
crime’
Thames valley Police United ‘Working with our communities to reduce crime, disorder and fear as Milton Keynes Police (2011)
Kingdom the leading, caring and professional police service’
San Diego Police United States ‘We are committed to working together, within the department, in a City of San Diego (2011)
Department 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’
4 Managing Fear of Crime
Policing Fear of Crime 55

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 mir-
rored the spiral of decay described in Wilson and Kelling’s (1982) broken
windows thesis. Unchecked disorder was seen to be leading to more signif-
icant 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 quality-
of-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 con-
tinue to guide the patrol and investigative work of the NYPD: timely accurate
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 zero-
tolerance approach is that it is based on community-oriented policing (COP).
Bratton lists three elements he considers critical to the success of COP: part-
nership, 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 out-
come 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 commu-
nity 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 bur-
dened 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 solv-
ing and prevention that Bratton (1996) argues are central to COP. Hence,
by using traditional policing methods and excessive force at the neighbour-
hood 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 bru-
tality 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 envi-
ronment 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 char-
acteristics 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 environ-
ment 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, eliminat-
ing 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 environ-
ments 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, surveil-
lance 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, pave-
ment 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 minimiz-
ing escape routes, crime generators and fear generators (Kitchen, 2002). Another
British approach, New Urbanism, draws on Jacobs’s (1961) works. New urban-
ism 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 success-
fully 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 simi-
lar 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 per-
ceived 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 influ-
ence 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 effec-
tiveness 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 poten-
tial 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 39 13 61 28

Burglary 34 18 47 26
Rubbish/litter 44 41 35 35
Theft of/from cars 26 15 52 34
Noise nuisance 18 10 25 14
Youths hanging around 17 8 49 20
Vandalism 17 11 44 23
Drunks 16 10 16 7
Robbery 13 6 29 13
Street lighting 10 4 20 4
Being pestered 5 3 12 9

tempers her optimistic suggestions by emphasizing that if lighting is to be an effec-

tive 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 pri-
vate 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 inextrica-
bly 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 exacer-
bate 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 separa-
tion 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). Wilson-
Doenges (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, territorial-
ization 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 secu-
rity 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 appropri-
ate 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 peace-
ful 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 cen-
tres 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 fundamen-
tal 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 com-
munities 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 rel-
evant 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 neighbour-
hoods. 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 exam-
ination 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 win-
dows 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 investi-
gations 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 los-
ing 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 67
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_5,

C Springer Science+Business Media, LLC 2012
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 phys-
ical 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 evalu-
ation 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 per-
ceptions 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 dan-
ger, 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 prob-
ability of someone being victimized (Ferraro and LaGrange, 2000; Skogan, 1999).
The distinction between perceptions of risk and threat of victimization is cumber-
some. 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 strate-
gies 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: B. Concern about crime C. Fear for others’

crime or safety to others victimization
assessments
Personal D. Risk to self: safety E. Concern about crime F. Fear for self-
of self to self: personal victimization
intolerance
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 percep-
tions 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, con-
cern 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 differ-
ent 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 tap-
ping 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 associ-
ated 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 ques-
tions. 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).

2 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.
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 disor-
der, 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 acknowl-
edged 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 theo-
ries 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 poten-
tially 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 (per-
sonal 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, tai-
loring 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 sig-
nificant 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 percep-
tions 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 neigh-
bourhood (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 dam-
age 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 dif-
ferent 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 neigh-
bourhoods 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 compari-
son 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 fluctu-
ate dramatically within urban neighbourhoods (LaGrange and Ferraro, 1987). Fisher
and Nasar (1995) argue strongly that much extant research is limited because stud-
ies 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) fur-
ther 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 exam-
ple 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 sig-
nificant 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 sin-
gle most serious domestic problem (from a list of 10) that you would like to see
government do something about’. Another simpler version involves asking respon-
dents 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, peo-
ple’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 becom-
ing 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 ques-
tion 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 influ-
enced 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, point-
ing out the importance of crime specificity (Clark, 2003). By referring to a specific
crime and eliciting these personal considerations, emotion-based measures effec-
tively 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 cer-
tain 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 stag-
nation’. 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 asso-
ciated 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 strate-
gies 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 lim-
ited 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 individ-
ual coping strategies or collective actions. Individual coping strategies are diverse
and extensive. In terms of protection against property crime, people adopt ‘target-
hardening 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 prob-
lem. 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 asso-
ciated 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 neighbour-
hood 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 compar-
ison 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 avoid-
ance 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 pre-

clude 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 pro-
vide 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 cog-
nitive 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 word-
ing 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 broad-
level analyses. In contrast, behavioural approaches to measuring fear of crime
hold the potential to overcome much of the subjectivity and ambiguity inher-
ent 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 spa-
tially 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 multidi-
mensional 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 numer-
ous 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 indi-
vidual 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 differ-
ent 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 dis-
advantage. 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 diffi-
cult 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 aggrega-
tion 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 81

Fig. 5.1 The effect of the A) B)

MAUP
6
2
6 3
7 6

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 sig-
nificantly 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 spa-
tial 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 dis-
trict (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 collec-
tive avoidance hotspots and concentrations of crime and disorder. It was evident that
there existed many avenues for future research. In particular, the underlying moti-
vations 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 under-
lying 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 geo-
graphic 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 spatiotem-
poral 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 deter-
mining 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 ‘land-
marks’ 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 compre-
hend 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 dis-
tinct 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 ‘appear-
ance, 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 trig-
ger 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 prop-
erties 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 evaluat-
ing 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 per-
ceive 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 regard-
ing 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 desig-
nated 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 cer-
tain 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 condu-
cted 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 par-
ticular 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 peo-
ple 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 respon-
dents 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 partic-
ular 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) out-
line 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 loca-
tion 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 incor-
porating 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 limita-
tions, 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 associ-
ated 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 con-
ventional 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 crim-
inal 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 influ-
ential 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 adop-
tion by police departments will increase rapidly in future because the technology
is simultaneously becoming cheaper and more powerful. One of the major advan-
tages 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 map-
ping 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 spa-
tiotemporal 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 spatiotem-
poral 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 analy-
ses 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 com-
monly used cognitive and affective approaches to measuring fear of crime. Some
of these limitations can be overcome by using behavioural measures and it is pos-
sible 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 soci-
ety, 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 rea-
soning 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: impli-
cations 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 methodolog-
ical 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 concep-
tual 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 three-
dimensional 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 epidemi-
ology 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 neighbour-
hoods. Los Angeles, CA, University of California Press.
Skogan, W. G. (1999). Measuring what matters: crime, disorder and fear. Measuring what mat-
ters: 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 behav-
iors – 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 envi-
ronmental 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 contempo-
rary 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 bear-
ing 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 spa-
tially 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, includ-

ing 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 pat-
terns 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 95
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_6,

C Springer Science+Business Media, LLC 2012
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 approx-
imately 80 km south of Sydney on the east coast of Australia (see Fig. 6.1).

( Cronulla

( Campbelltown

( Picton

° ( Bulli
Legend
!
. Study Site
( Regional Centers
LGA Boundary
!
. Wollongong CBD
Coastline
Wollongong Roadnetwork

Tasman Sea
0 5 10 20 ( Shell Harbour
Kilometers

Fig. 6.1 The location of Wollongong

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 sec-
tors 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 impor-
tant 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, retail-
ing, 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 eco-
nomic growth and diversification is anticipated in areas such as retail and wholesale
trade, transport, telecommunications, business services, metal industries, commu-
nity 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 considera-
tions. 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 poten-
tial 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 crime-
related 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 consid-
eration. 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
l St
t
250
ighS
l
Meters B Rd
treet

pto
l Street

Smith St
Da St Victoria S
w on t
Ne s
Keira S

bin
Ro
St

Hercules
Denison

Corrima

St Market S
t
Loftus St
Rawson Court Ln
St
Crown Street Globe Ln
Harbour St
ve

Burelli S
St

Osborne McCabe
A

St Stewart S treet
t

t
Kenny S
Gladstone

Atchison
t
Auburn S

Park Georg
Legend e St
Coastline Bank St
Main Roads Ellen St
St

St

Minor Roads
Church

Kembla

Rail Line
Retail Core
Glebe St
Government Offices
West of Railway
South of Burelli Street
East of Corrimal Street West St Beach S
t
Public Parks

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 domi-
nated 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 entertain-
ment 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 indus-
trial 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 memo-
rial 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 high-
est 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 – non-
dwelling, 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 – 11 3, 041 1,615.2

dwelling
Break and enter – 23 1, 786 948.6
non-dwelling
Motor vehicle theft 22 1, 676 890.2
Steal from person 21 490 260.3
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 poten-
tial 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) General factors known to influence fear of crime

(2) Questions on emotional levels of fear in relation to activity diaries
(3) Questions on protective behaviour in relation to activity diaries
(4) Vignettes assessing emotional levels of fear in hypothetical situations
(5) 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 neigh-
bourhood 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 ques-
tion 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) accord-
ing 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-English-
speaking 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 crime-
related 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 neighbour-
hood 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 stan-
dard 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 differ-
ent 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 respon-
dents 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 neighbour-
hoods 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 raster-
based 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 weight-
ing 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 val-
ues 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 Type of physical disorder Weighting factor

1 Cigarettes or cigars in street gutter 1 (not very visible)

2 Garbage or litter in street or sidewalk 2 (moderately visible)
3 Empty beer bottles visible in street 3 (visible)
4 Tagging graffiti 4 (quite visible)
5 Gang graffiti 4 (quite visible)
6 Political message graffiti 4 (quite visible)
7 Graffiti painted over 2 (moderately visible)
8 Abandoned cars/glass from smashed 5 (very visible)
windscreens
9 Abandoned/boarded-up houses 5 (very visible)
10 Lack of exterior maintenance 3 (visible)
11 Vandalism to buildings 5 (very visible)
12 Vandalism to public structures 5 (very visible)
13 Condoms on sidewalk 3 (visible)
14 Needles/syringes/methadone capsules on 5 (very visible)
sidewalk
15 Evidence of homeless people 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 sys-
tematically 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 vari-
ation 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 recod-
ing social disorder after 19:00, does not sufficiently account for potential temporal
variation. For example, Thomas and Bromley (2000) describe how the decentral-
ization 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 func-
tions 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 aban-
doned 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 con-
ducted 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 pre-
cise 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 6 The Wollongong Study

Table 6.3 Types of social

disorder recorded in the
physical disorder assessment
(after Sampson and
Raudenbush, 1999)
Disorder number Type of social disorder
1 Noise
2 Truancy
3 Loitering
4 Public insults
5 Prostitution
6 Panhandling
7 Adults fighting or arguing in a hostile
manner
8 Public drinking/public drunkenness
9 Loud parties
10 Street harassment of women
11 Street harassment of elderly
12 Drug dealing
13 Homeless or mentally ill people
14 Public urination

A number of authors describe certain venues that act as generators of social dis-
order, 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 accu-
racy 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 Bars
2 Bottle shops
3 Night clubs
4 Pornographic theatres
5 Massage parlours
6 Adult shops
7 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 band-
width 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 compari-
son 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 dis-
order 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 The Wollongong Study

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. 6.13 Diagrammatic representation of the potential expansion and eventual linking of two
crime or disorder hotspots in relation to a collective avoidance concentration

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. 6.14 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 com-
pare 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 dis-
order 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

30
Frequency

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 cate-
gories 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

80
Frequency

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

80
Frequency

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 neighbour-
hood 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 vio-
lence 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 victim-
ization 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

80
Frequency

60

40

20

0
0 1 2 3 4 5 6
0 = none, 1 = Deliberate use of a weapon, 2 = attack or assault,
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-of-
crime 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

90

70
Frequency

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 3.00 (3.05) [2.35] 1.24 (1.28) [1.29] 4.00 not very afraid
Car 1.95 (2.53) [3.27] 0.95 (1.15) [1.23] 2.00 quite afraid
Party 1.38 (1.73) [3.77] 0.66 (0.86) [1.08] 1.00 very afraid
Bus stop 2.06 (2.53) [2.30] 1.00 (1.14) [1.14] 2.00 quite afraid
Telephone 3.06 (3.18) [1.96] 1.18 (1.19) [1.32] 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 neighbour-
hood 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

80
Frequency

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

Percentage of respondents Percentage of respondents

adopting avoidance behaviour adopting avoidance
Area during the day behaviour at night

CBD 39.31 81.20

Neighbourhood 18.38 64.10

be seen that in both the CBD area and the neighbourhoods of respondents the per-
centage of the sample adopting avoidance behaviour increases substantially after
dark. GIS-based analysis of avoidance behaviour in the neighbourhoods is not pre-
sented, 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 notice-
ably 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
Meters

treet
Keira S
Crown Street

Burelli S
McCabe treet
Park

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 cen-
tred around the Piccadilly complex in west Crown Street extends eastwards to
occupy most of Crown Street to effectively link up with a collective avoidance con-
centration 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
Meters

treet
Keira S
Crown Street

Burelli S
McCabe treet
Park

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
Meters

treet
Keira S
Crown Street

McCabe Burelli S
treet
Park

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 The Wollongong Study

250

treet
Meters

l Street
Keira S

Corrima
Crown Street

Burelli S
McCabe treet
Park

Legend
Coastline
Main Roads
Minor Roads
Rail Line
Core avoidance hotspot
Intensity of Physical Disorder Hotspot

Low Intensity High Intensity

Fig. 6.24 Physical disorder hotspots, based upon weighted data

Weighting multiplied by level

1800
1500
1200
900
600
300
0
Tagging graffiti

Vandalism to public
Vandalism to
Empty beer bottles

Abandoned/boarded

Graffiti painted over

Lack of exterior
Garbage or litter in

Cigarettes or cigars

cars/glass from …

homeless people
maintenance
visible in street

buildings
in street gutter

Evidence of
Abandoned
structures
up houses
street

Fig. 6.25 Ranking of different types of disorder recorded in the physical disorder assessment
Research Setting 125

treet
250
Meters

l Street
Keira S

Corrima
Crown Street

Burelli S
McCabe treet
Park

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 concen-
tration 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 com-
plex, 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
126 6 The Wollongong Study

250

treet
Meters

l Street
Keira S

Corrima
Crown Street

Burelli S
McCabe treet
Park

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)

250
treet

Meters
l Street
Keira S

Corrima

Crown Street

Burelli S
McCabe treet
Park

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

250

treet
Meters

l Street
Keira S

Corrima
Crown Street

Burelli S
McCabe treet
Park

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)

250
treet

Meters
l Street
Keira S

Corrima

Crown Street

Burelli S
McCabe treet
Park

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 The Wollongong Study

250

treet
Meters

l Street
Keira S

Corrima
Crown Street

Burelli S
McCabe treet
Park

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 loiter-
ing 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 fre-
quencies 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, panhan-
dling, 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

20
Frequency

15

10

0
Loitering

Noise
Public drinking/ public

Public insults

Homeless or mentally ill people

Truancy

Prostitution

Panhandling
Adults fighting or arguing in a

Loud parties

Street harassment of women

Street harassment of elderly

Drug dealing

Public urination
drunkenness

hostile manner

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 avoid-
ance and the primary elements of the broken windows theory, namely physical
disorder, social disorder and crime itself. The general levels of avoidance were cre-
ated 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 avoid-
ance 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 col-
lective 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

30
Frequency

20

10

Loud parties
Loitering

Noise

Drug dealing
Public insults

Homeless or mentally ill people

Adults fighting or arguing in a hostile
Public drinking/ public drunkenness

Street harassment of women

Prostitution

Panhandling
Truancy
Public urination

Street harassment of elderly

manner

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 over-
lap 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 overlap-
ping of the general collective avoidance area and two intense hotspots located along
Gladstone Avenue and along Crown Street, approximately 100 m west of the junc-
tion 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

250

treet
Meters

l Street
Keira S

Corrima
Crown Street

Burelli S
McCabe treet
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
8
Number

6
4
2
0
Massage parlors
Pornographic

Adult shops
dispensaries
Night clubs

Bottleshops
Bars

Methadone
theatres

1 2 3 4 5 6 7

Fig. 6.35 The number of clubs, bars and nightclubs in the CBD area
132 6 The Wollongong Study

250

treet
Meters

l Street
Keira S

Corrima
CrownStreet

Burelli S
McCabe treet
Park

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 avoid-
ance 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 gen-
eral 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

250

treet
Meters

l Street
Keira S

Corrima
Crown Street

Burelli S
McCabe treet
Park

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 subse-
quent 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 increas-
ingly 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 sur-
rounding vicinity, at all times of the day, suggests that the working population in
134 6 The Wollongong Study

250

treet
Meters

l Street
Keira S

Corrima
Crown Street

Burelli S
McCabe treetx
Park

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

250

eet
Meters

l Street
tr
Keira S

Corrima
Crown Street

Burelli S
McCabe treet
Park

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 phys-
ical 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 The Wollongong Study

250

treet
Meters

Street
Keira S

l
Corrima
Crown Street

Burelli S
McCabe treet
Park

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 Strong Strong Strong Strong

Mall Weak Weak Strong Medium
McCabe Park Strong Strong Weak 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 work-
break 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 fea-
tures in the park, improving lighting and the creation of seating areas that were inti-
mate 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 poten-
tial 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 improve-
ments 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 facili-
ties 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 natu-
ral 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 bro-
ken 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 suc-
cess 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 work-
shops 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 concentra-
tion 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 follow-
ing 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’ cul-
ture 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 win-
dows 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 poten-
tially 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 popu-
lation 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 respon-
dents 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 manda-
tory 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 rel-
atively 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 08:30 231

Home between 08:30 and 17:30 106
Home after 17:30 229
Work before 09:00 119
Work between 09:00 and 17:00 232
Work after 17:00 135
Travelling in vehicle before 17:00 221
Travelling in vehicle after 17:00 199
Travelling by foot before 17:00 164
Travelling by foot after 17:00 102
Travelling to bank 17
At bank 30
Work break 86
Mandatory activities 46
Recreational activities before 17:30 41
Recreational activities after 17:30 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 classifi-
cation 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 pro-
tective 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 adopt-
ing 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 differ-
ent 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 per-
centage 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 differ-
ent 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
 142

Table 6.10 Average levels of emotion-based fear and percentages of respondents showing a degree of fear, adopting a protective measure and adopting more
than one protective measure for the 16 situations arising from the classification of the activity diary data

Percentage of
Percentage of Percentage of respondents adopting
Number of Average level of respondents showing a respondents adopting a more than one
Situation observations emotion-based fear degree of fear (i.e. <3) protective behaviour protective behaviour

Home before 08:30 231 4.71 3.90 46.3 6.9

Home between 08:30 and 17:30 106 4.60 4.71 50.0 8.5
Home after 17:30 229 4.48 6.55 46.7 11.8

Work before 09:00 119 4.00 12.61 40.4 10.1

Work between 09:00 and 17:00 232 3.90 16.40 44.8 10.3
Work after 17:00 135 3.61 23.7 48.1 11.1
Travelling in vehicle before 17:00 221 4.48 3.17 49.3 9.5
Travelling in vehicle after 17:00 199 4.22 9.55 60.1 16.1

Travelling by foot before 17:00 164 4.00 14.20 48.2 7.9

Travelling by foot after 17:00 102 3.10 45.10 73.5 22.5

Travelling to bank 17 2.71 58.82 76.5 17.6

At bank 30 3.47 20.0 53.3 10.0

6

Work break 86 3.84 16.30 52.9 8.0

Mandatory activities 46 4.33 13.04 34.7 8.7

Recreational activities before 17:30 41 4.30 12.20 41.5 4.8

Recreational activities after 17:30 53 4.56 3.78 41.5 12.2

The Wollongong Study
Research Setting 143

were recorded for situations where respondents were engaged in mandatory activ-
ities (34.7%), at work before 09:00 (40.4%), recreational activities before 17:30
(41.5%) and recreational activities after 17:30 (41.5%).
The percentage of respondents adopting more than one protective behaviour in
the different situations was generally between 8 and 12%. The situations where
the percentages were highest were while respondents were travelling on foot after
17:00 (22.5%), travelling to the bank (17.6%) and travelling in a vehicle after 17:00
(16.1%).

Summary of Results of Activity Diary Analysis

The general results from the activity diary analysis reveal that, when investigated
in relation to the daily routines, the percentage of respondents showing a degree of
emotion-based fear varies considerably. In most situations the percentage of respon-
dents showing a degree of fear was relatively low, between 4 and 15%, but in certain
situations this was as high as 59%. The highest percentages of respondents showing
a degree of fear were for situations when they were travelling on foot after 17:00,
travelling to the bank, at the bank or at work after 17:00. The highest percentages
of respondents adopting a protective behaviour (60–77%) or more than one type of
protective behaviour (16–23%) were recorded for situations where they were travel-
ling in a vehicle after 17:00, travelling on foot after 17:00 or travelling to the bank.
The situations that were segmented by time of day showed increasing percentages
of respondents indicating a degree of fear or adopting protective measures with later
times of the day.

Discussion of Activity Diary Analysis: The Discrepancy Between

Emotion-Based Fear in Relation to Daily Routines and Global Measures
of Fear
In relation to daily routines, the percentages of the sample showing a degree of fear
were found to be substantially lower than the percentages recorded using global
measures. In this study, for most of the situations resulting from the classification
of the activity diary data, 4–16% of survey respondents showed a definite degree of
fear. The percentage of respondents in this sample that showed a degree of fear using
the global measure ‘how safe do you feel walking alone after dark in the area around
your home?’ was 55%. Other studies using this measure typically report 30–45%
(e.g. Borooah and Carcach, 1997; Fishman and Mesch, 1996; Mirrlees-Black and
Allen, 1998; Michalos and Zumbo, 2000), and as much as 50–70% (e.g. Nair et al.,
1993; Joseph, 1997) of the sample showing a degree of fear.
On the surface, this discrepancy between emotion-based fear assessed in rela-
tion to daily routines and assessed using the standard global measure would seem
to suggest that global measures produce inflated percentages of samples showing a
degree of fear. Such a conclusion would support some of the broad criticisms made
of global measures. In general, global measures would appear to be hypothetical for
many of the survey respondents, as suggested by a number of authors (e.g. Ferraro
144 6 The Wollongong Study

and LaGrange, 1987; Walker, 1994; Hollway and Jefferson, 1997). In particular, it
would appear to support the criticism that global measures fail to assess how people
engage with fear of crime in their daily routines (Tulloch, 1998). Further, it would
reinforce arguments such as those proposed by Walklate (1998) that the results from
broad-scale victimization surveys using global measures give rise to somewhat sen-
sational and unrealistic statements relating to fear causing many people to become
prisoners in their own homes.
However, as noted by Ferraro (1995), the standard global measure of fear taps
more into cognitive judgements of risk on a personal level. Aligning this with the
fact that avoidance behaviour is primarily an attempt by individuals to reduce their
risk of being exposed to victimization (Skogan and Maxfield, 1981), it is likely that
the standard global measure better assesses avoidance behaviour than emotional lev-
els of fear. The results from this study, in terms of the percentage of respondents who
were adopting avoidance behaviour in their neighbourhoods, support this sugges-
tion. In relation to the specific question and cognitive mapping procedure regarding
avoidance behaviour in respondents’ neighbourhoods, 64.1% of the sample in this
study indicated that they were adopting avoidance behaviour in their neighbour-
hood at night. This is approximately 10% higher than the percentage of the sample
showing a degree of fear in response to the global measure.
It appears that a more direct assessment of avoidance behaviour, in relation to
the daily routines of respondents, is likely to reveal a higher proportion of people
who are restricted by their fear of crime than when using the standard global mea-
sure. This is in line with, and lends credence to, the assertion that fear of crime
leads many people to become prisoners in their own homes (e.g. Joseph, 1997).
Further, it emphasizes the appropriateness of Pantazis’s (2000) linking of the pat-
terns associated with avoidance behaviours to current debates on poverty and social
exclusion which focus on people’s ability to participate in activities that others take
for granted. Thus, the global measure of fear, from these perspectives may be more
relevant to the avoidance behaviour of survey respondents and less hypothetical than
a number of authors have suggested (e.g. Hollway and Jefferson, 1997; Tulloch,
1998).
Riger et al. (1982) suggested that 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. The low percentages of
respondents showing a degree of emotion-based fear recorded in this study for situ-
ations when respondents were at home may be a reflection of their ability to adopt
avoidance behaviours. In the home environment, people are likely to have greater
flexibility to act on their judgements of risk and, therefore, the potential to reduce
their exposure to perceived victimization. The result, it appears, are lower levels
of emotion-based fear. This conclusion contrasts with observations made by Liska
et al. (1988) who hypothesized that the constraining of behaviour to safe places and
times of the day may act to reduce levels of fear but found that constrained behaviour
was associated with increased fear.
Other situations where people could have more potential to adopt avoid-
ance behaviours may be while engaging in mandatory and recreational activities.
Research Setting 145

Mandatory activities can be conceptualized as having a fixed location and time

(Golledge and Stimson, 1997). However, the definition of mandatory activities
adopted for the classification of activity diaries in this study is similar to that used by
Kwan (2000b) and included activities such as school, shopping and appointments. It
could be argued that the selection of where to engage in these activities could stem,
in part, from judgements of risk. The same logic applies to recreational activities.
If people experience a crime or something that may influence their judgement of
risk in these situations, they could choose to do them elsewhere. This could involve
a change to a different time or a different location, for example another shopping
area, school or sports field. These possible avoidance behaviours may explain the
low percentages of the sample showing a degree of fear in relation to mandatory
and recreational activities.
In situations where respondents have less potential to adopt avoidance behaviour,
such as travelling on foot after work hours in the CBD, the results show higher
percentages of the sample with a degree of fear and higher percentages adopting
one or more protective behaviours. These are situations where the demands set by
occupational schedules necessitate exposure to risk (Riger et al., 1982). A number of
studies have found respondents more likely to adopt protective behaviours in more
fearful locations (e.g. Teske and Arnold, 1991; Nasar et al., 1993; Nasar and Jones,
1997). Nasar and Jones (1997) describe such situations as being characterized by a
climate of fear. In relation to daily routines, it appears that particular situations also
have a climate of fear.

Integrating the Key Spatiotemporal Findings with Police and

Community Initiatives in Wollongong: The Degree of Institutional
Involvement

A number of authors have argued that 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). Warr (2000)
notes that it is remarkable how there is no systematic evidence on the financial
impact of fear of crime on retail business. The collective avoidance maps of the
Wollongong CBD area and analysis of the activity diary data from the survey pro-
vide strategic information in this regard. Areas adjacent to, or enveloped by, areas
of collective avoidance are highly likely to be at a disadvantage compared to busi-
nesses in areas that are not in close proximity to areas of collective avoidance. In the
case of the CBD of Wollongong, the times of most relevance regarding the impact
of collective avoidance behaviours on retail businesses are the collective avoidance
behaviours during core business hours (i.e. between 9 am and 5:30 pm). Thus, in
cases where collective avoidance concentrations do not show a strong degree of
overlap with other elements of the broken windows thesis, councils and business
chambers logically have a greater degree of involvement in terms of managing fear
of crime than police services.
146 6 The Wollongong Study

Table 6.11 The degree of spatiotemporal overlap between elements of the broken windows thesis
and suggested degree of institutional involvement

Degree of overlap with elements of disorder and

crime Institutional response, in
order of suggested degree
Location PDA SDA1 SDA2 Crime of involvement

Piccadilly Strong Strong Strong Strong Police/Council/Business

Chamber
Mall Weak Weak Strong Weak Business Chamber/Council/
Police
McCabe Park Strong Strong Medium Medium Council/Police

Researchers such as Warr (2000) suggest that there are sound reasons for treating
crime and fear of crime as distinct social problems. However, since the emergence of
fear of crime as an area of social research there has been the recognition that crime
and fear of crime are linked through the role of protective and avoidance behaviours
(e.g. PCLEAJ, 1967; Wilson and Kelling, 1982; Skogan, 1990). Theories such as the
broken windows thesis provide the clearest formalization of the links between crime
and fear of crime (e.g. Wilson and Kelling, 1982; Kelling and Coles, 1997). Thus,
from a management perspective, it is important to recognize crime and fear of crime
as distinct but interlinked social problems. The application of cognitive mapping
techniques to fear of crime and the subsequent GIS-based analysis in this project
have provided a platform to examine the collective nature of avoidance behaviour
in the Wollongong CBD area to the spatial distribution of crime. Table 6.11 revisits
the degree of spatiotemporal overlap between collective avoidance and the other
elements of the broken windows thesis in Wollongong and provides an additional
column outlining the suggested degree of institutional involvement.
The distribution of the collective avoidance hotspots in relation to the hotspots
of crime and disorder provide useful information in terms of strategic intervention
early in the broken windows cycle. To date, many strategies designed to reduce
crime, disorder and fear have focused on the later stages of the cycle. For example,
the well-known zero-tolerance policing strategy in New York was based upon the
assumption that the aggressive policing of disorder would reduce crime and fear of
crime (Bratton, 1995, 1996). In part, the focus on the latter stages of the broken
windows cycle may be due to police services lacking tools to identify where and
when people are afraid of crime, recognized as key information for fear-reduction
strategies (NCAVAC, 1998). In the case of the Wollongong Study, the core spa-
tiotemporal findings (i.e. the ‘where and when’ information for fear of crime in
the CBD) were published in an article in the Professional Geographer (Doran and
Lees, 2005) and presented at a number of City Centre Revitalization Strategy (WCC,
2003) workshops. The findings were integrated directly with a Crime Prevention and
Community Safety Plan implemented in 2007 (WCC, 2007). Many of the strategic
actions in the plan drew on the suggested degree of institutional involvement for
key areas of the CBD and reinterpreted the recommendations to define the role of
specific agencies as well as funding streams at local, state and federal levels. In
Research Setting 147

many instances, the NSW Police Service was listed as a contributing agency but,
interestingly, often as a supporting agency rather than the lead institution. The coun-
cil was able to draw on several sets of information, including the results from the
Wollongong study, to better define its role and that of other agencies, as outlined in
the quotation below from the crime prevention plan:
Clearly there is a place for Wollongong City Council to take more of a lead role in enhancing
safety throughout the LGA and specifically in the CBD. Council’s Wollongong City Centre
Revitalisation Strategy states a vision for the future in which: ‘Wollongong is an inno-
vative and prosperous regional city with a vibrant, cosmopolitan and safe city centre. . .’
(Wollongong City Centre Revitalisation Strategy Overview 2005). Safety in this context
refers to the three basic CPTED principles of: natural surveillance, access control and own-
ership (CPTED guidelines) and (Wollongong City Centre DCP 2005). Apart from CPTED
guidelines, as yet there is not a plan, strategy or integrated set of policies within Council that
spell out crime prevention strategies. Yet there have been numerous surveys and consulta-
tions done throughout the LGA in which residents have given high priority to community
safety and crime prevention (Doran and Lees, 2005), (Social Data Research Project 2004)
and (Social Community Plan 2002–2006) (WCC, 2007: 43).

Building upon this recognition, the crime prevention plan put forth a number of
intervention strategies that drew upon different sources of data – the fear mapping
outputs from the Wollongong study being one of the layers of information. The
action tables listed in the plan form part of the safe communities compact 2007–
2010. Some examples of specific actions are described below.
‘Don’t Stall in the Mall’ was an action which aimed to ‘. . . reduce anti social
behaviour in Wollongong’s CBD, especially abusive language, harassment and evi-
dence of drug taking and drug dealing. . .[and also] reduce fear of crime by visitors
and workers in the Wollongong Mall’ (WCC, 2007: 13). The rationale for this action
drew on several sources of information:
Wollongong Local Area Command regularly ‘tasks’ the CBD making arrests for drug deal-
ing and abusive behaviour. Bruce Doran’s report “Investigating the Spatiotemporal Links
between Disorder, Crime and Fear of Crime” 2005, indicates Wollongong’s Mall is a ‘hot
spot’ for avoidance behaviour as a result of the community’s fear of crime. A recent safety
survey in December by Wollongong Police also indicated fear of crime is a direct result of
anti social behaviours witnessed, including abusive language, harassment and urinating in
public (WCC, 2007: 13).

Figure 6.41 has been extracted from the crime prevention plan. It illustrates how
the council has reinterpreted the generic recommendations outlined in Table 6.11 to
identify specific lead and partner agencies as well as funding sources in addressing
drug-related disorder which was more prevalent during weekdays.
The ‘City Centre Street Camera CCTV Program’ was an initiative designed for
the whole CBD area and had the stated objective ‘To reduce crime in Wollongong’s
CBD, namely assault, drug trafficking, theft and vandalism. . .’ (WCC, 2007: 26).
The rationale for this programme drew upon the results from the Wollongong study
that were published in Doran and Lees (2005), social data research projects con-
ducted by the council in 2004 and 2005 as well the fact that ‘. . . assault, drug
trafficking, theft and vandalism in this area is also well known to local police who
148 6 The Wollongong Study

regularly patrol the CBD’ (WCC, 2007: 26). The action table below (Fig. 6.42) out-
lines the role of key agencies and funding streams, both local and federal, that were
identified for the CCTV program.

Fig. 6.41 Don’t stall in the mall – action table (source: WCC, 2007:13)

Fig. 6.42 City centre street camera CCTV program – action table (source: adapted from WCC,
2007: 26)
Research Setting 149

Other recommendations based on the findings of the Wollongong study were less
formally integrated into planning and design initiatives undertaken as part of the
City Centre Revitalization Strategy which was implemented after the conclusion of
the research. The fear mapping and activity diary data were presented at workshops
and discussed in relation to different design options that had been put forth by a
landscape design consulting group who had been recruited by the Wollongong City
Council (Irwin et al., 2003). While it is difficult to determine whether these ini-
tiatives have resulted in reducing fear of crime and collective avoidance behaviour
across the CBD area, it would appear that the situation has improved. Follow-up
research would be needed to investigate this issue.

Assessments of Techniques and Approaches Developed

in Wollongong Study
Cognitive mapping, in combination with GIS proved a successful means to inves-
tigating collective avoidance behaviour. Essentially, the principles of cognitive
mapping have been well established for some time (e.g. Downs and Stea, 1973;
Downs, 1977; Walsh et al., 1981; Olson and Bialystok, 1983). The application of
cognitive mapping to perceptions of neighbourhood service use, objective physical
and social conditions, perceived neighbourhood and dangerous areas was studied by
Walsh et al. (1981). However, at the time of the study, GIS was in its infancy and
the researchers could do relatively little in terms of spatial collation and analyses in
comparison to the tools available today. The true emergence and recognition of GIS
as a distinct science began in the early 1990s (Goodchild, 1992). Since that time GIS
has evolved significantly (Longley et al., 2001). However, Walsh et al. (1981) were
nonetheless able to determine a strong degree of neighbourhood consensus based on
a range of usage variables, despite individuals having uniquely defined patterns.
More recently, Rengert (1995) used GIS-based techniques to investigate the per-
ceptions of 24 community service recruits being trained to work in the inner city of
Philadelphia. The author used a thematic mapping approach, where the study area
was broken into 16 nearly equal areas and the recruits were asked to identify and
rank these areas according to how dangerous they thought they were. The results
were discussed in relation to hotspots of violent crime and the recruits’ knowledge
of the area. A limitation acknowledged by the author was that the approach used to
assess fear did not allow for a clear understanding of whether the areas identified as
dangerous were also being avoided. A further limitation of Rengert’s (1995) study
is that the division of the study site into 16 areas imposes artificial boundaries that
may not accurately reflect the actual areas that people avoid in relation to their fear
of crime. The cognitive mapping methodology developed in this study overcomes
these limitations, first by specifically measuring avoidance behaviour and, second,
by allowing respondents to outline the actual areas they avoid, rather than supplying
them with predetermined regions. When combined using a grid-based analysis in a
GIS, the results provided new insights into collective avoidance at the community
150 6 The Wollongong Study

level. In turn, this provided a framework with which to investigate spatiotemporal

aspects of the broken windows theory and strategic information for the institutions
involved in addressing fear of crime.
The procedure used to assess social disorder and the resulting maps suggest that
the temporal aspects of disorder, while relatively more time consuming to record, are
important considerations. Previous block assessments of social disorder have been
based on data collection between specific time periods. For example, Perkins and
Taylor (1996) assessed social disorder between 17:00 and 20:00 on weeknights and
12:00 and 20:00 on weekends. Sampson and Raudenbush (1999) assessed social dis-
order between 07:00 and 19:00. Both Perkins and Taylor (1996) and Sampson and
Raudenbush (1999) acknowledge the potential limitations of their studies in terms of
capturing temporal variation in patterns of social disorder. In the case of this study,
assessing social disorder on weekdays and weekends for time segments spanning 24
hours revealed a high degree of variation in social disorder. The finer-scale results
were instructive when comparing social disorder to patterns of collective avoidance
behaviour. For example, the strong degree of overlap between the collective avoid-
ance of the Crown Street Mall on weekdays and social disorder in the mall area on
weekends, mostly at night, suggests a link between the behaviour of the ‘pub and
club’ culture on weekends and avoidance behaviour on weekdays by people working
in the CBD area. This link would not have been evident if social disorder had been
assessed between only between 07:00 and 19:00 or 12:00 and 20:00. Thus, the effort
taken in this study to investigate the temporal dimensions of social disorder, accord-
ing to time of day as well as day of week, is justified and underlines the importance
of testing and adapting existing techniques used to assess neighbourhood disorder.
The mapping of physical disorder proved useful in terms of highlighting the areas
where concentrations were evident. The inclusion of a weighting system for differ-
ent types of physical disorder was also useful, in that it reduced the number of
intense hotspots to show the areas where physical disorder was of greatest concern.
In general, this emphasizes the ability of GIS-based analyses to provide the institu-
tions responsible for managing crime, disorder and fear with relevant information.
For example, analysis of police-recorded incidents of graffiti in the Wollongong
LGA shows the area to be the worst area in the state on the basis of all graf-
fiti incidents (Fitzgerald, 2000). In terms of addressing this problem, being able
to highlight specific areas where graffiti is concentrated is of particular value for
managers.
The activity diary analysis was built on the micro-level behavioural studies of
Fisher and Nasar (1992, 1995) and Nasar and Jones (1997). The approaches adopted
by these researchers enabled investigations of emotional levels of fear and protective
behaviour in specific spatial and temporal contexts. The use of activity diaries in this
study provided a means for doing this on a larger scale, namely a working popula-
tion in a medium-sized city. As such, activity diaries were successful in identifying
particular times and activities where a ‘climate of fear’ was evident. One poten-
tial advantage of activity diaries over the methods used by Fisher and Nasar (1992,
1995) and Nasar and Jones (1997) is that they do not require survey respondents to
walk a particular route at a specified time. Activity diaries, by recording the actual
References 151

daily routines of survey respondents, may therefore capture more typical situations.
Further, the diary approach enabled an investigation of a range of activities, rather
than only one (i.e. walking). The analysis showed emotional levels of fear and pro-
tective behaviour to be most noticeable not only when respondents were travelling
by foot but also in other situations such as when respondents were at work between
certain hours.

