ArticlePDF Available

Examining the extent of repeat and near repeat victimisation of domestic burglaries in Belo Horizonte, Brazil

Authors:

Abstract

Substantial research suggests that a burglary event is a useful predictor of burglaries to the same or nearby properties in the near future. To date, the research that has suggested this predictive quality has been based on studies that have focused on crime patterns in western industrialised countries, such as the UK, USA and Australia. These studies have in turn informed the design of effective burglary reduction programmes that have a specific focus towards countering the risk of repeats and near repeats. This current study adds to the existing research knowledge by examining whether patterns of burglary repeats and near repeats are evident in Belo Horizonte, a large Brazilian city. Domestic dwellings in Brazilian cities, as typified by those in Belo Horizonte, are quite different to dwellings in western countries- many city-dwelling Brazilians live in apartments in high rise buildings, most houses and apartment blocks are surrounded by high perimeter fencing, and a reasonable proportion of dwellings are irregular self-constructed houses. As a consequence, a different infrastructure of domestic living may result in differences in patterns of domestic burglary when compared to patterns in western countries. The research identifies that the extent of repeat and near repeat patterns in the city of Belo Horizonte are lower than those in comparable western urban contexts. We discuss the implications of these findings and how they impact on the translating of practice on crime prevention and crime prediction to the urban Latin American context.
Chainey and da Silva Crime Sci (2016) 5:1
DOI 10.1186/s40163-016-0049-6
RESEARCH
Examining the extent ofrepeat andnear
repeat victimisation ofdomestic burglaries
inBelo Horizonte, Brazil
Spencer Paul Chainey1* and Braulio Figueiredo Alves da Silva2
Abstract
Substantial research suggests that a burglary event is a useful predictor of burglaries to the same or nearby properties
in the near future. To date, the research that has suggested this predictive quality has been based on studies that have
focused on crime patterns in western industrialised countries, such as the UK, USA and Australia. These studies have
in turn informed the design of effective burglary reduction programmes that have a specific focus towards counter-
ing the risk of repeats and near repeats. This current study adds to the existing research knowledge by examining
whether patterns of burglary repeats and near repeats are evident in Belo Horizonte, a large Brazilian city. Domestic
dwellings in Brazilian cities, as typified by those in Belo Horizonte, are quite different to dwellings in western coun-
tries—many city-dwelling Brazilians live in apartments in high rise buildings, most houses and apartment blocks
are surrounded by high perimeter fencing, and a reasonable proportion of dwellings are irregular self-constructed
houses. As a consequence, a different infrastructure of domestic living may result in differences in patterns of domes-
tic burglary when compared to patterns in western countries. The research identifies that the extent of repeat and
near repeat patterns in the city of Belo Horizonte are lower than those in comparable western urban contexts. We
discuss the implications of these findings and how they impact on the translating of practice on crime prevention
and crime prediction to the urban Latin American context.
Keywords: Repeat victimisation, Near repeat victimisation, Crime prediction, Crime prevention, Policing, Burglary,
Boost account, Flag account, Foraging theory
© 2016 Chainey and da Silva. This article is distributed under the terms of the Creative Commons Attribution 4.0 International
License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any
medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons
license, and indicate if changes were made.
Background
Repeat victimisation is the empirically observed pattern
of a person or other target (e.g., a building) being sub-
ject to victimisation a number of times (Farrell and Pease
1993; Polvi etal. 1991). Near repeat victimisation is the
observed finding that targets near to a recent incident
are also at a heightened risk of being victimized (Bowers
etal. 2004). ese patterns of repeat and near repeat vic-
timisation have been observed for a range of crime types,
including domestic burglary (Johnson and Bowers 2004a;
Pease 1998; Johnson etal. 2007), vehicle crime (Johnson
etal. 2009), and shootings (Haberman and Ratcliffe 2012;
Ratcliffe and Rengert 2008).
e empirical findings from research into the patterns
of repeats and near repeats has led some commenta-
tors to suggest that recent incidents provide a powerful
indicator for predicting where and when crime is likely
to take place (Bowers et al. 2004; Johnson and Bowers
2004b; Pease 1998; Skogan 1996). In turn, these observed
patterns of repeats and near repeats have resulted in
many police agencies designing crime prevention pro-
grammes, in particular for burglary, to counter the pre-
dicted heightened risk of further incidents following an
initial offence with reported successes including reduc-
tions in burglary of 27% in Trafford (UK) (Fielding and
Jones 2012) and 66% in Edmonton (Canada) (UCL 2014).
Additionally, several software companies have drawn
from the research findings into repeats and near repeats
Open Access
*Correspondence: s.chainey@ucl.ac.uk
1 Department of Security and Crime Science, University College London,
35 Tavistock Square, London, England, UK
Full list of author information is available at the end of the article
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 2 of 10
Chainey and da Silva Crime Sci (2016) 5:1
to design applications for predicting crime, for example
PredPol (2013) and HunchLab (Azavea 2015).
To date, the analysis of repeat and near repeat patterns
of crime has been applied in many western countries,
including Europe (Bernasco 2008; Bowers and John-
son 2005), the USA (Haberman and Ratcliffe 2012; Rat-
cliffe and Rengert 2008; Wells etal. 2011), and Australia
(Townsley et al. 2003) but has received very little ana-
lytical attention in non-western contexts, such as Latin
American countries. In this paper we describe research
that examined whether patterns of repeat victimisation
and near repeat victimisation of domestic burglary are
evident in Belo Horizonte, Brazil—a typical Brazilian
city. We hypothesise that patterns of burglary repeats and
near repeats are evident in Belo Horizonte, but the extent
of these patterns are likely to be different to those found
in western countries. We consider whether differences
in patterns of burglary repeats and near repeats are due
to contextual urban infrastructure differences between
Latin American countries and western countries. If pat-
terns of burglary repeat and near repeat victimisation
are less evident in Latin American countries, this would
indicate that crime prediction software tools and police
strategies for countering repeats and near repeats that
have been developed to suit western contexts may be less
suitable in Latin American countries for supporting bur-
glary prevention.
SectionRepeats, near repeats, theories that underpin
their patterns, and crime prevention initiatives designed
to counter these patterns” of the paper describes patterns
of repeat and near repeat victimisation from a range of
studies in western countries to offer a benchmark against
which the research findings can be compared. In
Repeats, near repeats, theories that underpin their pat-
terns, and crime prevention initiatives designed to coun-
ter these patterns” section we also describe the theory
that underpin the patterns of repeat and near repeat vic-
timisation and describe several burglary prevention pro-
grammes that have been designed to counter the risk of
repeats and near repeats. In “Policing, burglary, and
crime reporting in Brazil” section we describe burglary
patterns in Brazil and consider whether crime reporting
and recoding practices for our study area may have an
impact on our research findings. SectionStudy area and
the domestic living infrastructure in Brazilian cities
describes our study area of Belo Horizonte, including
details about the city’s urban infrastructure and typical
domestic security arrangements.1 Section“Method: data,
and spatial-temporal analysis of domestic burglary
1 e review of Brazilian urban infrastructure, domestic living and common
burglary prevention efforts allows us to draw conclusions from the analysis
findings and consider the practical replication of burglary prevention efforts
that have been applied in western contexts.
describes the method used for examining the extent of
burglary repeat and near repeat victimisation in Belo
Horizonte. In “Results: analysis of repeat victimisation
and near repeat victimisation”section the results are
described, and in “Discussion and implications” section
the results are discussed, along with the implications of
these findings on crime prediction and burglary preven-
tion in Brazil and elsewhere in Latin America. Conclu-
sions from the research are provided in “Conclusions
section.
Repeats, nearrepeats, theories thatunderpin their
patterns, andcrime prevention initiatives designed
tocounter these patterns
Patterns ofrepeat andnear repeat victimisation
Repeat victimisation is the concept of a person or build-
ing being subject to victimisation a number of times.
Research into repeat victimisation has shown that, over-
all, risk doubles following a victimisation, and that
repeats occur swiftly after the initial incident (Farrell and
Pease 1993; Polvi etal. 1991). Table1 shows the extent of
burglary repeat victimisation has been recorded to be
between 7 and 33% of all burglaries in several western
and developed countries.2 Interviews with offenders have
also supported the empirical observations of patterns of
repeat victimisation, with Ericsson (1995), for example,
finding that 76% of offenders interviewed returned to a
number of houses to burgle them 2–5 times.
Near repeat victimisation is the observed finding that
targets near to a recent incident are at a heightened risk
of being victimised. e level of risk to neighbouring tar-
gets is lower than the risk of victimisation to the recent
victimised target, and decays with distance from this
original target. Similar to repeat victimisation, this
heightened risk to neighbouring targets decays over time
(Bowers etal. 2004). Listed in Table2 are examples of the
extent of burglary near repeat victimisation sourced from
police agencies in the UK and New Zealand (from analy-
sis conducted within police agencies)3 and show, when
defined as burglaries committed within 200m and 7days
of an originator offence, near repeats can account for up
to a quarter of all burglaries. Table2 also shows variation
exists between locations, but that in general, near repeats
2 While the time-window for measuring repeat victimisation is known to
have an effect on results (Farrell etal. 2002), the studies selected for illustra-
tion in Table1 all used a 1year time-window, therefore, enabling compari-
son. Results from studies at national and local levels are provided in order
to illustrate a range of burglary repeat victimisation levels, and to illustrate
how levels at the local level may vary when compared to national levels of
burglary repeat victimisation.
3 To date, while many studies have reported statistical evidence of near
repeat patterns (for example, Bowers and Johnson 2005; Haberman and
Ratcliffe 2012; and Townsley etal. 2003), very few have recorded the extent
to which near repeats account for all crime.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 3 of 10
Chainey and da Silva Crime Sci (2016) 5:1
within 200 m and 7 days of an originator offence
accounted for at least 12% of all burglaries.
Theories thatexplain the patterns ofrepeats andnear
repeats
e reasons why repeats and near repeats occur can prin-
cipally be explained by the boost account and optimal
foraging theory, and the flag account. e boost account
refers to an offender deciding to return to the same loca-
tion, boosted by the success of previous crime commis-
sion (Bowers and Johnson 2004; Pease 1998). is boost
principle also applies to near repeats, based on the idea
that the offender is boosted to return because of their
familiarity with the area (following the initial offence),
the means of breaking in and the layout of the neighbour-
ing properties are likely to be similar, and the neighbours
are likely to have possessions worth stealing, similar to
those stolen in the initial burglary (Chainey 2012).
Optimal foraging theory provides a means of explain-
ing why the boost account occurs (Bowers and Johnson
2004; Johnson etal. 2009). is approach likens offend-
ers to foraging animals. As a forager, an offender makes a
trade-off between the rewards of crime commission that
are most obvious and immediately available (i.e., return-
ing to commit a repeat), and the effort and risks that will
be expended in seeking better opportunities. Once an
area has been exhausted of the best theft opportunities,
the forager moves on (i.e., seeking to commit burglary
elsewhere).
e flag account suggests there is some enduring qual-
ity about the target that highlights (flags) its high level of
vulnerability to would-be offenders (Pease 1998).4 Differ-
ent to the boost account, the flag account suggests a
repeat offence is likely to be committed by a different
offender than who committed the initial offence. In addi-
tion, the flag account suggests the time between an initial
offence and a repeat is more likely to be random, rather
than following swiftly after an initial incident.
In practice, a combination of boost, optimal forag-
ing and flag theories are likely at play in explaining why
repeats and near repeats occur. For instance, the flag
characteristics of a property may initially attract an
offender because it is seen as an easier target, with the
risk of future burglary being boosted following an initial
incident. Foraging behaviour then helps explain why an
offender may return to the same or nearby locations for a
short period after a previous offence to carry out a spate
of further offences.
4 In terms of burglary, these qualities could include the property being sit-
uated at the end of a terrace, which has an alley running along the back,
and the property appearing to have poor door and window security—all
of which are signals for easy property access opportunities to a would-be
offender.
Table 1 The extent of burglary repeat victimisation
asreported froma range ofstudies inwestern anddevel‑
oped countries
For each study the time measurement window for analysing repeat victimisation
was 1year
Location ofstudy andsource Proportion ofburglary
thatwas repeat victimi-
sation (%)
Birmingham, England: 2011
(BBC 2012)30
England and Wales: 2012/13
(ONS 2013)14
Japan: 1989
(Farrell and Bouloukos 2001)22
Newcastle, England: 2010
(Safe Newcastle Partnership 2011)15
Netherlands: 1996
(Farrell and Bouloukos 2001)21
Northumberland County, England: 2009
(Northumberland Community Safety Partner-
ship, 2009)
7
South Auckland, New Zealand: 2014
(Chainey and Silva 2015)10
Sweden: 1996
(Farrell and Bouloukos 2001)13
USA: 1996
(Farrell and Bouloukos 2001)33
Table 2 The extent of burglary near repeat victimisation
as reported from a range of analysis studies conducted
bypolice agencies inthe UK andNew Zealand
For each study, data for 1year was used to measure the extent of near repeats
Measure ofnear inspace andnear
intime tooriginator incident,
andlocation ofstudy andsource
Proportion ofburglary that
was nearrepeat victimisa-
tion (%)
Within 200 m and 7 days 14
Within 200 m and 14 days 23
Herefordshire, UK (West Mercia Police
2012)
Within 300 m and 10 days 16
Kettering, UK (Northamptonshire Police
2013)
Within 200 m and 7 days 23
Newcastle, UK (Chainey 2014)
Within 100 m and 1 day 1
Within 100 m and 7 days 5
Within 200 m and 7 days 12
Auckland, New Zealand (New Zealand
Police 2014a)
Within 100 m and 1 day 2
Within 100 m and 7 days 5
Within 200 m and 7 days 12
Wellington, New Zealand (New Zealand
Police 2014b)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 4 of 10
Chainey and da Silva Crime Sci (2016) 5:1
Crime prevention initiatives designed tocounter the risk
ofrepeat andnear repeat victimisation
e key finding from studies that have examined repeat
victimisation is that the risk of a second burglary occur-
ring is substantially higher than the risk of the first bur-
glary, but that this risk of a second burglary decays over
time. is finding presents opportunities for crime pre-
vention by attempting to counter the risk of the second
and other subsequent burglaries to the same property.
One of the first crime prevention initiatives that took
advantage to prevent domestic burglary by countering
the risk of repeats was the Kirkholt Burglary Prevention
Project in Rochdale, UK (Forrester etal. 1988). After an
analysis that identified the high level of repeat victimisa-
tion on the Kirkholt housing estate, crime prevention
efforts, including improvements in dwelling security,
were targeted to those properties that had previously
experienced burglary repeats. e result was an 80 %
reduction in burglary repeat victimisation, which con-
tributed to an overall reduction of 53% in burglary across
the Kirkholt estate (Forrester etal. 1988). Several bur-
glary prevention initiatives that have since followed the
Kirkholt experience5 have similarly focused on prevent-
ing the heightened risk of further burglaries occurring
(following an initial incident).6
e key finding from research into near repeat victimi-
sation is that the risk of burglary to properties near to a
recently burgled property is significantly higher than the
risk to those properties further away, but that the risk of
burglaries to these neighbouring properties decays over
time. is finding presents opportunities for burglary
prevention by attempting to counter the risk of burglary
to nearby properties. One of the first initiatives designed
to specifically counter both repeats and near repeats was
introduced by Greater Manchester Police in Trafford, UK
in 2011. In addition to the targeting of a crime prevention
officer to recently burgled properties, officers from the
local policing team would conduct door-to-door visits to
neighbouring houses on the day after the initial burglary,
informing residents about the recent burglary, reassuring
them over their likely risk of burglary, and providing
practical crime prevention advice to prevent these neigh-
bouring residents being burgled (Fielding and Jones 2012;
5 In a systematic review of repeat victimisation prevention, Grove et al.
(2012) further identified the promising opportunities for burglary preven-
tion by indicating that initiatives that were designed to prevent repeats
experienced an average reduction in burglary of 22%.