General Summary of the Wollongong Study

In general, the findings from the Wollongong study demonstrate that a GIS-based
approach in conjunction with techniques from behavioural geography, in this case
cognitive mapping and an activity diary analysis, does indeed have the potential
to deliver much needed, localized, information on fear of crime. The individual
cognitive mapping outputs provided the building blocks for a collective analysis
of avoidance behaviour. In turn, this enabled a spatiotemporal investigation of fear
of crime in relation to patterns of crime as well as social and physical disorder.
The activity diary analysis provided additional context and insights into protective
behaviour and emotional levels of fear among people working in the CBD area. It
was possible to integrate this body of spatial and temporal information with several
strategic initiatives that took place at the conclusion of the Wollongong study such
as the City Centre Revitalisation Strategy (WCC, 2003) and associated land use
planning changes (Irwin et al., 2003). The core findings from the Wollongong study
were published in an article in the Professional Geographer (Doran and Lees, 2005)
which was integrated directly with a Crime Prevention and Community Safety Plan
implemented in 2007 (WCC, 2007). The Wollongong study also opened a number
of avenues for future research. For example, while the cognitive mapping exercise
considered the degree of avoidance by individual respondents, it was clear that other
questions could be asked in relation to underlying motivations for avoidance – were
people responding more strongly to different types of disorder? How could fear
mapping contribute towards high-visibility policing strategies in a densely popu-
lated inner-city area? Given that Wollongong is a regional city with a number of
unique spatial features and a relatively isolated CBD population, would the approach
used be transferable to other contexts such as a large inner-city area? These and
other questions were explored in the Kings Cross study which is described in the
next chapter.

References
Allen, J. (2002). NSW Recorded crime statistics 2001: regional analysis of crime trends NSW
Bureau of Crime Statistics and Research.
Australian Bureau of Statistics (ABS). (2003). 2001 Census basic community profile and snapshot:
Wollongong. Canberra, ABS.
Bennett, T. (1991). “The effectiveness of a police-initiated fear-reducing strategy”. The British
Journal of Criminology 31(1): 1–14.
152 6 The Wollongong Study

Borooah, V. and C. Carcach (1997). “Crime and fear. Evidence from Australia”. The British
Journal of Criminology 37(4): 635–657.
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.
Chang, K. (2010). Introduction to geographic information systems. New York, NY, McGraw Hill.
Covington, J. and R. B. Taylor (1991). “Fear of crime in urban residential neighbourhoods: impli-
cations between – and within – neighbourhood sources for current models”. The Sociological
Quarterly 32(2): 231–249.
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.
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.
Drummond, W. J. (1995). “Address matching: GIS technology for mapping human activity
patterns”. Journal of the American Planning Association 61(2): 240–251.
Fagan, S. (2009). “St George Illawarra Dragons.” Retrieved April 4th, 2011, from http://www.
rl1908.com/Clubs/St-George-Illawarra.htm.
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 (1987). “The measurement of fear of crime”. Sociological Review
57: 70–101.
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.
Fishman, G. and G. S. Mesch (1996). “Fear of crime in Israel: a multidimensional approach”.
Social Science Quarterly 77(1): 76–89.
Fitzgerald, J. (2000). Graffiti in New South Wales. Sydney NSW Bureau of Crime Statistics and
Research (BOCSAR).
Gibson, C. L., J. H. Zhao, et al. (2002). “Social integration, individual perceptions of collective
efficacy, and fear of crime in three cities”. Justice Quarterly 19(3): 537–564.
Golledge, R. D. and R. J. Stimson (1997). Spatial behavior: a geographic behavior. New York,
NY, Guilford Press.
Goodchild, M. F. (1992). “Geographical information science”. International Journal of
Geographical Information Systems 6: 31–45.
Gupy, M., G. Lacombe, et al. (2000). Wollongong Cultural Industries Audit. Wollongong, NSW,
Wollongong City Council.
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.
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.
Illawarra Regional Information Service (IRIS). (2002). Community Values Survey: Executive
Summary Wollongong, IRIS.
Illawarra Regional Information Service (IRIS). (2004). Profile Illawarra March 2001. Wollongong,
IRIS.
References 153

Irwin, J., M. Wilson, et al. (2003). Wollongong City Centre Public Domain Masterplan.
Wollongong, Wollongong City Council.
Joseph, J. (1997). “Fear of crime among black elderly”. Journal of Black Studies 27(5): 698–717.
Kelling, G. L. and C. M. Coles (1997). Fixing broken windows: restoring order and reducing crime
in our communities. New York, NY, Touchstone.
Keuleers, B. and G. Wets (2001). Using association rules to identify patterns of activity diary data.
80th Annual Meeting of the Transportation Research Board. Washington, D.C.
Krahn, H. and L. W. Kennedy (1985). “Producing personal safety: the effexts 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 three-
dimensional 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.
Lee, H. (1997). A corporate presence: the economy, 1828–1945. A history of Wollongong. J. Hagan
and A. Wells (Eds.). University of Wollongong Press, Wollongong: 53–70.
Levine, N. A. (2002). Crimestat II: a spatial statistics program for the analysis of crime incident
locations. Washington, DC, National Institute for Justice.
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.
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.
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.
Michalos, A. C. and B. D. Zumbo (2000). “Criminal victimization and the quality of life”. Social
Indicators Research 50: 245–295.
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.
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.
NSW Bureau of Crime Statistics and Research (BOSCAR). (2004). Crime hotspots in NSW 2002.
Sydney, BOCSAR.
Oc, T. and S. Tiesdell (1997). Safer city centres: reviving the public realm. London, Chapman.
Olsen, E. P. (2003). Wollongong City Centre Access and Movement Study Final Stage 2A Report:
Principles, Policies and Draft Structure Plans. Brisbane, Eppell Olsen and Partners.
Olson, D. R. and E. Bialystok (1983). Spatial cognition: the structure and development of mental
representations of spatial relations. Hillsdale, NJ, Lawrence Erlbaum.
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.
President’s Commission on Law Enforcement and Administration of Justice. (1967). The challenge
of crime in free society. Washington, DC, United States Government Printing Office.
154 6 The Wollongong Study

Rengert, G. F. (1995). Comparing cognitive hotspots to crime hotspots. Crime analysis through
computer mapping. C. R. Block, M. Dabdoub and S. Fregly (Eds.). Police Executive Research
Forum, Washington, DC: 33–47.
Riger, S., M. T. Gordon and R. K. LeBailly (1982). “1982. Coping with urban crime:
women’s use of precautionary behaviors”. American Journal of Community Psychology 10(4):
369–386.
Sampson, R. J. and W. B. Groves (1989). “Community structure and crime: testing social-
disorganization theory”. American Journal of Sociology 94(4): 774–802.
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).
Skogan, W. G. (1990). Disorder and decline: crime and the spiral decay in American neighbour-
hoods. 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.
Stephens, D. W. (1999). Measuring what matters. Measuring what matters: proceedings from the
police research institute meetings. R. H. Langworthy (Ed.). National Institute of Justice Office
of Community Oriented Policing Services, Washington, DC.
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, Germany: 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.
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.
van der Wurff, A., L. van Staalduinen, 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.
Walker, M. A. (1994). “Measuring concern about crime”. The British Journal of Criminology
34(3): 366–702.
Walklate, S. (1998). “Crime and community: fear or trust?”. The British Journal of Sociology
49(4): 550–569.
Walsh, D. A., I. K. Krauss, et al. (1981). Spatial ability, and environmental knowledge, and envi-
ronmental 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. (2000). “Fear of crime in the United States: avenues for research and policy.” Criminal
Justice 4: 452–489.
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.
Wollongong City Council (WCC). (1999). Wollongong Safe Women Project 1999. Wollongong,
Wollongong.
Wollongong City Council (WCC). (2003). Wollongong City Centre Structure Plan: A Strategy for
the Revitalisation of the Wollongong City Centre Wollongong, Wollgongong City Council.
Wollongong City Council (WCC). (2007). Wollongong City Council Crime Prevention and
Community Safety Plan. Wollongong, Wollongong City Council.
Chapter 7
The Kings Cross Study

Background to the Kings Cross Study

In 2003, then NSW Police Superintendent Dave Darcy of the Kings Cross Local
Area Command implemented a Community Safety Mapping Project in order to
gain an appreciation of fear of crime in his command area and thereby have the
knowledge necessary to reduce it. Previously, such information on public percep-
tions of crime had been obtained through community meetings and, to a lesser
extent, on safety audits. However, the police were cognizant that it was unlikely
that the small numbers of attendees accurately represented the views, experiences
and perceptions of the general community. Thus, the community safety survey was
conducted to gain a wider and more objective understanding of the nature, scope
and causes of fear of crime in the local community (Darcy, 2003).
Using the latest hand-held computer technology and geopositioning software,
residents and visitors to Kings Cross were approached on the street and asked to
identify sites in the Local Area Command where they felt unsafe and safe. The
survey respondents were also asked which environmental cues triggered them to
feel unsafe or safe. A large sample of participants was obtained, which allowed the
spatial analysis of fear experienced by different socio-demographic groups (Darcy,
2003).
The enthusiasm and innovation shown by the police in instigating this fear map-
ping project was unprecedented in NSW, however the validity of the maps was
unfortunately limited by the techniques used to measure and visualize the spatial
fear data. First, this was because a traditional global approach to measuring fear of
crime was employed. Chapter 5 has established how such measures are problematic.
Second, the fear maps disproportionately represented the areas people felt unsafe in.
This resulted from the method used to interpolate the survey point data into the grid
data, which was necessary in producing a series of thematic maps showing areas
where certain numbers of individuals felt unsafe. Additionally, the coarse cell size
of 100 m2 meant a micro-scale analysis of the study site could not occur.
The fear mapping study presented in this chapter was carried out in Kings Cross
following the Police Community Safety Mapping Project in 2004. This latter study
employed the theoretically valid approach to measuring and mapping fear that was

B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series 155
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_7,

C Springer Science+Business Media, LLC 2012
156 7 The Kings Cross Study

explored in the earlier Wollongong study while meshing it with the innovative addi-
tions of the Police example in determining the factors associated with fear of crime,
as described in the following sections.

Goals of the Kings Cross Study

In line with the practical needs for problem-oriented policing and evidence-based
crime prevention, the Kings Cross study aimed to answer the following questions in
a spatially explicit manner:

1. Are people afraid of crime?

2. When are people afraid of crime?
3. Why are people afraid of crime?
4. Where are people afraid of crime?
5. Who is afraid of crime?

In terms of methodological improvements for the field of fear-of-crime research, the

study further developed unique visual-diagnostic mapping technique used to explore
the capacity fear mapping has for providing new and useful information that is not
revealed through traditional cognitive statistical approaches.

Research Setting

The study surveyed people in Kings Cross, an inner Sydney City district in the
state of New South Wales. Kings Cross loosely refers to Sydney’s 24-hour adult
entertainment precinct, which encompasses a disproportionate number of strip clubs
and associated brothels, licensed bars, clubs, cafes and restaurants and backpacker
accommodation. Travel organizations advertise Kings Cross as ‘the premier des-
tination for visitors’, featuring ‘a wild mixture of prostitution and crime, with
stylish restaurants and hotels’ (Australian Explorer, 2005), and ‘more than two
hundred of the city’s finest restaurants, bars and cafes’ (Tourism NSW, 2005).
These services are focused on the infamous Darlinghurst Road, a 200-metre-
long strip known colloquially as Sydney’s ‘dirty half mile’ (Butel and Thompson,
1984; Ellis and Stacey, 1971). Darlinghurst Road also accommodates Australia’s
first and only Medically Supervised Injecting Centre, established in May 2001
(MSIC).

Geographic Location

An area approximately one square kilometre, with Darlinghurst Road as its centre,
was chosen as the specific study site. It is surrounded by the suburbs of Darlinghurst,
Research Setting 157

Fig. 7.1 Street map of the Kings Cross study site

Wolloomooloo, Potts Point, Rushcutters Bay and Elizabeth Bay. This area was
selected in consultation with the NSW Police to include areas of high and low crime
to allow comparison between fear in high- and low-crime areas (See Fig. 7.1 for
study site map and Figs. 7.2, 7.3, 7.4, 7.5 and 7.6 for photos of Darlinghurst Road,
which were taken in 2004).
158 7 The Kings Cross Study

Fig. 7.2 Darlinghurst Road, Kings Cross. Looking north from Bayswater Road junction. Notice
the building architecture and road works reflect the history of Kings Cross and the current
development interest in the region

Fig. 7.3 Darlinghurst Road, Kings Cross, looking east. Notice the adult entertainment premises
and their resident spruikers
Research Setting 159

Fig. 7.4 Darlinghurst Road, Kings Cross. Restaurants and cafes are encouraging alfresco dining

Fig. 7.5 The fountain and Fitzroy Gardens. A popular tourist attraction

Historical Background

Kings Cross has historically fluctuated between periods where it has been regarded
as Sydney’s premier dining and entertainment district, and periods of economic
depression, poverty and slum conditions. Kings Cross has similarly been subject
to constant changes in demography with the resident population being defined
160 7 The Kings Cross Study

Fig. 7.6 Springfield Avenue,

off Darlinghurst Road. One of
the more developed laneways
in the area

by groups ranging from wealthy upper-class communities, refugees, returning

servicemen, European immigrants and bohemian artists.
During the first half of the nineteenth century Queens Cross (now Kings Cross)
was considered a wealthy area, being occupied by the community’s upper-class res-
idents who lived on large estates and in mansions (Whitaker, 2002). The depression
of the 1840s prompted residents to subdivide and sell off their land for the con-
struction of terraces (Butel and Thompson, 1984; Whitaker, 2002). By the 1850s
the area was renowned for its cheap housing, slum conditions and violence (Butel
and Thompson, 1984). The area had become the home of the infamous gang, the
‘Darlinghurst Push’ (ESNA, 2002). Residences were subject to further subdivision
in the 1870s following the imposition of heavy land taxes and by the 1880s ter-
race housing and ribbon development in the area became fashionable again (Butel
and Thompson, 1984). During this decade the area saw a large influx of European
migrants (Butel and Thompson, 1984; Whitaker, 2002; Lumby, 2005).
In 1905 Queens Cross was renamed Kings Cross and the region became popu-
lar for dining and entertainment (Whitaker, 2002). Gangs, mobsters, violence and
shootings became common from 1916 (Butel and Thompson, 1984; Ellis and Stacey,
1971). During the 1920s, prostitution was obvious and Darlinghurst Road gained
Research Setting 161

its nametag as ‘the dirty half mile’ (Butel and Thompson, 1984; Ellis and Stacey,
1971). Public outcry in the 1930s saw the gangs eradicated and the arrival of a
bohemian presence (Butel and Thompson, 1984). By this time, flats became promi-
nent, as many terraces had deteriorated or were turned into boarding houses (Butel
and Thompson, 1984; Lumby, 2005). An influx of refugees in the 1930s was closely
followed by US servicemen, throughout World War II (Lumby, 2005). Butel and
Thompson (1984) suggest that the ‘the growth of night clubs and strip clubs, black
market trading and rampant prostitution’ dated from World War II. Hence, World
War II reportedly changed Kings Cross with the local residents being unhappy with
corruption in the area and the presence of US servicemen (Ellis and Stacey, 1971).
Cheap rents in the late 1940s encouraged an influx of more immigrants and again
in the 1950s, Kings Cross entered a phase during which it was regarded as a sophis-
ticated and cosmopolitan gathering place for diners and tourists (Ellis and Stacey,
1971; ESNA, 2002). Kings Cross became famous for its ‘live theatre, good restau-
rants and cafes and intellectual and artistic activities’ (Lumby, 2005). More people
immigrated from Europe and the Mediterranean regions under the government’s
policies and settled in the area, adding to the culinary diversity (Lumby, 2005). In
the late 1960s demand for terrace houses returned and prices in the area increased
(Butel and Thompson, 1984).
Despite this, during the 1950s the area became known as a ‘red light’ district,
with the growth of ‘home’ brothels (ESNA, 2002). Residents were complaining of
harassment, assault and robbery and by 1967 a police station was built in Kings
Cross (Ellis and Stacey, 1971). In 1969, police made 11,624 arrests, an average of
over 31 per day. These arrests included charges of assault, robbery, drunkenness,
murder, prostitution, possession of drugs, vagrancy, obscene exposure, gaming and
betting, receiving stolen goods and indecent language (Ellis and Stacey, 1971). Ellis
and Stacey (1971) note that a period of crime followed the ‘invasion’ of American
servicemen on leave from the Vietnam War from 1967 until 1970. In the early 1970s
Kings Cross became the centre of heroin supply and use in Australia. By the late
1970s street prostitution was apparent (ESNA, 2002; Van Beek, 2004). This coin-
cided with a decriminalization of the offences of loitering and soliciting for the
purposes of prostitution in 1979 (ESNA, 2002). Public outcry in 1983 forced the
amendment of the Prostitution Act to prohibit soliciting for prostitution in residen-
tial streets, however this law was not enforced (ESNA, 2002). The area was regarded
by Executive Chief Superintendent, Ken Chapman, as being ‘volatile’, with assaults
‘happening all over the place, both day and night’ (Zadel, 1989).
Nevertheless, Kings Cross continued to be Sydney’s premier tourist district into
the 1970s and 1980s (Whitaker, 2002). Since the late 1990s, tourism in the area has
declined and a majority of the hotels have been converted into apartments (Whitaker,
2002). Kings Cross’s diverse history is evident in the heterogeneity of its current
physical and demographic profile (Ellis and Stacey, 1971). Streets contain buildings
from different eras, and a diverse range of people from differing socio-demographic
backgrounds occupy those buildings (CoSC, 2005d). This is illustrated in the next
section on the demographic characteristics of the region. Economically, Kings Cross
is still defined by its 24-hour adult entertainment services including strip clubs and
162 7 The Kings Cross Study

associated brothels, licensed bars, cafes and restaurants, and backpacker accom-
modation (Jochelson, 1997). Travel organizations advertise Kings Cross as ‘the
premier destination for visitors’, featuring ‘a wild mixture of prostitution and crime,
with stylish restaurants and hotels’ (Australian Explorer, 2005), and ‘more than two
hundred of the city’s finest restaurants, bars and cafes’ (Tourism NSW, 2005).

Demographic Characteristics
This diverse history of Kings Cross is evident in the heterogeneity of its current
demographic profile. On average the inner-east is a wealthy area, with residents
having a mean weekly individual income double that of the average for both NSW
and Australia ($700–$799 compared to $300–$399; ABS, 2002a, 2002b). However,
constituting this average are some of the highest income earners in Australia and
some of the most socially disadvantaged.
Kings Cross still has a high proportion of visiting tourists and a very transient
population. On the night of the Australian census, 23% of the census sample were
visiting Kings Cross from overseas or elsewhere in Australia. In comparison, only
1% of the population in Sydney on the night of the census were visiting from areas
outside of Sydney. Of the population over the age of 5 years, 66% occupied a dif-
ferent address five years ago. Reflective of broader Sydney, Kings Cross has a large
number of overseas-born residents (ABS, 2002a, 2002b).
Like many high crime and high fear-of-crime regions, Kings Cross is a high-
density inner-city area undergoing rapid gentrification (Grennan, 2001; Darcy, 2003;
Knox, 2003). In 2001, Kings Cross comprised Australia’s most densely populated
area (the inner-east suburbs of Elizabeth Bay, Potts Point, Rushcutters Bay and
Woolloomooloo). Here, fear of crime has the potential to affect a large number
of people in a relatively small area. There are practical and policy implications
of being able to map fear of crime in such areas. For example, already the local
community, police and council would be interested in reducing crime and fear of
crime. Thus there was increased likelihood that the maps produced in this study,
which show where and why people avoid areas, could be used in policy, planning
and practice. As an area undergoing gentrification, the redesign of the environment
to manage fear and crime and to promote the use of public space is particularly
relevant.

Crime
Crime has been high in Kings Cross since the 1800s and the region is renowned
for this association (Butel and Thompson, 1984; Jochelson, 1997; Whitaker, 2002).
The NSW Bureau of Crime Statistics and Research (BOCSAR) has identified Kings
Cross as an inner-Sydney hotspot of assault and robbery. Table 7.1 shows the num-
ber and rate for each of the following offences: assault, robbery and ‘other offences
against the person’ for the Kings Cross LAC between 1999 and 2004 (the five years
 Research Setting

Table 7.1 Number and rate per 100,000 population of selected offences recorded by NSW police in the Kings Cross local area command: 1999–2004

1999 2000 2001 2002 2003 2004

No. Rate No. Rate No. Rate No. Rate No. Rate No. Rate

Assault 796 2786.5 874 3059.6 1063 3721.2 982 3437.7 947 3315.1 943 3301.1
Robbery total 326 1141.2 346 1211.2 492 1722.3 296 1036.2 289 1011.7 211 738.6
Robbery without 210 735.1 232 812.2 328 1148.2 206 721.1 193 675.6 153 535.6
a weapon
Robbery with a 15 52.5 9 31.5 17 59.5 7 24.5 8 28.0 8 28.0
firearm
Robbery with a 101 353.6 105 367.6 147 514.6 83 290.6 88 308.1 50 175.0
weapon not
firearm
Other offences 9 31.5 12 42.0 14 49.0 14 49.0 16 56.0 15 52.5
against the
person

Source: NSW Bureau of Crime Statistics and Research, ref: tm05-3505.

163
164 7 The Kings Cross Study

prior to interviewing). To provide a comparison, Table 7.2 shows the number and
rate of these offences for NSW between 2001 and 2004. For each year during the
five-year period prior to interviewing (from 2000 to 2004), the rate of assault and
‘other offences against the person’ in The Kings Cross LAC was three times the rate
of those offences NSW-wide (BOCSAR, 2005b). However, it is important to note
that the crime trends in Kings Cross for these offences and other selected offences,
like ‘stealing’ and ‘break and enter’, are all stable or decreasing for the period from
2000 to 2004 (see Table 7.3).
BOCSAR statistics also show that out of the 80 police LACs in NSW, Kings
Cross has consistently been ranked in the top 10 for assault, robbery and ‘other
offences against the person’ from 2002 through to 2004 (see Table 7.4). In 2003, the
year prior to interviewing, Kings Cross was ranked 6th for assault, 4th for robbery
and 6th for ‘other offences against the person’. Comparatively, the statistics indicate
Kings Cross is a hotspot for these specific crimes.
At the time of interviewing, specific problems with crime in Kings Cross were
largely alcohol and drug related, such as assault, the use of and dealing of illicit
drugs, prostitution and vandalism: or associated with socially disadvantaged youth
who are being introduced into criminal enterprises (Darcy, 2005). Prostitution,
homelessness, organized crime and antisocial behaviour were also considered the
norm (Darcy, 2005).

Fear of Crime

In addition to crime, fear of crime is recognized as a problem in the study site.

The Australia-wide fear-of-crime study by Tulloch et al. (1998a, 1998b) identi-
fied Kings Cross as having a reputation for crime and fear of crime. For example,
their Tasmanian survey respondents most commonly mentioned Kings Cross as the
most dangerous place in Australia. However, even their Sydney respondents men-
tioned Kings Cross as a dangerous suburb (Tulloch et al., 1998a, 1998b). Many of
their respondents acknowledged that they have never been to Kings Cross and were
fearful of crime due to presumed crime levels and police corruption.
In line with this reputation, many travel organizations advise visitors to be careful
in Kings Cross (Sydney Online, 2005; Travel Online Australia, 2005). Nevertheless,
it is also recognized as a place of excitement and pleasure. For example, one
organization warns visitors to be careful ‘especially at night, as people do get
mugged here. . . spruikers outside nightclubs: they can be intimidating and aggres-
sive’ (Sydney Online, 2005). In contrast others comment that ‘it is well policed
and there’s rarely more trouble than a few drunks having a fight’ (Travel Online
Australia, 2005).
Results from the aforementioned Police Community Safety Mapping Project,
implemented in Kings Cross in 2003, also revealed a high level of fear of crime
(Darcy, 2003). Of 603 respondents, 62% stated they felt unsafe in the Kings
Cross Local Area Command (Darcy, 2005). Due to methodological inconsistency,
it is difficult to compare this level of fear with those evident in other regions
 Research Setting

Table 7.2 Number and rate per 100,000 population of selected offences recorded by NSW police in NSW: 2001–2004

1999 2000 2001 2002 2003 2004

No. Rate No. Rate No. Rate No. Rate No. Rate No. Rate

Assault – – – – 69,165 1051.9 72,279 1089.5 72,419 1083.8 70,122 1041.7

Robbery total – – – – – – – – – – – –
Robbery without a weapon – – – – 8055 122.5 6614 99.7 6270 93.8 4973 73.9
Robbery with a firearm – – – – 992 15.1 805 12.1 793 11.9 666 9.9
Robbery with a weapon not – – – – 5229 79.5 3486 52.5 2971 44.5 2609 38.8
firearm
Other offences against the – – – – 994 15.1 1130 17.0 1270 19.0 1414 21.0
person

Source: NSW Bureau of Crime Statistics and Research, BOCSAR (2005a).

165
166 7 The Kings Cross Study

Table 7.3 Trends in selected offences for the Kings Cross local area command: 2003–2004 and
2000–2004

Annual percentage Average annual percentage

Offence category change 2003–2004 change 2000–2004

Assault Stable Stable

Sexual assault Stable Stable
Indecent assault, act of indecency Stable Stable
and other sexual offences
Robbery without a weapon Down by 20.7% Down by 9.9%
Robbery with a weapon not a Stable Down by 16.9%
firearm
Break and enter – dwelling Down by 31.7% Down by 13.6%
Break and enter – non-dwelling Stable Down by 13.9%
Motor vehicle theft Stable Down by 8.7%
Steal from motor vehicle Down by 16.0% Down by 14.0%
Steal from retail store Stable Up by 10.8%
Steal from dwelling Stable Down by 4.1%
Steal from person Down by 22.0% Not calculated
Malicious damage to property Stable Stable

Source: NSW Bureau of Crime Statistics and Research, BOCSAR (2005b).

of Australia.1 In 2004, Burgess (2004) conducted a fear-of-crime survey using

crime-specific avoidance-based measurement questions and produced a number of
preliminary avoidance maps. When comparing areas of fear with sites of robbery
and assault, Burgess (2004) found that there were areas of the study site where crime
and fear of crime appear to coincide. However, there were regions of discrepancy,
where levels of fear appeared lower or higher than the occurrence of crime would
justify when using those spatially congruent areas as a point of reference (Burgess,
2004). This mismatch is largely attributed to the presence of environmental cues
that trigger fear of crime, which were not examined in this initial study (Burgess,
2004). Darcy’s survey found that junkies/homeless people were the top response for
triggering those respondents to feel unsafe (see Table 7.5). Both of these studies will
be discussed more throughout this chapter.

1 The ABS’s 2006 General Social Survey found that less than half (48%) of people reported feeling
safe or very safe walking alone in their local area at night (ABS, 2006). The ABS’s 2005 Crime
and Safety Survey found that 4% (day) and 8% (night) of respondents felt unsafe or very unsafe
when at home alone during the day and night respectively (ABS, 2005). The ABS’s 2002 Crime
and Safety Survey found that 4% (day) and 10% (night) of respondents felt unsafe or very unsafe
when at home alone during the day and night respectively (ABS, 2002a).
 Table 7.4 Number, rate and ranking per 100,000 population and ranking of selected criminal incidents recorded

Assault 2002 2003 2004

Research Setting

Local area command Number Rate Ranking Number Rate Ranking Number Rate Ranking

Castlereagh 668 5134.9 1 647 4973.5 2 752 5780.6 1

City Central 1445 4988.6 2 1606 5544.4 1 1460 5040.4 2
The Rocks 601 4861.3 3 598 4837.0 3 602 4869.4 3
Kings Cross 982 3437.7 6 947 3315.1 6 943 3301.1 6

Robbery 2002 2003 2004

Local area command Number Rate Ranking Number Rate Ranking Number Rate Ranking

Redfern 614 1763.4 2 559 1605.4 2 543 1559.4 1

City Central 1445 4988.6 2 1606 5544.4 1 1460 5040.4 2
Surry Hills 270 1172.5 4 279 1211.6 3 267 1159.5 3
Kings Cross 296 1036.2 5 289 1011.7 4 211 738.6 4

Other offences against the 2002 2003 2004

person

Local area command Number Rate Ranking Number Rate Ranking Number Rate Ranking

The Rocks 9 72.8 2 9 72.8 2 15 121.3 1

City Central 22 76.0 1 14 48.3 8 31 107.0 2
Orana 25 43.7 7 36 62.9 3 44 76.9 3
Kings Cross 14 49.0 5 16 56.0 6 15 52.5 9

Source: NSW Bureau of Crime Statistics and Research, ref: sew05-3525.

167
168 7 The Kings Cross Study

Table 7.5 Top 10 responses

accounting for triggering Reason for feeling unsafe Percent (%)
respondents to feel unsafe
Junkies/Homeless 18
Prostitutes 13
Spruikers/Intoxicated persons 8
Dark laneways 7
Vulnerabilities 6
Intimidation 6
Lighting 5
Lack of cleanliness 5
Laneways 4
Loitering 4

Source: Darcy, 2003.

Methods

Interviewing Approach

With the intention of sampling residents of, and visitors to, Kings Cross, respon-
dents were recruited in a public street setting. Interviews were primarily conducted
along the main streets of Darlinghurst Road, Macleay Street and Victoria Street,
in Fitzroy Gardens and in front of Wolloomooloo Police Station. Safety concerns
for the interviewers meant that few of the interviews were conducted in the back-
streets of Kings Cross. This also prevented door knocking as a recruitment option.
Interviewers moved from site to site at various times of day. At each site, the clos-
est individual was approached and asked if he/she would participate. If this person
declined, the next closest individual was approached until someone agreed to partic-
ipate. The survey was conducted in April–May 2004, between the hours of 7 am and
6 pm. These hours allowed the interviews to cover the temporal shifts in the demo-
graphic groups occupying public spaces in Kings Cross. For example, it allowed for
surveying of workers entering or leaving the study area (present from 7 am to 9 am
and 4:30 pm to 6 pm), the elderly (present from 9 am to 11:30 am) and intravenous
drug users and dealers (present from 11:30 am onwards). Standardized interviewing
methods were used because the presence of an interviewer overcomes many of the
problems with postal or self-administered surveys.

Survey Design and Questions

The survey was adapted from Doran and Lees (2003) and consisted of a ques-
tionnaire and mapping section where respondents mapped areas they avoided. The
questionnaire, comprising a series of closed questions, was included to gain relevant
Methods 169

socio-demographic information about the respondents. This included the respon-

dents’ gender, age, housing tenure type, residential status, experience of previous
victimization and income. Questions aimed to assess levels of social integration,
confidence in the police and fear of crime were also included in this section.
For comparative purposes, a global measure of fear of crime and a crime-specific
avoidance-based measure of fear of crime were presented. The global measure of
fear was based on the frequently used question ‘have you felt fearful or afraid
when walking alone in your neighborhood’ (Ditton, 2000; Pantazis, 2000). The geo-
graphic reference was changed from ‘neighborhood’ to Kings Cross to allow for a
comparative analysis. The final question read as ‘Have you ever felt fearful or afraid
when walking alone in Kings Cross? (yes or no)’.
The crime-specific avoidance-based question immediately followed in the map-
ping section of the survey. Respondents were provided with a map of the study
site and asked if they avoided any areas because they were afraid of being robbed,
beaten or attacked, first during the day and second during the night. As Kings Cross
is a definite hotspot for recorded incidents of assault and robbery, these specific
crimes were chosen because they are relevant to the research setting, are personal in
nature and can be distinctly conceptualized. If the respondents answered positively
to this crime-specific avoidance-based measure of fear, they were asked to illustrate
those areas they avoided on the map provided. Surveys where the respondents did
specify areas that they avoided on the provided map were considered as answering
positively to the question.
Respondents were then asked how hard they tried to avoid each area and what
environmental cues triggered their fear of being robbed, beaten or attacked. When
respondents answered how hard they avoided each area, they could choose only
one of five potential answers ranging from ‘very hard’, ‘quite hard’, ‘don’t know’,
‘not very hard’ to ‘not hard at all’. The options were displayed on a cue card and
corresponded to a numerical value, based on an ordinal Likert Scale. These values
are referred to as ‘avoidance hardness’ weights. A second question on environmental
cues was then presented to the respondents. Respondents were given another cue
card with a list of 16 social and physical environmental cues (see Table 7.6 for
cues and explanations). Multiple environmental cues could be selected and different
answers could be given for each avoided area that was defined.
The environmental cues were chosen due to their relevance to the research set-
ting and extensive reference throughout the fear-of-crime literature. Highlighting
this is important because Phillips and Smith (2003) criticize that researchers can
intentionally or unintentionally label certain groups as ‘incivil’ from the outset of
their studies. These groups are typically those who are disadvantaged and differ-
ent, whom the public stereotype as being involved in crime and become scapegoats
symbolically linked to crime (Blalock, 1967; Kelling and Coles, 1997). Such groups
include those used in this survey, including homeless people, sex workers and drug
users.

Table 7.6 List and explanations of the social and physical environmental cues used in the survey
Social factors
Drug usersa
Spruikersb
Homeless peoplec
Intoxicated personsd
Sex workerse
Gangsf
Loitering peopleg
Pedestrian absenceh
Physical factors
Poor street lightingi
Vandalismj
Rubbish/syringesk
Rundown/abandoned buildingsl
Offensive/degraded shopsm
Areas to hiden
170
Explanation
Intravenous drug users, users of other illicit drugs and drug dealers
Employees, usually of adult entertainment venues, who encourage pedestrians to buy tickets and enter their premises
People living in community shelters or on the street
People who have consumed alcohol or appear dunk
People who engage in sexual acts for money, also known as prostitutes
Groups of people, who generally appear menacing or who elicit feelings of concern in pedestrians
People who appear to have no purpose for being where they are; ‘up to no good’
Lack of other pedestrians
Explanation
Absence or lack of adequate street lighting
Property damage by vandals, for example graffiti or broken windows
Litter on streets/thoroughfares that is not in bins, and equipment used in drug injection
Vacant buildings or those that are dilapidated or in a state of disrepair
Shops that offend or appear dilapidated
Places where an attacker could seek refuge, for instance hiding behind bushes
7 The Kings Cross Study
 Methods

Table 7.6 (continued)

Blocked escapeo An area where escape would be difficult in the event of an attack, for instance stairways
Lanewaysp Small thoroughfares, generally pedestrian only although some are big enough for one-way traffic
a Other sources that examine drug users: Covington and Taylor (1991), Darcy (2003), Perkins and Taylor (1996), Skogan (1990), Perkins et al. (1993) cited in
Ross and Mirowsky (1999)
b Other sources that examine spruikers: Darcy (2003)
c Other sources that examine homeless people: Darcy (2003), Perkins et al. (1993) cited in Ross and Mirowsky (1999)
d Other sources that examine intoxicated persons: Covington and Taylor (1991), Darcy (2003), Skogan (1990), Ross and Mirowsky (1999)
e Other sources that examine sex workers: Darcy (2003), Perkins et al. (1993) cited in Ross and Mirowsky (1999)
f Other sources that examine gangs: Perkins and Taylor (1996), Rohe and Burby (1988) and Perkins et al. (1993) cited in Ross and Mirowsky (1999),

Skogan (1990)
g Other sources that examine loitering people: Darcy (2003), Skogan (1990), Ross and Mirowsky (1999)
h Other sources that examine pedestrian absence: Jacobs (1961), Loukaitou-Sideris (1999), Samuels and Judd (2002)
i Other sources that examine poor street lighting: Darcy (2003), Fisher and Nasar (1995)
j Other sources that examine vandalism: Lewis and Maxfield (1980), Perkins et al. (1993) cited in Ross and Mirowsky (1999), Skogan (1990)
k Other sources that examine rubbish/syringes: Covington and Taylor (1991), Darcy (2003), Doeksen (1997), [LaGrange, Ferraro and Supancic (1992); Perkins

et al. (1993) cited in Ross and Mirowsky (1999)], Skogan (1990), Taylor and Covington (1993)
l Other sources that examine rundown/abandoned buildings: Covington and Taylor (1991), Doeksen (1997), Perkins and Taylor (1996), [Ferraro and Supancic

(1992); Gates and Rohe (1987); LaGrange, Lewis and Maxfield (1980); Rohe and Burby (1988) cited in Ross and Mirowsky (1999)], Skogan (1990)
m Other sources that examine offensive/degraded shops: Darcy (2003)
n Other sources that examine areas to hide, or concealment: Rondeau et al. (2005), Fisher and Nasar (1992, 1995), Nasar and Jones (1997)
o Other sources that examine blocked escape: Rondeau et al. (2005), Fisher and Nasar (1992, 1995)
p Other sources that examine laneways: Darcy (2003), Fisher and Nasar (1995)
171
172 7 The Kings Cross Study

Spatial Data Visualization

Using ESRI’s ARC suite of GIS programs, this spatial avoidance data was used to
produce three different styles of fear maps. These consisted of two-dimensional
(2D) ‘collective-avoidance’ maps showing how many respondents avoided each
area, 2D ‘avoidance hardness’ maps showing the extent to which the respon-
dents tried to avoid each area and three-dimensional (3D) maps that concurrently
displayed the ‘collective-avoidance ’ and ‘avoidance hardness’ data.
Each individual survey map was digitized and imported in the GIS as separate
layers. Attributes were assigned to the layers so that each area avoided by a respon-
dent could, for example, display the sex, age and income of the respondent and the
corresponding environmental cues that triggered fear in the respondent.
For the 2D ‘collective-avoidance’ maps, each area avoided by an individual was
assigned a value of one (1). The individual maps were then aggregated to a single
map that illustrated the number of respondents avoiding each area of the study site.
The resulting accumulative values were then classified as a percentage of the total
number of respondents in that socio-demographic group. This allows for comparison
between the socio-demographic groups. The process is shown in Fig. 7.7.

Fig. 7.7 A visual representation of the process of aggregating the individual maps to produce
the 2D ‘collective-avoidance’ and ‘avoidance hardness’ maps. The number of respondents in the
example is 10, with two of these people employing avoidance behaviours
Results and Discussion 173

To create the 2D ‘avoidance hardness’ maps, the same process was replicated
using the ‘avoidance hardness’ values for each avoided area. However, to show the
average degree of ‘avoidance hardness’ for each area, the accumulative values in the
output maps were then divided by the total number of people avoiding areas in that
socio-demographic group. This returned the values to the original Likert Scale with
values from one to five, the Avoidance Hardness Index. This process is also shown
in Fig. 7.7.
The 2D ‘collective-avoidance’ and ‘avoidance hardness’ maps were then visu-
alized in 3D to better illustrate areas of fear according to the number of people
avoiding them and the extent to which those people tried to avoid the areas.
The unclassified ‘collective-avoidance’ maps were first displayed as the eleva-
tion (z-factor) of the land, thereby representing the number of people avoiding each
area. Higher land indicated areas avoided by more people than lower land. In order
to allow easy comparison between different maps, particularly those displaying the
patterns of avoidance adopted by different socio-demographic groups, population
percentile bands were added to the maps (white horizontal lines). The population
percentile bands were inspired from contour lines on a topographic map; however
instead of showing height above sea level, they indicate the proportion of respon-
dents that was avoiding each area. The population percentile bands were placed at
5% intervals (i.e. at several heights representing increments in the numbers of people
equal to 5% of the total sample population in that category). Each band therefore
indicates a 5% increase or decrease in the number of avoiding respondents. The
colour of the land shows how hard the respondents tried to avoid each area.
The 2D ‘avoidance hardness’ layers were then draped over the 3D collective-
avoidance’ maps. Thus, the colour of the maps shows how hard the respondents tried
to avoid each area. This adds an extra dimension that can be analyzed in conjunction
with the population numbers. Fig. 7.8 visually represents this 3D process using a 2D
graph that signifies a cross-section of a 3D map.