6 A commonly used prevention tactic involves the deployment of crime
prevention officers to burgled homes within 24 h of the burglary occur-
ring (Chainey, 2012). e tactic of deploying a crime prevention officer to
a recently burgled property aims to respond in a manner that is timely to
when the risk of a repeat incident is at its highest, and that any immediate
improvements in security and requests for residents to be extra vigilant will
resonate for several days after the visit of the crime prevention officer.
Chainey 2012).7 e impact of the crime prevention ini-
tiative to counter burglary repeats and near repeats in
Trafford was a reduction in burglaries of 42% in the areas
that were targeted.8 Similar burglary prevention tactics to
those introduced in Trafford have been implemented by
other police forces in the UK, USA, and Canada, and
have contributed to similar reductions (UCL 2014).9
Policing, burglary, andcrime reporting inBrazil
For the twenty-six states in Brazil, each has a Military Police
agency (Policia Militar) that acts as both the front line of law
enforcement and the agency that responds to incidents of
crime. While a militaristic approach to policing, with a focus
towards repression and enforcement has tended to be the
tradition in Brazilian police practice, in more recent years an
appreciation and adoption of prevention-focused, problem-
oriented and community policing principles have begun to
be adopted by Brazil’s Policia Militar (Beato Filho 2008).
Although violent crime is of a particular concern in Brazil
with rates exceeding those of many western countries,10
Brazil experiences relatively low levels of domestic bur-
glary. In 2012, the recorded domestic burglary rate in Brazil
was 11 burglaries per 100,000 population, compared to the
United States (494 per 100,000 population), Australia (659
per 100,000 population), New Zealand (886 per 100,000
population), and England and Wales (402 per 100,000 pop-
ulation) (UNODC 2014). e low domestic burglary rate in
Brazil is though comparable to most rates experienced in
other Latin American countries such as Colombia (47 bur-
glaries per 100,000 population), Mexico (95 per 100,000
population), Panama (9 per 100,000 population), and Para-
guay (28.5 per 100,000 population) (UNODC 2014).
Like in western countries, the under-reporting of crime
is a problem in Brazil. Results from the 2010 Brazil Vic-
timization Survey11 state that only 33% of burglary vic-
7 e timing of the visits to nearby properties was chosen to coincide
with when the previous burglary took place, acting as a possible deterrent
(through the presence of a police officer in high visibility uniform) to any
returning offender.
8 e overall burglary reduction across Trafford was 27% (Chainey 2012).
9 West Yorkshire Police in Leeds modelled a burglary prevention initiative
on the practice from Trafford and experienced a 48% reduction in burglary
(Professional Security 2012).
10 Crime and violence are among the main concerns of Brazilians (CNT/
SENSUS 2010). e issue of violence is also reflected in changes in the hom-
icide rate in Brazil, increasing from 12 homicides per 100,000 inhabitants in
1980 to 30 homicides per 100,000 inhabitants in 2012 (SIM/DataSUS 2014).
In the most recent Brazilian Victimisation Survey, 60% of respondents
stated that safety and crime problems were becoming worse in the country,
affecting their sense of security and increasing their fear of crime (Silva and
Beato Filho 2013).
11 e Brazil Victimisation Survey is completed by visiting residents in their
home. e sample unit of the survey is the individual, conducted on persons
aged 16 or over and resident in a town of over 15,000 inhabitants. e sur-
vey involved conducting 78,008 interviews and was considered representa-
tive of Brazil to a confidence level of 95 and 0.4% margin of error.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 5 of 10
Chainey and da Silva Crime Sci (2016) 5:1
tims reported the crime to the police (Silva and Beato
Filho 2013). is is, however, comparable to a reporting
rate for domestic burglary of 36% in England and Wales
(ONS 2015),12 and, therefore, suggests comparisons
between patterns observed in police recorded burglary
data between Brazil and western countries (in particular
the UK) are feasible. While additionally there are some
concerns over the quality and completeness of crime data
recorded by the state police agencies in Brazil (as noted
by Murray etal. 2013), the crime data we use in the pre-
sent study is for the state of Minas Gerais which is rated
as having high quality police recorded crime data (Fórum
Brasileiro de Segurança Pública 2011).
Study area andthe domestic living infrastructure
inBrazilian cities
e city of Belo Horizonte in Brazil was the chosen area
of study. is is because recorded crime data required
for the analysis were available, the authors’ knowledge
of the city, and because Belo Horizonte is a city that is
representative of urban living in Brazil. Belo Horizonte is
in the southeastern region of Brazil, is capital of the state
of Minas Gerais, and is Brazil’s third largest metropolitan
area with a population of 5.5 million (Brookings 2012).
An important contextual difference between Brazil-
ian cities and many cities in western countries relates to
urban living. In Brazil, eight of the country’s cities feature
in the top fifty cities in the world with the most high rise
buildings, many of which are for residential purposes. In
contrast, only New York City and Chicago are the two US
cities included and only London is included as the single
UK city in the same top 50. Belo Horizonte is ranked 15th
in the world for cities with the most high rise buildings
(Emporis 2015).
A third of the population of Belo Horizonte live in
apartments (in high rise buildings), 9 % of the city’s
inhabitants live in favelas (irregular settlements), with the
remainder mainly living in semi-detached or detached
houses (IBGE 2010). In addition to the differences in
housing context are the security features that are com-
mon to homes in Brazilian cities, typified in Belo Hori-
zonte. Results from the 2010 Brazil Victimisation Survey
showed that over 55% of respondents use security mech-
anisms in their homes, such as reinforced bars, guard
dogs and high fences in order to increase protection
against domestic burglary (CRISP/Datafolha/Ministério
da Justiça 2011). e fortification of domestic proper-
ties includes fencing around apartment blocks as well as
around semi-detached and detached properties (Caldeira
12 Based on an estimated 560,000 incidents of domestic burglary in a dwell-
ing determined by the Crime Survey of England and Wales and 204,136
domestic burglaries recorded by police forces in England and Wales, for the
year ending September 2014 (ONS 2015).
2000; Paixão 1991). Indeed, results from a recent victimi-
sation survey conducted in Belo Horizonte revealed that
more than a half of respondents stated that their homes
or apartments had high walls or fences over two metres
high, and that 14% had installed electric fences. 42% of
respondents from the survey also stated they had metal
bars installed across the windows of their residence, 37%
reported having extra locks on doors and locks on their
windows, and 13% had installed a burglar alarm in their
homes (CRISP 2006).
In addition to the physical security features that are
built into many Brazilian homes to help protect resi-
dents from burglary, many apartment blocks in Brazil
have a security guard stationed at the entrance to the
building, for all hours of the day. is is a feature that is
not exclusive to just apartment blocks that house the
wealthy, but is common to all types of apartment
blocks. 31% of respondents to the Belo Horizonte vic-
timisation survey also stated they had seen the pres-
ence of private security in their neighbourhood (CRISP
2006). In addition to the presence of security officers is
the presence of domestic staff in Brazilian homes. Most
residents with at least an average level of income in
Brazil, particularly families, employ at least one domes-
tic helper (Brazil Business 2014). is means that when
the home owners are at work during the day, these
trusted domestic staff occupy their home, and in-
doing-so provide informal security through their
presence.13
Method: data, andspatial‑temporal analysis
ofdomestic burglary
Recorded crime data
Recorded crime data on domestic burglary for 2012 to
2014 were provided by the Policia Militar de Minas Ger-
ais (PMMG) for the city of Belo Horizonte. e crime
data included the address of the home that had been bur-
gled, and the date and time of the offence. Between 2012
and 2014 there were 19453 recorded domestic burglaries
in Belo Horizonte.
PMMG employ a robust geocoding process to pinpoint
within a computer-based mapping environment the exact
location of the offence.14 An assessment of the geocoding
accuracy of the domestic burglary data determined the
13 13% of women working in the metropolitan region of Belo Horizonte are
employed as domestic servants (DIEESE 2013).
14 e geocoding process utilises the address as recorded in the crime
record to determine the spatial coordinates for the offence. In addition,
PMMG use Global Positioning System technology and satellite imagery to
help locate the position of a burglary offence occurred when committed to
homes in favelas.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 6 of 10
Chainey and da Silva Crime Sci (2016) 5:1
data to be at least 95% accurate15 and sufficient in quality
for the purposes of the research.
Measuring repeat victimisation
Repeat incidents of burglary were identified using a
two stage approach. e first stage identified repeats by
selecting those records where the geographic coordinates
for the burglary were the same as that for another offence.
e address details for each of these offences were then
checked to determine that each offence corresponded to
the same dwelling, and to remove any that did not (e.g.,
where the same geographic coordinates referred to two
different apartments within a high rise building). e sec-
ond stage identified repeats based on the same text string
recorded in the address field of the crime record. is
allowed for addresses to be corroborated between the
two selection approaches. e combination of the two
approaches resulted in a list of all recorded burglaries at
addresses where more than one burglary had occurred.
e list of repeat incidents of burglaries was then ana-
lysed to determine the number of addresses that had
experienced more than one burglary, the number of
repeats (not including the first incident), and the propor-
tion of repeats against the total number of burglaries. e
analysis of burglary repeats was conducted for the whole
3 year dataset and for each 1 year period over the
3years.16 Further analysis was conducted to determine if
the pattern of repeats was statistically significant, using
the Near Repeat Calculator (Ratcliffe 2009).
Measuring nearrepeat victimisation
Near repeats were measured using the Near Repeat Cal-
culator (Ratcliffe 2009). is near repeat analysis involved
examining the distance and time between burglaries to
determine if the pattern of near repeats was statistically
significant. e spatial bandwidth in the Near Repeat Cal-
culator was set to 100m and five bands were applied. e
temporal bandwidth in the Near Repeat Calculator was
set to 7days and four bands were applied. Analysis was
also conducted to determine the number of offences com-
mitted within 100m and 7days, within 200m and 7days,
and within 300 m and 7 days of an originator offence.
e spatial and temporal bandwidths and the number of
bands were chosen as they provided a means for compar-
ing the extent of near repeats in Belo Horizonte to several
15 Following the geocoding accuracy measurement process described by
Chainey and Ratcliffe 2005.
16 is analysis approach allowed for the effects of the time-window for
measuring repeats to be assessed for its impact on the extent of repeat vic-
timisation in Belo Horizonte, and to allow comparison to the levels of repeat
victimisation reported in Table1 where each study was based on measuring
repeats using one year of crime data.
of the results from near repeat victimisation studies for
other areas in the world (as reported in Table2).
Results: analysis ofrepeat victimisation andnear
repeat victimisation
Repeat victimisation
Table3 shows that during the 2012 to 2014 period, 1226
homes were known to have experienced more than one
burglary, and accounted for 2894 burglaries in total,
equivalent to 14.9% of all recorded domestic burglaries
in Belo Horizonte. ere were 1668 repeat domestic bur-
glaries in Belo Horizonte between 2012 and 2014, equat-
ing to 8.6% of all recorded domestic burglaries during
this period.
In any 1year, 201 to 341 repeat burglaries occurred
in Belo Horizonte, accounting for 5.4% of all recorded
domestic burglaries in 2012, 4.8% in 2013 and 3.2% in
2014 (see Table3c). is level of burglary repeats experi-
enced in Belo Horizonte was lower than levels of burglary
repeats in western countries reported in Table1 (where
the range of repeat victimisation from these previous
studies was 7–33%).
An analysis of the statistical significance of repeat bur-
glaries (see Table4) revealed that for each year, the pat-
tern of repeat victimisation was significant (p = 0.05)
within 0–7 days of an originator burglary offence. e
Table 3 The extent ofdomestic burglary repeat victimisa‑
tion inBelo Horizonte, Brazil
2012–2014 2012 2013 2014
(a) Number of locations that experienced a repeat burglary
n locations 1226 295 293 189
(b) Proportion of burglary that took place at locations that experienced
more than one burglary
n 2894 636 620 390
% 14.9 10.0 9.0 6.2
All burglary 19453 6349 6854 6250
(c) Proportion of burglaries that were repeats
n 1668 341 327 201
% 8.6 5.4 4.8 3.2
Table 4 The statistical signicance ofburglary repeat vic‑
timisation inBelo Horizonte, Brazil
p=0.05. Temporal bandwidth 7days, four bands
2012 2013 2014
0–7 days p 0.05 p 0.05 p 0.05
8–14 days p 0.05 p 0.05 n.s.
15–21 days n.s. n.s. p 0.05
22–28 days p 0.05 n.s. n.s.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 7 of 10
Chainey and da Silva Crime Sci (2016) 5:1
pattern of repeat victimisation was not statistically sig-
nificant (p=0.05) for each year between 8–14, 15–21
and 22–28days of an originator burglary. ese results
indicate that in Belo Horizonte, a repeat burglary is more
likely to occur swiftly after (and within 7days) of a previ-
ous burglary offence, rather than at any other time.
Near repeat victimisation
Table5 shows the pattern of near repeats in Belo Hori-
zonte was statistically significant (p0.05) for fourteen
of the twenty different spatial and temporal bands. All of
the six bands within 0 to 21days and within 1 to 200m of
originator incidents were statistically significant and dis-
played the highest levels of risk in comparison to other
bands.
Table 6 shows the number and proportion of near
repeats (of all domestic burglary) for different spatial and
temporal bands. Although the pattern of near repeats
was statistically significant for most of the spatial and
temporal bands that were closest (in space and time) to
originator incidents, less than 6% of all recorded domes-
tic burglaries were near repeats within 200 m and 7days
of an originator offence. ese levels of near repeat bur-
glaries experienced in Belo Horizonte were much lower
than levels of near repeat burglaries found from analy-
sis in the UK and New Zealand (as reported in Table2),
where the comparable levels (for burglaries within 200m
and 7days of an originator offence) were between 12 and
23% of all domestic burglaries.
Discussion andimplications
e analysis of burglary for the city of Belo Horizonte
has revealed that patterns of repeat and near repeat vic-
timisation are statistically significant, with these patterns
following the common typology of the greater level of
risk being soon after an initial burglary offence, and addi-
tionally for near repeats, the highest level of risk being
to those properties that are closest to where the previ-
ous burglary offence took place. However, the analysis
has also revealed that levels of burglary repeats and near
repeats were much lower than those found from studies
in western countries. For example, the level of burglary
repeats in Belo Horizonte in 2014 was half of that meas-
ured in the rural UK county of Northumberland and a
ninth of the levels of repeats for the city of Birmingham
(UK). Similarly, the extent of burglary near repeats in
Belo Horizonte was no more than half the levels of those
found from studies in the UK and New Zealand.