Results and Discussion

Sample Characteristics

Slightly more males (53%) than females (47%) were interviewed. This is consistent
with the male to female ratio in Sydney’s inner-east. The majority of respondents
were in the middle-age group (46% aged between 30 and 59). Almost 25% of
respondents were over the age of 60 and 30% of respondents were between the
ages of 18 and 29. While the percentage of respondents in the later age group is
consistent with the local demographic (32%2 ), there is a much higher proportion of
respondents aged over 60 in the research sample. In line with this, there is a lower

2 Thirty-two percent of the inner-east population over the age of 18 years.

174 7 The Kings Cross Study

Fig. 7.8 A visual representation of a 3D map cross-section, showing how the 2D maps were
combined for 3D display. The number of respondents in the example is 400. The maximum number
of avoiding respondents in the cross-section is approximately 150 (33% of the total)
Results and Discussion 175

18%
14.96% 15.22%
16%
14%
12% 11.29% 10.76% 11.02%
Percent

10% 8.14%
8.40% 7.87%
8% 7.09%

6% 4.72%
4%
2% 0.52%
0%
18−23 24−29 30−35 36−41 42−47 48−53 54−59 60−65 66−71 over 72 No
Age distribution of respondents answer

Fig. 7.9 Age distribution of respondents

proportion of respondents in the 30–59 age group in the sample. Only 1% of the
respondents did not state their age (see Fig. 7.9).
In terms of housing tenure, the majority of the sample were owner-occupiers
(45%) and non-owner-occupiers (36%). Less than 10% of the respondents lived
in government housing or were staying in backpacker accommodation and 2% of
the sample lived in a community shelter. Overall these percentages are consistent
with general housing tenure in the inner-east; however more of the sample are
owner-occupiers in comparison to the local demographic (36%). Only 1% of the
respondents did not answer the question on housing tenure.
Slightly more visitors to, than residents of, Kings Cross were interviewed. This
is contrary to the ratio of residents to visitors in the inner-east on census night (77%
to 23% respectively). Of those visitors to Kings Cross who were interviewed, 76%
were Australian residents and 22% were visiting from overseas. Less than 3% per-
cent of the visiting respondents did not answer this part of the residency question.
Of those residents of Kings Cross who were interviewed, the majority (50%) had
resided in the area for more than five years. Of the remainder of the residents,
approximately 20% had lived in the area less than 1 year, 10% for 1–2 years and
15% for 3–5 years; 5% of the respondents who were residents did not answer this
part of the question.
The majority of respondents had not been a victim of crime in the 12-month
period prior to interviewing (66%) (see Fig. 7.10 and Fig. 7.11); 35% of the
respondents had been victims of crime. Of these respondents, their experience of
victimizations had comprised of attack or assault (27%), threats of force or violence
(19%), theft or attempted theft (32%) and damage to property (20%) (see Fig. 7.11).
Only 2% had been victims of a crime involving the use of a weapon. The majority
of respondents had not ever been victims of crime in Kings Cross (64%); 20% of
the respondents who had been victims of crime in Kings Cross reported that crime
to the police.
176 7 The Kings Cross Study

70% 65.62%

60%

50%
Percent

40%

30%

20%
9.19% 11.02%
6.56% 6.82%
10%
0.79%
0%
Use of a weaponAttack or assault Threats Theft / Damage to Not applicable
force/violence attempted theft property
Experience of victimization

Fig. 7.10 Respondent distribution according to experience of victimization

35% 32.06%

30%
26.72%
25%
19.08% 19.85%
Percent

20%

15%

10%

5%
2.29%
0%
Use of a weaponAttack or assault Threats Theft / attempted Damage to
force/violence theft property
Use of a weapon

Fig. 7.11 Distribution of those respondents who had been victimized, according to experience of
victimization

The majority of respondents (85%) indicated they thought that they or their
neighbours would call the police if they saw someone being assaulted in Kings
Cross. Only 10% answered negatively and 6% did not answer this question; 64% of
the respondents indicated they were either ‘very confident’ or ‘quite confident’ in
the police. Only 18% of the respondents were ‘not very confident’ or ‘not confident
at all’ in the police.
A large proportion of respondents fell into the lowest income category (28%).
The two middle-income categories comprised of 55% of the respondents, while less
Results and Discussion 177

than 10% of respondents were in the top-income bracket. This result is very different
from typically high income levels earned by inner-east residents (for example 4%
and 21% in the lowest- and highest-income categories respectively). The result is
more consistent with the income levels earned by residents of greater Sydney: 8%
of the respondents did not answer the question on income.

People Are Afraid of Crime in Kings Cross

A sample of 399 survey respondents was obtained. A total of 18 surveys were

excluded from analysis because more than two general survey questions or the
mapping section was unanswered. The study results verify that people are afraid
of crime in Kings Cross, with 35% of the respondents indicating that they had felt
fearful or afraid when walking alone in Kings Cross (see Table 7.7). More specif-
ically, 36% (day) and 66% (night) of the respondents indicated that they avoided
areas of the study site because they were afraid of being robbed, beaten or attacked
during the day and night respectively. The finding that fear of crime is greater dur-
ing the night than the day (nearly double that during the day) is consistent with the
results from other fear-of-crime studies that attribute this increase in fear to the onset
of darkness.
These findings are also consistent with the earlier study in Wollongong, where
Doran (2004) found that 54.7% of respondents felt either ‘not very safe’ or ‘not safe
at all’ when walking alone in the city, and that 39.31% (day) and 81.20% (night)
respondents actually avoided parts of the CBD during the day and night respectively.
Given the fact that crime levels are higher in Kings Cross than in Wollongong, it is
somewhat surprising that avoidance levels, in terms of overall percentage of avoid-
ing respondents, are higher for Wollongong than Kings Cross. This finding reflects
the signal crimes perspective, which concludes that fear of crime depends on the sit-
uational context of the study region and the presence of environmental cues within
it. Regardless, the levels of fear found in Kings Cross are high and it is possible that
they are underestimated given the probability that the most fearful members of soci-
ety could not be interviewed due to the street-based interviewing approach. While
this study does not suggest that these research findings will be true for the wider
population, they do indicate a large proportion of residents of, and visitors to, Kings
Cross could be avoiding parts of the region due to fear of crime.
Forty-six percent of the avoiding respondents tried either very hard or quite hard
to avoid the areas in which they were afraid of being robbed, beaten or attacked
during the day. This figure increased to 57% during the night; 34% of the respon-
dents did not try very hard or did not try hard at all to avoid those areas during
the day. This figure decreased to 30% during the night. For both the day and night,
only 4% of respondents did not know how hard they tried to avoid those areas; 22
respondents did not answer the question. This accounted for 16% and 9% of the
respondents during the day and night respectively (see Table 7.8).
178 7 The Kings Cross Study

Table 7.7 Number and percent of respondents indicating fear of crime

Yes No No answer Total

No. % No. % No. % No. %

Global measure: 133 34.91 244 64.04 4 1.05 381 100.00

Have you ever felt
fearful or afraid when
walking alone in Kings
Cross?

Crime-specific 138 36.22 243 63.78 0 0.00 381 100.00

avoidance-based
measure: Do you avoid
any areas shown on this
map of Kings Cross,
because you are afraid
of being robbed, beaten
or attacked, during the
day?

Crime-specific 252 66.14 129 33.86 0 0.00 381 100.00

avoidance-based
measure: Do you avoid
any areas shown on this
map of Kings Cross,
because you are afraid
of being robbed, beaten
or attacked, during the
night?

Table 7.8 Number and percent of respondents by degree of avoidance hardness

Day Night

No. % No. %

Very hard 40 28.99 86 34.13

Quite hard 24 17.39 57 22.62
Don’t know 5 3.62 11 4.37
Not very hard 30 21.74 54 21.43
Not hard at all 17 12.32 22 8.73
No answer 22 15.94 22 8.73
Total responses 138 100.00 252 100.00

The Dissonance Between Traditional Global Measures and Crime-Specific

Avoidance-Based Questions
By employing a survey comprising of a global measure of fear and a crime-specific
avoidance-based question, it was not only possible to assess fear-of-crime levels,
but also possible to compare unspecified cognitive approaches to measuring fear
Results and Discussion 179

of crime with specified behavioural approaches. This section discusses the differ-
ences between the results of the two measurement approaches and discusses the
implications of these differences for future fear-of-crime research.
One would expect respondents declaring that they have not felt fear of crime
to not avoid areas because they were afraid of being robbed, beaten or attacked.
Likewise, one would expect respondents affirming fear of crime in Kings Cross
to avoid areas because they have felt such fear. If this were true, the responses
would coincide. However, agreement occurs in only 68% of cases during the day
and in 60% of cases during the night. This exposes a degree of dissonance, which in
turn provides some useful insights and asserts the appropriateness of crime-specific
avoidance-based questions in measuring fear of crime.
There are two scenarios where there is dissonance between the results from
the crime-specific avoidance-based question and a global measure of fear. The
first scenario consists of respondents answering positively to the crime-specific
avoidance-based question and negatively to the global measure of fear. The sec-
ond scenario involves the converse. While these discordant results could be due to
respondents answering survey questions dishonestly or misinterpreting questions,
possible arguments are discussed for each scenario.

Scenario One: Fearful People Choosing Not to Avoid

The first scenario consists of respondents stating that they have felt fearful or afraid
when walking in Kings Cross and yet do not avoid any areas because they fear
being robbed, beaten or attacked. Approximately 16% of respondents answered in
this manner during the day and 3% during the night. Three arguments support this
scenario.
First, it is possible the fear experienced in Kings Cross was either not due to fear
of crime at all or not specifically due to fear of being robbed, beaten or attacked.
Having never felt fearful of being robbed, beaten or attacked there would be no
motive to avoid areas. This reinforces the notion that global questions about fear
can provide misleading results. It is imperative that questions are crime-specific
when querying respondents about fear.
The second rationale is respondents were fearful of being robbed, beaten or
attacked, however not to the extent whereby avoidance was warranted. Benefits of
using the areas would outweigh the negative responses felt when evaluating the sit-
uation. This argument could account for lower levels of dissonance during the night
(13.4% down from 16.3% during the day). The percentages could indicate that fear
does not warrant avoidance behaviour during the day, but does at night. This demon-
strates that avoidance behaviours produce more informative results than cognitive
assessments of fear, as suggested by Warr (2000) and Burnett (1976).
Finally, it may be that respondents were previously fearful of being robbed,
beaten or attacked. The question put forward was ‘have you ever felt fearful. . .?’.
This implies respondents could answer affirmatively having once felt fearful, even
if they no longer felt that fear. Theoretically, if people no longer felt fearful there
180 7 The Kings Cross Study

would be no reason for avoidance behaviour. This suggests that the area, or respon-
dents’ perceptions of the area, might have improved (Couclelis, 1998). In a dynamic
area like Kings Cross this occurrence could be exaggerated. Additionally, it indi-
cates that global questions may not reflect fear of crime at the time of surveying. As
in the avoidance-based question, survey items should specifically ask respondents
about current feelings of, or responses to, fear.

Scenario Two: Fearless People Avoiding Areas

This scenario involves respondents stating they have not felt fearful of crime when
walking in Kings Cross and yet do avoid areas due to fear of being robbed, beaten
or attacked. Approximately 16% and 37% of respondents answered in this manner
during the day and night respectively. Four arguments support this scenario. Each
assumes respondents do not feel fearful because they avoid the areas in which they
feel such fear. This reinforces the benefit of studying behavioural responses to fear
of crime, rather than emotional statements.
It is important to re-emphasize that respondents state they have never felt fearful
when walking in Kings Cross. This elicits questions of why these people originally
avoided areas because they felt fearful of being robbed, beaten or attacked if they
never felt fear in the first place. The first two arguments infer respondents never
visited the areas they avoid and only assume that they would fear being robbed,
beaten or attacked if they resided there.
This situation requires respondents to associate areas with fear based on sec-
ondary information provided by friends, family or media (Romer et al., 2003;
Weitzer and Kubrin, 2004; Killias and Clerici, 2000; Koskela and Pain, 2000;
Valentine, 1989). This supports theories implying that indirect-victimization is
largely responsible for heightened levels of fear (Katz et al., 2003). Consequently,
the respondents’ spatial knowledge and patterns of avoidance are based on the
reputation of the area, not personal experiences. If this is the case, surveys that
consequently investigate the influence of certain environmental cues that act to trig-
ger fear in the area may be flawed, as the respondents will not have actually seen
the area themselves. Therefore, it may be beneficial that future investigation ask
respondents whether they have actually been into or seen the areas in question.
The second line of reasoning entails respondents observing the areas they avoid
from a distance. Individuals may view a laneway from another street and perceive it
as dangerous in which they would feel fearful of being robbed, beaten or attacked. In
this case, the avoided areas should only be as large as the field of vision from various
vantage points used to assess the area. If the avoided area is larger, the respondents
could be associating the region with environmental cues that are not present. As
mentioned above, this could hamper any investigations of environmental cues that
act to trigger fear of crime. Hence respondents should be asked if they have seen all
of the areas that they avoid.
Another case involves respondents having once visited the areas they avoid. That
they did not feel fearful within those areas could imply they were not walking alone
and felt less vulnerable to attack. For example accompaniment by others or driv-
ing a vehicle could provide a sense of security. Further studies could investigate
Results and Discussion 181

such circumstances or protective actions enabling people to feel secure in normally

avoided areas. Similarly, people may be walking alone and employ other protective
actions which stopped them from feeling fear.
Finally, the results show that the dissonance between the statistical and spatial
results is greater for the night than the day. Respondents may automatically think
of daytime hours, or the time of interviewing, when contemplating if they have felt
fear. This demonstrates the practicality of asking respondents about fear during the
day and night, as did the crime-specific avoidance-based question.
As expected, given the literature on global measures of fear of crime, many
respondents answered the crime-specific avoidance-based and global fear-of-crime
survey questions differently. This dissonance demonstrated the ambiguity inher-
ent in global measures and reinforces appropriateness of using crime-specific
avoidance-based questions.

People Avoid Specific Areas of Kings Cross Due to Fear of Crime

This study specifically examines where people are afraid of being robbed, beaten
or attacked in Kings Cross. By developing a visual-diagnostic technique for map-
ping collective avoidance, the study found there are common patterns of avoidance
throughout the study site, and that fear of crime in Kings Cross occurs in noticeable
hotspots. The evidence that specific micro-level hotspots of fear do exist in this high-
crime area is consistent with other fear mapping examples, for instance Doran and
Lees’s (2003) and Fisher and Nasar’s (1992, 1995) studies. Information regarding
the whereabouts of these fear hotspots in Kings Cross has useful implications for
policy and planning in the region, especially given that the City of Sydney Council
aims to identify safety issues and crime ‘hotspots’ and design strategies to improve
crime rates and safety (CoSC, 2006m). This section of the chapter examines over-
all avoidance patterns in the study site, as well as the implications of the maps for
policing and CPTED.

Mapping Reveals Three Fear-of-Crime Hotspots

The overall avoidance maps reveal a radical change in elevation in the centre of the
fear maps, which indicate an enormous divide in collective avoidance either side
of William Street (Figs. 7.12, 7.13, 7.14, 7.15 and 7.16). There is a much larger
proportion of respondents avoiding the study site north of William Street. The three
peaks north of William Street show that collective avoidance in the study site is
greatest in three distinct hotspots, referred to as hotspots A, B and C (see Fig. 7.16).
Avoidance predominates

• over central Woolloomooloo (hotspot A);

• in the street block bordered by Victoria Street, Orwell Street and Darlinghurst
Road (hotspot B); and
• in the street block bordered by Darlinghurst Road, Bayswater Road, Roslyn Street
and Ward Avenue (hotspot C).
182 7 The Kings Cross Study

Fig. 7.12 Two-dimensional collective-avoidance map depicting areas where all respondents
avoided because they were afraid of being robbed, beaten or attacked during the DAY
Results and Discussion 183

Fig. 7.13 Two-dimensional collective-avoidance map depicting areas where ALL RESPON-
DENTS avoided because they were afraid of being robbed, beaten or attacked during the NIGHT
184 7 The Kings Cross Study

Fig. 7.14 Three-dimensional fear map depicting areas of fear according to the number of respon-
dents avoiding them and the extent to which those respondents tried to avoid the areas because they
were afraid of being robbed, beaten or attacked during the DAY

Avoidance in the study site was greatest in the Kings Cross neighbourhood, which
locates the fear hotspot peaks B and C. These hotspots are discussed before
peak A.

Kings Cross: Hotspots B and C

Victoria Street, Orwell Street and Darlinghurst Road are frequently the bound-
aries for the second area (hotspot B). Streets within this area include Earl Place,
Earl Street, Llankelly Lane, Mall Place and Springfield Avenue. Hotspot B often
extends to include Orwell Lane, Hughes Lane, Hughes Place and Hughes Street.
Darlinghurst Road, Bayswater Road, Roslyn Street and Ward Avenue border hotspot
C, which encompasses Kellett Way, Kellett Street and Mansion Lane. The slopes
of the peaks defining hotspots A and B are quite high, suggesting that there is
a large change in the number of avoiding respondents over a relatively small
area.
The fear maps show that up to 15% and 30% of all respondents avoided hotspots
A and B during the day and night respectively. With a history of crime and disor-
der in the Kings Cross neighborhood, the City of Sydney Council has made plans
to encourage opportunities for surveillance throughout the area. This is proposed
Results and Discussion 185

Fig. 7.15 Three-dimensional fear map depicting areas of fear according to the number of respon-
dents avoiding them and the extent to which those respondents tried to avoid the areas because they
were afraid of being robbed, beaten or attacked during the NIGHT

by maintaining the mix of land uses so there is 24-hour surveillance along major
streets; lighting pedestrian pathways with vandal-proof fixtures; having public open
spaces that are well lit and have clear sight lines; signage describing pathways and
facilities, for example taxi ranks and bus stops; provision for Help Points; active
uses and frontages around public open spaces; building entry points that are readily
identifiable, clearly visible and well lit; and buildings that are designed to minimize
entrapment spots and have openings in all walls that have frontage to a public area
(AJC, 2006).
Kings Cross Town Centre and Plaza are situated on Springfield Avenue near
its intersection with Earl Place. These venues are in the heart of fear hotspot B
and, have been identified by Council as key sites for redevelopment with the estab-
lishment of active edges and outdoor eating to activate the street life and provide
opportunities for natural surveillance (AJC, 2006). Those streets forming hotspot
C have not specifically been addressed in terms of redevelopment opportunities. If
combating fear of crime is an objective, these streets should also be a priority for
gentrification.
186 7 The Kings Cross Study

Fig. 7.16 The Kings Cross study site depicting Blocks A, B and C, which make up the three
common fear-of-crime hotspots

Woolloomooloo: Hotspot A
The avoidance maps show that up to 11% (day) and 29% (night) of all respondents
avoided central Woolloomooloo (hotspot A) during the day and night respectively.
The coverage and definition of hotspot A varies greatly for each environmental
Results and Discussion 187

cue (discussed later). Generally speaking, the City of Sydney intends to maintain
the quiet residential atmosphere that currently defines the Woolloomooloo neigh-
bourhood (AJC, 2006). However, the results from the fear mapping show that this
environment should not be maintained as is if fear of crime and avoidance are to be
addressed.
The City of Sydney Council does plan to landscape the rail corridor and viaduct,
which are sites of particularly pronounced avoidance levels in peak A (AJC,
2006).This will allow for greater spatial definition, increased amenity, more activ-
ity along its edges and improved surveillance (AJC, 2006). There are also plans
to redevelop a series of vacant lots on the western edge of the study site (AJC,
2006). However, the research results show that overall avoidance is low along the
affected streets and that people are not avoiding these areas because the presence
of rundown/abandoned buildings trigger their fear of crime. It is therefore likely
that the reduction of fear of crime will require alternative CPTED strategies in
Woolloomooloo.
The population percentile bands and elevation of the land show collective avoid-
ance throughout the study site increases during the night. In all of the maps the
average avoidance hardness did not exceed ‘not very hard’. In the each of the maps,
the jump in the extent to which respondents try to avoid areas (from ‘not hard at all’
to ‘not very hard’) generally coincides with the onset or middle of the three major
avoidance peaks.

Safe Areas and Cognitive Barriers

William Street
The fear maps show that William Street acts as a divide separating areas that trig-
ger high levels of avoidance (north-side) from those areas that do not (south-side).
Brantingham and Brantingham (1993) propose that people include perceptual edges
in their cognitive maps to indicate the spatial limits of high-crime areas. The avoid-
ance maps demonstrate that such edge effects exist and that they influence how
hotspots of crime and fear are cognitively defined, with people commonly perceiv-
ing William Street as a physical barrier between safe and unsafe areas. The maps
also provide evidence that such phenomena do affect peoples’ spatial choices and
behaviours. Brantingham and Brantingham (1993) suggest that edges have increased
levels of crime because they represent the limits of territoriality and the surveillance
or identification of strangers who may commit crime. However, given the fact that
visitors frequent the suburbs to the north and south of William Street, it is unlikely
that this is the cause of the change in fear and avoidance levels. More likely, it is
perceived changes in crime levels, environmental cues, land-use types and demo-
graphic characteristics that influence levels of fear and avoidance. For example,
those suburbs immediately to the north of William Street have higher crime lev-
els, more signs of disorder, a higher density of adult entertainment premises and
lower average incomes than the suburbs to the south.
188 7 The Kings Cross Study

Darlinghurst Road
Similarly, the fear maps establish that during the day and night respectively, 5%
and 20% fewer respondents avoid Darlinghurst Road than the surrounding areas.
This establishes that Darlinghurst Road is considered a distinctly safe thorough-
fare through fear hotspots A and B. This finding is interesting given that most
signs of disorder, robbery incidents and assault incidents in Kings Cross occur
along Darlinghurst Road. In line with this, Brantingham and Brantingham (1993)
comment that crime is concentrated along high-activity pathways, like Darlinghurst
Road, that are frequented by people travelling to work, shopping centres and enter-
tainment premises. The research results suggest that although Darlinghurst Road
is high in crime, the fact that it is a major pathway with the attractions of trans-
port, shopping and entertainment may mean people’s levels of fear of crime are
not great enough to warrant avoidance behaviour. It could also indicate the pres-
ence of signs of safety or cues that counteract feelings of fear. Further research into
why such high-crime pathways may have low levels of fear will help with CPTED
efforts. Nevertheless, the research results show that Kings Cross is losing a very
large number of potential customers and the presence of crime and disorder means
that Darlinghurst Road should remain a focus for CPTED and police attention. The
City of Sydney has recognized this and taken action to encourage ‘better quality’
entertainment and retail activities so that Kings Cross continues to be attractive to
local and global visitors (AJC, 2006).

Victoria and MacLeay Streets

Victoria and MacLeay Streets are also defined on the fear maps as areas where col-
lective avoidance is considerably lower than in the surrounding areas, although not
to the same extent as along Darlinghurst Road. Victoria Street separates hotspots
A and B. These streets are both largely residential with a few specialty stores and
coffee shops. They generally do not have the same level of signs of physical and
social disorder evident along Darlinghurst Road and have not been targeted for
gentrification by the City of Sydney.

Exploring the Underlying Reasons for Fear of Crime

While fear and crime can be spatially coincident this is not always the case (Smith,
1987; Nelson et al., 2001). Commonly, research has ascertained that fear in certain
areas is higher than warranted by actual rates of victimization (Liska et al., 1988;
Painter, 1996). The occurrence of this spatial discrepancy has led researchers to
query the validity of crime survey questions (Taylor and Hale, 1986). This section
of the chapter provides alternative explanations for the failure of fear and crime
to match spatially. It first looks at the coincidence between crime and fear of crime
and then explores those environmental cues that trigger people to feel afraid of being
robbed, beaten or attacked in Kings Cross.
Results and Discussion 189

The Presence of Crime

Not surprisingly, most robbery and steal-from-person incidents occur in close prox-
imity to the entertainment strip, forming a well-defined hotspot running along
Darlinghurst Road between its intersections with William and McCleay Streets
(Fig. 7.17). The side streets of Bayswater Road, Kellet Street, Kellet Way, Rosyln

Fig. 7.17 Sites of robbery and steal-from-person incidents in the Kings Cross study site, in the
six-month period prior to interviewing (October 2003 to April 2004)
190 7 The Kings Cross Study

Street, Earl Place and Springfield Plaza are also contained within the hotspot. These
areas provide the focus for police deployment with high-visibility uniformed foot
patrols supported by specialist undercover police.
In Kings Cross, there are areas of the study site where crime and fear of crime
appear to coincide. However, there are regions of discrepancy where levels of fear
appear either lower or higher than the occurrence of crime would justify (when using
those spatially congruent areas as a point of reference).

A Positive Association Between Crime and Fear

In the northern half of the study site (above William Street) a positive association
between crime and fear of crime appears. Areas with a high density of assault or rob-
bery correspond to areas that many respondents avoid (fear hotspots B and C, which
are avoided by 10–15% of respondents during the day and 30–35% of respondents
at night). Similarly, in areas north of William Street where the incidence of assault
and robbery is high but more sporadic than in the aforementioned areas, there has
been a proportionate decrease in fear (with 5–10% and 15–30% of the respondents
avoiding these areas during the day and night respectively). These areas are referred
to as ‘secondary’ areas of fear and crime. However with regard to the area of the
study site south of William Street, the correlation between fear and crime is not as
apparent.

A Spatial Mismatch Between Crime and Fear

There is a comparable amount of crime between the southwestern quadrant of the
study site, bounded by and encompassing William Street and Victoria Street, and the
‘secondary’ areas north of William Street. Despite this, considerably fewer people
avoid the area (less than 5% during the day and less than 15% during the night). Fear
of crime in this southwestern quadrant appears lower than expected given the level
of crime3 . In the southeastern quadrant of the study site, bordered by William Street
and Victoria Street, there are no cases of assault or robbery. As few people avoid
this area during the day (less than 5% of respondents), there is congruence between
levels of fear and crime. However, a discrepancy exists during the night, where up
to 15% of the respondents avoided the area. When comparing levels of avoidance
in the northern half and southwestern quadrant, this percentage appears higher than
warranted by spatial crime statistics.
Nevertheless, an increase of approximately 10% of respondents avoiding an area
during the night than during the day is consistent throughout most of the study site.
This suggests that darkness triggers approximately 10% of the population to feel

3 Using the levels of fear experienced by the respondents in the northern half of the study site as a
benchmark.
Results and Discussion 191

more fearful4 . In areas where the increase is proportionately greater, for example
either side of Darlinghurst Road, it is likely that night-time social changes also play
a role in heightened levels of fear (Koskela, 1999). For example, the presence of
intoxicated persons may increase during the night. Time affects one’s perception of
events and phenomena (Allen and Kautz, 1985 cited in Block, 2000; Tversky and
Taylor, 1998). As intoxicated persons were found to be a significant environmental
cue during the night, and most liquor outlets are on Darlinghurst Road, this propo-
sition is likely. Nonetheless, spatial analysis of environmental cues is needed for
conclusive hypotheses.

Accounting for the Mismatch Between Crime and Fear

This part presents possible explanations for the mismatch between crime and fear of
crime in Kings Cross. First, researchers argue that unreported crimes can justify high
levels of fear in low-crime areas (Koskela and Pain, 2000). Given that approximately
half of the respondents who had been victims of crime in Kings Cross reported it to
the police, this argument may account for spatial mismatches. The crime map shows
only the crimes sites that respondents felt fear towards (robbery and assault). Other
forms of crime may occur and trigger fear of being robbed, beaten or attacked, thus
crime in general may predominate in the heavily avoided areas. Furthermore, the
spatial crime data encompassed the six-month period prior to interviewing. If data
over a larger time period was presented, the analysis may show that levels of crime
were more strongly associated with avoidance patterns.
As discussed in the previous section of the chapter, while spatial cognition is
an ongoing process, a person can formulate a cognitive map after a brief encounter
with an environment. Hence, a person can take a static ‘snapshot’ of an environ-
ment and immediately choose to avoid it (Block, 2000). Downs and Stea (1973)
advocate spatial behaviour is habitual. Given this, it is likely people would not
return to environments they initially avoided and their current avoidance patterns
are the result of ‘snapshots’ produced some time ago. The avoided areas would
therefore not coincide with current crime hotspots. As Kings Cross is undergoing
rapid change, this may be an important factor explaining the mismatch between fear
and crime.
Cognitive maps are regarded as incomplete and distorted, due to the subjectivity
of spatial cognition (Downs and Stea, 1973). People overestimate distances, partic-
ularly the spatial extent of conceptually important areas (Day, 1976). An area one
fears and avoids could be considered conceptually important. Thus, respondents
may have exaggerated the extent of avoided areas, explaining why fear of crime
may be exaggerated in low-crime areas.

4 Researchers agree that people exhibit more fear after dark. Physically, the reduction in visibility
and recognition abilities, and the creation of blind spots, shadows and potential places of entrap-
ment could play a role (Fisher and Nasar, 1995; Nasar and Jones, 1997; Painter, 1996; Samuels
and Judd, 2002).
192 7 The Kings Cross Study

Block (2000) states that people overestimate short distances and underestimate
long distances. This may have occurred when the respondents illustrated the avoided
areas. It was observed that many respondents roughly circled the areas they avoided,
rather than going into specific detail. This encourages more emphasis in the analysis
to be placed on the central regions (peaks) of avoided areas, rather than their periph-
eries (lower-lying areas). Additionally, sharp edges between the areas of high and
low avoidance indicate the respondents collectively perceive those streets as spe-
cific fear boundaries (namely William Street, Darlinghurst Road, MacLeay Street
and Victoria Street). In contrast, when the land’s slope is gentle it indicates the
respondents were less accurate in drawing or that collective avoidance is gradual
and generalized. This could account for why hotspots B and C, which have defined
edges, coincide with a high density of crime incidents, yet hotspot A, which has
much softer boundaries, is an area with more dispersed incidents of crime.
As discussed in Chapter 3, people experience fear of crime when they encounter
certain environmental cues. These cues may be present in areas where there is no
crime. Examination of the study site during interviewing revealed that both physical
and social environmental cues occurred in a higher density north of William Street.
This is consistent with areas of higher avoidance levels. Thus, it is likely the pres-
ence of signs of disorder and incivility do influence the mismatch between crime
and fear of crime. While this study did not examine the actual presence of environ-
mental cues, it did investigate the perceived presence of environmental cues, which
is equally as likely to provoke fear of crime. This is discussed further next.

Environmental Cues Triggering Fear of Crime

This study is primarily exploratory research into the environmental cues that trigger
people to feel afraid of being robbed, beaten or attacked in Kings Cross. Knowing
which environmental cues are most likely to trigger fear of crime has promising
implications for policy and planning, for example, allowing limited public resources
to be validly directed towards combating the most pertinent environmental cues that
trigger fear.
Table 7.9 lists the 16 social and physical incivilities according to the degree to
which they triggered an avoidance reaction. Three columns are shown in the table
for the day and night. The first column shows the percentage of the respondents indi-
cating that the particular cue was a factor in triggering their fear of being robbed,
beaten or attacked. The percentage is derived from the proportion of respondents
adopting avoidance behaviour during the day (n = 138) and night (n = 252), not
the sample as a total (n = 381). Ranks were assigned to each environmental cue
according to this percentage, with the environmental cue ranked ‘1’ triggering fear
of crime in the largest proportion of respondents. These ranks are shown in the sec-
ond column and were used to order the environmental cues from highest to lowest.
Additional ranks are listed in the third column, which were allocated after exam-
ining the fear maps for each environmental cue and ordering the cues according to
the proportion of respondents avoiding each area of the study site. This later rank
reflects the levels of avoidance triggered by each of the environmental cues.
Table 7.9 List and rank of environmental cues by percent and density of avoiding respondents indicating that the environmental cue triggered feelings of fear
of crime and avoidance behaviour

Day Night Change

Environmental cue % (n = 138) Rank by % Rank by density % (n = 252) Rank by % Rank by density % (n = 381)
Results and Discussion

Top
Drug users 63.77 1 3 64.29 1 2 19.42
Gangs 56.52 2 1 56.35 2 1 16.80
Intoxicated persons 53.62 3 2 55.16 3 3 17.06
Middle
Laneways 44.93 4 4 49.60 5 6 16.54
Rubbish/syringes 44.20 5 7 48.41 6 4 16.01
Loitering people, 42.75 6 6 45.63 7 8 14.70
Areas to hide 39.13 7 5 42.46 8 7 13.91
Poor street lighting 35.51 8 10 52.38 4 5 21.78
Blocked escape 34.78 9 8 35.71 10 9 11.02
Pedestrian absence 33.33 10 9 36.51 9 10 12.07

Low
Homeless people 32.61 11 12 30.56 12 14 8.40
Rundown/abandoned buildings 25.36 12 14 32.14 11 11 12.07
Vandalism 24.64 13 13 28.57 13 12 9.97
Spruikers 21.74 14 15 24.60 15 15 8.40
Offensive/degraded shops 21.01 15 11 25.40 14 13 9.19
Sex workers 17.39 16 16 19.84 16 16 6.82

Note: Percent is based on the proportion of the sample adopting avoidance behaviour during the day (n = 138) and night (n = 252), not the sample as a total
(n = 381).
193
194 7 The Kings Cross Study

Table 7.9 is divided into three sections grouping the top, middle and low-
est ranked environmental cues. These groups were identified when examining the
avoidance maps. However, the groups also roughly parallel natural breaks that can
be observed when looking at the avoidance percentages for each environmental
cue. Despite arranging the environmental cues in a slightly different sequence, the
rank by avoidance density and the rank by avoidance percentage categorize each
environmental cue into the same group. Drug users, gangs and intoxicated per-
sons made up the top group, with the highest proportion of respondents avoiding
each area.
According to the avoidance percentages, the top three environmental cues for
both the day and night were drug users, gangs and intoxicated persons; 64% of the
respondents who avoided areas of the study site indicated that the presence of drug
users was a factor in triggering their fear of being robbed, beaten or attacked during
the day and the night. For gangs, these numbers were 57% (day) and 56% (night).
For intoxicated persons, these numbers were 54% (day) and 55% (night). Laneways,
areas to hide, loitering people, rubbish/syringes, blocked escape, pedestrian absence
and poor street lighting made up the middle group, while rundown/abandoned
buildings, homeless people, offensive/degraded shops, vandalism, spruikers and sex
workers made up the lowest group. The presence of sex workers was the environ-
mental cue triggering fear of crime in the least amount for respondents for both the
day (17%) and night (20%).
The last column in Table 7.9 shows the change in percentage of respondents
avoiding areas of the study site from the day to the night. This percentage is derived
from the whole sample (n = 381). Percent-wise, the most pronounced temporal
variation between the day and night fear maps was found with poor street lighting
(19%). An assessment of the fear maps also found a marked increase in avoidance
between the day and night for all of the 16 environmental cues. This was most pro-
nounced for drug users, intoxicated persons, gangs, rubbish/syringes and poor street
lighting. The temporal variation found in the avoidance maps is quite consistent
with that found by an examination of the percent-based changes in the proportion of
respondents who stated each environmental cue triggered their fear of crime.

Mapping the Perceived Presence of Disorder and Incivility

To illustrate differences in the proportion of respondents who avoided the study
site because of the different environmental cues, four cues are shown here using
3D visualization. Two of the top-three ranked environmental cues, drug users and
gangs, were chosen for 3D mapping. Percent-wise, drug users ranked above gangs
by 7% during the day and 8% at night. While this is not a statistically significant
difference, it is interesting because the assessment of the fear maps ranked gangs
well above drug users in terms of the proportion of respondents avoiding each area.
The maps also revealed slight areawise differences in the patterns of avoidance for
these two environmental cues.
In contrast to these highly ranked environmental cues, sex workers were ranked
lowest according to avoidance density and avoidance percentage and are also shown
Results and Discussion 195

here in 3D for comparative purposes. Being the highest and lowest ranked environ-
mental cues, drug users and sex workers were additionally chosen for 3D mapping
to show patterns of avoidance adopted by the male and female respondents, and
visitors and residents.
Overall, the maps show that each of the environmental cues triggered the respon-
dents to adopt very similar patterns of avoidance. As discussed earlier, avoidance
is generally heightened in the northern half of the study site, specifically in the
street blocks making up hotspots A, B and C. While the patterns of avoidance were
reasonably consistent across all of the environmental cues, the areas-to-hide avoid-
ance maps were more spatially general than the others. The areas-to-hide map was
ranked in the middle grouping in Table 7.9. Areas to hide are also shown using
3D visualization to provide a comparison with the maps of the other environmental
cues.
The avoidance hardness maps for each of the environmental cues during the night
were quite consistent. The average avoidance hardness weighting for the majority
of the study site, regardless of the environmental cue in question, was ‘quite hard’.
Many of the avoidance hardness maps for the night also had smaller areas where
the avoidance hardness weighting was in the middle (mid/don’t know) or top (‘very
hard’) of the Avoidance Hardness Index. The ‘mid’ areas generally ran along, or
centred over, Darlinghurst Road. The ‘very hard’ areas were generally situated on
the outer regions of the study site. For the day, there is a little less consistency
between the avoidance hardness maps for each environmental cue. Most of the envi-
ronmental cues have avoidance hardness weightings ranging from ‘mid’ to ‘quite
hard’. A few of the avoidance hardness maps for the day also have areas where the
avoidance hardness weighting was in the ‘not very hard’ category.