An initial indication of differences between experi-
ences of domestic burglary in Brazil and experiences of
domestic burglary in the UK, USA and other western
countries was illustrated by the differences in domes-
tic burglary rates between these countries. e under-
reporting of crime is an issue in Brazil, like it is in many
other western countries, however, the similarity in crime
reporting levels between Brazil and England and Wales
suggests the under-reporting of crime in Brazil is unlikely
to fully explain the differences in domestic burglary rates.
In addition, the assessment of the police crime record-
ing standards of the data used in the research suggests
that confidence can be placed in the findings, and that a
real difference does exist in the levels of repeats and near
repeats experienced in Belo Horizonte in comparison to
similar studies in western countries.
e domestic housing infrastructure in Brazilian cities
is very different to that in many western countries. Many
more domestic properties in Brazil have situational pre-
vention measures, such as perimeter fences and security
guards, implemented as standard to improve domestic
safety. While the current research study has not statisti-
cally examined the relationship between differences in
housing infrastructure in Belo Horizonte and western
countries, we offer a theoretical reason for explaining
the differences in repeats and near repeats in relation
to these international contextual differences in housing
infrastructure.
e primary theories that explain repeat and near
repeat victimisation are the boost account and foraging
theory, and the flag account. For a property to be singled
Table 5 Spatial and temporal bands for which near
repeats ofburglaries were statistically signicant inBelo
Horizonte, Brazil
0–7days 8–14days 15–21days 22–28days
1–100 m 1.42
p 0.05 1.18
p 0.05 1.21
p 0.05 n.s.
101–200 m 1.24
p 0.05 1.18
p 0.05 1.11
p 0.05 n.s.
201–300 m 1.15
p 0.05 n.s. n.s. 1.09
p 0.05
301–400 m 1.11
p 0.05 1.11
p 0.05 1.07
p 0.05 n.s.
401–500 m 1.18
p 0.05 1.09
p 0.05 1.07
p 0.05 n.s.
Table 6 The proportion ofnear repeats forthree dierent
denitions ofnear inspace andnear intime
Near repeat denition Number ofnear repeats
andproportion ofall burglary
Within 100 m and 7 days 242 (2.2 %)
Within 200 m and 7 days 657 (5.8 %)
Within 300 m and 7 days 1430 (12.7 %)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 8 of 10
Chainey and da Silva Crime Sci (2016) 5:1
out by an offender as a suitable target for an initial bur-
glary requires that property to typically display some
enduring characteristic that makes it more vulnerable
than others. In Belo Horizonte, the opportunities to com-
mit burglary are considered to be lower due to the higher
levels of in-built situational crime prevention that for-
tify homes from would-be offenders. In addition, within
favelas, the close proximity of dwellings and an often
high level of social capital can ward off burglars (Villa-
real and Silva 2006). Furthermore, the common style of
living in high rise apartment blocks in Brazilian cities
such as Belo Horizonte, may also act in naturally limiting
the opportunities for burglary. For instance, it is antici-
pated to be more difficult for an offender to determine if
a home is unoccupied (and, therefore, a potential target
for burglary) if it is on a level above the ground floor. It
is, therefore, likely that the combination of a greater level
of situational crime prevention and more high rise apart-
ments in a Brazilian urban setting results in a lower prev-
alence of flag account opportunities that explain burglary
offending, and as a result lower levels of burglary repeat
victimisation that are associated with the flag account
explanation.
In Brazil, when a home is burgled it has typically
required the offender to overcome the perimeter fencing
that is present, the metal bars across doors and windows,
the security guard at the entrance to the apartment block
or domestic staff who are present in the dwelling, both
on entry and exit. is comes with risks and extra effort
to overcome, and although an offender may have escaped
undetected after committing an initial offence, the expe-
rience of this recent offence is likely to increase the vigi-
lance of the home owners and the other people who are
present (i.e., security and domestic staff) in preventing
a repeat offence from occurring soon. is suggests that
the boost account for explaining burglary repeats is likely
to be less prevalent in the Brazilian context. Foraging
behaviour, involving seeking additional nearby oppor-
tunities, may also be limited due to the risks and efforts
required to overcome the in-built situational crime pre-
vention measures that are present at the nearby proper-
ties. Combined, if the limited opportunities in utilizing
boosts and further foraging following the commission
of a previous burglary offence influences offender deci-
sion-making, then it is likely that fewer repeats and near
repeats will occur swiftly after an initial offence.
Impressive results in crime reduction quickly gain
interest across the international police community. How-
ever, the sharing of good practice on what has worked
to reduce crime requires not only an understanding of
the tactics and initiatives that were applied, but also an
appreciation of whether the context in which replication
is to take place is likely to produce similar results. is
means that tactics and strategies that have been used
elsewhere to predict and prevent domestic burglary by
countering patterns of repeats and near repeats will only
have a high level of impact if the extent of repeats and
near repeats account for a large proportion of all domes-
tic burglaries.
Effective burglary prevention programmes such as
those in Manchester and Edmonton displayed high levels
of burglary repeats and near repeats prior to the imple-
mentation of the initiatives to counter the risk of repeats
and near repeats. In areas where the levels of repeats and
near repeats are not as high, the impact of the same tac-
tics and strategies are likely to have less impact (i.e., there
are fewer burglaries that can be countered using tactics
to prevent repeats and near repeats). is means that
burglary prevention programmes that focus on reduc-
ing repeat and near repeat victimisation are likely to have
less of an impact in reducing burglary in Belo Horizonte
and in other Latin American cities where levels of repeats
and near repeats are low. As an estimate for Belo Hori-
zonte, taking the example of 2014 where burglary repeat
victimisation and near repeat victimisation accounted
for 3.2 and 5.8% respectively of all burglaries, a crime
prevention programme designed specifically to counter
repeats and near repeats may only yield an overall bur-
glary reduction of 9%. In addition, the results from Belo
Horizonte also suggest that predictive policing software
that includes algorithms for predicting burglary based
on the patterns of repeats and near repeats may not be as
effective in Latin American countries where domestic liv-
ing and housing infrastructure differ greatly to the west-
ern cities on which the software has been designed.
At present, victimisation surveys in Brazil do not probe
on experiences of repeat victimisation, so little else is
available other than recorded crime data that allows for
analysis of the extent of these experiences. Also, to date,
research in Brazil has not been conducted that has
involved interviewing offenders about their decision-
making in selecting properties to burgle and whether the
concepts of the boost and flag accounts feature in this
decision-making. is type of questioning of offenders is
an obvious area for future research that will allow the
examination of whether these contextual differences
between Brazil (and other Latin American countries) and
western countries influence offender decision-making.17
Additional areas for future research could involve
17 In particular, little is known about the differences in the levels of
attempted burglary between Latin American and western countries, and
the impact a failed attempt at committing a burglary has on boosting the
offender to return to the same property or seek other burglary opportunities
nearby. In the Latin American context where the successful commission of
burglary may be lower, examining the impact of failed attempts to commit
burglary on how it then influences future offending behavior would benefit
from further research.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 9 of 10
Chainey and da Silva Crime Sci (2016) 5:1
analysis of burglary near repeat patterns in relation to
differences in burglary rates, housing density and hous-
ing type, and research that aims to distinguish whether it
is the preponderance of high rise buildings, or whether
the presence of situational domestic security (such as
perimeter fencing, security guards, or the presence of
domestic staff) are the reasons for lower rates of burglary
and lower levels of repeat and near repeat victimisation
in Brazilian cities.
Conclusions
Policing agencies often draw from the successful practice
of others, and apply this practice to the crime problems
they experience. However, rather than just applying what
has worked for someone else in reducing crime, police
decision-makers must also determine how the practice
works and if it is applicable to their context. Patterns of
repeat victimisation and near repeat victimisation have
been observed in many studies conducted in western
countries, with these patterns considered to provide a
practical means for police agencies to predict and pre-
vent further burglaries from occurring.
is research has shown that the extent of burglary
repeat and near repeat victimisation in Belo Horizonte
was much lower than observed in similar studies in cit-
ies in western countries. e lower rates of burglary in
Brazil, and the lower levels of burglary repeats and near
repeats in Belo Horizonte suggest there is an impor-
tant contextual difference between the commission of
and opportunties for burglary in Brazil when compared
to western countries. We argue that this contextual dif-
ference is likely to be due to differences in domestic liv-
ing and housing infrastructure, with dwellings in Brazil
tending to be designed or purposefully adapted to pro-
vide higher levels of situational domestic security when
compared to dwellings in cities in western countries.
While further research into understanding how differ-
ences in the domestic living infrastructure in Brazil influ-
ences offender decision-making would be useful, the
study illustrates the importance of examining whether
the crime patterns on which successful crime preven-
tion practice is based are similarly present where this
practice is being considered for replication. In terms of
applying the prevention practice for reducing burglary
by predicting where burglaries are likely to occur (based
on patterns of repeat and near repeat victimisation), the
current study shows that in a Brazilian urban context
the same practice will likely yield lower levels of burglary
reduction.
Authors’ contributions
Spencer Chainey is the lead author of this article. Spencer conceived the idea
of the research, planned its design, conducted the majority of the analysis
and drafted the manuscript. Braulio Silva sourced the recorded crime data,
provided comment on the research results, and drafted sections relating to
the contextual characteristics of Brazil and Belo Horizonte. Spencer has visited
Belo Horizonte on several occasions, met with the Policia Militar of Minas Ger-
ais (with Braulio) to support the research and discuss the results. Braulio has
approved the final version of the article. We are accountable for all aspects of
the work and ensure that questions related to the accuracy or integrity of any
part of the work will be appropriately investigated and resolved. Both authors
read and approved the final manuscript.
Author details
1 Department of Security and Crime Science, University College London, 35
Tavistock Square, London, England, UK. 2 Department of Sociology, Federal
University of Minas Gerais, Belo Horizonte, Brazil.
Acknowledgements
The research (involving a visit to Belo Horizonte to conceive the research,
examine police recorded crime data, conduct preliminary analysis and
conduct field visits to areas of different housing types) was supported by an
award from the Santander Universities Research Catalyst Awards for 2014. The
completion of the research and manuscript preparation was supported by
each of the author’s universities.
Competing interests
The authors declare that they have no competing interests.
Received: 2 July 2015 Accepted: 27 January 2016
References
Azavea. (2015). Next generation predictive policing. https://www.hunchlab.com.
Accessed May 27, 2015.
BBC. (2012). Could ‘predictive policing’ help prevent burglary? http://www.bbc.
co.uk/news/uk-19623631 Accessed May 27, 2015.
Beato Filho, C. C. (2008). Compreendendo e Avaliando Projetos de Segurança
Publica. Belo Horizonte: Editora UFMG.
Bernasco, W. (2008). Them again? Same-offender involvement in repeat and
near repeat burglaries. European Journal of Criminology, 5, 411–431.
Bowers, K. J., & Johnson, S. D. (2004). Who commits near repeats? A test of the
boost explanation. Western Criminology Review, 5, 3.
Bowers, K. J., & Johnson, S. D. (2005). Domestic burglary repeats and space-
time clusters: The dimensions of risk. European Journal of Criminology, 2(1),
67–92.
Bowers, K. J., Johnson, S., & Pease, K. (2004). Prospective hotspotting: The future
of crime mapping? British Journal of Criminology, 44(5), 641–658.
Brazil Business. (2014). Rules for domestic workers in Brazil. http://thebrazilbusi-
ness.com/article/rules-for-domestic-workers-in-brazil. Accessed May 27,
2015.
Brookings. (2012). Belo Horizonte metropolitan area profile. http://www.
brookings.edu/about/programs/metro//~/media/Research/Files/
Papers/2012/11/30-metro-brazil-economy/30-brazil-profiles-en/Belo-
Horizonte.pdf. Accessed May 27, 2015.
Caldeira, T. P. (2000). A cidade de muros: Crime, segregação e cidadania em São
Paulo. São Paulo: EDUSP.
Chainey, S. P. (2012). JDI Brief: Predictive mapping (predictive policing). London:
UCL. http://discovery.ucl.ac.uk/1344080/. Accessed May 27, 2015.
Chainey, S. P. (2014). Examining the extent to which hotspot analysis can support
spatial predictions of crime. Ph.D. thesis, Department of Security and Crime
Science, University College London.
Chainey, S. P., & Ratcliffe, J. H. (2005). GIS and crime mapping. London: Wiley.
Chainey, S. P., & Silva, B. (2015). The predictability of domestic burglaries in
Belo Horizonte: translating repeats, near repeats and hotspot analysis to
a Brazilian context. Presentation at the international crime and intel-
ligence analysis conference, Manchester, 2015. http://www.ucl.ac.uk/
jdi/events/int-CIA-conf/icia-15/ICIAC15_6B_SCHAINEY.pdf. Accessed
May 27, 2015.
CNT/SENSUS. (2010). Pesquisa de Opinião Pública NacionalRodada 100.
Relatório de cruzamentos. 25 a 29 de janeiro de 2010. Available at http://
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Page 10 of 10
Chainey and da Silva Crime Sci (2016) 5:1
www.cnt.org.br/Imagens%20CNT/PDFs%20CNT/Pesquisa%20CNT%20
Sensus/2010/100%20Cruzamento.pdf. Accessed March 30, 2011.
CRISP. (2006). Pesquisa de Vitimização na Região Metropolitana de Belo Horizonte.
Belo Horizonte: UFMG.
CRISP/Datafolha/Ministério da Justiça. (2011). Pesquisa Nacional de Viti-
mização. http://www.crisp.ufmg.br/wp-content/uploads/2013/10/
Relat%C3%B3rio-PNV-Senasp_final.pdf. Accessed March 30, 2015.
DIEESE—Departamento Intersindical de Estatísticas e Estudos Socioeconômi-
cos. (2013). Relatório: Emprego Doméstico no Brasil. http://www.dieese.
org.br/estudosetorial/2013/estPesq68empregoDomestico.pdf. Accessed
February 25, 2015.
Emporis. (2015). Cities with the most high-rise buildings. https://en.wikipedia.
org/wiki/List_of_cities_with_the_most_high-rise_buildings. Accessed
May 27, 2015.
Ericsson, U. (1995). Straight from the Horse’s Mouth. Forensic Update, 43, 23–25.
Farrell, G., & Bouloukos, A. (2001). International overview: A cross-national
comparison of rates of repeat victimization. In G. Farrell & K. Pease (Eds.),
Repeat victimization. Crime prevention studies (Vol. 12). New York: Criminal
Justice Press.
Farrell, G., & Pease, K. (1993). Once bitten, twice bitten: Repeat victimisation and its
implications for crime prevention, Crime Prevention Unit Paper 46. London:
Home Office.
Farrell, G., Sousa, W. H., & Lamm Weisel, D. (2002). The time-window effect in
the measurement of repeat victimization: A methodology for its meas-
urement and an empirical study. Crime Prevention Studies, 13, 15–27.
Fielding, M., & Jones, V. (2012). Disrupting the optimal forager: Predictive risk
mapping and domestic burglary reduction in Trafford, Greater Manches-
ter. International Journal of Police Science and Management, 14(1), 30–41.
Forrester, D., Chatterton, M., & Pease, K. (1988). The Kirkholt Burglar y Pre vention
Project, Rochdale, Crime Prevention Unit Paper 13. London: Home Office.