The Perceived Presence of Drug Users: Exploring the Maps

Drug users were the top environmental cue triggering fear of crime in terms of
total number of survey respondents adopting avoidance behaviours. The avoidance
maps for all respondents illustrate that they avoided large areas of the study site
because of drug users, and therefore avoidance is high over the entire study site (see
Fig. 7.18). The fact that avoidance is high and fairly generalized mirrors hypothe-
ses made by other researchers who imply drug users are connected with crime and
therefore create fear of crime (Skogan, 1990). This also lends support for the sig-
nal crimes and disorder/incivilities hypotheses. For example, crime is frequently
considered a means of financing drug use and drug users might appear threaten-
ing and unpredictable, therefore triggering fear of crime (Tulloch et al., 1998a,
1998b). In addition, drug users may not even be construed as rational criminals
because they do ‘crazy things that are unnecessary and violent’ (Skogan, 1990).
What is more, an amplification effect could cause drug users to trigger high levels
of fear and avoidance. With the amplification effect, there is a cumulative impact
of numerous weak related signals in close proximity with one another. With regard
to drug users, these weak signals could be signs of drug use like syringes or people
offering drugs.
196 7 The Kings Cross Study

Fig. 7.18 Areas where the respondents stated the presence of DRUG USERS triggered their fear
of being robbed, beaten or attacked – during the day and night

Meanwhile, the avoidance mapping also introduces some more enlightening

and useful information about public fear of crime in Kings Cross. Despite the
fact that some people avoid large areas, others were spatially sensitive, identify-
ing specific areas where drug users triggered their fear of crime. These fear hotspots
Results and Discussion 197

reflect the three peaks displayed in the overall avoidance maps, which is not sur-
prising, considering that drug users are the primary cue triggering fear of crime.
However, what is noteworthy for strategies aimed at combating fear of crime is
an assessment of these areas of fear in terms of actual presence of drug users or
dealers.
In relation to drug users, avoidance in the study site was highest in hotspot C,
which was avoided by 33 respondents during the day and 76 respondents during the
night. Hotspot C begins at William Street, Darlinghurst Road, Roslyn Gardens and
Baroda Street. However, avoidance climbs more steadily over Bayswater Road and
Roslyn Street to peak over Kellett Street, where the rear exit from the Medically
Supervised Injecting Centre (MSIC) is located. Avoidance on Darlinghurst Road
was also greatest at the front entrance to the MSIC. These results indicate that the
public could be avoiding the area because of an awareness of the existence of the
MSIC and a perception or knowledge that drug users frequent its surrounds. This
further reflects a conjecture that the public are attentive to the presence of drug
users, especially those who are about to use drugs, or may have recently used and
are intoxicated upon leaving the MSIC. With such high fear levels, the need for
planning could be recommended to offset this fear and avoidance. A response could
take the form of boosting the social infrastructure available to the drug users, for
example expanding and improving the recovery area in the MSIC so that the users
do not have to loiter in the public streets when they are intoxicated.
While Darlinghurst Road visibly divides hotspots B and C on the drug users
maps, half of the respondents who avoided Darlinghurst Road did so because the
presence of drug users triggered their fear. This result reflects the reality of a
large number of drug users in the immediate locality. For example, a 1999 study
revealed 90% of ambulance overdose call outs occurred with a 300-metre radius of
Darlinghurst Road (Van Beek, 2004). This points towards the recommendation for
improved social infrastructure to help reduce the number of drug users in the area,
such as increased facilities to help users overcome their drug addiction and more
robust policing of the drug trade.
Fear hotspot A is apparent on the drug users avoidance maps. Avoidance is
heightened in two places during the day, one over Sydney Place and the other over
the Forbes and Judge Street walkways, which are avoided by a maximum of 22 and
24 respondents respectively. During the night, avoidance peaks at 65 respondents in
two places, one between Windeyer Street and Rae Place and the other above Kings
Cross Railway Station. Sydney Place was identified as a fear hotspot in a prelimi-
nary study comparing areas of fear with sites of recorded crime. A mismatch over
this area alerted the police to, and substantiated, intelligence that drug dealing was
regularly occurring in the area. This displays the value of the avoidance mapping
method as a police tool. The fact that drug users triggered fear of crime in this
area should be addressed. This is largely because of the location of Plunkett Street
Public School, situated next to Sydney Place, and the subsequent problems arising
from dealing that could impact on school children. Drug-related crime should addi-
tionally be addressed because community members may not be accessing social
infrastructure in the middle of fear hotspots, which includes a public vegetable
198 7 The Kings Cross Study

garden, playground, BBQ setting and tennis court. However of note, drug users
trigger fear and avoidance in this area predominantly during the night and much
less so during the day when these facilities would be in use and school children
about.
The avoidance mapping further identified the presence of drug users caused peo-
ple to avoid hotspot B, specifically Earl Place and Springfield Mall (avoided by
a maximum of 28 respondents during the day and 72 respondents at night). It is
acknowledged that the street-based drug scene in Australia is focused in Springfield
Mall and Plaza, where drug supply and use is concentrated (CoSC, 2005e; NSW
Legislative Assembly Hansard, 2001; Van Beek, 2004). Likewise, The City has
already acknowledged the ongoing incidence of antisocial behaviour in this vicinity
and is already taking action to discourage loitering and improve the public amenity,
something these results confirm is necessary to help reduce fear of crime (CoSC,
2005g). As the Village Centre and a hub for retail activity, pedestrian attendance in
Springfield Mall is particularly important to the vitality of Kings Cross. Therefore,
the social programmes mentioned in Chapter 4, which are designed to ‘activate the
street life’ and could potentially deter the occurrence of drug dealers or users, are
recommended.
Also of note is the apparent lack of a relationship between fear of crime and
the location of syringe bins for drug users. In some areas fear of crime reflects the
location of syringe bins and in others it does not. The maps show that there are
some syringe bins located in high-fear areas, which makes it appear fear of crime
could be related to the presence of syringe bins and drug users making use of them.
However, there are also locations with syringe bins and low fear of crime. This could
be brought about by a number of reasons. First, it could mean that the respondents
did not go to those locations for reasons other than fear of crime. It could also
be that the respondents were not aware of the location of the syringe bins or the
presence of drug users near them. In contrast, the respondents may have been aware
of the syringe bins and drug users yet did not find their presence threatening in those
locations. If this were the case, it could suggest that people react differently to signs
of disorder, for example drug users, when there is evidence that the disorder is being
monitored and managed, like through the establishment of syringe bins. This may
have implications for policy and planning and therefore looking at the secondary
effects of disorder management on fear of crime may be an area worthy of future
research. Lastly, the low levels of fear around certain syringe bins could mean that
drug users do not operate near these bins. This last possibility has also implications
for The City, and it may be worthwhile that the bins are assessed in terms of the
appropriateness of their location.
Another dissonance between the avoidance maps and the actual location of drug-
related crime occurs in Roslyn Street. Avoidance is low along Roslyn Street despite
previous reports that drug dealers regularly congregated there (NSW Legislative
Assembly Hansard, 2001). However, Roslyn Street is considered to be a major
venue for cannabis, rather than heroin dealing (Darcy, Personal communication,
17/3/04). Therefore this dissonance is not surprising considering the fact that the
Results and Discussion 199

respondents were asked whether the presence of junkies,5 rather than drug users
in general, triggered their fear of being robbed, beaten or attacked. Nevertheless,
this uncovers an opening for research focusing on how different types of drugs and
drug-related activity differentially affect fear of crime and avoidance.
Some additional avenues for further research have come about from analyzing
the drug users avoidance map. By breaking down the drug users category into
subgroups more information on public fear of crime can be obtained. For exam-
ple delineating drug users as drug dealers, known drug users or people who look
like drug users could potentially provide some more useful results with regard to
policy and planning. The police could target areas where drug dealers are perceived
to operate. Areas where drug users are located, when they do not spatially coin-
cide with drug dealers, could be improved through the provision of better social
infrastructure. Furthermore drawing on the signal crimes perspective, future survey
questions could be more qualitative when asking respondents how drug users trigger
their fear of crime. This might include asking if the drug users must be outwardly
verbally or physically threatening to trigger fear or if drug users trigger fear even
when they appear quiet and unobtrusive.

The Perceived Presence of Gangs: Exploring the Maps

The avoidance maps illustrate that gangs triggered the respondents to avoid large
areas of the study site, and therefore avoidance was high overall (see Fig. 7.19).
Aggregate avoidance was greatest over hotspots B and C. For hotspot B this was
particularly the case along Earl Place, which is avoided by a maximum of 34 and 72
respondents during the day and night respectively. Peak C begins at William Street,
Darlinghurst Road, Roslyn Gardens and Baroda Street. However, avoidance climbed
more steadily immediately over Bayswater Road and within the area bounded by
Roslyn Street and Ward Avenue to peak over Kellett Way and Kellett Street (avoided
by 34 and 75 respondents during the day and night respectively). Hotspot A is evi-
dent, however not clearly defined. During the day, avoidance reaches a maximum
of 29 respondents in the areas immediately surrounding Sydney Place and Stephen
Street. During the night, avoidance reached a maximum of 65 avoiding respondents
along Windeyer Street and the western side of Victoria Street where it intersects
Earl Street.
This finding that avoidance is high due to the perceived presence of gangs could
suggest that fear of crime experienced by these respondents is irrational given the
current objective levels of risk from gang-related crime. Yet, coming to this conclu-
sion would be premature without considering the historical and situational context
of the area. Kings Cross does have a history of gangs and gang-related organized
crime, knowledge of which assists in our understanding of the high levels of fear.
For example the Razor Gang wars, which occurred over the control of prostitution,

5 Junkies are colloquially known as intravenous drug users, users of other illicit drugs and drug
dealers (Darcy, Personal communication 12/3/04).
200 7 The Kings Cross Study

Fig. 7.19 Areas where the respondents stated the presence of GANGS triggered their fear of being
robbed, beaten or attacked – during the day and night
Results and Discussion 201

‘sly grog’ and cocaine were prominent in Kings Cross between the 1920s and 1930s
and to a lesser extent until the 1960s (SESIAHS, 2005).6 ‘Bikie’ gangs, linked to the
illicit drug trade, have further been associated with Kings Cross from the 1950s.7
Kings Cross has more recently seen instances of gang-related organized drug crime
in the 1980s and 1990s (Wood, 1997).8 This history could account for the respon-
dents’ fear of crime due to gangs at the time of interviewing. This implies that public
fear of crime could be based predominantly on past events and possibly an outdated
reputation of the area as ‘gangland’. In this case public perception, rather than the
reality of actual gangs, would be responsible for fear and avoidance and could be tar-
geted in fear-reduction strategies. This may have implication for previous research,
for example after finding that fear of crime in Glasgow was not reduced following
street lighting enhancements, Nair et al. (1993) concluded that fear of crime is an
intractable and resistant problem. An historical assessment of the area and the effect
of past experiences or levels of disorder on current levels of fear and avoidance
behaviours may have assisted their study.
Comparing the research results with ideas about the presence of gangs can lead
to further observations about public fear of crime. For instance, the results indi-
cate it may not be the frequency with which the public see or hear about gangs but
rather, drawing on the signal crimes perspective, the denotative meaning, connota-
tive meaning and signal value of gang-related crime. The fact that avoidance due to
the presence of gangs was high and quite generalized in Kings Cross suggests gangs
are an environmental cue with a high signal value, probably because the crimes they
denote are severe. Gang-related crimes are often conceptualized as being violent, are
perceived as serious in nature, connote increased risk and elicit greater fear (Clark,
2003). This may also account for Katz et al. (2003) finding that fear of gangs in
Arizona was equally high for residents of high- and low-gang areas and not simply
a consequence of objective threat or prior experiences of victimization. Katz et al.
(2003) explain this finding by insinuating that residents of high-gang areas do not
have increased levels of fear because they recognize that gangs are more likely to
victimize other gang members rather than them. However, this explanation does not
account for the high levels of fear experienced by residents of low-gang areas, which
may be a result of the possible high signal value of gangs and gang-related crime.

6 The Razor Gangs’ Kellett Street riot of 1929, reportedly one of Sydney’s most vicious riots, is
now acknowledged in The City’s pavement signage on the corner of Kellett Street and Bayswater
Road (SESIAHS, 2005).
7 The history of such motorcycle groups and their prominent members is acknowledged on a tree
plaque in Kings Cross.
8 Various newspaper articles published after the interviewing for this study also indicate that gang-
related crime still occurs in Kings Cross. For example, there was the fatal gang shooting of a
member of the Bandidos Motorcycle Club in Chapel Street during April 2006 and the shooting of a
Kings Cross club bouncer by a Hells Angel bikie member in February 2006 (Braithwaite and Baker,
2007; Cummings, 2007). Referring to a police crackdown on bikies through Operation Ranmore,
the Gangs Squad Commander Scott Whyte reported that ‘by week three [of the operation] there
was no evidence of bikies . . . in Kings Cross. That’s something that hasn’t been seen for many
years’ (Braithwaite, 2007).
202 7 The Kings Cross Study

Furthermore, as the avoidance maps indicate, there is a strong behavioural effect

resulting from this environmental cue. This is consistent with Lane and Meeker’s
(2000) findings from their qualitative survey based in California, which found that
respondents reported that they modified their behaviour when entering gangland
during the day and completely avoided certain streets at night to keep away from
gangs. Despite the increase in avoidance during the night that was documented by
Lane and Meeker (2000), it is interesting to note that the difference in avoidance lev-
els during the day and night on the gangs avoidance maps is comparatively less than
that for most of the other environmental cues. This suggests that the temporal com-
ponent to people’s fear of crime due to the presence of gangs is not very prominent,
which could indicate that respondents think gang-related victimization is equally as
likely during the day as at night. Alternatively, the findings could suggest that peo-
ple’s fear of being a victim of serious gang-related crime is severe enough to warrant
avoidance during the day too. These results reflect the premise that gang culture is
based on and contributes to public fear and intimidation (Lane and Meeker, 2003).
Skogan (1990) indicates that gangs can be perceived as ranging from casual
groups of loitering youths to more threatening groups engaging in public drink-
ing, drug use or harassment of passers-by, and to real organized ‘fighting squads’.
Interpretation of the research results could be assisted by further investigations into
public perception of gangs and determining whether the public conceive gangs in
Kings Cross as only representing members of organized criminal groups or also as
groups of people loitering in a localized area. If the latter is true, the high levels
of fear and avoidance are understandable considering the reputation Kings Cross
has as a ‘hang out’ for groups of threatening people who may appear to be gang
members. For example, a restaurant owner in Kellett Way has reported that gangs
of young men harass pedestrians in the area (NSW Legislative Assembly Hansard,
2001). Occasional media exposure of criminal incidents occurring in Kings Cross,
for example the gang rape of a lady in 1979, may also contribute to the fear of crime
triggered by gangs and the consequent avoidance reaction. However despite this,
these severe criminal incidents are not necessarily a frequent or ongoing occurrence.

The Perceived Presence of Sex Workers: Exploring the Maps

Sex workers triggered the least fear of crime of all of the environmental cues, in
terms of total number of survey respondents adopting avoidance behaviours. The
avoidance maps reflect this by displaying low avoidance density during the day
and night in comparison to the other 15 environmental cues (see Fig. 7.20). The
low avoidance density is expected, as sex workers are not frequently discussed in
fear-of-crime literature as being threatening in their own right.
During the day avoidance was very low and constant over the study site, with
peaks A, B and C barely evident. Despite being relatively low, avoidance is slightly
higher along Darlinghurst Road (avoided by 5–7 respondents) and in the street block
that makes up hotspot C (a maximum of 11 respondents) than in the remainder of the
study site. As Darlinghurst Road is the primary location where daytime sex workers
solicit potential clients, this result indicates fear of crime and avoidance reflects the
Results and Discussion 203

Fig. 7.20 Areas where the respondents stated the presence of SEX WORKERS triggered their
fear of being robbed, beaten or attacked – during the day and night

actual presence of sex workers and commercial brothels. Similarly, the avoidance
in hotspot C reflects the location of a sex workers’ outreach centre and the MSIC
on Darlinghurst Road. It is widely acknowledged street prostitution is intimately
connected with drug use and dealing, and therefore it is likely respondents could
link the presence of sex workers with the MSIC (ESNA, Undated).
204 7 The Kings Cross Study

Avoidance in these two areas and over the remainder of the study site increases
during the night. During the night, Darlinghurst Road provides a small partition
between hotspots B and C, being avoided by 22–26 respondents. Avoidance over
hotspot C increased from 11 during the day to 32 respondents during the night. This
increase is to be expected given the increase in number of sex workers operating in
the afternoon and evening hours. However avoidance increased more notably around
Forbes Street near William Street, and along Darlinghurst Road south of William
Street. These two sites fall within another known area for sex workers, which is
bounded by Bourke, William, Victoria and Burton Streets (Perkins, 1991). While
avoidance was also slightly heightened south of William Street, along Victoria Street
and in an area east of Darlinghurst Road, these streets are not documented as having
a presence of sex workers. In converse, avoidance is not particularly noteworthy
along Bourke and Liverpool Streets, where illegal street prostitution is regarded as
a problem (ESNA, 2002).
However despite the low overall fear levels and slight avoidance hotspots in
some areas, the avoidance maps for sex workers do reveal large avoided areas that
are surprisingly general. Drawing from the disorder/incivilities thesis, this could be
because the presence of sex workers can be considered indicative of a breakdown
in standard social norms and possibly law enforcement, and therefore may well
instigate public perceptions of crime and fear of crime. In line with this, ESNA9
states prostitution can be accompanied by heroin trafficking in the streets; the reck-
less disposal of syringes; large volumes of litter; the use of laneways as places
for sex; open-air toilets and ‘shooting galleries’; the menacing presence of ‘min-
ders’ who may be involved in drug dealing and ‘turf wars’; the verbal and physical
abuse of residents by the sex workers; the frequent accosting of female residents by
clients, who assume that all females in the streets are sex workers; and the threat-
ening behaviour and presence of young, often drunk, men who are attracted to the
street as clients or voyeurs (ESNA, 2002). Following on, prostitution has also been
associated with crime, for example there have been numerous rapes and assaults
of street sex workers by their clients (ESNA, 2002). It is perhaps these potential
accompanying factors, which may also independently signify crime, that account
for the generalized fear of crime triggered by sex workers.
As mentioned in the previous section, the City’s Adult Entertainment and Sex
Industry Premises Development Control Plan 2006 aims to reduce any negative
side effects arising from the presence of adult entertainment and sex industry
premises (CoSC, 2006h).10 For example, the plan permits the location of sex indus-
try premises in areas that optimize the safety and security of staff and visitors. Sex

9 ESNA also describes multiple other dangers and nuisances arising from the presence of street
sex workers (ESNA, 2002).
10 This includes commercial brothels, local business brothels, safe house brothels for street work-
ers, escort agencies offering sexual services, restricted premises and sex on premises venues,
bondage and discipline venues, swingers clubs and the like, but does not include private sex
workers home-business premises (CoSC, 2006h).
Results and Discussion 205

industry sites should further ensure that the premises do not have an adverse impact
on the character or amenity of the area and neighbouring properties (CoSC, 2006h).
However, ESNA maintains the City’s plan does not cover the large number of home-
business brothels that do not require development approval and do create adverse
effects (ESNA, 2002). Moreover, Sections 19 and 19A of the Summary Offences
Act 1988 declare street prostitution and curb crawling ‘near or within view from a
dwelling, school, church or hospital’ as an offence (ESNA, 2002). ESNA similarly
argues that this act is not enforced and therefore street prostitution is left uncon-
trolled (ESNA, 2002). ESNA hopes for a unified effort between the community,
police and government to remove prostitution from residential zones to mixed-
use, commercial and industrial zones. While the results from the avoidance maps
show sex workers are by no means a priority environmental cue that triggers fear
of crime, the fact that avoidance still occurs, especially in residential zones, favours
the suggestions presented by ESNA.

The Perceived Presence of Areas to Hide: Exploring the Maps

The avoidance patterns, adopted by respondents who indicated areas to hide trig-
gered their fear of crime, were extremely general during the day (see Fig. 7.21).
As in the other avoidance maps, avoidance is comparatively low south of William
Street and along Darlinghurst Road and Macleay Street. However, unlike in the
other avoidance maps, minimal differentiation was made between peaks A, B and C
and the remainder of the study site north of William Street. With surprisingly little
spatial variance of avoidance during the day, it is difficult to see how an analysis
of this avoidance map can affect policy and planning. This is disappointing given
that the fear mapping studies conducted by Nasar et al. (1993); Fisher and Nasar
(1995) and Nasar and Jones (1997) found that specific areas with low prospect, high
concealment and blocked escape, which included areas to hide, were significantly
associated with increased levels of fear of crime. Fisher and Nasar (1995) propose
that potential victims may feel greater exposure to risk and loss of control over their
immediate environment in these areas and therefore experience heightened fear of
crime. While this may be true, the results from the current study could suggest that
the levels of fear of crime triggered by areas to hide in Kings Cross may not warrant
micro-level changes in avoidance.
Nevertheless with no known reports or studies investigating the whereabouts of
areas to hide in Kings Cross, it is difficult to compare these fear maps with the exis-
tence of areas to hide in reality. An assessment of the areas may or may not reveal
that the respondents’ perceptions reflect the physical environment. If their percep-
tions reflect the physical environment, the avoidance mapping will have been useful
for identifying areas where development should focus on the removal of hiding and
entrapment spots. If not, this could indicate that respondents do not successfully
visualize the areas they avoid when considering this environmental cue. It could
also indicate that the thought of areas to hide affects public fear and avoidance
behaviours even when areas to hide are not present in reality.
206 7 The Kings Cross Study

Fig. 7.21 Areas where the respondents stated the presence of AREAS TO HIDE triggered their
fear of being robbed, beaten or attacked – during the day and night

Fear Experienced by Different Socio-Demographic Groups

Table 7.10 shows that only one (the respondent’s sex) of the five socio-demographic
variables was associated with fear of crime. There was no significant association
between fear of crime and age, housing tenure type, income or residential status. The
respondent’s sex was strongly related to fear of crime (p = 0.0057), with females
Results and Discussion 207

Table 7.10 Chi-square analysis of socio-demographic variables and fear of crime

Percent feeling Percent adopting Percent adopting

unsafe or afraid avoidance during avoidance during
Socio-demographic variable (Global measure) the day the night

Sex p = 0.0057 p = 0.0001 p = 0.0001

Male 28.86 25.37 62.77
Female 42.53 44.89 82.1

Age (over 18 years) p = 0.0526 p = 0.2712 p = 0.1031

18–29 42.11 29.82 69.03
30–59 36.00 40.34 71.93
60+ 25.58 29.89 57.47

Housing tenure p = 0.3689 p = 0.6242 p = 0.3162

Government housing 50.00 26.47 55.88
Non-owner-occupier 35.33 37.5 71.08
Owner-occupier 31.11 35.56 68.94
Backpacker 34.48 31.03 57.14
Community shelter 37.5 12.5 62.5

Residential status p = 0.8829 p = 0.1092 p = 0.074

Resident 35.63 40.23 73.56
Visitor 34.9 30.05 62.57

Income (per annum) p = 0.4864 p = 0.8427 p = 0.0646

$0–$15,599 33.02 33.02 58.49
$15,600–$41,599 41.53 37.82 76.27
$41,600–$77,999 32.97 30.77 69.32
$78,000 or more 33.33 39.39 63.64

Note: While percentages are shown in this table, note that the chi-square analysis calculations
were based on the raw data numbers and not percentages

being more fearful than males. The results show that 43% of females and 29% of
males indicated that they had felt fearful when walking in Kings Cross. In terms
of avoidance, 45% (day) and 82% (night) of women avoided one or more areas in
Kings Cross during the day and night respectively, as opposed to 25% (day) and
63% (night) of men.

Mapping Avoidance Adopted by Selected Groups

The avoidance maps also provided insights into how different demographic groups
experience fear of crime and react to different environmental cues. This was demon-
strated by exploring the drug users and sex workers avoidance maps according to
the residential status and sex of the respondents. In contrast to the chi-square analy-
sis, which indicated that residents of and visitors to Kings Cross have similar levels
of fear, the avoidance maps revealed that they adopted very different avoidance pat-
terns in response to the perceived presence of sex workers and drug users. While
the chi-square analysis showed that the female respondents had much higher levels
208 7 The Kings Cross Study

of fear than the male respondents, the avoidance maps illustrated that overall pat-
terns of avoidance for the males and females were more similar than expected given
the chi-square result. This exploration of the avoidance maps demonstrated that fear
mapping can provide new information concerning public fear of crime that is not
apparent through traditional statistical methods.

Fear of Crime Between Residents and Visitors

The results show that the proportion of residents (36%) and visitors (35%) who
had been afraid when walking in Kings Cross is very similar. This study provides a
specific analysis of fear of being robbed, beaten or attacked. Statistically, the resi-
dents of the inner-east are very slightly more fearful of these crimes, with 40% and
74% of residents avoiding areas during the day and night respectively, as opposed
to 30% and 63% of men. The chi-square analysis further revealed that no signifi-
cant relationship was found between residential status and fear of crime. However,
Figs. 7.22, 7.23, 7.24 and 7.25 indicate that there were very different patterns of
avoidance adopted by visitors and residents in response to both sex workers and
drug users triggering their fear of crime.

Residents Use Local Knowledge to Avoid Specific Areas

Figures 7.22, 7.23, 7.24 and 7.25 show that the residents were limited in their avoid-
ance. During the day, few residents adopted avoidance behaviour due to the presence
of sex workers, with only small parts of the study site actually being avoided (see
Fig. 7.24). Aggregate avoidance was at its limit alongside Darlinghurst Road, which
was avoided by three to four residents. This area reflects the actual whereabouts of
daytime sex workers. During the night, aggregate avoidance was greatest in hotspot
C over Kellett Way (avoided by 8–15 residents), hotspot B (avoided by 9–12 res-
idents) and Darlinghurst Road (avoided by 8–10 residents) (see Fig. 7.25). Those
specific areas with strip clubs, brothels and x-rated shops had pronounced levels of
avoidance among the residents at night.
Residents also avoided localized areas due to the presence of drug users (see
Figs. 7.22 and 7.23). For example, the steep slope of the land around peak C illus-
trates that the residents may have identified the MSIC and avoided a specific area
around it during the day and night. Avoidance was specifically heightened for the
residents in Kellett Street and Kellett Way (where drug users enter and exit the
MSIC), which were avoided by 14 and 39 residents during the day and night respec-
tively. Avoidance was also heightened over Sydney Place in hotspot A (avoided
by 9 and 31 residents during the day and night respectively), over Butler Stairs
and in Walla Mulla Park (where syringe bins are located and where drug users
are known to congregate). Increased avoidance at these sites indicates an intimate
knowledge of the presence of drug users. Similarly, during the day the residents
avoided Darlinghurst Road nearly as much as they did the neighbouring street block
Results and Discussion 209

Fig. 7.22 Areas the RESIDENTS and VISITORS stated the presence of DRUG USERS triggered
their fear of being robbed, beaten or attacked during the DAY

(avoided by 6 residents during the day and 20–27 during the night). This further
points towards a residential awareness of drug users on Darlinghurst Road and sug-
gests that during the day residents may be traversing through and patronizing areas
other than their own neighbourhood centre.
210 7 The Kings Cross Study

Fig. 7.23 Areas the RESIDENTS and VISITORS stated the presence of DRUG USERS triggered
their fear of being robbed, beaten or attacked during the NIGHT

Visitors Indiscriminately Avoid Large Areas

Visitors to Kings Cross account for a majority of those respondents who indicated
that the presence of sex workers and drug users triggered their fear of crime. The
visitors also avoided a much broader area than the residents, and were more spa-
tially general in their avoidance patterns. This was evident on the maps for both
Results and Discussion 211

Fig. 7.24 Areas the RESIDENTS and VISITORS stated the presence of SEX WORKERS
triggered their fear of being robbed, beaten or attacked during the DAY

the sex workers and drug users (see Figs. 7.22, 7.23, 7.24 and 7.25), although an
increase in avoidance over Darlinghurst Road was evident on the sex workers map,
indicating that they too identified this as the primary location for sex workers. The
visitors did not have heightened avoidance of Darlinghurst Road due to the pres-
ence of drug users during the day, which was unlike the residents maps (4 and 19
212 7 The Kings Cross Study

Fig. 7.25 Areas the RESIDENTS and VISITORS stated the presence of SEX WORKERS
triggered their fear of being robbed, beaten or attacked during the NIGHT

visitors avoided the sides of Darlinghurst Road during the day and night respec-
tively). Instead the visitors had increased avoidance over hotspots B and C during
the day (both avoided by a maximum of 8 visitors) and hotspot A during the night
(avoided by 12 visitors). However, avoidance in these hotspots was not well defined
in comparison to avoidance in the surrounding area (see Figs. 7.24 and 7.25).
Results and Discussion 213

This is the same for the avoidance by visitors due to the perceived presence of
drug users. While avoidance for the visitors was greatest in hotspots A (avoided by
up to 16 visitors during the day and 35 during the night) and B (avoided by 21 and
40 visitors at Earl Place during the day and night respectively), neither peak A nor
peak B was visible in comparison to levels of avoidance in the surrounding area.
Peak C was avoided by a maximum of 19 and 38 visitors during the day and night
respectively (see Figs. 7.22 and 7.23).
Overall these results suggest that residents’ fear of crime is more consistent
with the reality of risk of victimization than visitors’ fear of crime. This is likely
because residents are familiar with their environment, as inferred in other studies
(see Ferraro and LaGrange, 2000; Gray and O’Conner, 1990; Gilchrist et al., 1998).
The results support the proposal that knowledge of actual crime rates, or in this case
the presence of drug users, can affect perceptions of risk. Knowledge also plays
a role in reducing fear of crime (Garofalo, 1981). It consequently strengthens the
argument that community involvement in combating crime and the dissemination
of crime-prevention information through newsletters and meetings could be effec-
tive in further reducing fear of crime. Similarly, the results also mirror studies that
propose residents with a strong sense of place attachment to their home and neigh-
bourhood perceive fewer incivilities and have lower fear of crime (Brown et al.,
2003).11 The results additionally insinuate that visitors, who may be less familiar
with the environment, are likely to avoid a larger area than is perhaps warranted by
the risk of victimization in order to reduce their fear levels. If this is the case, it
is difficult to judge whether something should be done to change the extent of this
method of risk minimization. However, the results could also mean that visitors’ fear
of crime and consequently patterns of avoidance are based more on the reputation of
the area rather than on the actual presence of drug users. In this situation, improving
the reputation of Kings Cross could help reduce fear of crime in the region.

Fear of Crime Between Men and Women

A larger proportion of women were fearful, with 43% of females and 29% of males
indicating they had been afraid when walking in Kings Cross. These percentages
confirm previous trends12 (Gray and O’Conner, 1990). This study provides a spe-
cific analysis of fear of being robbed, beaten or attacked. Statistically, women are
more fearful of these crimes, with 45% and 76% of women avoiding areas during
the day and night respectively, as opposed to 25% and 58% of men. The chi-square

11 Those residents who have strong place attachments are often those who have resided in the
neighbourhood through times of decline and high residential turnover, are older and own their own
homes (Brown et al., 2003).
12 Women’s comparative fearfulness is the most consistent result within the research field (see
also: Gibson et al., 2002; Hanson et al., 2000; Kanan and Pruitt, 2002; Toseland, 1982; Riger, 1978
[Braugart et al., 1980; Lebowitz, 1975 cited in Clarke and Lewis, 1982]; [Anderson and Leitch,
1996; Mirrlees-Black et al., 1996; Pain, 1993 cited in Gilchrist et al., 1998]; LaGrange and Ferraro,
1989; Gray and O’Conner, 1990; Pain, 2000).
214 7 The Kings Cross Study

analyses found that the sex of the respondent was significantly associated with fear
of crime (p = 0001),
Overall, the spatial analysis confirms that more women than men adopt avoidance
behaviour because of fear of these crimes in specific areas (see Figs. 7.26, 7.27, 7.28
and 7.29). On average, the female respondents also tried harder to avoid those areas.

Fig. 7.26 Areas that MALES and FEMALES stated the presence of SEX WORKERS triggered
their fear of being robbed, beaten or attacked during the DAY
Results and Discussion 215

Fig. 7.27 Areas that MALES and FEMALES stated the presence of SEX WORKERS triggered
their fear of being robbed, beaten or attacked during the NIGHT

The difference is best indicated by looking at the maps indicating that the presence
of sex workers triggered fear of crime (see Figs. 7.26 and 7.27). In addition to the
differences in overall avoidance levels, the avoidance maps point out some subtle
differences in the patterns of avoidance due to the presence of sex workers. Those
few males that adopt avoidance behaviour do so mainly along Darlinghurst Road
216 7 The Kings Cross Study

Fig. 7.28 Areas the MALES and FEMALES stated the presence of DRUG USERS triggered their
fear of being robbed, beaten or attacked during the DAY

(avoided by a maximum of 5 male respondents during the day and 10 during the
night). The majority of the remainder of the study site was avoided by less than three
male respondents during both the day and the night (no male respondents avoided
the south and north east of the study site during the day).
For the female respondents, avoidance was also greatest along Darlinghurst Road
during the day and night due to the perceived presence of sex workers (avoided
Results and Discussion 217

Fig. 7.29 Areas that MALES and FEMALES stated the presence of DRUG USERS triggered their
fear of being robbed, beaten or attacked during the NIGHT

by 5 and 16 female respondents during the day and night respectively). However,
during the night avoidance was very broad and generalized, extending into much of
peak B and particularly peak C (avoided by 9 female respondents during the day and
23 at night). These slight differences may also reflect that the females were being
more cautious than the males by avoiding much larger areas.
218 7 The Kings Cross Study

The avoidance maps for the drug users also show differences and similarities in
the avoidance behaviours adopted by the two sexes (see Figs. 7.28 and 7.29). During
the day, avoidance patterns adopted by the male and female respondents are varied.
The female respondents were quite spatially general in their avoidance patterns.
While avoidance was highest in the three main hotspots (hotspot A was avoided by
a maximum of 12 females, hotspot B by 17 and hotspot C by 21), the slope of the
surface shows that avoidance was relatively constant throughout the study site. This
is different from the avoidance adopted by the male respondents during the day, who
have more clearly defined levels of increased avoidance over hotspots A, B and C.
For the male respondents, aggregate avoidance during the day was greatest in peak
B, which was avoided by 16 males over Earl Place. Hotspots A and C were avoided
by a maximum of 11 and 12 males respectively. An interesting difference is that
Fitzroy Gardens seems to portray a distinctly safe area on the avoidance map for the
female respondents. This could reflect propositions that signs of ‘naturalness’ and
vegetation, for women at least, help to create a sense of safety and reduce fear of
crime (Kuo and Sullivan, 2001; Nasar, 1998).
However, despite these daytime differences in avoidance, during the night the
avoidance patterns adopted by the male and female respondents in response to the
presence of drug users are more similar. For the males, aggregate avoidance dur-
ing the night was greatest in hotspot C, which is defined by two peaks. The first
of these illustrates a quick increase in avoidance from approximately 20–25 males
over Bayswater Road. The second illustrates another sharp increase in avoidance
to 38 males directly over Kellett Way. A peak of 36 avoiding males also defines
hotspot B, predominantly along Earl Place. Aggregate avoidance reached a maxi-
mum of 28 males over Woolloomooloo, with the gradient of this peak being quite
gradual. Between 17 and 28 males avoided Darlinghurst Road.
It is clear from the night maps that fear triggered by drug users is not solely a
women’s problem. The spatiality of men’s fear and patterns of avoidance are not
overly different from women’s.

Women’s Heightened Fear of Crime

Increased fear of crime among women contrasts statistics showing they are less
likely to be victimized than men (Nelson et al., 2001; Hanson et al., 2000; Mirrlees-
Black and Allen, 1998). This ‘fear-victimization paradox’ prompted the following
explanations for women’s heightened fear (Clarke and Lewis, 1982; Hanson et al.,
2000; LaGrange and Ferraro, 1989).
Accounting for women’s increased fear of crime is a number of proposed phys-
ical vulnerabilities. These are that women may believe that they do not have the
physical strength or self-defence skills necessary to resist or flee such an attack
by a drug user (Garofalo, 1981; Gilchrist et al., 1998; Gray and O’Conner, 1990;
Riger, 1978; Smith and Tortensson, 1997; Toseland, 1982; Will and McGrath, 1995).
Consequently, women may also consider themselves a more attractive target than
men and feel additionally fearful (Gilchrist et al., 1998). Furthermore, women may
feel like they are more likely to be injured than their male counterparts during
Results and Discussion 219

victimization (Garofalo, 1981; Riger, 1978; Smith and Tortensson, 1997). In con-
trast, men are socialized to believe they can physically resist and recover from an
attack and therefore could have lower levels of fear than women (Gilchrist et al.,
1998; Smith and Tortensson, 1997). It is not surprising that fear of crime during the
day was higher for the female than the male respondents. As mentioned previously,
fear triggered by drug users could arise because they are often regarded as being
threatening, unpredictable and likely to attack and rob passers-by.
Drawing on notions of social vulnerability, women are said to fear male-
dominated areas like ‘red light’ districts (Koskela, 1999). This is partly attributed
to the idea that women can be socialized to fear strangers, men and public spaces
(Gilchrist et al., 1998).13 Hale (1996) similarly comments that women have been
socialized to feel dependent on men and powerless in society, particularly in such
male-dominated areas (Katz et al., 2003). A number of researchers further suggest
this powerlessness causes women to perceive they have less control over their per-
sonal space and the public domain (Pain, 2000; 1991; Pain, 1993 cited in Gilchrist
et al., 1998; Toseland, 1982). These vulnerabilities consequently elicit feelings of
being at greater risk of victimization than men. Thus women have elevated levels of
fear of crime. With the large number of brothels and other adult services targeting
male clients, Kings Cross can be considered a male-dominated area that could there-
fore provoke fear in women.14 This could be particularly the case for those specific
sites in which sex workers are perceived to be present. Thus these sites would trigger
heightened fear of crime in the female respondents.
Women’s heightened fear of being robbed, beaten or attacked could originate
from the inclusion of sexual assault in these crimes, the one type of crime that
women are more likely to experience than men (Hanson et al., 2000). Women’s
fear of sexual assault, particularly rape, is high because such crimes are perceived
to be extremely serious and relatively likely, particularly in environments with sex
workers (Braumer, 1978 cited in Gray and O’Conner, 1990; Warr, 1985). Given
the point made in the previous section that sex workers are often accompanied by
‘menacing’ males, it is expected that sex workers would trigger fear of being robbed,
beaten or attacked among women (ESNA, 2002). When thinking about sex work-
ers, sexual acts and ‘menacing’ males the female respondents could become fearful
of being sexually assaulted. They would therefore be inclined to adopt avoidance
behaviours. In contrast, men do not generally consider themselves likely victims of
sexual assault. Their fear is lower and therefore it is expected their levels of avoid-
ance would be lower than that of the female respondents, as was the case in this
study (Hanson et al., 2000; Gilchrist et al., 1998; Riger, 1978).
Women may be more emotionally affected by crime than men due to additional
indirect experiences of victimization, through personal communication or media

13 For further information on women’s increased social vulnerabilities see Katz et al., 2003; Pain,
2000; Gilchrist et al., 1998; and Toseland, 1982.
14 Additionally, 87.5% of the ‘persons of interest’ in assaults in inner Sydney are male (Briscoe
and Donnelly, 2001).
220 7 The Kings Cross Study

reports involving female victims (Ditton and Duffy, 1982 and Stanko, 1987 cited in
Koskela, 1999; Winkel and Vrij, 1990 cited in Gilchrist et al., 1998). This may be as
traumatic as direct victimization (LaGrange and Ferraro, 1989). Women frequently
experience low-level victimization, for example verbal harassment, which reminds
them of their susceptibility to attack and intensifies their fear (Junger, 1987 and
Stanko, 1990 cited in Gilchrist et al., 1998; Smith and Tortensson, 1997).
Women’s elevated levels of collective avoidance may be because women are
more likely than men to exercise this reaction. There are differences in the pre-
cautions men and women employ to prevent victimization (Warr, 1985). Avoidance
behaviours are more commonly employed by women (Pain, 2000; Valentine, 1989;
Warr, 1985). This may be due to a decreased willingness to take risks, which may
account for women’s low victimization rates (Hale, 1996 in Katz et al., 2003;
Koskela, 1999; Valentine, 1992 cited in Koskela, 1999).