Fórum Brasileiro de Segurança Pública. (2011). Anuário brasileiro de segurança
pública [Annual Brazilian report on public security]. São Paulo, Brasil: Fórum
Brasileiro de Segurança Pública.
Grove, L. E., Farrell, G., Farrington, D. P., & Johnson, S. D. (2012). Preventing repeat
victimisation: A systematic review. Stockholm, Sweden: The Swedish Coun-
cil for Crime Prevention.
Haberman, C. P., & Ratcliffe, J. H. (2012). The predictive policing challenges
of near repeat armed street robberies. Policing: A Journal of Policy and
Practice, 6(2), 151–166.
IBGE—Instituto Brasileiro de Geografia e Estatística. (2010). Censo Demográfico
2010Belo Horizonte. http://www.cidades.ibge.gov.br/. Accessed May
27, 2015.
Johnson, S. D., Bernasco, W., Bowers, K. J., Elffers, H., Ratcliffe, J. H., Rengert, G.
F., & Townsley, M. (2007). Space–time patterns of risk: A cross national
assessment of residential burglary victimization. Journal of Quantitative
Criminology, 23(3), 201–219.
Johnson, S. D., & Bowers, K. J. (2004a). The stability of space-time clusters of
burglary. British Journal of Criminology, 44(1), 55–65.
Johnson, S. D., & Bowers, K. J. (2004b). The burglary as clue to the future: The
beginnings of prospective hotspotting. European Journal of Criminology,
1(2), 237–255.
Johnson, S., Bowers, K. J., Birks, D., & Pease, K. (2009). Predictive mapping: Accu-
racy, units of analysis and the environmental backcloth. In D. Weisburd, W.
Bernasco, & G. Bruinsma (Eds.), Putting crime in it’s place: Units of analysis in
geographical criminology. New York, NY: Springer.
Murray, J., de Castro Cerqueira, D. R., & Kahn, T. (2013). Crime and violence in
Brazil: Systematic review of time trends, prevalence rates and risk factors.
Aggression and Violent Behavior, 18(5), 471–483.
New Zealand Police. (2014a). Analysis of domestic burglary repeats and near
repeats in Auckland Central and Counties Manukau Central, Wellington:
New Zealand Police [Restricted].
New Zealand Police. (2014b). Analysis of domestic burglary repeats and near
repeats in Wellington and Kapiti-Mana. Wellington: New Zealand Police
[Restricted].
Northamptonshire Police. (2013). Problem profile of domestic burglary in Ketter-
ing. Northampton: Northamptonshire Police [Restricted]
Northumberland Community Safety Partnership. (2009). Strategic assessment of
crime and community safety. Morpeth: Northumberland County Council.
[Restricted]
ONS—Office for National Statistics. (2013). Crime Statistics, Focus on
Property Crime, 2012/13. http://www.ons.gov.uk/ons/rel/crime-stats/
crime-statistics/focus-on-property-crime–2012-13/index.html. Accessed
May 27, 2015.
ONS—Office for National Statistics. (2015). Crime Statistics for England and
Wales, year ending September 2014. http://www.ons.gov.uk/ons/rel/crime-
stats/crime-statistics/year-ending-september-2014/index.html. Accessed
April 9, 2015.
Paixão, L. A. R. (1991). Segurança privada, direitos humanos e democracia:
Notas preliminares sobre novos dilemas politicos. Novos Estudos CEBRAP,
31, 131–141.
Pease, K. (1998). Repeat victimization: Taking stock, Home Office Police Research
Group, crime detection and prevention series, paper 90. London: Home
Office.
Polvi, N., Looman, T., Humphries, Ch., & Pease, K. (1991). The time-course of
repeat burglary victimization. British Journal of Criminology, 31, 411–414.
PredPol. (2013). Predict crime in real time. http://www.predpol.com/. Accessed
January 22, 2014.
Professional Security. (2012). Tackling burglary. http://www.professionalsecu-
rity.co.uk/news/case-studies/tackling-burglary/. Accessed January 22,
2014.
Ratcliffe, J. H. (2009). Near Repeat Calculator (version 1.3). Temple University,
Philadelphia, PA and the National Institute of Justice, Washington, DC.
http://www.jratcliffe.net/software/. Accessed May 27, 2015.
Ratcliffe, J. H., & Rengert, G. F. (2008). Near-repeat patterns in Philadelphia
shootings. Security Journal, 21, 58–76.
Safe Newcastle Partnership. (2011). Strategic assessment of crime and commu-
nity safety. Newcastle-upon-Tyne: Newcastle City Council. [Restricted]
da Silva, B. F. A., & Beato Filho, C. C. (2013). Ecologia social do medo: avaliando
a associação entre contexto de bairro e medo de crime. Revista Brasileira
de Estudos de População, 30, 155–170.
SIM/DataSUS. (2014). Ministério da Saúde. Sistema de Informação sobre Mortali-
dade. Brasília-DF. www.datasus.gov.br Accessed January 22, 2014.
Skogan, W. G. (1996). The Decade’s most important criminological insight.
National Institute of Justice: Research in Action. Washington: Department
of Justice.
Townsley, M., Homel, R., & Chaseling, J. (2003). Infectious burglaries. A test of
the near repeat hypothesis. British Journal of Criminology, 43(3), 615–633.
University College London. (2014). Forecasting the time and place of crime hot-
spots. https://www.ucl.ac.uk/impact/case-study-repository/forecasting-
crime-hotspots. Accessed May 27, 2015.
UNODC—United Nations Office on Drugs and Crime. (2014). UNODC interna-
tional burglary, car theft and housebreaking statistics. https://www.unodc.
org/unodc/en/data-and-analysis/statistics/data.html. Accessed May 27,
2015.
Villarreal, A., & Silva, B. F. A. (2006). Social cohesion, criminal victimization and
perceived risk of crime in Brazilian neighborhoods. Social Forces, 84(3),
1725–1753.
Wells, W., Wu, L., & Ye, X. (2011). Patterns of near-repeat gun assaults in Hou-
ston. Journal of Research in Crime and Delinquency, 49, 186–212.
West Mercia Police. (2012). Near repeat analysis of domestic burglary in Hereford-
shire. Worcester: West Mercia Police [Restricted].
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
... Since we have no specific focus in investigating these general spatial effects, we opted for this subsampling approach. 7 It is necessary to distinguish the burglary risks between houses and apartments because they are significantly different residence settings in Brazil (Chainey & da Silva, 2016). For urban permanent private residences, houses represent 85% of all dwellings in the country, while apartments represent 12.5%, although a greater proportion of apartment dwelling can be observed in the larger cities (Chainey & da Silva, 2016). ...
... 7 It is necessary to distinguish the burglary risks between houses and apartments because they are significantly different residence settings in Brazil (Chainey & da Silva, 2016). For urban permanent private residences, houses represent 85% of all dwellings in the country, while apartments represent 12.5%, although a greater proportion of apartment dwelling can be observed in the larger cities (Chainey & da Silva, 2016). When analyzed with family income, the apartments and gated communities represent the most common typology for families with higher income (Triana et al., 2015), and most of apartments in Brazil have private security and/or CCTV paid monthly by its residents. ...
Article
Although income inequality has been often pointed out as an important cause of crime, it is yet unclear how spatial patterns of income in a city can explain its geography of crime. In this study, we apply a model to test the influence of income inequality on the spatial concentration of residential burglaries in the city of Campinas, Brazil. Following criminological theory, our model decomposes income inequality into two hypothetical effects: that of local income, which determines how attractive residences are to burglary, and exposure to poverty, where poverty boosts criminal motivation through economic hardship. Our study reveals that higher local income is indeed significant and positively associated to higher burglary risk, but that exposure to poverty does not increase risk. Therefore, higher income areas more surrounded by poor areas do not feature a particularly increased burglary risk if compared to other higher income areas, contrary to what could be expected from some criminological frameworks such as relative deprivation and strain theories. Instead, our findings suggest that the geography of residential burglaries can be explained by the distribution of burglary opportunities, that is, of where the most profitable targets are. To conclude, we compare our findings to other existing studies.