Men’s Abnormally High Fear of Crime

While men’s fear of crime is comparatively lower than women’s, a large proportion
of men are fearful and adopt avoidance behaviours. This contradicts usual statistical
reports that men’s fear is low (Gilchrist et al., 1998; Smith, 1987; Warr, 1985).
It is theorized that men believe they can physically resist an attack or that
such societal values of invulnerability prevent them from admitting to feeling fear
(Clemente and Kleiman, 1977; Crawford et al., 1990 cited in Pain, 2000; Gilchrist
et al., 1998; Smith and Tortensson, 1997; Young, 1997 cited in Nelson et al., 2001).
This study supports this assumption. Only 28.5% of males indicated they had felt
fearful in Kings Cross and yet a maximum of 58% avoided areas due to fear of
being robbed, beaten or attacked. This suggests men fail to admit to fear when
asked directly and do when avoidance is the focus. This may also account for men’s
fear being comparatively high in qualitative studies enabling them to talk more
about their reactions, rather than definitively admitting to fear (Gilchrist et al., 1998;
Pain, 2000).
Regardless, 28.5% of men indicating they had felt fearful is a high proportion.
This insinuates men may be changing as a group and are more willing to admit their
vulnerabilities, which could be unique to this area. In contrast, male respondents
may have been aware of high victimization risks and without a comparative study
these assumptions cannot be justified.
It is clear from the statistics and spatial maps that fear is not only a women’s
problem. The male and female respondents collectively avoided the same three
areas in the study site (peaks A, B and C), showing that spatiality of men’s fear
and patterns of avoidance are not overly different from women’s (see Figs. 7.26,
7.27, 7.28 and 7.29). The implication of this is that some study methodologies
fail to identify fear in males quite as well as this survey. This signals the need
for ‘more sensitive qualitative understanding’ of men’s fear, particularly through
spatial analyses (Gilchrist et al., 1998). Additionally, it stipulates that strategies
aimed at combating fear should do so from a variety of perspectives (Gilchrist
et al., 1998).
Integrating the Fear Mapping Results with Policy and Community Crime. . . 221

Integrating the Fear Mapping Results with Policy

and Community Crime and Fear-of-Crime Prevention

The study finds that people are afraid of crime in Kings Cross and provides confir-
mation that people do react through avoidance when they experience this fear. The
avoidance levels in Kings Cross are high, with 36% and 66% of the respondents
avoiding at least one area of the study site during the day and night respectively.
The literature suggests that these elevated levels of fear could have numerous nega-
tive social, physical and economic consequences for Kings Cross. For instance, the
economic growth of the area may be being hindered through reduced numbers of
business patrons. Fear of crime and avoidance may also be disrupting neighbour-
hood cohesion and the sense of community among Kings Cross residents. Likewise,
it could be creating opportunities for disorderly behaviour and serious crime because
of the potential reduction in natural surveillance and social control, which could fur-
ther encourage greater levels of fear and avoidance. While it is not definite that this
will occur, the potential for these and other problems seem to warrant action being
taken to address both crime and fear of crime in Kings Cross.
The avoidance mapping has successfully revealed three main fear hotpots in
Kings Cross. It has pinpointed specific streets that have comparatively high or low
levels of avoidance. In doing so, the maps have provided useful knowledge for pol-
icy, planning and practice and allowed specific management recommendations to be
made based on the research findings.

Addressing Crime

The NSW Police are the primary agency responsible for maintaining social order
and preventing, detecting and investigating crime. Efforts made by the NSW Police
to reduce crime and fear of crime in Kings Cross have been much assisted though the
information gained from the fear-of-crime maps produced in this study. For exam-
ple, avoidance was higher than expected around Sydney Place given the level of
reported crime in the area. Further investigation alerted the police to unreported drug
dealing in the area. The avoidance maps therefore assisted the police in identifying
areas of disorder that they were previously unaware of and focusing intelligence and
foot patrols on those areas.
The combating of fear of crime through the identification and targeting of fear
hotspots in this way can be worked into a variety of the policing models adopted
by different police departments across the world. Of key relevance is the problem-
oriented policing, zero-tolerance, community-oriented and reassurance policing
models. By doing so, police can not only target fear of crime in its own right, but
also possibly predict future locations of crime and prevent crime. This is particu-
larly important as police continue to move from traditional retrospective policing
to proactive and preventative policing. Likewise, it encourages the recognition of
community needs and response to disorder in addition to violations of criminal law.
222 7 The Kings Cross Study

Local government also has a legislative role in responding to crime and

community safety concerns. The City of Sydney therefore complements the work of
the NSW Police in reducing and preventing crime and fear of crime: and works with
community stakeholders to build and strengthen the community, prevent community
harm and enhance quality of life (CoSC, 2006n). This work is outlined in the Safe
City Strategy, which aims to improve actual and perceived safety across the City of
Sydney and provides a framework to direct the work of the City of Sydney Council
in addressing priority crime and safety issues (CoSC, 2006n).
The City of Sydney Council has acknowledged that Kings Cross has high rates
of crime and is carrying out community safety audits to identify issues in the phys-
ical environment that may impact negatively on public safety in the area. These
safety audits are performed in partnership with the police and other stakeholders
(CoSC, 2006n). They are used to identify areas for high-visibility patrolling with
the City Rangers and NSW Police in key entertainment precincts during peak times
for assaults (CoSC, 2008)
The City of Sydney Council has also responded by providing residents with
high quality, targeted information on how to respond to or prevent crime (CoSC,
2006j, 2006k, 2006l). This includes a series of safety fact sheets and a fridge magnet
with key safety contacts, which are developed and disseminated in partnership with
the NSW Police and the NSW Attorney General’s Department Crime Prevention
Division. As part of this education campaign, the council is also focusing on build-
ing closer community networks, for example through a series of ‘Good Neighbour’
BBQs in conjunction with the NSW Police and NRMA Insurance. These ‘meet and
greet’ occasions also provide residents with further information to reduce their risk
of personal and property crime. These initiatives are not only designed to reduce
crime, but also improve a sense of personal capability and improve perceptions of
safety (CoSC, 2006n).
The City of Sydney Council’s approach to urban design is based on CPTED prin-
ciples with the aim of preventing crime and encouraging neighbourhood interaction
in order to maximize casual surveillance so that people feel safer at all times of the
day and night. To promote CPTED, ‘Safer By Design’ trainings are provided to city
planners and other staff by the NSW Police. The council has also developed a series
of CPTED guidelines and checklists that are used when planning its own develop-
ments and when considering private development applications. A key component of
the guidelines is the notification of relevant development applications to the NSW
Police. In this way police can provide comments on the development from a crime-
prevention perspective prior to the application being considered by council (CoSC,
2006n).
The City of Sydney Council is also continuing to provide a 24-hour monitored
closed circuit television (CCTV) network, largely as a preventative measure and as
a substitute for natural surveillance. It is hoped that the presence of CCTV cameras
may deter criminal activity, as any potential offender may be fearful of being iden-
tified and prosecuted, and assist the police to apprehend and convict perpetrators of
crime. CCTV cameras have been installed along Darlinghurst Road, Kings Cross
(CoSC, 2008).
Integrating the Fear Mapping Results with Policy and Community Crime. . . 223

The City of Sydney Council recognizes that crime and fear of crime can be
reduced by engaging residents in community activities. As a result, the council is
providing an extensive network of recreation and learning centres that host activi-
ties ranging from sports, arts, music, fitness, self-defence and street safety advice,
employment programmes, sexual health education, nutrition courses, adult educa-
tion, youth programmes and children’s services, and provide targeted recreation and
diversionary programmes for children and young people. The council offers these
latter programs so that children and young people in particular can engage in legiti-
mate activities as a diversion from risk-taking behaviour (CoSC, 2006n). Recreation
and learning is being promoted at a number of venues in the study site including
Mary McDonald Centre (Corner of Bourke and Charles Streets), Juanita Nelsen
Centre (Corner Nicholson & Dowling Streets), Sydney Place and Walla Mulla Park,
which are all in Woolloomooloo.
Furthermore, the City of Sydney Council is working with the community on a
programme of cultural events which could include outdoor film screening, string
quartets, jazz in the square, carols in the park and street theatre. A City East village
website has also been set up to help people in the community connect with each
other and to provide local information and links to services (CoSC, 2007).
The City of Sydney Council is further partaking in advocacy for increased public
transport services to Kings Cross, thereby providing the public with better services
and encouraging intoxicated persons to leave the public spaces at the end of the
evening. This includes more general bus and train services, additional nightrider
(late night) bus services from the entertainment area and increasing after-hour taxi
ranks (CoSC, 2007).

Targeting Pertinent Signs of Disorder and Incivility

The research finds that environmental cues did trigger fear of crime and avoidance
in Kings Cross. The results also indicated that different environmental cues have
different signal values and triggered different levels of fear of crime, as suggested
by the signal crimes perspective. Generally, this knowledge allows fear-reduction
efforts to focus on those environmental cues that trigger the most fear of crime.
This section briefly discusses each of the 16 environmental cues examined in this
study.15 The environmental cues are discussed in order from the most to the least
likely to trigger fear of crime according to the percent-based rank listed in Table 7.9.
The different strategies used to address the different environmental cues are also
discussed.

15 The environmental cues are not discussed with reference to findings from previous studies exam-
ining the link between environmental cues and fear of crime. This is because comparisons between
studies are problematic due to vastly varying research methods. For instance it is unlikely the
environmental cues would have the same effect on the formless levels of fear examined in most
studies, as on the levels of avoidance examined in this research. Therefore comparing the results is
somewhat ineffective.
224 7 The Kings Cross Study

Top-Ranked Cues
In Kings Cross, drug users, intoxicated persons and gangs triggered the highest lev-
els of fear of crime. Consequently, they have also been the focus of many strategies
aimed at reducing crime and fear of crime.

Reducing Levels of Intoxication

It is not surprising that intoxicated persons16 triggered the most fear of crime, as
Kings Cross has a distinct history of problems associated with these social groups.
In fact, the City of Sydney Council readily acknowledges that a large proportion of
crime, disorder and fear stems from alcohol intoxication and that areas within and
surrounding Kings Cross have serious drug- and alcohol-related issues that compro-
mise the ability of public spaces to be attractive places ‘to work, live and recreate’
(AJC, 2006; CoSC, 2006d, 2006e, 2006f).
Since the time of interviewing and release of the research findings, the City of
Sydney Council has developed specific policies and plans in order to deal with drug-
and alcohol-related issues, for instance the ‘Drug and Alcohol Strategy 2006–2011’
and the draft liquor bill 2005 (CoSC, 2006g, 2005c). First, the council has imple-
mented the City of Sydney Street Drinking Strategy to respond to the specific issues
generated and faced by chronic alcohol-addicted groups and individuals. This strat-
egy incorporates various components. Through this strategy, the City of Sydney
Council is redesigning public space to reduce harmful alcohol consumption via
good landscaping, lighting and change of use. This includes providing public toilets
in entertainment precincts and in areas which have a high level of public drinking
activity (CoSC, 2008).
The City of Sydney Council has also developed the City’s Street Drinking
Strategy 2006–2011. Key hotspots for street drinking have been identified in
Woolloomooloo (Talbot Lane, Bourke Street Park, Tom/Uren Square and Walla
Mulla Park) for the improvement of health and community services responses to
street drinkers and reduce the negative social impact of street drinking (CoSC,
2007).
The City of Sydney Council is continuing to prohibit alcohol consumption in
high-conflict public spaces (spaces with high pedestrian volumes or adjacent res-
idential uses) via the establishment of Alcohol Free Zones (AFZs) on roads and
footways and Alcohol Prohibited Areas in public parks. The council has identified
additional locations for Alcohol Free Zones and has extended the time period in
which those AFZs are effective (CoSC, 2006c). Alcohol Free Zones in the study
site include

16 Intoxicated persons triggered fear of crime in 54% (day) and 55% (night) of the avoiding
respondents.
Integrating the Fear Mapping Results with Policy and Community Crime. . . 225

Effective until December 2014

• Macleay Street from Darlinghurst Road to Wylde Street

• Victoria Street from Darlinghurst Road to McDonald Lane
• Brougham Street from William Street to Brougham Lane
• Brougham Street from Brougham Lane to Rae Place
• Brougham Street from Rae Place to Cowper Wharf Road
• Brougham Lane from Victoria Street to Brougham Street
• Dowling Street from northern side of railway viaduct to Sydney Place
• Francis Street from Riley Street to College Street
• Liverpool Street from Yurong Street to College Street
• Kings Lane from Riley Street to Crown Street
• Little Burton Street

Effective until November 2011

• Darlinghurst Road from Bayswater Road to Liverpool Street

• Victoria Street from Bayswater Road to Burton Street
• Barncleuth Lane from Roslyn Street to Barncleuth Square
• Barncleuth Square from Darlinghurst Road to Ward Avenue
• Bayswater Road from Victoria Street to Kings Cross Road
• Darlinghurst Road from Victoria Street to Ward Avenue (includes road closure
adjacent to Fitzroy Gardens)
• Goderich Lane from Pennys Lane to Ward Avenue
• Kellett Place cul-de-sac
• Kellett Street from Bayswater Road to Ward Avenue
• Kellett Way from Kellett Street to Roslyn Street
• Kings Cross Road from Victoria Street to Roslyn Street
• Mansion Lane cul-de-sac
• Pennys Lane from Kings Cross Road to Bayswater Road
• Roslyn Street from Darlinghurst Road to Kings Cross Road
• Ward Avenue from Barncleuth Square to Kings Cross Road
• Earl Place from Earl Street to Springfield Avenue
• Earl Street from Victoria Street to Orwell Street
• Hughes Lane from Orwell Street to Hughes Street
• Hughes Place cul-de-sac
• Hughes Street from Victoria Street to Macleay Street
• Llankelly Place from Darlinghurst Road to Orwell Street
• Orwell Lane from Orwell Street to Hughes Street
• Orwell Street from Macleay Street to Victoria Street
• Springfield Avenue from Darlinghurst Road (Springfield Plaza) to Spring-
field Mall
• Springfield Mall from Llankelly Place to Earl Street
• All streets and laneways bounded by Cathedral Street, Forbes Street, Palmer
Street and William Street including Bourke Street, Burrahpore Lane, Corfu
Street, Cross Lane, Egan Place, St Kilda Lane, Talbot Place and William Lane
226 7 The Kings Cross Study

• Dowling Street from Pring Street to Cowper Wharf Road

• Forbes Street from Cowper Wharf Road to William Street
• Nicholson Street from Dowling Street to Bourke Street
• Sydney Place bounded by Stephen Street and the Hills Stairs, McElhone Street
• Cathedral Street from Forbes to Bourke Street

The City of Sydney Council is also taking a comprehensive approach to combating

alcohol-related harm, antisocial behaviour and crime on and immediately around
licensed premises (CoSC, 2006a). For example, the council has partnered in an
Accord with Licensed Premises, which aims to ‘reduce alcohol related crime and
antisocial behavior in and around licensed premises and to improve the perception of
safety’ (CoSC, 2006b). Council support to Liquor Accords includes contributing to
the development of the NSW Government patron education campaign, about appro-
priate behaviour and disseminating the materials across licensed premises (CoSC,
2008). Licensed premises with high levels of assaults and other issues are targeted
in joint operations with NSW Police and the Office of Liquor Gaming and Racing
(CoSC, 2010a).
In an attempt to further curb alcohol-fuelled violence on and around licensed
premises, the City of Sydney Council further introduced a 12-month freeze on liquor
licenses in Kings Cross in June 2009. This freeze meant that no new premises
could obtain a liquor license and that existing license holders could not change
their premises boundaries, increase their venue capacity or alter hours of operation
(CoSC, 2010a).

Minimizing Harm from Drug Use and Dealing

It is not surprising that drug users17 and intoxicated persons18 triggered the most
fear of crime, as Kings Cross has a distinct history of problems associated with these
social groups. For instance, the East Sydney Neighbourhood Association (ESNA)
identifies one of the main problems for pedestrians as being ‘personal risk from
drug dealers and associated criminal elements, drugged street sex workers and their
minders’ (ESNA, 2002).
While the supply of illicit drugs is a crime, and ultimately the responsibility of
the NSW Police, the City of Sydney Council is aiming to reduce this crime by
making it a requirement of premises with a nightclub license to include a Harm
Minimization Plan, which will reduce the likelihood of drug dealing occurring
(COSC, 2006j). The council has also activated public space for legitimate uses,
such as entertainment and outdoor footway dining to discourage drug dealing and
continues to provide a responsive safety camera CCTV network, enabling the NSW
Police to apprehend drug dealers (CoSC, 2008).

17 Drug users triggered fear of crime in 64% (day) and 64% (night) of the avoiding respondents.
18 Intoxicated persons triggered fear of crime in 54% (day) and 55% (night) of the avoiding
respondents.
Integrating the Fear Mapping Results with Policy and Community Crime. . . 227

Understanding Public Perceptions of Gangs

The results strongly indicate that the perceived presence of gangs19 in Kings Cross
also triggered large numbers of people to feel afraid of being robbed, beaten or
attacked and adopt avoidance behaviour. This avoidance reaction was somewhat
unanticipated because neither The City nor the police appear to have publicly rec-
ognized gang activity as a current problem in the Kings Cross area. While The City
and police do accept gangs as an environmental cue signifying the threat of criminal
victimization, they chiefly act to deal with gang-related problems in other parts of
Sydney. In contrast, a review of Australian newspaper articles printed in the three
years following the interviewing does show that information regarding gang-related
crime in Kings Cross is occasionally published.20 The research results suggest that
in order to reduce public fear of crime, action is needed to address gangs in Kings
Cross. Nevertheless, as gangs are not considered part of the current Kings Cross
environment, a deeper look at public perception of gangs may be necessary.

Middle-Ranked Environmental Cues

Revitalization of Laneways and Streets
Laneways21 were top of the middle-ranked environmental cues that triggered fear
of crime in the respondents.22 Prior to this study the City of Sydney Council did
focus on the management of laneways in development plans and policies, for exam-
ple in the ‘City of Sydney Policy for the Management of Laneways in Central
Sydney’23 (CoSC, 1999). Only one of the laneways located in the Kings Cross study
site, McElhone Stairs, is discussed in this policy. References to the management
of McElhone Stairs were limited to stating that the steps were to be retained with
enhanced pedestrian activity.
However since the time of this study, the City of Sydney Council has realized
that laneways can be the location of criminal and antisocial behaviour and evoke
fear of crime. In response to this, the council has begun a Laneways Revitalization
Strategy to reclaim Sydney’s laneways so that they become safe, lively and more
pedestrian friendly (CoSC, 2006n). Laneways will be reactivated through art and
cultural exhibitions, and small businesses such as cafes and bars, specialist retail,
fashion or galleries (CoSC, 2009). In addition, physical improvements will be made
to selected lanes. Works may include improved lighting, better paving, removal of

19 Gangs triggered fear of crime in 57% (day) and 56% (night) of the avoiding respondents.
20 See Braithwaite (2007; Braithwaite and Baker, 2007; Cummings, 2007).
21 Laneways triggered fear of crime in 45% (day) and 50% (night) of the avoiding respondents.
22 According to Darcy’s (2005) study in 2003, laneways were ranked as the equal least common
reason for making the public feel unsafe. Dark laneways were ranked as the equal fourth most
common reason for making the public feel unsafe.
23 Generally, this policy states that laneways will be improved with appropriate lighting (CoS,
1999). This plan may help reduce the extent to which laneways trigger public fear of crime.
228 7 The Kings Cross Study

high planter walls to improve visibility, improved landscaping, planting new trees,
the provision of additional space for outdoor dining, and the installation of traffic-
calming measures to improve pedestrian amenity and respond to business activation
(CoSC, 2010b).
Several streets and laneways within the study site have been targeted for revi-
talization. These include Sydney Place, Forbes Street, Talbot Lane, Bourke Street,
Harmer Street and Cathedral Street, which are all located in the fear hotspot A over
Woolloomooloo (CoSC, 2010c). It is hoped that improvements to these streets, and
others in Woolloomooloo, will increase pedestrian traffic through Woolloomooloo
by residents of the area and visitors travelling to Kings Cross, Potts Point, East
Sydney and Darlinghurst. Increased wayfinding signage will also facilitate this
(CoSC, 2010c). Earl Street and Earl Place, making up fear hotspot B, have also
been targeted for a complete upgrade (CoSC, 2011a).

Safe Collection of Rubbish, Syringes and Injecting Equipment

Rubbish/syringes24 ranked quite high as an environmental cue that triggered the
respondents to feel afraid of crime.25 The City of Sydney Council has recognized
that a poorly maintained environment can impact negatively on people’s percep-
tions of safety and security. The council has a street cleaning service that ensures
the flushing and cleaning of footpaths, malls and plazas and the emptying and steam-
cleaning of street furniture, litter and recycling bins and cigarette ash cylinders on
a regular basis (CoSC, 2006n). In addition, the council is educating Department
of Housing residents on waste management and the upkeep of their residential
premises (CoSC, 2010c).
Similarly, the City of Sydney Council has acknowledged that the presence of
discarded syringes and other drug-injecting equipment poses a serious risk to public
health and safety and has developed a Syringe Management Plan 2005–2010. This
document outlines key strategies designed to reduce the number of syringes inappro-
priately discarded by drug users in public spaces (CoSC, 2005b, 2006g).26 The plan
outlines how the council provides a fast, coordinated and efficient response to pub-
lic injecting and the associated waste; installs community sharps bins in response to
injecting hotspots; and educates residents and businesses on what to do if they find a
discarded syringe, via the ‘Who To Call’ card which promotes the free call 24-hour
Needle Clean Up Hotline. With regards to the sharps bins, 62 community sharps
bins have been placed throughout the Sydney LGA, many of which are located in
the Kings Cross study site (CoSC, 2006g).

24 Rubbish/syringes triggered fear of crime in 44% (day) and 48% (night) of the avoiding
respondents.
25 According to Darcy’s (2005) study in 2003, lack of cleanliness was ranked as the equal second
least common reason for making the public feel unsafe.
26 This plan also demonstrates The City’s ‘commitment to public health, harm reduction and the
improvement of safety and cleanliness of the public domain for the entire community – residents,
visitors and workers alike’ (CoSC, 2006g).
Integrating the Fear Mapping Results with Policy and Community Crime. . . 229

Removal of Areas to Hide

Areas to hide27 are placed towards the top of the middle group of environmental cues
that trigger people’s fear of being robbed, beaten or attacked. A review of council
plans and policies found little reference to addressing areas within public spaces that
can be used as hiding or entrapment places. One recommendation was found in the
‘South Sydney Plan 1997’, in which The City refers to Green Square Town Centre
and states that external lighting needs to make potential hiding spots visible. The
research results suggest that the council also needs to consider how areas to hide
trigger fear of crime in Kings Cross.

Moving Loitering People

The presence of loitering people28 is placed in the middle group of environmental
cues that trigger people to feel afraid of crime.29 NSW has the power to move loi-
tering people on and regularly does so in the Kings Cross Area (Snowball et al.,
2008). The City of Sydney Council also identifies loitering people as a problem
in Kings Cross and targets them through various policies. For example, because
loitering people can cause ‘disturbance’, the council has disallowed people loiter-
ing outside sex industry premises through the ‘City of South Sydney Sex Industry
Policy 2000’. The presence of loitering people is also restricted through planning, as
the council requires that loitering people be addressed by businesses in development
applications. For instance, the council stipulates that the licensee and staff of a pro-
posed licensed hotel in Potts Point ‘take all reasonable steps to ensure that there is
no loitering by persons seeking admittance to the premises’ (CoSC, 2005f).30 Other
approaches focusing on loitering, for example that occurring in Springfield Plaza,
involves the occurrence of concerts as ‘part of an overall program to activate the
space in a very positive way’ (CoSC, 2006e). The results suggest these actions may
also be sensible for addressing fear of crime.

27 Areas to hide triggered fear of crime in 39% (day) and 43% (night) of the avoiding respondents.
28 Loitering people triggered fear of crime in 43% (day) and 46% (night) of the avoiding
respondents.
29 According to Darcy’s (2005) study in 2003, loitering people were ranked as the equal least
common reason for making the public feel unsafe.
30 Similarly, The City also demanded that another application be approved for Kings Cross if
the management or licensee of the premises is responsible for ensuring loitering patrons do not
detrimentally affect the amenity of the neighbourhood (CoSC, 2006e). The management was also
held responsible for ‘the control of noise, loitering and litter generated by patrons of the premises
and shall ensure that people leave the premises and area in an orderly manner’ (CoSC, 2006e).
230 7 The Kings Cross Study

Ensuring Accessibility and Minimizing Blocked Escape

Blocked escape31 was also placed in the middle group of environmental cues that
triggered people to feel afraid of crime in Kings Cross. In the ‘South Sydney
Plan 1997’, The City recommends that building designs, particularly entry points,
minimize the presence of entrapment spots (another term for blocked escape). In
reference to one city area, Green Square Town Centre, The City also states that
entrances to public open space should encourage pedestrian use and also provide
visual security through the establishment of clear sight lines (CoSS, 1997). The
public domain is to be designed to ensure there are no dead ends or similar blocked
escape routes. With regard to fear of crime, the research results indicate blocked
escape can adequately be addressed through such plans.

Lower Ranked Environmental Cues

Better Street Lighting
During the night,32 poor street lighting ranked towards the top of the middle-ranked
environmental cues that triggered fear of crime in the respondents, and towards the
bottom of this category during the day33 . The City of Sydney Council34 recognizes
that good street lighting is a measure which has the effect of enhancing feelings of
safety not only for its visual effect of improving overall surveillance but also as a
means of attracting people to a site. Street lighting in its various forms is therefore
a central consideration of the council and is being addressed through ‘The City of
Sydney Exterior Lighting Strategy’ (CoSC, 2000). The first objective of this strat-
egy is ‘to improve the illumination of the City of Sydney at night to ensure public
safety, public enjoyment, architectural appreciation, and night-time entertainment’.
Additional aims are to illuminate public and pedestrian areas ‘to a standard that
provides a safe and comfortable visual environment’, and ‘to a level that will reduce
the risk of crime to people and property’ (CoSC, 2000). As part of City Lights
Strategy, the council has outlined target minimum levels for public domain lighting
throughout the city and standardized lighting infrastructure. The council is further
targeting the upgrade of street lighting in identified hot-spot locations, laneways and
Department of Housing streets (CoSC, 2006n, 2010c).

31 Blocked escape triggered fear of crime in 35% (day) and 36% (night) of the avoiding
respondents.
32 Poor street lighting triggered fear of crime in 36% (day) and 52% (night) of the avoiding
respondents.
33 The fact that poor street lighting also triggered a fear-of-crime response during the day could
suggest this issue is a very salient issue in the minds of the respondents, that some streets like
small laneways need street lighting during the day, or that there were problems with the survey
interviewing procedure.
34 According to Darcy’s (2005) study in 2003, lighting was ranked as the equal second least
common reason for making the public feel unsafe.
Integrating the Fear Mapping Results with Policy and Community Crime. . . 231

Activation of Public Spaces for Pedestrians

Pedestrian absence35 was at the bottom of the middle-ranked environmental cues
that triggered people to feel afraid of crime. An assessment of many City of Sydney
plans and policies indicates that most of The City objectives are either directly or
indirectly aimed at encouraging pedestrian use of and activity in public spaces.
Similar to the Laneways Revitalisation Strategy, the City of Sydney Council has
been focusing on public parks and spaces for activation through buskers, outdoor
events, markets and organized sporting competitions. These activities are designed
to increase the legitimate use of open spaces and reduce opportunities for criminal
activity (CoSC, 2006n).
Several parks in the study site have also been identified for activation and
improvements. For example, the City of Sydney Council is renewing Walla Mulla
and Bourke Street Parks as part of the Woolloomooloo Improvements Plan to ensure
that they are safe, attractive and accessible. Key improvements include the building
of green walls, upgrade of toilet facilities, removing and lowering retaining walls to
improve visibility and emergency vehicle access, installing new lights to make the
park safer, thinning thick vegetation to increase visibility, installing new furniture
and better paving and reducing clutter in the park to increase open space. Bourke
Street Park will also feature the creation of a secure space for a future community
garden and exercise equipment (CoSC, 2011c).
Upgrade to those parks which are located under the railway viaduct has also been
identified by the City of Sydney Council as a priority. These include Sir John Young,
Tom Uren Square, Forbes Street Park and Talbot Lane. These areas will also feature
options for innovative lighting solutions and public art programmes (CoSC, 2007).
Fitzroy Park and Lawrence Hargrave Reserve in Kings Cross proper are also
being revitalized. Improvements to these parks include making a space for markets
within the park, providing a better play space among the gardens, improving public
toilets and making the landscape more attractive. Pedestrian access to Lawrence
Hargrave Reserve is also being improved and a community garden area and fruit
tree orchard installed (CoSC, 2011b, 2011c).

Maintenance of Rundown and Abandoned Buildings

Of the lower-ranked environmental cues, rundown/abandoned buildings36 triggered
the most fear of crime in the respondents. The City realizes that vacant building
sites and buildings have a negative effect on ‘the quality of the public domain, and
on businesses and residents surrounding these sites’ (CoSC, 2001). In the ‘Central

35 Pedestrian absence triggered fear of crime in 33% (day) and 37% (night) of the avoiding
respondents.
36 Rundown/abandoned buildings triggered fear of crime in 25% (day) and 32% (night) of the
avoiding respondents.
232 7 The Kings Cross Study

Sydney Development Control Plan 1996’37 The City aims to improve the appear-
ance of such sites and, where practicable, ensure that ongoing temporary active uses
or landscaping at the street frontage is provided.38 A review of more recent council
plans and policies developed after the study site has not shown additional refer-
ence to the maintenance of rundown or abandoned buildings. The results from this
research therefore indicate that The City could do more in ensuring these objectives
are met, thereby reducing fear of crime.

Supporting Homeless People

The presence of homeless people39 was ranked in the lower group of environmen-
tal cues that trigger people to feel afraid of crime.40 The City of Sydney Council
is implementing various initiatives to support homeless persons in the area. Three
major services for homeless people include the Homeless Persons Information
Centre, the Homelessness Brokerage Program and the City Street Outreach Service
(CoSC, 2006e). The council has the only dedicated homelessness unit in Australia
and has developed a Homelessness Strategy, which features components that fos-
ter affordable housing, increase accommodation and support options for people
who are homeless and advocate for improved mental health, drug and alcohol
and other essential services in the area. The council is working with key stake-
holders to provide this assistance, including the St. Vincent de Paul Society and
the NSW Department of Housing. In particular, the council is working with the
Woolloomooloo Homelessness Coordination group which involves local homeless
services, police, mental health and drug and alcohol services. Similarly, the council
is beginning initiatives to engage homeless people to gain skills via working on local
projects for example, redeveloping park or garden areas, providing cleaning services
or contributing to other community-based activities (CoSC, 2007). These social ser-
vices are not directed at reducing fear of crime, but rather at assisting homeless
people to find accommodation and live independently. However, it is possible that a
reduction in fear of crime is an unexpected benefit, as suggested in Chapter 4.

37 Consolidated in 2001.
38 The plan states ‘it is important that construction sites and vacant sites present an attractive
appearance to the streets and public areas in order to enhance the amenity of Central Sydney’
(CoSC, 2001).
39 Homeless people triggered fear of crime in 33% (day) and 31% (night) of the avoiding
respondents.
40 As mentioned earlier drug users/homeless were ranked as the most common reason for making
the public feel unsafe in Darcy’s (2005) study in 2003. The results from this study suggest that the
drug users/homeless category in Darcy’s study was ranked high because of the existence of drug
users within it.
Integrating the Fear Mapping Results with Policy and Community Crime. . . 233

Removing Vandalism and Graffiti

Vandalism41 also ranked in the lower group of environmental cues that trigger peo-
ple’s fear of crime. The City of Sydney Council recognizes that public spaces and
streets that are clean and in good condition increase the public’s confidence in the
safety of the city. Under their ‘Strategic Plan 2006–2009’ the council therefore
ascertains that public assets are to be maintained as ‘clean, accessible, safe, aes-
thetic, fit for purpose, and meet community needs’ (CoSC, 2006i). The City states
that, ‘high levels of community ownership of public domains, parks and facilities
reduce the incidence of vandalism and through timely reporting assists in proactive
maintenance of City assets’ (CoSC, 2006i). In this document, little more is said on
how the City plans to achieve this goal. However other policies, such as the Aerosol
Art and Graffiti Resolution (CoSC, 2003) and Graffiti Management Policy (CoSC,
2004) target specific types of vandalism (graffiti) and are more thorough in terms of
outlining strategies. These policies state that in addition to regular street cleaning,
waste collection and litter control, the council also monitors for signs of vandalism
and graffiti. For example, graffiti ‘hotspots’ are inspected every 24 hours and graffiti
removed within 24 hours of identification, or when consent from the building owner
or resident is obtained. The remainder of the LGA is inspected every five days and
graffiti removed within 24 hours of identification (CoSC, 2004, 2006n).

Offensive/Degraded Shops
The presence of offensive/degraded shops42 was not a key environmental cue trig-
gering people’s fear of crime. Offensive/degraded shops can take a variety of forms.
First, the City of Sydney Council recognizes that the public can consider music and
crowd noise from shops, particularly licensed premises, as offensive or a nuisance.
With this in mind, The City puts in place noise restrictions and actively supports the
investigation of noise complaints by other parties like the Department of Gaming
and Racing Legal and Licensing Section (CoSC, 2006h). Second, The City recog-
nizes that shops advertising or displaying products associated with sexual behaviour
can also be offensive. The ‘City of South Sydney Sex Industry Policy 2000’ refers
to Section 578E of the Crimes Act 1900, in restricting the terms of selling or dispos-
ing of these products (CoSS, 2000).43 Similarly, under the objectives of the Central
Sydney Local Environment Plan 1996,44 the impact of premises which degrade the

41 Vandalism triggered fear of crime in 25% (day) and 29% (night) of the avoiding respondents.
42 Offensive/degraded shops triggered fear of crime in 21% (day) and 25% (night) of the avoiding
respondents.
43 The act documents that ‘any person who carries on, or who is engaged in, the business of selling
or disposing of products to which this section applies must not: Advertise, or cause another person
to advertise, in any manner the nature of that business, or exhibit or display any such products: (i)
to a person who has not consented to or requested the exhibition or display, or (ii) in a manner so
that they can be seen from outside the premises of the business by members of the public’ (CoSS,
2000).
44 Consolidated in 2005.
234 7 The Kings Cross Study

amenity of Central Sydney, such as brothels, restricted premises and late opening
pubs, will be minimized. This involves the assurance that these premises ‘are not
concentrated together, and that their cumulative impact is assessed’ (CoSC, 2005a).
The results of this research do not indicate that any additional strategies focusing on
fear of crime in relation to offensive/degraded shops need to be implemented.

Spruikers
The presence of spruikers45 was the second lowest-ranked environmental cue
triggering people’s fear of crime in Kings Cross.46 The City is officially in oppo-
sition to spruikers. Sex-industry related premises, strip clubs and other properties
require development consent from The City to undertake spruiking activities (CoSC,
2006a). Approval may be subject to the applicant’s devising and complying with
a stringent Spruiker Management Plan (CoSC, 2005g). For example under their
Spruiker Management Plan, Playbirds International directs spruikers not to use
swearing or offensive language, spruik at a volume that other pedestrians are dis-
turbed, act as a physical barrier to pedestrians, touch potential customers, approach
disinterested persons and more (CoSC, 2005g). The results of this research indicate
in terms of fear of crime the City of Sydney Council’s attention to the behaviour of
spruikers is sufficient.

Sex Workers
The presence of sex workers47 was the lowest-ranked environmental cue trigger-
ing the respondents’ fear of being robbed, beaten or attacked.48 The City’s Adult
Entertainment and Sex Industry Premises Development Control Plan 2006 controls
the development and operation of sex industry premises (CoSC, 2006h). The plan
is designed to minimize any negative impact arising from these premises (CoSC,
2006h). The results from this study indicate the city is doing a good job in manag-
ing the presence of sex workers in terms of their impact on peoples’ fear of crime.
Conversely, ESNA argues that illegal brothels, street prostitution and curb crawling
are a significant problem in the area (ESNA, 2002).

45 Spruikers triggered fear of crime in 22% (day) and 25% (night) of the avoiding respondents.
46 As mentioned earlier spruikers/intoxicated persons were ranked as the third most common rea-
son for making the public feel unsafe in Darcy’s (2005) study in 2003. The results from this study
suggest that the spruikers/intoxicated persons category in Darcy’s study was ranked high because
of the existence of intoxicated persons within it.
47 Sex workers triggered fear of crime in 18% (day) and 20% (night) of the avoiding respondents.
48 According to Darcy’s (2005) study in 2003, sex workers were ranked as the second most com-
mon reason for making the public feel unsafe. The contrasting result from this study is noteworthy,
however not necessarily surprising given the different research approaches taken in the two studies.
Integrating the Fear Mapping Results with Policy and Community Crime. . . 235

Sensitively Addressing the Environmental Cues

Despite arguing that the highly ranked environmental cues should be targeted when
designing and implementing fear-reduction strategies, it is recognized that there is
a need for sensitivity. There are two ends of a continuum that focus on the target-
ing of social incivilities as a form of social control. One end is dominated by the
thoughts of ‘communitarians’ or ‘universalists’ who propose the rigorous mainte-
nance of social control and the need for all fear-inducing environmental cues to be
eradicated from affected communities, for the common good (Kelling and Coles,
1997). At the other end, the ‘rights’ activists argue that people involved in so-called
disorderly behaviour are being made scapegoats and inappropriately marginalized
through social control (at the expense of their fundamental liberties and rights to
express themselves) (Kelling and Coles, 1997; Pain, 2001).
The ‘rights’ activists argue that social control and the maintenance of public order
can involve the subordination of these groups to the norm, their retribution and/or
their estrangement from the community (Bauman, 2000; Hubbard, 2003; Kelling
and Coles, 1997). Using the example of homeless people, Kelling and Coles (1997)
agree that this ignores the fact that most homeless are decent, responsible and law-
abiding people, with individual emotional, psychological and physical needs that
need to be considered (Gold and Revill, 2003; Kelling and Coles, 1997; Phillips and
Smith, 2003). While not prejudging what is or is not ‘incivil’, the results from this
study have empirically identified drug users and gangs as definite signs of disorder.
However, this research does not recommend police and other authorities indiscrim-
inately target these social groups in an insensitive manner. In the same token, while
stating a large number of people were fearful and avoided Kings Cross because of
these environmental cues, other people may be drawn to the area for the sense of
excitement that encountering these environmental cues may bring.

Targeted Intervention
The avoidance maps illustrated that fear of crime triggered by different environ-
mental cues is expressed through different patterns of avoidance. For instance,
comparing the avoidance maps for sex workers and drug users revealed the very
different levels and patterns of avoidance. This indicates that the situational context
of environmental cues also plays a role in whether they trigger fear of crime and
how that fear manifests through avoidance.
Accordingly, it is also noteworthy that the ‘significant’ environmental cues are
dense along Darlinghurst Road, where collective avoidance is relatively low. It may
be that the aforementioned cues may only trigger fear when they are present in
laneways, and not major thoroughfares. Despite the proposed higher incidence of
social environmental cues along Darlinghurst Road during the night, fear is much
lower than in the surrounding areas. Street lighting is adequate along this main
street and may diminish fear triggered by the other cues (poor street lighting was
ranked fourth during the night; see Table 7.2). Additionally, Darlinghurst Road into
MacLeay Street is the major thoroughfare north of William Street. Thus, while the
236 7 The Kings Cross Study

environmental cues may be present and trigger fear, this may not be enough to deter
people from using the main street.
The environmental cues avoidance maps can be used to compare fear of crime
with actual presence of the environmental cues. If there is a spatial match between
perceptions and reality, the avoidance maps can enable fear-reduction strategies to
target the pertinent environmental cues in the most appropriate areas. This could
increase the chances of successfully combating fear of crime while minimizing
resource expenditure. As mentioned above, in instances of a spatial mismatch, the
avoidance maps could potentially alert the government and police to areas of dis-
order that they were previously unaware of. Similarly, the finding that avoidance
did not always increase around council syringe bins might indicate drug users
are not using some bins and that the bins could be relocated to more suitable
locations.
In other instances, differences between the perceptions and reality of environ-
mental cues may provide information about the content, effect and signal value
of those signs of disorder and incivility. The gangs avoidance maps indicated that
avoidance is very high and generalized throughout the study site despite the fact
that gangs are not considered to be currently operating in the area. This is use-
ful information as it clarifies the need for attempting to combat fear of crime by
altering public perceptions of disorder. Providing the public with information that
gang-related crime in King Cross is rare and unlikely to do this.
The results further provide new information that different socio-demographic
groups experience different levels of fear of crime and, more importantly, adopt
different patterns of avoidance. Realizing social groups react differently to fear of
crime means fear-reduction strategies can target more fearful groups, like visitors
rather than residents. For example, investing in advertising that encourages visitors
to Kings Cross and improves the reputation of the area may be more successful
in decreasing local fear of crime than crime reports distributed in Kings Cross
community meetings.
Similarly, the fear mapping results for the male and female respondents provided
new information that fear of crime is not solely a women’s problem, does affect
a large proportion of men and that both men and women adopt similar patterns of
avoidance. This is interesting for the field of research, as traditional analyses of
fear of crime using global measures indicate that fear of crime is low among male
populations.

A Snapshot from the Future

The following photos of the areas within the study site were taken in 2007, two
years after the time of interviewing (Figs. 7.30, 7.31, 7.32, 7.33, 7.34, 7.35).
Integrating the Fear Mapping Results with Policy and Community Crime. . . 237

Fig. 7.30 Sydney place, Woolloomooloo. Featuring public amenities including the tennis courts,
play ground, community garden, graffiti art and laneways
238 7 The Kings Cross Study

Fig. 7.31 Council signs to control disorder around Sydney place, Woolloomooloo

Fig. 7.32 Public and private CCTV around Sydney place, Woolloomooloo
Integrating the Fear Mapping Results with Policy and Community Crime. . . 239

Fig. 7.33 William Street, looking east

Fig. 7.34 A laneway in Woolloomooloo

240 7 The Kings Cross Study

Fig. 7.35 The railway viaduct in Woolloomooloo

Assessments of Techniques and Approaches Developed in the Kings

Cross Study

The Survey and Interviewing Procedure

By employing a largely quantitative survey, the data gained in this study could be
easily and quickly acquired and analysed. This was beneficial to the study and the
development of an avoidance mapping technique. However, it also meant that many
of the advantages of qualitative research could not be harnessed. For example if the
Integrating the Fear Mapping Results with Policy and Community Crime. . . 241

survey was qualitative, more information could be gained on the respondents’ expe-
riences of fear of crime, their mechanisms to cope with it, their patterns of avoidance
and responses to those environmental cues that trigger their fear. This would be very
valuable information. Therefore the presence of open-ended questions in a future
survey is recommended.
While the street-based survey approach was rigorous and necessary to ensure the
safety of the interviewers, it had one limitation. By interviewing respondents in the
public arena, the interviewers were unable to include the most fearful people in the
research sample. Such people were those too afraid to leave their residences, referred
to in the literature as ‘prisoners of their own homes’ (Joseph, 1997; Stephens, 1999).
This raises the possibility that fear-of-crime surveys carried out in high-crime areas
may be harder to manage than other residential surveys. Postal questionnaires are
thus recommended to elicit a sample truly representative of those members of the
public who are really afraid and confined to their homes. However, as intended in
the research approach chapter, this survey was not intended to be representative of
the regional demographic.
A similar limitation of the survey relates to the chosen sampling region and style.
Sampling was primarily restricted to the main roads located within the study site.
While this was suitable for the purposes of this study, which was to develop an
informative technique for visualizing and mapping spatial fear-of-crime data, it has
some drawbacks that are mentioned. First, had sampling been conducted outside of
the study site boundaries or in the smaller roads within the study site, the results
may have been different. By carrying out most of the sampling within the study
site or along main roads in the region, the sample could have been biased towards
those people who are not afraid of crime and do not avoid the area, particularly the
main streets. For example, the sampling would not include those people who were
actually avoiding the entire Kings Cross area because they were afraid of crime.
In contrast, if surveying were to occur over a more expansive street setting it is
likely that reported avoidance levels would have increased all over the study site,
particularly in the main streets where interviewing primarily took place. Had a postal
survey been conducted this may have been pronounced, providing the methods for
completing the mapping section of the questionnaire were clearly presented.

Cognitive Mapping
Any potential limitations with the resulting avoidance maps, in terms of their accu-
racy, largely rest in the fact that avoidance mapping is dependent upon cognitive
mapping. Due to the subjective nature of spatial cognition, one’s cognitive map can
be regarded as incomplete, distorted, schematized, augmented and overly simpli-
fied (Downs and Stea, 1973; Nasar, 1998). Consequently, cognitive maps are not
representative of reality. Nor are the individual avoidance maps the respondents
drew necessarily representative of their own cognitive maps. Therefore, the result-
ing avoidance maps cannot be regarded with absolute authority when it comes to
policy and planning, but merely an informative guide to be used in triangulation
with other information sources.
242 7 The Kings Cross Study

People have a tendency to overestimate distances, particularly the spatial extent

of familiar and conceptually important areas (Day, 1976). An area worthy of being
avoided because one fears being robbed, beaten or attacked could be considered
as a conceptually important area. Specifically, Block (2000) states that people
usually overestimate short distances and underestimate long distances. This may
have occurred when the respondents illustrated the areas that they avoided. It
was observed that many survey respondents roughly illustrated the areas that they
avoided, rather than going into specific and careful detail. While these possibilities
do not discount the spatial maps produced, they can act to encourage more emphasis
on the analysis to be placed on the central regions of those avoided areas, rather than
on their peripheries. One approach to test validity of the resulting avoidance maps
would be to present the respondents with different looking maps of the same area
or with maps that have a different scale, and compare the results. However, with the
presence of such obvious fear hotspots, safe thoroughfares and cognitive barriers
between safe and unsafe areas, which clearly represent consistent public behaviour,
it is likely this is not necessary.

General Summary of the Kings Cross Study

The Kings Cross study continued to develop a technique for avoidance mapping to
provide a spatiotemporal investigation into people’s fear of being robbed, beaten or
attacked. It explored the environmental cues that trigger people’s fear and their con-
sequent avoidance reaction. By doing this, the research tests the hypothesis that the
spatial visualization of avoidance can provide new information concerning public
fear of crime.
The 2D maps confirmed that all of the environmental cues triggered fear of crime
and that avoidance levels were consistently higher during the night than the day.
They also illustrated that each of the environmental cues triggered different levels
of avoidance. The perceived presence of drug users, intoxicated persons and gangs
triggered the highest levels of avoidance. The perceived presence of sex workers
triggered the lowest levels of avoidance. The avoidance maps further revealed that
the environmental cues triggered distinct patterns of avoidance, showing obvious
fear hotspots, as well as streets perceived to be safe thoroughfares through those
fear hotspots. Likewise, many of the avoidance maps displayed streets that act as
cognitive barriers separating seemingly safe and unsafe areas. This information pro-
vides some new spatially sensitive insight into how people react to fear of crime
through avoidance. It additionally provides an evidence base that can be used by
police and governments when allocating resources to specific environmental cues in
those critical fear hotspots.
The 2D avoidance maps for four environmental cues were selected for fur-
ther exploration using three-dimensional (3D) mapping. These were drug users,
sex workers, areas to hide and gangs. The 3D avoidance maps exposed micro-
scale differences in patterns of avoidance between these environmental cues. An
References 243

exploration of the avoidance reaction adopted by different socio-demographic

groups in response to drug users and sex workers was additionally carried out.
Separate maps were produced for men and women, and residents of, and visitors
to, Kings Cross.

References
Allen Jack and Cottier Inc (AJC). (2006). Draft master plan report: Urban design study – City
East and Darlinghurst / Surry Hills, Allen Jack and Cottier Inc. for City of Sydney Council,
Sydney.
Australian Bureau of Statistics. (2002a). 2001 Census basic community profile 2011 (poa2011),
Australian Bureau of Statistics, Commonwealth of Australia, Canberra.
Australian Bureau of Statistics. (2002b). 2001 Census basic community profile Sydney (ucl
171400), Australian Bureau of Statistics, Commonwealth of Australia, Canberra.
Australian Explorer. (2005). Kings Cross, Australian Explorer, Available at http://www.
australianexplorer.com/Sydney_kings_cross.htm. Accessed: 18/10/05.
Australian Bureau of Statistics (ABS). (2005). The 2005 Crime and Safety Survey (cat no. 4506.0),
Australian Bureau of Statistics, Commonwealth of Australia, Canberra.
Australian Bureau of Statistics (ABS). (2006). General Social Survey (cat no. 4159.0), Australian
Bureau of Statistics, Commonwealth of Australia, Canberra.
Bauman, Z. (2000). Social uses of law and order. Criminology and social theory. D. Garland and
R. Sparks (Eds.). Oxford University Press, Oxford.
Blalock, H. (1967). Toward a theory of minority group relations. New York, NY, Wiley.
Block, R. (2000). Cross national comparison of victims of crime: victim surveys of twelve
countries. The fear of crime. J. Ditton and S. Farrall(Eds.). Ashgate, Aldershot: pp. 13–37.
Braithwaite, D., 2007, August 23: ‘Raids target robbery and bikie gangs’, Sydney Morning Herald.
Braithwaite, D. and J. Baker, August 11, 2007: ‘Elusive, but they’re no pimpernels’, Sydney
Morning Herald.
Brantingham, P. L. and P. J. Brantingham (1993). “Nodes, paths and edges: considerations of the
complexity of crime and the physical environment”. Journal of Environmental Psychology 13:
3–28.
Briscoe, S. and N. Donnelly (2001). “Temporal and regional aspects of alcohol related violence
and disorder”. Alcohol Studies Bulletin 1: 1–14.
Brown, B., D. D. Perkins and G. Brown (2003). “Place attachment in a revitalizing neighborhood:
individual and block levels of analysis”. Journal of Environmental Psychology 23(3): 259–271.
Burgess, M. (2004). Fear of crime: spatiotemporal patterns of avoidance in Kings Cross. Canberra,
Australian National University.
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:
pp. 23–50.
Butel, E. and T. Thompson (1984). King’s Cross album. Sydney, Atrand.
City of South Sydney Council (CoSS). (1997). Strategy for a sustainable city of South Sydney.
Sydney, City of South Sydney.
City of South Sydney Council (CoSS). (2000). City of Sydney adult entertainment and sex industry
premises development control plan 2006. Sydney, South Sydney Council Sex Industry Policy
2000.
City of Sydney Council (CoSC). (1999). The city of Sydney policy for the management of laneways
in central Sydney. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2000). The city of Sydney exterior lighting strategy. Sydney, City
of Sydney Council.
244 7 The Kings Cross Study

City of Sydney Council (CoSC). (2001). Central Sydney development control plan 1996 –
consolidated 2001. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2003). Aerosol art and graffiti. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2004). The city of Sydney graffiti management policy. Sydney,
City of Sydney Council.
City of Sydney Council (CoSC). (2005a). Central Sydney local environment plan 1996 consoli-
dated 2005. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2005b). City of Sydney syringe management plan 2005–2010.
Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2005c). Draft liquor bill 2005. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2005d). Kings Cross and surrounds, City of Sydney
Council, Available at: http://www.cityofsydney.nsw.gov.au/AboutSydney/CityLocalities/
KingsCrossandSurrounds.asp. Accessed: 18/10/05.
City of Sydney Council (CoSC). (2005e). Meeting minutes 05/12/05: Item 9 Development
application 33-35 Darlinghurst Road Potts Point, City of Sydney Council, Available
at: http://www.cityofsydney.nsw.gov.au/Council/documents/meetings/2005/Planning_051205/
05-12-05_PDTC_ITEM9.pdf Accessed: 23/8/06.
City of Sydney Council (CoSC). (2005f). Meeting minutes 08/05/06: Item 5 Development
application 111-139 Darlinghurst Road Potts Point, City of Sydney Council, Available at:
http://www.cityofsydney.nsw.gov.au/Council/documents/meetings/2006/Committee/Planning/
080506/08-05-06_PDTC_ITEM5.pdf. Accessed: 23/8/06.
City of Sydney Council (CoSC). (2005g). Response to community concerns, City of
Sydney Council, Available at: http://www.cityofsydney.nsw.gov.au/Community/documents/
CommunityForums/QA_KingsCross281105.pdf Accessed: 22/8/06.
City of Sydney Council (CoSC). (2006a). Accords with licensed premises, City of
Sydney Council, Available at: http://www.cityofsydney.nsw.gov.au/Community/Safety/
AccordWithLicensedPremises.asp. Accessed: 22/6/06.
City of Sydney Council (CoSC). (2006b). Alcohol and licensed premises, City of Sydney
Council, Available at: http://www.cityofsydney.nsw.gov.au/Community/Safety/AlcoholSafety.
asp#Downloads. Accessed: 22/6/06.
City of Sydney Council (CoSC). (2006c). Alcohol free zones, City of Sydney Council, Available
at: http://www.cityofsydney.nsw.gov.au/Community/Safety/AlcoholFreeZones.asp. Accessed:
22/6/06 and 03/03/11.
City of Sydney Council (CoSC). (2006d). Alcohol summit submission one: responsible supply and
consumption. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2006e). Alcohol summit submission two: alcohol related crime
and anti-social behaviour. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2006f). Alcohol summit submission three: alcohol related crime
and anti-social behaviour. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2006g). “Building a stronger and safer community”, A safer
community, 3.
City of Sydney Council (CoSC). (2006h). City of Sydney adult entertainment and sex industry
premises development control plan 2006. Sydney, City of Sydney Council.
City of Sydney Council (CoSC). (2006i). City of Sydney strategic plan 2006–2009. Sydney, City
of Sydney Council.
City of Sydney Council (CoSC). (2006j). Drug safety, City of Sydney Council, Available
at: http://www.cityofsydney.nsw.gov.au/Community/Safety/DrugSafety.asp. Accessed: 22/6/06
and 03/03/11.
City of Sydney Council (CoSC). (2006k). Education and safety campaigns, City of
Sydney Council, Available at: http://www.cityofsydney.nsw.gov.au/Community/Safety/
EducationAndCampaigns.asp. Accessed: 22/6/06 and 03/03/11.
City of Sydney Council (CoSC). (2006l). Flier: security and safety tips. Sydney, City of Sydney
Council.
References 245

City of Sydney Council (CoSC). (2006m). Safe city program, City of Sydney Council, Available
at: http://www.cityofsydney.nsw.gov.au/Community/Safety/SafeCityProgram.asp. Accessed:
22/6/06.
City of Sydney Council (CoSC). (2006n). Safe City Strategy 2007–2012, City of Sydney
Council, Available at: http://www.cityofsydney.nsw.gov.au/Community/documents/Safety/
SafeCityStrategyFullReport2007-2012.pdf. Accessed: 03/03/11.
City of Sydney Council (CoSC). (2007). Improvements plan Woolloomooloo: Draft key direc-
tions and actions, City of Sydney Council, Available at: http://www.cityofsydney.nsw.gov.
au/Development/documents/UrbanRenewals/WoolloomoolooPlan/CSWL_WORKBOOK_
FINAL030407.pdf. Accessed: 03/03/11.
City of Sydney Council (CoSC). (2008). Drug and alcohol strategy 2007–2010, City of
Sydney Council, Available at: http://www.cityofsydney.nsw.gov.au/Community/documents/
Safety/DAStrategy2007.pdf. Accessed: 03/03/11.
City of Sydney Council (CoSC). (2009). Cringing laneways to life, City of Sydney
Council, Available at: http://www.cityofsydney.nsw.gov.au/Development/documents/
CityImprovements/CBDLanewayUpgrade/InformationSheet_v2.pdf. Accessed: 03/03/11.
City of Sydney Council (CoSC). (2010a). Liquor Freeze, City of Sydney Council, Available
at: http://www.cityofsydney.nsw.gov.au/Community/Safety/LiquorFreeze.asp. Accessed:
03/03/11.
City of Sydney Council (CoSC). (2010b). Revitalisation Initiatives, City of Sydney
Council, Available at: http://www.cityofsydney.nsw.gov.au/Development/CityImprovements/
RoadsAndStreetscapes/CBDLanewayUpgrade/Revitalisation.asp. Accessed: 03/03/11.
City of Sydney Council (CoSC). (2010c). Woolloomooloo Improvements Plan, City
of Sydney Council, Available at: http://www.cityofsydney.nsw.gov.au/Development/
UrbanRenewalProjects/WoolloomoolooPlan.asp. Accessed: 03/03/11.
City of Sydney Council (CoSC). (2011a). Earl Street and Earl Place, Potts Point
Streetscape Upgrade, City of Sydney Council, Available at: http://www.cityofsydney.
nsw.gov.au/Development/CityImprovements/RoadsAndStreetscapes/EarlStreetPlace.asp.
Accessed: 03/03/11.
City of Sydney Council (CoSC). (2011b). Fitzroy Gardens and Lawrence Hargrave Reserve,
City of Sydney Council, Available at: (http://www.cityofsydney.nsw.gov.au/Development/
CityImprovements/ParksAndReserves/FitzroyGardensLawrenceHargravesReserve.asp.
Accessed: 03/03/11).
City of Sydney Council (CoSC). (2011c). Walla Mulla and Bourke Street Parks Upgrade
City of Sydney Council, Available at: http://www.cityofsydney.nsw.gov.au/Development/
CityImprovements/ParksAndReserves/WallaMullaParkRenewal.asp. Accessed: 03/03/11.
Clark, J. (2003). “Fear in fear-of-crime”. Psychiatry, Psychology and Law 102: 267–282.
Clarke, A. H. and M. J. Lewis (1982). “Fear of crime among the elderly”. The British Journal of
Criminology, Delinquency and Deviant Social Behaviour 22(1): 49.
Clemente, F. and M. B. Kleiman (1977). “Fear of crime in the United States: a multivariate
analysis”. Social Forces 56(2): 519.
Couclelis, H. (1998). Aristotelian spatial dynamics in the age of GIS. In spatial and temporal
reasoning in geographic information systems. M. Egenhofer and R. G. Golledge(Eds.). Oxford
University Press, New York, NY: pp. 109–118.
Covington, J. and R. B. Taylor (1991). “Fear of crime in urban residential neighbourhoods: impli-
cations between – and within – neighbourhood sources for current models”. The Sociological
Quarterly 32(2): 231–249.
Cummings, A. (2007). 17 June, 2007: ‘Police get tough on bikies for club chaos’, Sydney Morning
Herald.
Darcy, D. (2003). The highs and lows of dealing with disorder: The Kings Cross police experience,
Proceedings from the Graffiti and Disorder Conference, Brisbane, 18-19 August, 2003.
Darcy, D. (2005). “Policing the socially disadvantaged, the value of rekindling community polic-
ing in Wolloomooloo: a police commander’s perspective”. Current Issues in Criminal Justice
17(1): 144–153.
246 7 The Kings Cross Study

Day, R. A. (1976). “Urban distance cognition: review and contribution”. Australian Geographer
13: 193–200.
Ditton, J. (2000). “Crime and the city”. The British Journal of Criminology 40(4): 692–709.
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. (2004). A spatiotemporal investigation into fear of crime in Wollongong, New South
Wales, Australia. Canberra, Australian National University.
Doran, B. J. and B. G. Lees (2003). Using GIS to investigate spatio-temporal links between dis-
order, crime and the fear of crime. Proceedings from the Conference, Graffiti and Disorder
Conference, Brisbane, 18–19 August 2003.
Downs, R. M. and D. Stea (1973). Cognitive maps and spatial behaviour: processes and prod-
ucts. Image and environment. R. M. Downs and D. Stea(Eds.). Edward Arnold, Great Britain:
pp. 1–8.
Eastern Suburbs Neighbourhood Association (ESNA) (2002). East Sydney community strategic
plan. Sydney, Eastern Suburbs Neighbourhood Association Inc.
Eastern Suburbs Neighbourhood Association (ESNA). (Undated). Policy paper: location of street
worker houses. Sydney, Eastern Suburbs Neighbourhood Association Inc.
Ellis, R. and W. Stacey (1971). Kings Cross – Sydney. Sydney, Thomas Nelson.
Ferraro, K. F. and R. LaGrange (1987). “The measurement of fear of crime”. Sociological
Review 57.
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 micro level physical cues: exploring
the overlooked”. Journal of Research in Crime and Delinquency 32(2): 214–239.
Garofalo, J. (1981). “The fear of crime: causes and consequences”. Journal of Criminal Law and
Criminology 72(2): 839.
Gibson, C. L., J. H. Zhao, N. P. Lovrich and M. J. Gaffney (2002). “Social integration, individual
perceptions of collective efficacy, and fear of crime in three cities”. Justice Quarterly 19(3):
537–564.
Gilchrist, E., J. Bannister, J. Ditton and S. Farrall (1998). “Women and the ‘fear of crime’ –
challenging the accepted stereotype”. The British Journal of Criminology 38(2): 283–298.
Gold, J. R. and G. Revill (2003). “Exploring landscapes of fear: marginality, spectacle and
surveillance”. Capital and Class 80: 27.
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.
Grennan, H. (2001). 10 May, 2001: ‘Critical mass’, The Sydney Morning Herald.
Hale, C. (1996). “Fear of crime: a review of the literature”. International Review of Victimiology
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 1992 civil disturbances”. Journal of
Community Psychology 28(6): 607–623.
Hubbard, P. (2003). “Fear and loathing at the multiplex: everyday anxiety in the post-industrial
city”. Capital & Class 80: 51.
Jacobs, J. (1961). The death and life of great American cities. New York, NY, Vintage Books.
Jochelson, R. (1997). Crime and place: an analysis of assaults and robberies in inner Sydney.
Sydney, New South Wales Bureau of Crime Statistics and Research.
Joseph, J. (1997). “Fear of crime among black elderly”. Journal of Black Studies 27(5): 698–717.
Kanan, J. W. and M. V. Pruitt (2002). “Modeling fear of crime and perceived victimisation risk:
the (in)significance of neighborhood integration”. Sociological Inquiry 72(4): 527–548.
Katz, C. M., V. J. Webb and T. A. Armstrong (2003). “Fear of gangs: a test of alternative theoretical
models”. Justice Quarterly 20(1): 95–130.
References 247

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.
Knox, M. (2003). 19 March, 2003: ‘Turning point’, The Sydney Morning Herald.
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.
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 (2000). “Subcultural diversity and the fear of crime and gangs”. Crime
and Delinquency 46(4): 497–521.
Lane, J. and J. W. Meeker (2003). “Fear of gang the crime: a look at three theoretical models”.
Law and Society Review 37(2): 425–456.
Lewis, D. and M. Maxfield (1980). “Fear in the neighbourhoods: an investigation of the impact of
crime”. Journal of Research in Crime and Delinquency 17: 160–189.
Liska, A. E., A. Sanchirico and M. D. Reed (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.
Lumby, R. (2005). Kings Cross, Kings Cross Partnership Inc., Available at: http://www.
kingscrossonline.com.au/page.asp?e_page=371346. Accessed: 18/10/05.
Mirrlees-Black, C. and J. Allen (1998). Concern about crime: Findings from the 1998 British crime
survey, Home Office Research, Development and Statistics Directorate, London.
Nair, G., J. Ditton and S. Phillips (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 and M. Grannis (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 and C. Thomas (2001). “Identifying micro-spatial and temporal patterns
of violent crime and disorder in a British city centre”. Applied Geography 21: 249–274.
NSW Bureau of Crime Statistics and Research (BOCSAR). (2005a). Specific Crime
Information Tool, Available at: http://www.lawlink.nsw.gov.au/lawlink/bocsar/ll_bocsar.nsf/
vwFiles/NewSouthWales.xls/$file/NewSouthWales.xls. Accessed 12/9/2005.
NSW Bureau of Crime Statistics and Research, (BOCSAR). (2005b). Personal communication
with data provision, ref: tm05-3505, 12 September 2005.
NSW Legislative Assembly Hansard. (2001). Kings Cross policing. Sydney, NSW Legislative
Assembly.
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. (2001). “Gender, race, age and fear in the city”. Urban Studies 38(5/6): 899.
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.
248 7 The Kings Cross Study

Perkins, R. (1991). Working girls: prostitutes, their life and social control. Canberra, Australian
Institute of Criminology.
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.
Phillips, T. and P. Smith (2003). “Everyday incivility: towards a benchmark”. Sociological Review
51(1): 85–108.
Riger, S. (1978). “Women’s fear of crime: from blaming to restricting the victim”. Victimology
3(3/4): 274.
Romer, D., K. H. Jamieson and S. Aday (2003). “Television news and the cultivation of fear of
crime”. Journal of Communication 53(1): 88–104.
Rondeau, M. B., P. L. Brantingham and P. J. Brantinham (2005). “The value of environmental
criminology for the design professions of architecture, urban design, landscape architecture,
and planning”. Journal of Architectural and Planning Research 22(4): 294–304.
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.
Samuels, R. and B. Judd (2002). Public housing estate renewal: Interventions and the epi-
demiology of victimization, Proceedings from the Conference, Housing, Crime and Stronger
Communities Conference, Melbourne,
Skogan, W. G. (1990). Disorder and decline: crime and the spiral of decay in American
neighbourhoods. Los Angeles, CA, University of California Press.
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.
Snowball, L., M. Burgess. and B. Price (2008) ‘Trends in property and illicit drug-related crime in
Kings-Cross: An update’. Crime and Justice Bulletin, No. 120, Attorney General’s Department
of NSW, Sydney.
South Eastern Sydney and Illawarra Area Health Service (SESIAHS). (2005). “Living history”,
Cross Lights Kings Cross Drug and Alcohol Community Focus, June.
Stephens, D. W. (1999.). Measuring what matters. Measuring what matters: proceedings from the
police research institute meetings. R. H. Langworthy (Ed.). National Institute of Justice and
Office of Community Oriented Policing Services, Washington, DC: 55–64.
Sydney Online. (2005). Kings Cross, Sydney Online, Available at: http://www.Sydney.com.au/
kingsx.htm. Accessed: 18/10/05.
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.
Toseland, R. W. (1982). “Fear of crime: who is most vulnerable?”. Journal of Criminal Justice
10(3): 199.
Tourism NSW. (2005). Kings Cross region, Tourism NSW, Available at: http://kingscross.visitnsw.
com/HolidayLocation/TNSW500148CR.htm. Accessed: 18/10/05.
Travel Online Australia. (2005). Kings Cross region, Travel Online Australia, Available at: http://
www.Sydney.visitorsbureau.com.au/page2-04.html. Accessed: 18/10/05.
Tulloch, J., D. Lupton, W. Blood, M. Tulloch, C. Jennett and M. Enders (1998a). Fear of crime,
volume one. Canberra, Centre for Cross Cultural Research.
Tulloch, J., D. Lupton, W. Blood, M. Tulloch, C. Jennett and M. Enders (1998b). 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: pp. 155–166.
Valentine, G. (1989). “The geography of women’s fear”. Area 21(4): 385–390.
References 249

Van Beek, I. (2004). Some facts about the Sydney Medically Supervised Injecting Centre,
Sydney Medically Supervised Injecting Centre, Available at: http://www.sydneymsic.com/
studentsretional.htm. Accessed: 13/9/06.
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.
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.
Whitaker, A. M. (2002). Pictorial history: South Sydney. Alexandria, Kingsdear Books.
Will, J. A. and J. H. McGrath (1995). “Crime, neighborhood perceptions, and the underclass: the
relationship between fear of crime and class position”. Journal of Criminal Justice 23(2): 163.
Wood, T. H. J. J. (1997). Royal commission into the New South Wales police service: volume 1,
corruption. Sydney, The Government of the State of New South Wales.
Zadel, S. (1989). 30 March, 1989: Police hope to stem the Kings Cross crime rate, The Sydney
Morning Herald.
Chapter 8
Future Avenues for Fear Mapping: Potential
Applications and Improvements

Has Collective Avoidance Behaviour Changed in Wollongong

and Kings Cross?

While the Wollongong and Kings Cross studies were conducted over an eight-year
period, the research was cross-sectional in nature. In both cases, surveys were con-
ducted over a period of months and, as a result, the findings are relatively specific
to the time and context of the two projects. Further, many of the interpretations
and recommendations, such as those integrated into the crime prevention plan
for Wollongong (WCC, 2007), were based on the spatiotemporal implications of
behavioural responses to fear. One logical extension of this body of research would
be to explore temporal patterns of collective avoidance behaviour in more depth. At
the simplest level, this could involve rerunning fear mapping surveys in the study
areas several years after the initial research. As outlined in the concluding sections
of Chapter 6, the Wollongong CBD area underwent substantial landscape design
changes as part of a broader city centre revitalization strategy (WCC, 2005). This
process has resulted in significant changes to the character and physical layout of
many areas within the CBD. For example, run-down buildings have been replaced
in McCabe Park, including those shown in Figure 6.9. The council has also removed
many structures that were previously obscuring sight lines and reducing the potential
for natural surveillance (see Figs. 8.1 and 8.2 below for a comparison between 2003
and 2010). Other changes include the approval of high-density apartment complexes
that are a fusion of commercial and residential interests. Figure 8.3 provides an
example of one of these new buildings which has restaurants and cafes in the lower
section and with several floors of apartments above that overlook Crown Street.
Such buildings were designed with an aim to generate a more balanced mix of res-
idential and commercial land use in the CBD area (Irwin et al., 2003) and have
undoubtedly changed the character of the city core. A further activity which may
have influenced collective avoidance behaviour, and the ‘five o’clock flight’, are the
new ‘Market Day Fridays’ in Crown Street Mall. The idea for a produce market
arose from the workshops held during the early stages of the revitalization strategy
and the recognition of the ‘mall problem’ among a range of stakeholders (WCC,
2007). The markets are held weekly between January and July each year and offer

B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series 251
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7_8,

C Springer Science+Business Media, LLC 2012
252 8 Future Avenues for Fear Mapping: Potential Applications and Improvements

Fig. 8.1 A public walkway in McCabe Park in 2003

Fig. 8.2 The same walkway in 2010 after extensive modifications as part of the City Centre
Revitalization Strategy
Has Collective Avoidance Behaviour Changed in Wollongong and Kings Cross? 253

Fig. 8.3 One of many

mixed residential
apartment-commercial
buildings that have been
constructed in the
Wollongong CBD since 2003

‘. . .a selection of handmade jewellery, wooden albums, photography, leatherware,

handmade cards, soaps, toys, handmade cakes, biscuits, honey and farm fresh fruit
and veggies. There is also a selection of hot food stalls including traditional Turkish
cuisine, Asian inspired curries, Korean style food and Hungarian donuts’ (WCC,
2011).
Anecdotally, it would appear that the mall area now has a very different atmo-
sphere on Fridays compared to the time of the Wollongong study when it was a
focus for social disorder leading into the weekend. In general, it would appear that
the design and character of the CBD has improved significantly over the interven-
ing years – but has there been a corresponding change in the collective avoidance
behaviour among people living or working in the area? Have hotspots of collective
avoidance behaviour shrunk in some areas and expanded in others, or in the best
case scenario, have they disappeared entirely from some parts of the city? A follow-
up fear mapping project could be used to investigate such questions in Wollongong.
Similarly, Kings Cross has experienced significant change through ongoing City of
254 8 Future Avenues for Fear Mapping: Potential Applications and Improvements

Sydney order maintenance and gentrification strategies since the time of the study
presented in Chapter 7. It would be instructive to investigate the same questions
posed above in this context as well.
At a broader level, fear mapping could be used as part of monitoring and eval-
uation procedures to assess the impact of urban renewal and gentrification in other
locations. Closed-circuit televisionis being increasingly used in to address crime,
disorder and fear in Europe and North America (Hier et al., 2007). The effects of
CCTV programmes, while often designed to improve safety, may actually serve
to increase fear of crime in some individuals (Williams and Ahmed, 2009). Ditton
(2000) found that the instillation of CCTV cameras in Glasgow did not deliver an
improvement in feelings of safety among survey respondents. This was despite a
majority of the sample indicating that CCTV would make them feel safer. Such
findings point towards a strong need for rigorous monitoring and assessment regard-
ing such programmes. Given the relative simplicity and cost effectiveness of the
cognitive mapping procedure, as well as the increased adoption of GIS software
by government and non-government agencies, fear mapping projects could be con-
ducted by local authorities and integrated with urban renewal, gentrification and
CCTV programs.

Investigating Behavioural Responses in Relation to Different

Types of Crime

In order to overcome limitations associated with cognitive and emotional mea-

sures, in both the Wollongong and Kings Cross study, we used an approach that
was spatially explicit and specifically tapped fear of personal crime. The underly-
ing motivations for collective avoidance behaviour were investigated in detail in
the Kings Cross research and the outputs demonstrated considerable variation in
patterns across time and space according to different types of social and physical
disorder. The diverse results illustrate the complexity of fear of crime but they also
point towards the potential utility of investigating the spatial nature of behavioural
responses in relation to other types of crime. For example, a frequent observation
in the literature is that altruistic fear, namely fear on behalf of others, may be an
important issue for some people but one that is rarely investigated (e.g. Snedker,
2006; Tulloch, 2004; Warr, 2000). What are the likely behavioural responses to,
and consequences of, altruistic fear? One possible connection may relate to parents
placing restrictions on their children in terms of being out in neighbourhoods alone.
This, in turn, could have health-related implications in terms of physical inactiv-
ity and childhood obesity which is an emerging concern in many western countries
(e.g. Nitzko, 2010; Rahman et al., 2011). It would also be possible to investigate
what specific crimes trigger altruistic fear and to customize fear mapping exercises
to tap relevant types of fear. In a similar vein, recent studies looking into fear of
crime among university students suggest that the types of crime that students fear
Investigating Behavioural Responses in Relation to Different Types of Crime 255

most, and how those fears manifest themselves, may differ from other sectors of
society (e.g. Barberet and Fisher, 2009; Cubbage and Smith, 2009; Morrall et al.,
2010; Wu, 2010). Morrall et al. (2010: 827) argue that
[Students] are enculturated with values and codes of behaviour that are internally normative
for both the culture of their host organization and their subcultural lifestyle. Whether they
are attending inner-city or outer-city universities, students tend to congregate in specific
geographical locations both for leisure and residence (‘student ghettos’). . .These factors
are likely to shape the process of victimhood, reactions to actual or perceived risk of
victimization, and health effects of crime and the fear of crime.

Such observations are consistent with informal discussions that frequently took after
the two-step survey procedure used in the Wollongong study. In the case of young
men, who were under-represented in the sample, a number of participants stressed
that while they were not afraid of being robbed, beaten or attacked in the CBD
area, they were very concerned about theft when surfing at nearby beaches. One
respondent described how he was often very afraid about leaving his wallet and
car keys either on the beach or in his car after being a frequent victim of theft.
In his case, he felt compelled to change the times or places where he surfed but
he also made it clear that this was a trade-off with the desire to gain access to the
best surfing conditions. As such, it would seem that replicating fear mapping in
relation to different types of crime that are directly relevant to different subgroups
within society may be an avenue worthy of exploration. It would also seem that
there is scope for more innovative models to investigate behavioural responses to
fear. In relation to female students on university campuses, Cubbage and Smith
(2009) propose
. . .that women adopt a range of strategies to manage their fear, and take personal secu-
rity measures such as moving in groups, monitoring their environment, remaining alert
and aware of surroundings and other people, enrolling in self-defence courses and carrying
mobile telephones and mace spray. These strategies indicate that women generally main-
tain a positive attitude and refuse to assume the role of victim. Women display a sense of
territorial concern for the open space and maintain surveillance of their area.

These observations bear a striking similarity to behavioural theories describing

predator-prey interactions in terrestrial animals. A number of studies have inves-
tigated the effect of moonlight, a factor generally linked with increased exposure
to predation, on the foraging behaviour of small terrestrial mammals (e.g. Price
et al., 1984; Daly et al., 1992; Kotler et al., 1993). These studies have generally
found that animals respond to bright moonlight by reducing their activity levels and
concentrating their foraging in denser micro-habitats. This behaviour is frequently
attributed to a predator response, whereby animals balance the marginal value of
energy against the cost of predation (e.g. Price et al., 1984; Daly et al., 1992; Kotler
et al., 1993; Kotler et al., 1994; Rogowitz, 1997). This trade-off between the need
to access suitable food resources and the risk of predation is termed the ‘ecology of
fear’ by Brown et al. (1999). This approach to predator-prey interactions is based
on the assumption that fierceness is a property of the prey, rather than the predator
(Brown et al., 1999). In a similar sense, the ability of people to adopt avoidance
256 8 Future Avenues for Fear Mapping: Potential Applications and Improvements

behaviours in relation to their fear of crime is a proactive response to risk (Oc and
Tiesdell, 1997). However, as with the logic behind the ‘ecology of fear’, the adop-
tion of avoidance behaviour involves the loss of access to potential resources. In
the case of animals this is potentially detrimental to overall fitness and chances of
survival. In the case of people adopting avoidance behaviour in relation to their fear
of crime the cost can be expressed in terms of isolation, reduced potential to form
social ties with neighbours and increases in levels of mistrusts (Ross and Mirowsky,
2000).
A further parallel with predator-prey interactions relates to prey displaying higher
degrees of vigilance when they encounter a predator (Jarman and Wright, 1993).
Brown et al. (1999) suggest that as a predator becomes more abundant, prey should
become more vigilant. The results from the exploratory activity diary analysis con-
ducted as part of the Wollongong study suggest a similar pattern. In general, the
results demonstrate that people seem more likely to adopt protective measures in sit-
uations where they have less potential to adopt avoidance behaviours and, therefore,
are exposed to a relatively higher risk of victimization. The greater likelihood of
respondents adopting protective measures in relatively more fearful situations sup-
ports the results of studies into micro-level cues to fear of crime (e.g. Nasar et al.,
1993; Nasar and Jones, 1997). Interestingly, Jackson and Gray (2010) conducted
the first empirical study which differentiated fear into something that is dysfunc-
tional compared to something that is functional. They found that recently some of
the respondents took precautions that made them feel safer and neither the precau-
tions nor the worries reduced quality of life. The authors pose a counter argument
to the notion that fear of crime is an unqualified social ill and ask whether some
level of emotional response may comprise a natural defence against crime. In a field
where innovation is rare, it would seem that a transfer of well-developed logic from
predator-prey modelling to investigate the functional role of fear of crime may be
one useful future research direction.

Further Avenues for Investigating Links Between Fear, Crime

and Disorder
There is also potential to integrate fear mapping exercises with longitudinal studies
that have a spatial focus such as ‘The Crime Experiment’, an innovative study cur-
rently being undertaken by the Greater Manchester Police and Professor Lawrence
Sherman from Cambridge University. The aims of the experiment are spelt out in
the following quotation:
The idea is based on earlier American tests of policing crime hotspots which Professor
Sherman developed in the late 1980s. In 1987, he discovered that just 3% of the street
addresses in Minneapolis produced more than half of all calls to police. In parts of the
US, concentrating police on these streets has since successfully cut crime by two-thirds
within the hotspots. What remains unknown from US studies is how much this strategy
may encourage offenders to commit crimes at other locations. . . The idea of systematically
mapping police patrols so that they focus on a list of long-term pressure points has never
Further Avenues for Investigating Links Between Fear, Crime and Disorder 257

been subject to a controlled test in Britain. . . The experiment will divide 200 hotspots into
two groups. The first will be policed normally, but in the second, the police presence will
be intensified with officers stationed in pressure points for many more minutes during high-
crime periods. Researchers will then test the comparative effects over the course of about a
year, measuring the average change in crime over time in one group with that of the other.
(University of Cambridge, 2011).

Research of this nature will undoubtedly provide fresh insights into questions of
displacement and the effects of strategic policing at a local level. From a broken
windows perspective, such projects would also be an ideal setting to explore the role
of fear of crime in relation to patterns of crime and disorder over space and time.
As set out in Chapter 6, the spatial arrangement of collective avoidance hotspots
relative to concentrations of crime or disorder may facilitate the linking up of crime
or disorder hotspots over time. A recent study conducted over two years in Canada
used a very similar approach to ours to investigate spatial patterns of fear of crime
in relation to disorder within a high-crime community (Kohm, 2009). The author
used ‘perceptual mapping’ to look into areas where respondents felt unsafe and why,
different types of social disorder, frequency of victimization and the perceived crime
rate relative to other neighbourhoods. The key finding regarding spatial patterns of
fear was that
Perceptual mapping has considerable potential to shed new light on the micro-level spatial
dynamics of neighbourhood fear. As discussed above, the mapping exercise in this study
suggests that disorder is a powerful explanation for the spatial patterning of fear at the local
level. In addition, the spatial pattern of fear associated with disorder appears more focused
and arguably more accurate than spatial fear attached to crime.

Kohm (2009) emphasizes that while links between disorder, crime and fear are
problematic and that there are many justified criticisms of these relationships
(e.g. Harcourt, 1998; Harcourt and Ludwig, 2006), there is more than simply anec-
dotal evidence for disorder being linked to fear of crime. The author also draws
attention to the ongoing influence of the broken windows theory on policy initia-
tives aiming to address crime in North American central cities and cites O’Shea
(2006: 174) who argues that the theory is the most ‘. . .policy-influencing work in
the crime and place literature’. These trends alone, serve to justify spatially explicit
analyses of fear, crime and disorder.
However, the influence of the broken windows theory and zero-tolerance style
policing extends beyond Northern America, and beyond the city context. A
very topical issue in Australia at the current time is the federal government’s
response to the plight of Aboriginal people in the Northern Territory. In 2007, the
Australian Commonwealth Government introduced the Northern Territory National
Emergency Response Act (NTERA) which has become more commonly known as
‘the intervention’. This act is explicitly race-based in that it applies legislation to all
Aboriginal communities in the NT, termed ‘prescribed areas’ under the Act. This
controversial act was designed to positively impact the lives of Aboriginal people
in remote communities distributed throughout the Northern Territory. One of the
primary intervention measures took the form of welfare quarantining, or income
management, for all Aboriginal people receiving welfare payments. This measure
258 8 Future Avenues for Fear Mapping: Potential Applications and Improvements

was designed to reduce the proportion of social security payments spent on ‘antiso-
cial behaviours’ such as excessive gambling and alcohol consumption. The stated
rationale of this measure was to ‘protect children and make communites safer’
(AIHW, 2009; Commonwealth of Australia, 2007; FaHCSIA, 2009; Stevens and
Young, 2009). Therefore, it can be argued that the federal intervention draws upon
the logic of broken windows in identifying the control of disorder, both social and
physical, as a means of preventing longer-term detrimental impact on communities
across the Northern Territory.
In conjunction with the initiatives implemented as part of the federal interven-
tion, there have been major changes or additions to local council by-laws in remote
towns across the Northern Territory such as Alice Springs, Tennant Creek, Katherine
and Nhulunbuy. In the case of Alice Springs, the Management of Public Spaces By-
laws 2009 (ASTC, 2010) were put into effect on 1 February 2010. The Human
Rights Law Resource Centre (HRLEC, 2009) and the Tangentyere Council (2009)
have raised a number of serious concerns in submissions to the Alice Springs Town
Council when the by-laws were first proposed. Many of these concerns focused on
how the definition and localized policing of antisocial behaviour is likely to have a
disproportionate impact on vulnerable sectors of the community, particularly home-
less people, young people and Aboriginal people. Some of the provisions that drew
the most concern from the HRLRC and Tangentyere Council are outlined below in
the table below (Table 8.1).
Many of these concerns receive support in the context of broader literature on the
management of social disorder. For example, Noaks (2004) investigated the impact
of private security on a local community in the UK and found that the private com-
pany’s style of policing was based on targeting individuals, thereby jeopardizing
individual rights and civil liberties. In South Africa, Berg (2010) found that private
security companies interpreted by-laws in fluid ways when managing public space
and went on to make a strong argument for a greater understanding of the hierarchi-
cal structure of policing (i.e. the relationship between private and state-controlled
measures), as such measures are becoming progressively more spatially intrusive.
In the case of Alice Springs, the concerns expressed by the Tangentyere Council
(2009) and the HRLRC (2009) suggest that the measures enacted under the federal
intervention and the ASTC Management of Public Spaces By-laws, may have given
rise to rapid ecological change within the community and the displacement of social
and physical disorder among vulnerable sections of the community. According to a
number of authors (e.g. Sampson and Groves, 1989; Markowitz et al., 2001), com-
munities that experience rapid ecological change are more likely to show increases
in crime and fear of crime. In turn, fear of crime can potentially give rise to pro-
tective and avoidance behaviours, further atomization of the community and create
potential opportunities for crime (Doran and Lees, 2005; Kelling and Coles, 1997;
Taylor and Covington, 1993). As such, there is an urgent need to gain insights into
the spatial, temporal and social impact of the recent policies implemented in towns
across the Northern Territory like Alice Springs. A Geographic Information Systems
(GIS) and cognitive mapping-based approach is a potentially appropriate means of
investigating these issues.
Further Avenues for Investigating Links Between Fear, Crime and Disorder
Table 8.1 Summary of concerns raised in submissions to the ASTC regarding the management of
public spaces by-laws 2009
ASTC management of public
spaces by-laws 2009 clauses
• Clauses 49–50, 52–56 which
are provisions that create a
number of public space
offences
• Clause 57 which contains
provisions that criminalize
begging
• Clauses 80, 81 and 83 that
extend police powers to
council officers, particularly
‘move on’ powers
259
Concerns expressed
• ‘Given that the activities mentioned above are offences when
they are committed in public spaces, these offences will
disproportionately impact on people who are homeless’
(HRLRC, 2009)
• ‘The Proposed Bylaws will have a substantial impact on
people in public places in the Alice Springs area, many of
whom are Aboriginal’ (HRLRC, 2009)
• ‘Tangentyere is concerned that the by-laws indicated above
will have a significant impact on those individuals who are
forced into the condition of “sleeping rough” due to a lack of
appropriate accommodation’ (Tangentyere Council, 2009)
• ‘The implementation of these by-laws is a contradiction to
the Alice Springs Transformation Plan which recognizes that
homeless people are disadvantaged and “at risk”.’
(Tangentyere Council, 2009)
• ‘It is difficult to understand what will be achieved by fining
people, who have to resort to begging in the first
place. . . .The proposed fine system is a very harsh one which
we believe will disproportionately target Aboriginal . . .Even
if people can somehow pay their fine, their income will be
further reduced, and they may have to again resort to
begging’ (Tangentyere Council, 2009)
• ‘. . .given the financial status of the people who will be
targeted by the begging offence, it will be highly unlikely
that, if found guilty, the individual will have the ability to pay
any fine imposed on them. . . .Clause 57 also discriminates
on the basis of social status, as it disproportionately impacts
on persons who are living in poverty’ (HRLRC, 2009)
• ‘These provisions, in effect, enable Council officials to act in
the capacity of police officers, without having to undergo
law enforcement training’ (HRLRC, 2009)
• ‘These provisions effectively give an authorised officer the
power to forcibly remove a person from a park, garden or
reserve and to ban them from returning to such public place
for up to 6 h. . .The move on powers clearly target persons
sleeping or inhabiting public spaces, and to the extent these
provisions disproportionately affect Aboriginal people’
(HRLRC, 2009)
• ‘Tangentyere is concerned that this draft by-law will bring
those people in contact with the law who may be doing
nothing intrinsically wrong. The execution of this by-law is
likely to create circumstances that could lead to altercations
between people being targeted and the ASTC’ (Tangentyere
Council, 2009)
• ‘Tangentyere feels that these powers should be reserved for
the police. Police receive sufficient training and operate
under a transparent and well-established code of
conduct/procedural framework’ (Tangentyere Council, 2009)
260 8 Future Avenues for Fear Mapping: Potential Applications and Improvements

The federal intervention and its impact across the Northern Territory on the polic-
ing of disorder is used as an example here to emphasize the need for research
into fear of crime to extend beyond high-density urban environments in the United
States, the United Kingdom and Australia. There are increasing calls for this to
take place. For example, Morrall et al. (2010) highlight the need for fear-of-crime
research to be conducted in universities outside city conurbations. Karakus et al.
(2010) note that there is a growing interest in cross-cultural research in criminol-
ogy and validity of existing fear-of-crime models may be limited given this bias of
studies in western countries. The recent increase in fear-of-crime studies in non-
western contexts (e.g. Adu-Mireku, 2002; Karakus et al., 2010; Liu et al., 2009;
Zhang et al., 2009) strongly suggests that there will be a growing need for appro-
priate tools for cross-cultural means of communication in the field and, potentially,
approaches other than traditional survey techniques to investigate fear of crime. An
important contribution made by Kohm (2009) was to combine mapping exercises
on fear of crime with qualitative questions in an attempt to gain a more nuanced
understanding of spatial dimensions of fear. Similarly, Soini (2001) suggested that
combining mapping exercises with other tools such as interviews is an appropri-
ate means of obtaining more complex spatial information. This is reflective of a
broader attempt to more frequently incorporate qualitative approaches into GIS-
based research (Leszczynski, 2009). Cognitive mapping is often used in conjunction
with qualitative interviews in Participatory GIS (PGIS) research, which has been
particularly effective in aid and development applications (e.g. Bauer, 2009). Thus
it would seem that there is genuine potential to adapt cognitive mapping and tech-
niques from behavioural geography in general for use in these emerging areas of
fear-of-crime research in non-western and non-urban contexts. In a recent review
of behavioural geography, Argent and Walmsley (2009) note the vast contribution
of this area to the broader discipline of geography and emphasize that ‘. . . it high-
lighted the need to consider interrelationships between individuals, groups, society
and environment thereby bringing into prominence the ways in which shared envi-
ronmental meanings are contested and negotiated’. However, the authors also note
that behavioural geography has shifted from a ‘. . .cutting edge sub-discipline to a
branch of enquiry that is now much less prominent in mainstream human geog-
raphy’ and that it may increasingly find expression in behaviour-oriented research
of an interdisciplinary nature. While this trend represents a loss for geography in
general, it has been our experience that the GIS-based analyses presented in this
book were richer and more rigorous through drawing techniques and principles from
the field of behavioural geography. If researchers looking into fear of crime are to
address the calls and challenges of investigating the issue beyond densely popu-
lated city environments in western countries, it is likely that approaches based on
behavioural geography will make a valuable contribution.
Broken Windows Theory in the Transit Context 261

Broken Windows Theory in the Transit Context

Evidence for the broken windows theory is often found in transit environments (e.g.
Loukaitou-Sideris, 1999; Loukaitou-Sideris et al., 2002). Indeed, the development
and early focus of the theory has its roots in the New York Transit System (Jang
et al., 2008). The use of public transport is a discretionary activity for many people.
Discretionary activities, no matter how attractive they are as options, are likely to
be forgone if people feel their safety cannot be guaranteed (Cothran and Cothran,
1998). Thus, fear of crime, and avoidance behaviour in this context incurs a direct
cost to transit authorities through loss of ridership (Loukaitou-Sideris et al., 2002).
As argued elsewhere in this book and by others (e.g. NCAVAC, 1998), a key factor
in being able to address the negative aspects of fear of crime is an understanding
of where and when people are afraid of crime. Traditional survey tools are limited
in this respect, as they are not geographically structured. Beyond illustrating the
extent of fear of crime within administrative boundaries, such as census districts or
suburbs, traditional surveys cannot highlight the times and areas that people avoid
due to their fear of crime. Fisher and Nasar (1995) argue that this means much
extant research is limited because studies using standard 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. In turn, this presents a substantial
problem for the institutions responsible for managing fear of crime, including transit
authorities.
GIS began to be used widely in transportation research and management in the
late 1980s (Thill, 2000). Applications frequently involve using GIS as a decision
support system (DSS) in urban transportation policies (e.g. Arampatzis et al., 2004;
Horner and Grubesic, 2001). Often these applications focus on improving the effi-
ciency of services (e.g. Murray, 2003) or evaluating suitable locations for associated
facilities such as park-and-ride lots (e.g. Horner and Grubesic, 2001). O’Sullivan
et al. (2000) argue that the goal of any transport system is not mobility per se, but
access to facilities. Accessibility is considered to be a multifaceted concept which,
in part, includes safety in getting to and from the points of access on the transit sys-
tem to one’s intended destination (Murray, 2003). Loukaitou-Sideris et al. (2002)
emphasizes that perceptions of violence also cause loss of ridership and revenue
by impacting on people’s decisions to use public transportation. There exist a num-
ber of well-developed transport GIS approaches specifically tailored to investigate
the spatiotemporal nature of accessibility and interactions between origin board-
ing and alighting points as well as the transport network itself (e.g. Liu and Zhu,
2004; Yigitcanlar et al., 2006). Adapting such approaches and integrating them with
behavioural measures could further the understanding of fear of crime in transit sit-
uations and may prove a powerful means of informing strategy and policy designed
to improve accessibility. For example, the results from any such research could be
used in situational crime prevention in transit contexts and assistance in ranking
crime- and disorder-related problems along rail lines (e.g. Loukaitou-Sideris et al.,
2002).
262 8 Future Avenues for Fear Mapping: Potential Applications and Improvements

Fear Mapping and Advances in Spatial Technology

A final area where there are numerous potential applications for fear mapping
relates to rapid advances in spatial technology. Elwood (2009:256) writes eloquently
on emerging questions and linkages between GIS-based research and the rapid
development of geospatial technology:
In the world of geospatial technologies, change is afoot. In the past five years, we have seen
the emergence of a wide array of new technologies that enable an ever-expanding range
of individuals and social groups to create and disseminate maps and spatial data. Online
mapping platforms such as Google Maps and Google Earth, Microsoft’s Virtual Earth, or
Wikimapia allow users to add their own geographic information to web-based displays or
modify the contributions of others.

A fascinating series of applications are presented the book ‘Emotional Cartography:

Technologies of the Self’ which displays biometric measurements in relation to
geocoded locations (Nold, 2009). A visualization displaying heightened physiologi-
cal arousal at a busy traffic intersection (Nold, 2009: 13) is conceptually very similar
to experiments by Nasar and Jones (1997) who provided respondents with a dicta-
phone and asked to them verbally record feelings of safety or unsafeness and any
emotional reactions while walking a specified transect at night. In relation to fear
mapping, Kohm (2009) suggests that hard-copy mapping exercises could potentially
be shifted to a computer-based environment and eliminate the need to digitize off
paper. Similarly, Bugs et al. (2010) suggest that Web 2.0 applications facilitate the
creation of sophisticated mapping interfaces that improve participatory mapping and
strengthen interactions between the public and decision makers.
While there is no doubt merit in these claims, and many innovative visualization
avenues to explore, other researchers caution that when used for data collection,
the challenges of web-based citizen science are challenging and, to be effective,
websites must be easy to use, support a range of tasks and ensure data quality
(Newman et al., 2010). Given the general methodological challenges of obtaining
valid fear-of-crime data, web-based mechanisms for fear mapping may prove to be
a confounding influence rather than an improvement to traditional approaches if not
carefully implemented. Elwood (2009) describes a website based on Google Maps
that is used to display the attributes of ‘rotten’ neighbours. Expressed in terms of
different types of disorder such as noise, trash and nosiness, Elwood (2009) empha-
sizes that despite social concerns about such a site, it would generate spatial data that
is patchy and not necessarily representative of the communities where it is obtained
from. A further issue regarding the suitability of web-based fear-of-crime research
relates to the fact that many of the subgroups that generally exhibit higher levels of
fear, such as the elderly, people of non-Caucasian origin and the poor, may also be
subject to the ‘digital divide’ and often have relatively lower IT literacy or access
to the internet (Bruno et al., 2011). A greater contribution may lie in the applica-
tion of more sophisticated GIS-based models to the analysis of fear of crime. Index
models are a powerful means of conducting combinatory analyses (Chang, 2010).
They have been used extensively to investigate issues characterized by complex-
ity and produce outputs that explicitly support decision making in biophysical and
social settings (e.g. Carver, 1991; Jankowski et al., 2001; Jiang and Eastman, 2000;
References 263

Doran and Young, 2010). Algorithms such as G∗ and Geographically Weighted

Regressions (GWR) provide rigorous means of assessing the spatial significance of
locational attributes (e.g. Cromley and McLafferty, 2002; Fotheringham et al., 2006;
O’Sullivan and Unwin, 2010) and go far beyond describing the spatial concentration
of and intensity of phenomena, as is often the case with hotspot and overlay models.
In various studies, Kwan (1999a, 1999b, 2000a, 2000b) has demonstrated the use
of techniques for space–time visualization of activities and has investigated issues
such as employment-related restrictions among women working full or part-time. In
essence the impact of fear of crime is inherently geographic – it varies over time and
space. Kohm (2009) emphasizes the limited nature of literature covering the spatial
impact of fear of crime. It would seem there remain many openings for GIS-based
research to make valuable contributions to the understanding and management of
this pervasive problem, and many more ways in which spatial science can be used
to ‘put fear on the map’.
In the final analysis, it is our hope that the GIS-based approaches used to inves-
tigate fear of crime in the Wollongong and Kings Cross studies may contribute on
several levels. First, the studies provided locally specific information on where and
when people were afraid of crime in the study areas. In Wollongong, the research
findings and spatial outputs were integrated with crime prevention and city cen-
tre revitalization strategies. Being able to identify hotspots of collective avoidance
and overlaps with crime and different types of disorder, the outputs were useful in
defining the degree of institutional involvement and strategic inputs from different
agencies concerned with the management of crime, disorder and fear within the
CBD area. The approach developed in the Wollongong study was transferred and
refined for the research subsequently conducted in Kings Cross where there was
a fear mapping program already underway. The research team worked alongside
the NSW Police Service to gain a deeper underlying of motivations for avoidance
behaviour, as well as responses to different cues. The benefits of 3D visualization
were explored, as well as extending GIS-based analysis of fear of crime to a densely
populated inner-city area characterized by high crime rates. As outlined in this chap-
ter, there are a number of avenues to explore in both of the study sites and to extend
the approach more broadly. In terms of challenges facing fear-of-crime research at
the current time, there appear to be many opportunities for GIS-based behavioural
research to make valuable strategic and methodological contributions to the field.

References
Adu-Mireku, S. (2002). “Fear of crime among residents of three communities in Accra, Ghana”.
International Journal of Comparative Sociology 43: 153–168.
AIHW. (2009). Report on the evaluation of income mangement in the Northern Territory. Canberra.
Alice Springs Town Council (ASTC). (2010). Alice Springs Management Of Public Places By-
laws 2009. Alice Springs Alice Springs Town Council (ASTC).
Arampatzis, G., C. T. Kiranoudis, et al. (2004). “A GIS-based decision support system for planning
urban transportation policies.” European Journal of Operational Research 152(2): 465–475.
Argent, N. M. and D. J. Walmsley (2009). “From the inside looking out and the outside looking in:
whatever happened to ‘Behavioural Geography’?” Geographical Research 47: 192–203.
264 8 Future Avenues for Fear Mapping: Potential Applications and Improvements

Barberet, R. and B. S. Fisher (2009). “Can security beget insecurity? security and crime preven-
tion awareness and fear of burglary among university students in the East Midlands”. Security
Journal 22: 3–23.
Bauer, K. (2009). “On the politics and the possibilities of participatory mapping and GIS: using
spatial technologies to study common property and land use change among pastoralists in
Central Tibet”. Cultural Geographies 16: 229–252.
Berg, J. (2010). “Seeing like private security: evolving mentalities of public space protection in
South Africa”. Criminology & Criminal Justice 10: 287–301.
Brown, J. S., J. W. Laundre and M. Gurung (1999). “The ecology of fear: optimal foraging, game
theory, and trophic interactions”. Journal of Mammalogy 80: 385–389.
Bruno, G., E. Esposito, A. Genovese and K. L. Gwebu (2011). “A critical analysis of current
indexes for digital divide measurement”. Information Society 27: 16–28.
Bugs, G., C. Granell, O. Fonts, J. Huerta and M. Painho (2010). “An assessment of public partici-
pation GIS and web 2.0 technologies in urban planning practice in Canela, Brazil”. Cities 27:
172–181.
Carver, S. (1991). “Integrating multicriteria evaluation with GIS”. International Journal of
Geographical Information Science 3: 321–339.
Chang, K. (2010). Introduction to geographic information systems. New York, NY, McGraw Hill.
Commonwealth of Australia. (2007). Social Security and Other Legislation Amendment (Welfare
Payment Reform) In Act 2007 (No. 130).
Cothran, D. A. and C. C. Cothran (1998). “Promise or political risk for Mexican tourism”. Annals
of Tourism Research 25: 477–497.
Cromley, E. K. and S. L. McLafferty (2002). GIS and public health. New York, NY, Guilford Press.
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: 73–86.
Daly, M., P. R. Behrends, M. I. Wilson and L. F. Jacobs (1992). “Behavioural modulation of pre-
dation risk: moonlight avoidance and crepuscular compensation in a nocturnal desert rodent,
Dipodomys merriami”. Animal Behaviour 44: 1–9.
Department of Families, Housing, Community Services, and Indigenous Affairs (FaHCSIA).
(2009). New model of income management.
Ditton, J. (2000). “Crime and the city”. The British Journal of Criminology 40: 692–709.
Doran, B. J. and B. G. Lees (2005). “Investigating the spatiotemporal links between disorder,
crime, and the fear of crime”. Professional Geographer 57: 1–12.
Doran, B. and M. Young (2010). “Predicting the spatial distribution of gambling vulnerability: an
application of gravity modeling using ABS mesh blocks”. Applied Geography 30: 141–152.
Elwood, S. (2009). “Geographic information science: new geovisualization technologies – emerg-
ing questions and linkages with GIScience research”. Progress in Human Geography 33:
256–263.
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: 214–239.
Fotheringham, A. S., C. Brunsdon and M. Charlton (2006). Geographically weighted regression –
the analysis of spatially varying relationships. West Sussex, Wiley.
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: 291–389.
Harcourt, B. and J. Ludwig (2006). “Broken windows: new evidence from New York City and a
five-city social experiment”. University of Chicago Law Review 73: 271–320.
Hier, S. P., J. Greenberg, K. Walby and D. Lett (2007). “Media, communication and the estab-
lishment of public camera surveillance programmes in Canada”. Media Culture & Society 29:
727–751.
Horner, M. W. and T. H. Grubesic (2001). “A GIS-based planning approach to locating urban rail
terminals”. Transportation 28: 55–77.
References 265

Human Rights Law Resource Centre (HRLEC). (2009). Submission to the Alice Springs Town
Council on the Draft Alice Springs (Management of Public Places) Bylaws 2009 HRLEC.
Irwin, J., M. Wilson and M. Zenardo. (2003). Wollongong City Centre Public Domain Masterplan.
Wollongong: Wollongong City Council.
Jackson, J. and E. Gray (2010). “Functional fear and public insecurities about crime”. The British
Journal of Criminology 50: 1–22.
Jang, H., L. T. Hoover and B. A. Lawton (2008). “Effect of broken windows enforcement on
clearance rates”. Journal of Criminal Justice 36: 529–538.
Jankowski, P., N. Andrienko and G. Andrienko (2001). “Map-centred exploratory approach to
multiple criteria spatial decision making”. International Journal of Geographical Information
Science 15: 101–127.
Jarman, P. J. and S. M. Wright (1993). “Macropod studies at Wallaby Creek. IX. exposure and
responses of eastern grey kangaroos to dingoes”. Wildlife Research 20: 833–843.
Jiang, H. and J. R. Eastman (2000). “Application of fuzzy measures in multi-criteria evaluation in
GIS”. International Journal of Geographical Information Science 14: 173–184.
Karakus, O., E. F. McGarrell and O. Basibuyuk (2010). “Fear of crime among citizens of Turkey”.
Journal of Criminal Justice 38: 174–184.
Kelling, G. L. and C. M. Coles (1997). Fixing broken windows: restoring order and reducing crime
in our communities. New York, NY, Touchstone.
Kohm, S. A. (2009). “Spatial dimensions of fear in a high-crime community: fear of crime or fear
of disorder?” Canadian Journal of Criminology and Criminal Justice 51: 1–30.
Kotler, B. P., Y. Ayal and A. Subach (1994). “Effects of predatory risk and resource renewal on the
timing of foraging activity in a gerbil community”. Oecologia 100: 391–396.
Kotler, B. P., J. S. Brown and W. A. Mitchell (1993). “Environmental factors affecting patch use in
two species of gerbilline rodents”. Journal of Mammalogy 74: 614–620.
Kwan, M. (1999a). “Gender and individual access to urban opportunities: a study using space-time
measures”. The Professional Geographer 51: 210–227.
Kwan, M. (1999b). “Gender, the home-work link think and space-time patterns of nonemployment
activities”. Economic Geography 75: 370–394.
Kwan, M. -P. (2000a). “Interactive geovisualization of activity-travel patterns using three-
dimensional 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: 145–156.
Leszczynski, A. (2009). “Quantitative limits to qualitative engagements: GIS, its critics, and the
philosophical divide”. Professional Geographer 61: 350–365.
Liu, J. H., S. F. Messner, L. N. Zhang and Y. Zhuo (2009). “Socio-demographic correlates of fear of
crime and the social context of contemporary urban China”. American Journal of Community
Psychology 44: 93–108.
Liu, S. X. and X. A. Zhu (2004). “Accessibility analyst: an integrated GIS tool for accessibility
analysis in urban transportation planning”. Environment and Planning B-Planning & Design
31: 105–124.
Loukaitou-Sideris, A. (1999). “Hot spots of bus stop crime: the importance of environmental
attributes”. Journal of the American Planning Association 65: 395–412.
Loukaitou-Sideris, A., R. Liggett and H. Iseki (2002). “The geography of transit crime – documen-
tation and evaluation of crime incidence on and around the green line stations in Los Angeles”.
Journal of Planning Education and Research 22(2): 135–151.
Markowitz, F. E., P. E. Bellair, A. E. Liska and J. Liu (2001). “Extending social disorganization
theory: modeling the relationships between cohesion, disorder, and fear”. Criminology 39: 293.
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.
Murray, A. T. (2003). “A coverage model for improving public transit system accessibility and
expanding access”. Annals of Operations Research 123: 143–156.
266 8 Future Avenues for Fear Mapping: Potential Applications and Improvements

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: 291–323.
National Campaign Against Violence and Crime. (1998). Fear of crime – summary volume.
Canberra, NCAVAC.
Newman, G., D. Zimmerman, A. Crall, M. Laituri, J. Graham and L. Stapel (2010). “User-friendly
web mapping: lessons from a citizen science website”. International Journal of Geographical
Information Science 24: 1851–1869.
Nitzko, S. (2010). “Overweight and obesity in childhood and adolescence”. Praxis Der
Kinderpsychologie Und Kinderpsychiatrie 59: 831–851.
Noaks, L. (2004). “Diversification of British policing – the citizen experience”. Policing-an
International Journal of Police Strategies & Management 27: 264–274.
Nold, C. (2009). Emotional Cartography: Technologies of the Self. Nold, C. http://
emotionalcartography.net/ (last accessed).
Oc, T. and S. Tiesdell (1997). Safer city centres : reviving the public realm. London, Chapman.
O’Shea, T. (2006). “Physical deterioration, disorder and crime”. Criminal Justice Policy Review
17: 173–187.
O’Sullivan, D., A. Morrison and J. Shearer (2000). “Using desktop GIS for the investiga-
tion of accessibility by public transport: an isochrone approach”. International Journal of
Geographical Information Science 14: 85–104.
O’Sullivan, D. and D. J. Unwin (2010). Geographic information systems. Hoboken, NJ, Wiley.
Price, M. V., N. M. Waser and T. A. Bass (1984). “Effects of moonlight on microhabitat use by
desert rodents”. Journal of Mammalogy 65: 353–356.
Rahman, T., R. A. Cushing and R. J. Jackson (2011). “Contributions of built environment to
childhood obesity”. Mount Sinai Journal of Medicine 78: 49–57.
Rogowitz, G. L. (1997). “Locomotor and foraging activity of the White-tailed jackrabbit (Lepus
townsendii)”. Journal of Mammalogy 78: 1172–1181.
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.
Sampson, R. J. and W. B. Groves (1989). “Community structure and crime: testing social-
disorganization theory”. American Journal of Sociology 94: 774–802.
Snedker, K. A. (2006). “Altruistic and vicarious fear of crime: fear for others and gendered social
roles”. Sociological Forum 21: 163–195.
Soini, K. (2001). “Exploring human dimensions of multifunctional landscapes through mapping
and map-making”. Landscape and Urban Planning 57: 225–239.
Stevens, M. and M. Young (2009). “Betting on the evidence: reported gambling problems among
the indigenous population of the northern territory”. Australian and New Zealand Journal of
Public Health 33: 556–565.
Tangentyere Council. (2009). Submission to the Alice Springs Town Council in relation to the
Draft Alice Springs (Management of Public Places) By Laws 2009.
Taylor, R. B. and J. Covington (1993). “Community structural change and fear of crime”. Social
Problems 40: 374–395.
Thill, J. C. (2000). “Geographic information systems for transportation in perspective”.
Transportation Research Part C-Emerging Technologies 8: 3–12.
Tulloch, M. I. (2004). “Parental fear of crime – a discursive analysis.” Journal of Sociology 40(4):
362–377.
University of Cambridge. (2011). Manchester Police to test Cambridge “pressure points”.
Warr, M. (2000). “Fear of crime in the United States: avenues for research and policy”. Criminal
Justice 4: 452–489.
Williams, D. and J. Ahmed (2009). “The relationship between antisocial stereotypes and public
CCTV systems: exploring fear of crime in the modern surveillance society”. Psychology Crime
& Law 15: 743–758.
References 267

Woolongong City Council (WCC). (2005). Wollongong futures draft strategy report 2005.
Wollonong.
Wollongong City Council (WCC). (2007). Wollongong City Council Crime Prevention and
Community Safety Plan. Wollongong: Wollongong City Council.
Wollongong City Council (WCC). (2011). What’s On – Market Day Friday’s at Crown St Mall.
Wu, Y. N. (2010). “College students’ evaluation of police performance: a comparison of Chinese
and Americans”. Journal of Criminal Justice 38: 773–780.
Yigitcanlar, T., N. Sipe, R. Evan and M. Pitot. (2006). The land use and public transport accessibil-
ity index model. In International Geographical Union 2006 Brisbane Conference. Queensland
University of Technology.
Zhang, L. N., S. F. Messner, J. H. Liu and Y. A. Zhuo (2009). “Guanxi and fear of crime in
contemporary urban China”. The British Journal of Criminology 49: 472–490.
Index

Note: The letters ‘f’ and ‘t’ following the locators refer to figures and tables in the text.

A 218–220, 227, 242, 251, 253–256,

Activity diaries, 84–86, 96–98, 103, 105–106, 258, 261, 263
139, 141, 140t–142t, 143, 145, Ayers, I., 16, 18–19
149–151, 256
behavioural geography technique, 85 B
and daily routines, 85–86 Baker, T., 52–53, 86
diary method, 85 Balkin, S., 2–3
game-based method, 85 Bannister, J., 9
recall method, 85 Barberet, R., 255
time-space budgets, 86 Bauer, K., 260
Adams, R. E., 52–53 Beck, U., 31–32
Adu-Mireku, S., 260 Beck, V. S., 71
Age, 104, 115, 115f, 116, 162, 169, 172–173, Behavioural responses, economic impact of,
175f, 206, 207t 16–19
Ahmed, J., 254 avoidance behaviours, safety concerns, 17
Akers, R. L., 27 CCTV surveillance system, 18
Allen, G. L., 84 ‘hidden costs,’ 16
Allen, J., 3, 9, 38–39, 73, 79–80, 102, 143, ‘mall problem’ in Australia, 16–17
191, 218 protective behaviours, 17–18
Altruistic fear, 71–72, 254 protective measures and redistribution of
Anticipatory fear, 30, 70 crime between communities, 18–19
Anti-social behavior, 33, 57, 164, 198, ‘pub-and-club’ youth culture, 17
226–227, 258 ‘safety elasticity of demand,’ 17
Anxiety, 9, 27, 30–32, 35, 37, 43, 67, 70 spatial concentration of crime in poor
Areas to hide, 41, 170t–171t, 193t, 194–195, neighbourhoods, 18
205, 206f, 229, 242 work in evenings and at night, 17
Argent, N. M., 260 Behavioural theories, predator-prey
Arnold, H. R., 10, 18, 145 interactions, 255–256
Ashby, D. I., 51, 81, 86 Bennett, T., 15, 52–53, 98
Avoidance-based measure of fear, 78–79, 166, Berg, J., 258
169, 178t Bialystok, E., 149
Avoidance behavior, 9–12, 16–17, 19, 32, Blakely, E. J., 60
78–79, 82, 84–85, 87–88, 95, 103, Blalock, H., 35, 169
105–107, 112, 115, 119, 120t, 133, Blocked escape, 171t, 193t, 194, 205, 230
135, 144–147, 149–151, 172f, Block, R. A., 82–84, 191–192, 242
179–180, 188, 192, 193t, 194–195, Borooah, V., 2, 12, 15, 27, 53, 73, 104,
201–202, 205, 208, 214–215, 117, 143

B.J. Doran, M.B. Burgess, Putting Fear of Crime on the Map, Springer Series 269
on Evidence-Based Crime Policy, DOI 10.1007/978-1-4419-5647-7,

C Springer Science+Business Media, LLC 2012
270 Index

Bowers, K. J., 11, 86 “The Challenge of Crime in a Free Society”

Bowling, B., 13, 55 (paper), 1
Box, S., 2, 4, 9–10, 19, 80, 104 Chang, K., 106, 262
Braithwaite, D., 201, 227 Chiricos, T., 28, 35
Brantingham, P. J., 25, 40, 57, 82, 187–188 City Centre Revitalisation Strategy, 147, 151
Brantingham, P. L., 25, 40, 57, 82, 187–188 City of Sydney Council, 181, 184, 187,
Bratton, W. J., 11, 55–56, 87, 95, 146 222–224, 226–234
Brennan, A., 11 City Street Outreach Service, 232
Briscoe, S., 219 Clarke, A. H., 213, 218
British Crime and Disorder Act 1998, 58 Clark, J., 2, 26–27, 29, 37, 69–70, 76, 201
Broken windows hypothesis, 11, 12f, 13, 15, Clemente, F., 2–4, 220
19, 51–52, 55, 95–96, 112, 115, Clerici, C., 2, 28, 73, 180
129, 137–138, 145, 146t, 150, 170t, Cochran, J. K., 25, 33
257–258, 261 Cognitive assessments, 68, 179
Bromley, R., 3, 11, 17, 61, 80, 86, 109, 138 Cognitive barriers, 187, 242
Brown, B., 213 Cognitive mapping, 82–84, 95, 103, 105,
Brown, J. S., 255–256 133, 144, 146, 149, 151, 241, 254,
Brown, M., 3–4, 9, 16, 53, 60 258, 260
Bruno, G., 262 Cohen, L. E., 4, 25
Brunt, P., 17 Coles, C. M., 4, 15, 18, 52, 55, 87, 95, 112,
Bugs, G., 262 138, 146, 169, 235, 258
Burgess, M., 166 Collective avoidance, 13, 78, 82, 106, 106f,
107, 112, 114f, 119–122, 129–130,
Burnett, P., 83–84, 179
132–135, 137–139, 145–146,
Burrough, P. A., 86
149–150, 172–173, 181, 183f,
Bursik, R. J., 25, 33–34
187–188, 192, 220, 230, 235, 251,
Butel, E., 156, 160–162
253–254, 257, 263
Community concern hypothesis
C decline model, 36
Cameron, D., 53 neighbourhood, vitality/viability/quality
Carcach, C., 2, 12, 15, 27, 53, 73, 80, 104, of, 36
117, 143 Community oriented policing, 51–53, 56
Carlson, J., 68–70 Concern or worry about crime, 69, 75
Carvalho, I., 38, 77–78 Conklin, J. E., 2, 16, 36
Carver, S., 262 Contemporary criminology, 3
Cates, J. A., 27, 30, 82 Control signals, 41
Causes, fear of crime Cordner, G. W., 53
criminal opportunity and risk of Cothran, C. C., 17, 261
victimization theories, 25–26 Cothran, D. A., 17, 261
demographic theories, 26–31 Couclelis, H., 83–84, 180
environmental theories, 38–44 Covington, J., 3–4, 11–12, 31, 33–36, 73, 87,
social theories, 31–37 95, 104, 171t, 258
See also Individual entries Cozens, P., 25, 40–41, 57–58
CCTV Closed Circuit Television, 18, 41, 60, CPTED, see Crime Prevention Through
147–148, 148f, 222, 226, Environmental Design (CPTED)
238f, 254 Crank, J. P., 26–27, 29, 36, 38–39, 52
Central business district (CBD) area, 82, Cressey, D. R., 70
95–96, 98–101, 99f, 103–107, Crime Prevention And Community Safety
110–111, 119–120, 121f, 122, 122f, Plan, 96, 138–139, 146, 151
123, 128, 131f–132f, 133–134, Crime Prevention Through Environmental
137–139, 145–151, 177, 251, 253f, Design (CPTED), 56–59, 147, 181,
255, 263 187–188, 222
Chadee, D., 3, 86 Crime specificity, 74, 76, 78
Index
Crime-specific measure of fear, 74, 76, 166,
169, 178–179, 178t, 181
Criminal opportunity and risk of victimization
theories, 25–26
symbolic interactionism, 25, 43
variation in routine activities, 25
Criminal opportunity theory, 25
Cromley, E. K., 80, 263
Crowe, T. D., 57
Cubbage, C. J., 10, 40, 255
Cummings, A., 201, 227
D vulnerabilities hypothesis, 29–30
Daily routines, 74–75, 85–86, 96, 107, 139,
141, 143–145, 151
Daly, M., 255
Darcy, D., 155, 162, 164, 166, 168, 171t,
198–199, 227–230, 232, 234
Darkness, 58, 83, 177, 190
Davidson, N., 58–60
Davis, M., 18, 60
Day, R. A., 84, 191, 242
Dean, M., 31
Decline model, 36
‘Defensible space’, concept of, 57–58
Defining fear of crime
emotion, not cognition, 68–69
cognitive assessments, 68
concern or worry about crime, 69
emotion, definition, 68
negative emotional reactions, 68
perceived risk, definition, 68
perceptions of risk and threat of
victimization, 68–69
taxonomy of crime perceptions, 68t
individual terms ‘fear’ and ‘crime’, 67,
70–71
involves violation of criminal law, 70–71
social codes or ‘laws of morality’, 71
social concept of crime, 71
traditional jurisprudential definitions,
70
relation to emotional reaction/stimuli,
69–70
immediate danger, 70
one of primary human emotions, 69
symbols associated, 69–70
warnings, 69
types of fear of crime, 71–72
altruistic fear of crime, 71
personal and altruistic fear,
distinguishing, 71–72
271
subject of victimization (personal or
altruistic), 71
type of victimization (personal or
property), 71
DeFronzo, J., 77
Delinquent behaviour, 33
Demographic theories
indirect victimization hypothesis
interpersonal communication and
fear, 29
media and fear, 28
victimization hypothesis, 26–27
Department for Transport Urban Planning and
the Arts in Australia, 57
Dickinson, J., 28, 69–70
Dietz, A. S., 52–53
Dililio, J. J., 18
Ding, Y., 86
Disorder and decline hypothesis, 13–16
causal effect, 16
economic impact, 14–15
fear of personal and housing crime, 15
flow chart, 14f
links between serious crime and, 15
perceptions, 13–14
signs/impacts, 13
Disorder/incivilities hypotheses, 11–14, 36,
38–43, 195, 204
‘direct’ encounter, 39
disorder, definition, 39
incivilities, definition/warning signals,
38–39
‘less targeted’ encounter, 39
physical disorder, 39
signs of disorder or visible cues, 38
social disorder, 39
social incivilities, 39
‘soft’ crimes, 38
Disorder/incivilities, targeting pertinent signs
(King Cross study), 223–240
addressing environmental cues, 235
Council signs to control disorder, 238f
lower ranked environmental cues
better street lighting, 230–231
maintenance of rundown and abandoned
buildings, 231–232
offensive/degraded shops, 233–234
sex workers, 234
spruikers, 234
supporting homeless people, 232
middle-ranked environmental cues
272
Disorder/incivilities, targeting pertinent (cont.)
accessibility and minimizing blocked
escape, 230
moving loitering people, 229–230
removal of areas to hide, 229
revitalization of laneways and streets,
227–228
safe collection of rubbish/syringes/
injecting equipment, 228–229
Sydney place, 237f–238f
public/private CCTV, 238f
targeted intervention, 235–236
top-ranked cues
minimizing harm from drug use and
dealing, 226–227
reducing levels of intoxication,
224–226
Disorderly behaviour, 11, 12f, 221, 235
Disorder policing, see Zero-tolerance policing
Ditton, J., 3, 18, 35, 60, 67, 69, 73, 76, 78, 86,
169, 220, 254
Dixon, J., 53
Doeksen, H., 2, 19, 39, 41, 59, 171t
Donnelly, N., 219
Doran, B. J., 11–12, 40, 78, 80, 82, 86,
146–147, 151, 168, 177, 181, 258,
263
Douglas, M., 32
Downs, R. M., 82–84, 149, 191, 241
Drug users, 39, 168–171, 193t, 194–199,
207–211, 213, 216f–217f, 218–219,
224, 226, 228, 235–236, 242
Drummond, W. J., 110
E
Eastman, J. R., 262
‘Ecology of fear’, 255–256
Eco, U., 42
Egenhofer, M. J., 86
Ellison, C. G., 17, 77
Ellis, R., 156, 160–161
Elwood, S., 262
‘Emotional Cartography: Technologies of the
Self’, 262
Emotion-based measures of fear, 75–76
Emotion, definition, 68
Environmental cues, 13, 25, 41, 44, 58,
69–70, 84, 155, 166, 169, 170t,
172, 177, 180, 187–188, 191–192,
193t, 194–195, 202, 207, 223, 227,
229–233, 235–236, 241–242
lower ranked
better street lighting, 230–231
Index
maintenance of rundown and abandoned
buildings, 231–232
offensive/degraded shops, 233–234
sex workers, 234
spruikers, 234
supporting homeless people, 232–233
middle-ranked
accessibility and blocked escape, 230
moving loitering people, 229–230
removal of areas to hide, 229
revitalization of laneways and streets,
227–228
safe collection of rubbish/syringes/
injecting equipment, 228–229
sensitively addressing, 235
social/physical environmental cues used
in survey, Kings Cross study,
170t–171t
triggering fear of crime, 192–207, 193t
Environmental design, 56–61
CPTED, 56
British Department of Environment’s
Secured by Design scheme, 58
concept of ‘defensible space’, 57
controls, 58
primary goal, 57
criticisms, 60
environmental or order-related
improvements, 58
criticisms, 60
improved lighting, 58
‘fortress city’, concept of, 60
negative effects of protective measures, 60
situation in Sweden (case), 60
Environmental mobility restrictor, 10
Environmental theories, 38–44
disorder/incivilities hypothesis, 38–39
signal crimes perspective, 41–43
threatening and safe, 40–41
See also Individual entries
Erskine, H., 2
Ethnicity, 35, 104
Evans, D. J., 27
Ewald, U., 29, 31–32, 67, 70
Experience of victimization, 1, 26–27, 29, 117,
118f, 175–176, 176f
Exposure to risk, 30, 77, 145, 205
See also Vulnerability
F
Fagan, S., 97
Familial controls, breakdown in, 33
Index
Farrall, S., 3–4, 15, 26, 30, 69, 78, 86, 104–106,
118–119
Fear as a concern, see Social disorganization
hypothesis
Fear hotspots, 82, 181, 184–185, 188, 190,
196, 221, 228, 242
Fear mapping
beginning of, 84–85
advantages of studies, 85
‘denotative meanings’, or people’s
knowledge of a city, 84
pedestrian activity, 85
future avenues for, see Fear mapping,
potential applications and
improvements
integrating results, Kings Cross study,
221–243
Fear mapping, potential applications and
improvements
broken windows theory in transit context,
261
GIS as a decision support system
(DSS), 261
limited traditional survey tools,
261
transit environments, 261
collective avoidance behaviour with
Wollongong and Kings Cross,
251–254
fear mapping, use of, 253–254
‘Market Day Fridays’ in Crown Street
Mall, 251
mixed residential apartment-
commercial buildings, 253f
public walkway in MacCabe Park (2003
and 2010), 252f
investigating behavioural responses in
relation to crime, 254–256
complexity, 254
predator-prey interactions, comparison
with, 255–256
two-step survey procedure, 255
investigating links between fear,
crime/disorder, 256–260
‘antisocial behaviours’, 258
cognitive mapping and Participatory
GIS (PGIS), 260
‘The Crime Experiment’, 256
effects of strategic policing, 257
GIS and cognitive mapping-based
approach, use of, 258
influence of broken windows theory,
257
273
influence of zero-tolerance style
policing, 257
management of public spaces by-laws
2009, submissions to ASTC, 259t
need for research, 260
need for tools for communication, 260
‘perceptual mapping’, use of, 257
‘prescribed areas’, 257
spatial technology, 262–263
G∗ and Geographically Weighted
Regressions (GWR), 263
geospatial technology, 262
GIS-based approaches, contributions,
263
hard-copy mapping exercises, 262
index models, 262
space–time visualization of activities,
263
Fear of crime
activity diaries and daily routines, 85–86
See also Activity diaries
avoidance behaviours, 9–10
disorder/decline hypothesis, see Disorder
and decline hypothesis
endocrinic changes, 9
environmental mobility restrictor, 10
geographic information systems (GIS),
86–87
See also Geographic information
systems (GIS)
hypothesized links between the fear of
crime, disorder and crime, 11–13
individual reactions, 9–11
mapping, see Fear mapping
MAUP and ecological fallacy, effect of,
80–81, 81f
physiological changes, 9
protective or avoidance behaviours, 9
psychological perspective, 9
research, 30, 72, 76, 156, 179, 260,
262–263
spatial analyses, advantages of, 81–82
‘geodemographic’ analyses, 81
spatial cognition and cognitive mapping,
82–84
types of, 71–72
See also Defining fear of crime
Fear-reduction strategies, 26, 31, 61, 78, 88,
139, 146, 201, 235–236
Felson, M., 4, 25
Ferraro, K. F., 9, 11, 17, 25, 27, 29, 39, 67–76,
80, 143–144, 171t, 213, 218, 220
274 Index

Fisher, B. S., 9, 40, 74, 78, 81, 84–85, 88, Gottfredson, S. D., 32, 39, 57
105, 150, 171t, 181, 191, 205, Grabosky, P. N., 3, 9, 16, 53
255, 261 Graffiti, See Vandalism
Fishman, G., 4, 9, 143 Graham, J., 60
Fitzgerald, J., 150 Grasmick, H. G., 34
Fletcher, M., 27 Gray, D. E., 27, 77, 213, 218–219
‘Fortress city’, concept of, 60 Gray, E., 16, 60, 256
Fotheringham, A. S., 263 Greene, J., 11, 13, 52, 55–56
Fotheringham, S., 86 Green, G., 9
Freeman, S., 18 Grennan, H., 162
Freundschuh, S., 84 Greve, W., 76
Furstenberg, F. F. Jr., 1–2, 37, 67, 69, 75 Groves, W. B., 33–34, 98, 258
Fyfe, N., 9 Grubesic, T. H., 261
Gupy, M., 97
G
Gabriel, U., 76 H
Gall, O. R., 72 Hale, C., 3, 15, 76, 219–220
Gallup and Harris polls, 2 Hale, M., 2, 25–26, 29–30, 36, 39, 71, 188
G∗ and Geographically Weighted Regressions Hamermesh, D. S., 16–18
(GWR), 263 Hanson, R. F., 3, 27, 73, 213, 218–219
Gangs, 39, 109t, 109, 160–161, 170t–171t, Harcourt, B. E., 11–13, 16, 51–52, 56, 95, 257
193t, 194, 199, 200t, 201–202, 224, Harries, K., 87
227, 235–236, 240 Hartnagel, T. F., 2
Garofalo, J., 3–4, 11, 16, 19, 27, 36, 52, 67, Hatfield, E., 68–69
69–71, 73, 78, 213, 218–219 Helsley, R. W., 18, 60
Gender, 4, 43, 169 Herbert, D., 58–60
Geographic information systems (GIS), 79–80, Herrington, V., 11, 15, 38, 41, 53
82, 86–87, 95, 106, 109, 133, 146, ‘Hidden costs’, 16
149–151, 172, 254, 258, 260–263 Hier, S. P., 254
computer-based tools, 87 Hill, G. D., 11, 27
links between fear and occurrence of Hindelang, M. J., 78
crime, 87 Hirschfield, A., 11
NYPD’s CompStat procedure, 87 Hollway, W., 2, 4, 32, 37, 144
relationship between fear and victimization, Homelessness Brokerage Program, 232
87 Homeless Persons Information Centre, 232
set of tools, 86 Homicide, 17
usefulness, 86 Horner, M. W., 261
Geospatial technology, 262 Housing tenure, 169, 175, 206, 207t
Gibbons, S., 11, 16 The Human Rights Law Resource Centre
Gibson, C. L., 36, 134, 213 (HRLEC), 258
Gilchrist, E., 213, 218–220 Hunter, A., 34, 37, 39
Girling, E., 31 Husserl, E., 83
GIS, see Geographic information systems Hyman, J. M., 2
(GIS) Hypothetical questions, 74, 76, 78, 80, 103,
Glackman, W., 30 105, 143–144
Glensor, R. W., 51–52
Global measure of fear, 76, 85–86, 104, 119, I
143–144, 169, 178–179, 181 Immediate danger, 70
Gold, J. R., 81, 235 Income, 1, 30, 60, 80, 86, 104, 116, 116f, 162,
Goldstein, H., 51 169, 172, 176–177, 187, 206, 207t,
Golledge, R. G., 82–83, 85–86, 105, 139, 145 257, 259t
Gomme, I. M., 78 Indirect victimization hypothesis, 27–29
Goodchild, M. F., 80, 149 interpersonal communication and fear, 29
Index
media and fear, 28
‘non-victims’ and fear, 27
Individual reactions, 9–11
avoidance behaviours, 10
endocrinic changes, 9
environmental mobility restrictor, 10
negative feelings, 9
parallel between social exclusion and the
fear of crime, 11
physiological and psychological effects, 9
physiological changes, 9
“prisoners of fear” (old people), 11
protective, 9–10
Informal social control, 12–14, 12t, 14t, 33, 38
Innes, M., 13, 28, 41–43, 53, 69
Interpersonal communication
consequences
comparison to victim, 29
reputation, 29
relationship between fear of crime and
indirect victimization, 29
‘Intervention’, 257
Intoxicated person, 168t, 171t, 191, 193t, 194,
223–224, 226, 242
Investigating fear of crime
analysing fear-of-crime data
activity diaries and daily routines,
85–86
advantages of spatial analyses, 81–82
beginning of fear mapping, 84–85
GIS, 86–87
spatial cognition and cognitive
mapping, 82–84
defining fear of crime
fear in relation to emotional reactions
and stimuli, 69–70
fear is emotion, not cognition, 68–69
types of fear of crime, 71–72
measuring fear of crime
behavioural approaches to, 76–79
improvements through affective
approaches, 75–76
problems with cognitive approaches,
72–75
See also individual entries
Irwin, J., 96–97, 100, 137–138, 149, 151, 251
J background, historical
Jackson, J., 16, 60, 256
Jacobs, J., 19, 40, 57–58, 171t
Jang, H., 261
Jankowski, P., 262
Jarman, P. J., 256
275
Jefferson, T., 2, 4, 32, 37, 144
Jeffery, C. R., 57, 71, 83
Jenkins, J., 69
Jiang, H., 262
Jochelson, R., 162
Jones, K. M., 10, 38–40, 78, 83–86, 105, 116,
145, 150, 171t, 191, 205, 256, 262
Joseph, J., 3–4, 11, 80, 143–144, 241
Judd, B., 19, 40, 76, 78, 81, 171t, 191
K
Kanan, J. W., 213
Kaplan, S., 70, 83
Karakus, O., 3, 80, 260
Katz, C. M., 27–30, 35, 52, 73, 180, 201,
219–220
Keane, C., 4, 9–10
Kelling, G. L., 4, 11–15, 18–19, 34, 51–52, 55,
73, 87, 95, 112, 138, 146, 169, 235,
258
Kennedy, L. W., 10, 77, 105
Kenney, D. J., 52, 77
Keuleers, B., 139
Killias, M., 2, 28, 30, 73, 180
Kings Cross study
background, 155–156
Community Safety Mapping Project, 155
goals of, 156
integrating fear mapping results
addressing crime, 221–223
assessments of techniques/approaches,
240–242
targeting signs of disorder and incivility,
223–240
Kings Cross study, methods
interviewing approach, 168
social/physical environmental cues in
survey, 169t–170t
spatial data visualization
‘collective-avoidance’ maps, 172f, 173
2D ‘avoidance hardness’ maps,
172f, 173
3D collective avoidance’ maps, 173
survey design and questions, 168–172
crime-specific avoidance-based
question, 169
Kings Cross study, research setting
cheap rents in late 1940s, 161
destination for visitors, 162
Queens Cross (now Kings Cross), 160
‘red light’ district, 161
‘the dirty half mile’, 161
276
Kings Cross study, research setting (cont.)
crime, 162–164
inner-Sydney hotspot of assault and
robbery, 162
selected criminal incidents recorded,
167t
selected offences recorded by NSW
police, 163t, 165t
trends in selected offences, 166t
demographic characteristics, 162
fear of crime, 164–168
factor triggering to feel unsafe, 168t
geographic location
Darlinghurst Road, 158f–159f
fountain and Fitzroy Gardens, 159f
Springfield Avenue, 160f
street map of study site, 157f
travel organizations advertisements, 156
Kings Cross study, results and discussion
exploring reasons for fear of crime
areas, drug users triggered fear during
day, 209f, 216f
areas, drug users triggered fear during
night, 210f, 217f
areas, sex workers triggered fear during
day, 211f, 214f
areas, sex workers triggered fear during
night, 212f, 215f
environmental cues triggering fear,
192–207, 193t
fear between men and women,
213–218
fear between residents and visitors,
208
fear experienced by different groups,
206–207, 207t
local knowledge to avoid specific areas,
208
mapping avoidance by selected groups,
207–220
mapping perceived presence of
disorder/incivility, 194–195
men’s abnormally high fear, 220
perceived presence of areas to hide,
205–206, 206f
perceived presence of drug users,
195–199, 196f
perceived presence of gangs, 199–202,
200f
perceived presence of sex workers,
202–205, 203f
positive association, crime and fear, 190
presence of crime, 189–192
Index
sites of robbery and steal-from-person
incidents, 189f
spatial mismatch between crime and
fear, 190–192
visitors avoiding large areas, 210–213
women’s heightened fear, 218–220
people, afraid of crime
degree of avoidance hardness, 178t
indicating fear of crime, 178t
scenario 1: fearful people choosing not
to avoid, 179–180
scenario 2: fearless people avoiding
areas, 180–181
people avoiding specific areas
mapping, fear-of-crime hotspots,
181–182
safe areas and cognitive barriers,
187–188
sample characteristics, 173–177
age distribution of respondents, 175f
distribution, experience of
victimization, 176f
3D map cross-section, 174f
Kings Cross study, techniques and approaches
cognitive mapping, 241–242
survey and interviewing procedure,
240–241
Kirasic, K. C., 84
Kitchen, T., 58, 81
Kitchin, R. M., 82–84
Kleiman, M. B., 220
Knox, M., 162
Kohm, S. A., 257, 260, 262–263
Koskela, H., 3, 29, 40, 59–60, 83, 180, 191,
219–220
Kotler, B. P., 255
Kovecses, Z., 9
Krahn, H., 10, 77, 105
Krause, N., 36
Krivo, L. J., 18
Kubrin, C. E., 28, 180
Kuo, F. E., 40–41, 218
Kwan, M. -P., 86, 98, 105, 139, 145, 263
L
LaGrange, R. L., 9, 67–69, 71–75, 80, 144,
171t, 213, 218, 220
Land, K. C., 12, 28, 38–39, 67–68, 71–73,
75–76
Lane, J., 28, 30, 34–36, 71, 202
Laneways, 160f, 168t, 171t, 193t, 194, 204,
225, 227–228, 230–231, 235, 237f
Laub, J., 36, 73
Index 277

Lawton, B. A., 51 McLafferty, S. L., 80, 263

Lee, H., 99 McLaughlin, T. F., 77
Lees, B. G., 11–12, 40, 78, 82, 146–147, 151, Measuring fear of crime
168, 258 affective approaches
Lemanski, C., 81 emotion-based measures, 75–76
Levine, N. A., 111 improvements through, 75–76
Levitt, S. D., 16, 18–19 behavioural approaches
Lewis, D. A., 29, 38–39, 69, 72, 77–78, 171t avoidance-based measures, 78–79
Lewis, M. J., 213, 218 collective actions, 77
Lianos, M., 32 individual coping strategies, 77
Lighting, 17, 40, 58, 59t, 74, 137, 168t, protection-based measures, 76–78
170t–171t, 185, 193t, 194, 201, 224, self-protection, 76–77
227, 229–231, 235 socio-demographic groups, 78
Lindstrom, M., 36 problems with cognitive approaches
Liska, A. E., 2, 9–10, 16, 18–19, 25, 27, 29, global measures, approach and
73, 77, 133, 144, 188 criticism, 73–75
Liu, J. H., 260 value- or concern-based measures, 75
Liu, S. X., 261 Media and fear of crime
Loitering people, 170t–171t, 193t, 194, 229 cultivation, 28
Longley, P. A., 51, 81, 86, 149 exposure, 28
Loss of control, 30, 32, 205 interpersonal-diffusion, 28
See also Vulnerability resonance, 28
Loukaitou-Sideris, A., 11, 41, 95, 135, 171t, social comparison, 28
261 substitution, 28
Ludwig, J., 257 Medically Supervised Injecting Centre,
Lumby, R., 160–161 156, 197
Lupton, D., 3, 28, 31–32, 81 Meeker, J. W., 28, 30, 34–36, 71, 202
Lymes, D., 18, 60 Mennis, J. L., 83
Merry, E. E., 35, 73
M Mesch, G. S., 4, 9, 26, 30, 68, 72, 143
‘Mall problem’ in Australia, 16–17, 251 Miceli, R., 9, 25, 82
Management of Public Spaces By laws 2009, Michalos, A. C., 143
258 Millie, A., 11, 15, 38, 41, 53
Managing fear of crime Mirowsky, J., 10, 15, 19, 38–39, 60, 171t, 256
environmental design, 56–61 Mirrlees-Black, C., 3, 9, 38–39, 73, 79–80,
See also Environmental design 143, 213
policing, 51–56, 54t Mismatch between crime and fear
See also Policing fear of crime accounting for, 191–192
Markowitz, F. E., 15–16, 33–36, 98, 134, 258 spatial mismatch, 190–191
Martin, D., 86 Monmonier, M., 80
Mathieu, J. T., 67, 77 Moonlight, effect of, 255
MAUP and ecological fallacy, effect of, 80–81, Moran, L. J., 81
81f Morrall, P., 9, 255, 260
Mawby, R. I., 17, 29, 32, 67, 73, 75 Mukherjee, S., 27, 80
Maxfield, M. G., 9–10, 13, 15, 18, 26, 28–30, Murray, A. T., 18, 87, 261
36–37, 39, 52, 74–75, 77–79, 144, Mustaine, E. E., 77–78
171t
Mayhew, P., 3, 80 N
McCoy, V. H., 30 Nair, G., 4, 59, 74–75, 116, 137, 143, 201
McCullagh, M. J., 87 Nasar, J. L., 9–10, 19, 38–41, 74, 77–78,
McDonnell, R. A., 86 83–86, 88, 105, 116, 145, 150,
McGrath, J. H., 218 171, 181, 191, 205, 218, 241, 256,
McIntyre, J., 1–2 261–262
278 Index

Natural surveillance, 12, 19, 40, 57–58, 61, 87, Perceptions of risk and threat of
112–113, 136–137, 139, 147, 185, victimization, 68
221–222, 251 ‘Perceptual mapping’, use of, 257
Negative emotional reactions, 67–68 Perkins, D. D., 4, 12, 38–39, 87, 150, 171t, 204
Neighbourhood cohesion hypothesis, 36 Perloff, L. S., 30
Neill, W. J. V., 69 Personal and altruistic fear, distinguishing,
Nelson, A., 25, 81, 83, 87, 188, 218, 220 71–72
Newman, G., 262 Personal and housing crime, fear of, 15
Newman, O., 40, 57, 84 Peterson, R. D., 18
New South Wales Police, 53, 54t Phillips, T., 38–39, 169, 235
Nitzko, S., 254 Physical vulnerability, 30
Noaks, L., 258 Policing fear of crime, 54t
Nold, C., 262 case study: New York Police Department’s
Northern Territory National Emergency (NYPD) Policing Model, 55–56
Response Act (NTERA), 257 community-based initiatives, 52–53
Novak, K. J., 52 community-oriented or neighbourhood
NYPD zero-tolerance program of 1990s, 52 policing, 52
mobile citizen patrols, 52
O NYPD zero-tolerance program of
Oatley, K., 69 1990s, 52
OÇonner, M. E., 27, 213, 218–219 public cooperation with police, 53
Oc, T., 4, 9–11, 16–17, 57–58, 60–61, 69–71, SARA model (Scanning, Analysis,
84, 87, 95, 135, 145, 256 Response and Assessment), 51
Offensive shops, 170t–171t, 193t, 194, zero-tolerance policing, 51–52
233–234 Polk, K., 3–4, 9, 16, 53, 60
Olsen, E. P., 97, 134 Pollack, L. M., 57
Olsen, P., 86 Population turnover, 33–34, 37
Olson, D. R., 149 Potas, I., 70–71
Order-maintenance policing, see Poveda, T. G., 1–3
Zero-tolerance policing ‘Prescribed areas’, 257
Orleans, P., 83 President’s Commission on Crime and
Ortega, S. T., 30 consultants from every part of
O’Shea, T., 257 America (PCLEAJ), 1–2, 146
O’Sullivan, D., 80, 261, 263 Previous victimization, 43, 104, 169
Price, M. V., 255
P Problem oriented policing, 51–53, 156
Pain, R. H., 3, 11, 29, 32, 59–60, 81, 83, 86, Professional Geographer, 146, 151
180, 191, 213, 219–220, 235 Protective actions, 18, 76–77, 181
Painter, K., 2, 19, 25, 30, 39–40, 58, 61, Pruitt, M. V., 213
188, 191 ‘Pub and club’ culture, 17, 109, 138, 150
Pantazis, C., 10, 30, 37, 67, 73, 78, Public antisocial behaviour, 33
133, 169 See also Anti-social behavior
Paradoxical nature of fear of crime Public perception
discrepancies between official crime data of crime, 1, 52, 155, 204, 236
and surveys, 3 gangs, 201–202, 227
mismatch between fear and incidence of of security, 41
crime, 2
“Paradox of fear,” 2 R
Peak, K., 51–52 Racial diversity, 35
Pedestrian absence, 170t–171t, 193t, 194, 231 Rader, N. E., 27
Perceived risk, 29–30, 40, 58, 68–69, 73, Rahman, T., 254
75, 255 Ratcliffe, J. H., 87
Index 279

Raudenbush, S. W., 11, 16, 95, 107, 109–110,
150
Reassurance policing, 53, 221
Recreational grouping, 139–140
Reid, L., 3, 9, 75, 77–78, 80, 87
Reiss, A. J., 33, 70
Rengert, G. F., 149
Residential status, 30, 169, 206–208, 207t
Revill, G., 81, 235
Revitalization strategy
City Centre Revitalization Strategy, 96,
146, 149, 251, 252f
Laneways Revitalization Strategy, 227
Riger, S., 9, 144–145, 213, 218–219
Risk society hypothesis, 31–32
‘I am afraid’, statement, 31
state of ‘dangerization’, 32
‘unconscious’ anxieties, 32
Risk- victimization paradox, 87
Robinson, J. B., 39
Rogowitz, G. L., 255
Rohe, W. M., 171t
Romer, D., 28, 180
Ross, C. E., 10, 15, 19, 38–39, 60, 171t,
256
Rountree, P. W., 12, 28, 38–39, 67–68, 71–73,
75–76
Rubbish, 59t, 170t, 193t, 194, 228
Rundown/abandoned buildings, 39, 170t–171t,
187, 193t, 194, 231–232
Russo, J., 87
S See also Disorder/incivilities hypotheses
Sacco, V. F., 30
Safe cues, 41
‘Safety elasticity of demand’, 17
Salem, G., 29, 69, 72
Salmi, S., 53
Sampson, R. J., 11, 16, 33–34, 95, 98, 107,
109–110, 150, 258
Samuels, R., 19, 40, 76, 78, 81, 171t, 191
SARA model (Scanning, Analysis, Response
and Assessment), 51
Scarborough, B. K., 3
Schweitzer, J. H., 57–58
Scott, H., 75
Seriousness of consequences, see Vulnerability
Sex workers, 39, 169, 171t, 193t, 194–195,
202–205, 207–208, 210, 212f,
214f–215f, 216, 219, 226, 234–235,
242–243
Sholl, M. J., 82
Signal crimes perspective, 27, 38, 41–43, 177,
199, 201, 223
effect of signal, 42
interpretations and effect, 43
semiotics and signs, 41
‘signal crimes’ and ‘signal disorders’, 42
signal value, 42
situational relevance, 42–43
social semiotics and symbolic interactionist
sociology, 41, 43
social signs, 42
strong signal crimes, 42
strong/weak signal crimes, 42
Signals, definition, 44
Sims, B., 51–53
Situational crime prevention, 51, 57, 261
Skogan, W. G., 9–16, 18–19, 26–30, 36–39,
52–53, 60, 67–69, 74–79, 81, 87,
95, 110, 138, 144, 146, 171t, 195,
202
Smith, C. L., 10, 40, 255
Smith, L. N., 11, 27
Smith, P., 38–39, 169, 235
Smith, S. J., 3–4, 10–11, 16, 25, 53, 79, 188,
220
Smith, W. R., 2–3, 75–76, 218–220
Snedker, K. A., 254
Snowball, L., 229
Snyder, M. G., 60
Social change hypotheses, 31, 35, 37
diversifying racial heterogeneity, 37
Social codes or ‘laws of morality’, 71
Social concept of crime, 71
Social control, 12–13, 14f, 16, 19, 33–34, 36,
38, 40–41, 43, 57, 221, 235
Social disorganization hypothesis, 32–35
characters of, 34
community concern hypothesis
decline model, 36
neighbourhood, vitality/viability/
quality of, 36
criticism, 34
definition, 33–34
familial controls, breakdown in, 33
heterogeneity and rapid population, 33
social change hypothesis, 37
social control, concept of, 34–35
social disorganization, 33–34
social integration/neighbourhood cohesion
hypotheses, 36
social mistrust, 33
subcultural diversity hypothesis, 35
280
Social disorganization hypothesis (cont.)
fear-of-crime survey questions, 35
systemic social disorganisation model,
examination of, 34
unsupervised youth, 33–34
Social incivilities, 39, 58, 235
Social integration, 31, 35–37, 134, 169
Social integration/neighbourhood cohesion
hypotheses
collective efficacy, 36
social capital, 36
social integration, definition, 36
social support, definition, 36
Social mistrust, 33
Social psychology, 3
Social signs
‘content’ and connotative description, 42
effect of a signal, 42
‘expression’ and denotative description, 42
Social support, 36
Social theories explaining fear of crime,
31–37
risk society hypothesis, 31–32
social disorganization hypothesis,
32–35
community concern hypothesis, 36
social change hypothesis, 37
social integration/neighbourhood
cohesion hypotheses, 36
subcultural diversity hypothesis, 35
Social vulnerability, 30, 219
See also Vulnerability
‘Soft’ crimes, 38
Soini, K., 260
Sparks, R., 32, 70
Spatial analyses, 79, 82, 220
advantages of, 81–82
‘geodemographic’ analyses, 81
Spatial cognition and cognitive mapping,
82–84
definition, 83
mental state, 83
relevant ‘landmarks’, 82
spatial behaviour, 84
spatial choice, 83
Spelman, W., 9, 16, 19, 52
Spruikers, 158f, 164, 168t, 170–171, 193t, 194,
234
Stacey, W., 156, 160–161
Stafford, M. C., 72
Stanko, E. A., 32, 59–60, 67, 220
Stea, D., 82–84, 149, 191, 241
Stephens, D. W., 11–12, 38, 109, 241
Index
Stevens, M., 258
Steventon, G., 57
Stimson, R. J., 82–83, 85–86, 105, 139, 145
Stinchcombe, A. L., 30
Strange, W. C., 18, 60
Subcultural diversity hypothesis, 31, 35, 71
fear-of-crime survey questions, 35
Subject of victimization (personal or altruistic),
71
Sullivan, W. C., 40–41, 218
Sundeen, R. A., 67, 77
Sun, I. Y., 33–34
Sutherland, E. H., 70
Symbolic interactionism/interactionist
sociology, 25, 43
Symbols associated with crime, 67, 69
Syringes, 107t, 170t–171t, 193t, 194–195, 198,
204, 228, 236
T
Taxonomy of crime perceptions, 68t
Taylor, H. A., 83, 191
Taylor, R. B., 2–4, 11–13, 25–26, 29–31,
33–39, 73, 87, 95, 104, 150, 171t,
188, 258
‘Territoriality’, concept of, 14, 53, 57–58, 187
Teske, R. H. C. J., 10, 18, 145
Tewksbury, R., 76, 78
Thill, J. C., 261
Thomas, C. W., 2–3, 11, 17, 61, 80, 86, 109,
138
Thompson, E. E., 35–36
Thompson, T., 156, 160–162
Threatening and safe environments theories,
40–41
control signals, definition, 41
fear reducing features, 41
‘likeable features’, decrease the fear, 41
safe environments, 41
threatening physical/social environments,
40–41
time of day, 40
Tiesdell, S., 4, 9–11, 16–17, 57–58, 60–61,
69–71, 84, 87, 95, 135, 145, 256
Tortensson, M., 2–3, 75–76, 218–220
Toseland, R. W., 77, 82, 213, 218–219
Trends in fear of crime research, 3–4
“enduring frustration” for policy makers, 4
Tulloch, J., 3–4, 25, 28, 32, 52, 67, 69, 74–76,
81, 144, 164, 195
Tulloch, M., 3, 38–39, 76
Tversky, B., 83, 191
Index
U Warr, M., 2–3, 9, 11, 15, 17, 19, 26, 29, 67–72,
Unsupervised youth, 33–34
Unwin, D. J., 80, 263
V Webb, J., 34
Vacha, E. F., 77
Valentine, G., 3, 83, 180, 220
Van Beek, I., 161, 197–198
Vandalism, 59t, 77, 107, 117, 124f, 164,
170t–171t, 193t, 194, 233
Van der Wuff, A., 3
Vanetti, E. J., 84
Victimization, 2, 4, 10, 15, 18, 25–32, 37, 40,
43–44, 51, 68–69, 71–73, 78, 87,
104, 117, 118f, 144, 169, 175, 176f,
180, 201–202, 213, 218–220, 227,
256–257
Victimization hypothesis, 26–27
blame, 26
direct experience and fear, relationship, 26
direct victimization, 26
fear neutralization techniques
acceptance of responsibility or denial of
future vulnerability, 27
denial of injury or damage, 27
personal (direct) victimization, impact of,
27
‘theories of reality’, stages of emotional
loss, 26
type of victimization (personal or property),
71
victimization and fear, positive/negative
association, 26–27
victimization, influence of, 26–27
Vignettes, 103, 105, 118, 119t
Violation of criminal law, 70–71
Vrij, A., 220
Vulnerability
factors, 30
feelings of, 26
hypothesis, 29–30
physical vulnerability, 30
purist theorists, 30
social vulnerability, 30
trends, 30
W cognitive maps of avoidance,
Wagner, A. E., 58
Walker, M. A., 73, 144
Walklate, S., 4, 25, 32, 52, 144
Walmsley, D. J., 260
Walsh, D. A., 70–71, 84, 149
281
76–77, 80–81, 87, 135, 145–146,
179, 219–220, 254
Weatherburn, D., 18
Weitzer, R., 28, 180
Wets, G., 139
Whitaker, A. M., 160–162
White, P., 3, 80
Wikström, P. H., 60–61, 110
Williams, D., 254
Williams, P., 28, 69–70
Will, J. A., 218
Wilson-Doenges, G., 3, 9–10, 60, 80
Wilson, J. Q., 11–14, 19, 34, 38, 51–52, 55, 73,
87, 95, 138, 146
Winkel, F. W., 220
Wolfer, L., 52–53, 86
Wollongong City Council, 96–97, 100, 106,
138–139, 147, 149
Wollongong study
goals of, 95–96
industrial city, background, 97–98, 97f–98f
location of Wollongong, 96f
Wollongong study, methods
combinatory spatial analysis, 112–115
expansion and linking hotspots, 114f
expansion of hotspot in relation to
collective avoidance concentration,
113f
expansion of hotspot in relation to
partially overlapping collective
avoidance concentration, 114f
disorder assessment
graffiti in Crown Lane/MacCabe Park,
example, 108f
physical disorder, 107–109
social disorder, 109–110
types of physical disorder, 107t
types of social disorder, 110t
fear-of-crime survey and analysis
cognitive mapping of avoidance
behaviour, 105–106
combinatory process to collate
individual avoidance grids, 106f
factors known to influence, 104
GIS-based technique, collating
106–107
questions on emotional levels and
protective behaviour, 105
vignettes, emotional levels of fear in
hypothetical situations, 105
282
Wollongong study, methods (cont.)
spatial analysis of crime data, 110–112
types of venues with social disorder,
110t
typical view, Crown and Keira streets,
112f
typical view, Crown Street Mall area,
111, 111f
Wollongong study, research setting
Central Business District (CBD), 99–101,
99f
public seating area in Crown Street
Mall, 100f–101f
structure plan, 100
crime and fear of crime, 101–103, 102t
crime hotspots at LGA Level in NSW
(2002), 102–103
crime trends in Illawarra Region (2001),
101–102
discussion of spatial outputs, see
Wollongong study, spatial outputs
integrating findings with police/community
initiatives, 145–149
CCTV program – action table, 148,
148f
Crime Prevention and Community
Safety Plan, 146
Don’t stall in the mall – action table,
147, 148f
spatiotemporal overlap between broken
windows thesis and institutional
involvement, 146t
logic behind study site selection, 98
high rates of crime, 98
techniques and approaches, assessments,
149–151
Wollongong study, results, 115–133
distribution of collective avoidance
concentrations, 119–122
areas of CBD avoided in relation to fear
of crime, 121f–122f
respondents adopting avoidance
behaviour in neighbourhoods/CBD
area, 120t
responses to question regarding police,
120f
overlap between areas of avoidance
and crime hotspots, 133f
and social disorder hotspots for
weekdays/weekends, 135f–136f
and weighted physical disorder
hotspots, 134f
Index
overlap between collective avoidance
concentrations, physical and social
disorder, 129–133, 134f
overlap with elements of disorder and
crime, 136t
sample characteristics, 115–119, 115f
answers to global measure of fear
amongst respondents, 119f
experience of victimization among
respondents, 118f
income distribution of respondents,
116f
media perception of crime-related
issues among respondents, 117f
overlap between collective avoidance
concentrations, physical/social
disorder, 129–133
safety in relation to vignettes, 119t
spatiotemporal distribution of phys-
ical/social disorder/crime,
122–129
areas of CBD avoided after 19:00, 123f
crime hotspots within CBD, 132f
hotspots of social disorder on
weekdays/weekends, 127f–128f
hotspots of social disorder on weekends,
128f
number of clubs/bars/nightclubs in
CBD area, 131f
physical disorder hotspots, based upon
weighted data, 124f
ranking of different types of disorder,
124f
social disorder hotspots,
bars/clubs/adult entertainment
stores, 131f
social disorder on weekdays at
day/night, 125f–126f
social disorder on weekends at
day/night, 126f–127f
types of social disorder on
weekdays/weekends, 129f–130f
Wollongong study, spatial outputs
activity diary analysis, 143–145
categories from classification
procedure, 140t
coding system for protective
behaviours, 141t
data preparation, 139–141
diary data, 140t
emotional levels of fear/ protective
behaviour, 140–141, 142t
mandatory activities, 145
Index 283

results of, 143 Y

structural constraints and role Yarwood, R., 82, 87
obligations, 144 Yigitcanlar, T., 261
constraints on social interaction from Yin, P. P., 67
collective avoidance behaviour Yokohari, M., 41
Crown Street Mall Area, 138–139 Young, M., 258, 263
MacCabe Park Area, 135–138
Piccadilly Area, 133–135 Z
Women’s fear of crime, 59 Zadel, S., 161
Zero tolerance policing, 51, 55, 146
Wood, T. H. J. J., 201
community oriented policing, 51–53, 56
Wright, S. M., 256
problem oriented policing, 51–53, 156
Wu, Y. N., 255 reassurance policing, 53, 221
Zhang, L. N., 3, 260
X Zhu, X. A., 261
Xu, Y., 12, 51, 53 Zumbo, B. D., 143