... Part of our understanding of risky places results from extensive studies into who are the victims of crime and how this victimization can manifest in risky places, and two phenomena we identify in Chapter 3 were Repeat Victimization (RV) and 'Near Repeat Victimization' (NRV), which we found in the Global North and Global South (Chainey & da Silva, 2016). Whilst we have a good understanding of repeat victimization at risky places (Chapters 4-6), we rarely consider the specific set of circumstances that brought victims to that place. ...
... Part of our understanding of risky places results from extensive studies into who are the victims of crime and how this victimization can manifest in risky places, and two phenomena we identify in Chapter 3 were Repeat Victimization (RV) and 'Near Repeat Victimization' (NRV), which we found in the Global North and Global South (Chainey & da Silva, 2016). Whilst we have a good understanding of repeat victimization at risky places (Chapters 4-6), we rarely consider the specific set of circumstances that brought victims to that place. ...
... Studies of burglary patterns provided the first empirical validations of the near-repeat phenomenon (Bediroglu et al. 2018;Bernasco 2008;Chainey & Da Silva 2016;Hino & Amemiya 2019;Johnson et al. 2007;Johnson et al. 2009;Moreto et al. 2014;Short 2009;Townsley et al. 2003). Scholars have recently extended the nearrepeat concept to several crime types, including shootings, gun assaults, and hand grenade attacks (Loeffler et al. 2017;Ratcliffe & Rengert 2008;Sturup et al. 2017;Sturup et al. 2019;Wells & Wu 2012), insurgent attacks in Iraq (Townsley et al. 2008), ...
Article
Purpose Initiator crimes, also known as near-repeat crimes, occur in places with known risk factors and vulnerabilities based on prior crime-related experiences or information. Consequently, the environment in which initiator crimes occur might be different from more general crime environments. This study aimed to analyse the differences between the environments of initiator crimes and general crimes, confirming the need for predicting initiator crimes. Design/methodology/approach We compared predictive models using data corresponding to initiator crimes and all residential burglaries without considering repetitive crime patterns as dependent variables. Using random forest and gradient boosting, representative ensemble models and predictive models were compared utilising various environmental factor data. Subsequently, we evaluated the performance of each predictive model to derive feature importance and partial dependence based on a highly predictive model. Findings By analysing environmental factors affecting overall residential burglary and initiator crimes, we observed notable differences in high-importance variables. Further analysis of the partial dependence of total residential burglary and initiator crimes based on these variables revealed distinct impacts on each crime. Moreover, initiator crimes took place in environments consistent with well-known theories in the field of environmental criminology. Originality/value Our findings indicate the possibility that results that do not appear through the existing theft crime prediction method will be identified in the initiator crime prediction model. Emphasising the importance of investigating the environments in which initiator crimes occur, this study underscores the potential of artificial intelligence (AI)-based approaches in creating a safe urban environment. By effectively preventing potential crimes, AI-driven prediction of initiator crimes can significantly contribute to enhancing urban safety.
Article
Certain aspects of optimal forager theory (OFT), which is drawn from ecology, have shown positive results in predicting areas at risk of future domestic burglary offending. This led to police services developing analysis methods that embraced OFT to underpin their deployment of resources to prevent or reduce domestic burglary. There has been limited examination, using quantitative approaches, of how individual police services have implemented such crime reduction schemes. This study broadens this literature by qualitatively exploring OFT strategies within five police services. By interviewing participants involved in the programmes the study gathers views and perspectives of the implementations, identifying many positive by-products of the strategies. By contrast, factors affecting the implementation and application of the theoretical framework are also identified. Both good and bad are discussed in the context of their practical implications for police services globally looking to implement crime reduction plans that embrace OFT.
Article
Full-text available
Interview given by Professor Spencer Chainey, Professor of Security and Criminal Science at University College London, to Federal Police Chief Wellington Clay Porcino Silva, through the Teams platform, on September 16, 2022, detailing the relationship between Prof. Doctor Spencer Chainey and the Revista Brasileira de Ciências Policiais, as well as the researcher's research and relationship with the Federal Police and the results of the Intelligence-Oriented Policing Project, carried out with support from the IADB (Inter-American Development Bank) and the Brazilian Federal Police.
Article
Full-text available
Women in South and Southeast Asia encounter unique mobility barriers which are a combination of poor services by public transport modes and underlying patriarchal societal norms. Although international organisations provide guidelines for national policy makers to develop inclusive public transport systems, women’s mobility remains restricted and unsafe. This paper provides a critical review on women’s mobility barriers from built-environment to policy for public transport ridership. It includes three main aspects. Firstly, the key barriers encountered by women from poor service quality, sexual harassment and patriarchal societal norms. Secondly, the limitations in common methods adopted to measure these barriers. Finally, the effectiveness of international guidelines and national policies on women’s travel needs for public transport ridership. Findings revealed that women’s mobility barriers in South and Southeast Asian countries originate from the lack of adequate inclusive policies and protection laws from authorities. The underlying patriarchal societal norms form a toxic base, which allow for severe forms of sexual harassment to take place when riding public transport and for women to experience victim-blaming, if the incidents are reported. The paper concludes with knowledge gaps to assist practitioners and researchers to move toward safer journeys and development of inclusive public transport systems for women in developing countries.
Article
Full-text available
Crime is a threat to any nation’s security administration and jurisdiction. Therefore, crime analysis becomes increasingly important because it assigns the time and place based on the collected spatial and temporal data. However, old techniques, such as paperwork, investigative judges, and statistical analysis, are not efficient enough to predict the accurate time and location where the crime had taken place. But when machine learning and data mining methods were deployed in crime analysis, crime analysis and predication accuracy increased dramatically. In this study, various types of criminal analysis and prediction using several machine learning and data mining techniques, based on the percentage of an accuracy measure of the previous work, are surveyed and introduced, with the aim of producing a concise review of using these algorithms in crime prediction. It is expected that this review study will be helpful for presenting such techniques to crime researchers in addition to supporting future research to develop these techniques for crime analysis by presenting some crime definition, prediction systems challenges and classifications with a comparative study. It was proved though literature, that supervised learning approaches were used in more studies for crime prediction than other approaches, and Logistic Regression is the most powerful method in predicting crime.
Technical Report
Full-text available
Edição Especial do Anuário Brasileiro de Segurança Pública em que as informações são apresentadas para os últimos anos disponíveis da série e organizadas para as unidades da Federação. Os dados de cada UF são acompanhados por textos analíticos. A publicação também conta com um mapa das facções prisionais identificadas também por estado.
Book
Full-text available
This report pulls together a number of research results from a variety of sources, much of it carried out with Home Office support. The subject of the report is ‘repeat victimisation’ – the paper describes the extent to which victims or places are repeatedly subject to crime and speculates about the implications for prevention. In relation to some offences the repeated vulnerability of particular individuals is self evident – domestic violence is probably the most obvious example. But in relation to other crimes, such as domestic burglary, attacks on schools or car crime the extent to which repetition occurs is far from obvious but clearly shown in the report. Some of the research had been lying in the academic arena for a long time – but its practical significance for prevention and for policing had not been appreciated. The reduction of repeat victimisation in its several manifestations offers a challenge to the police and their partners in crime prevention. The report is intended to provoke discussion and preventive action across a wide field.
Article
This briefing note provides details of the predictive mapping approach initially implemented in Trafford, Greater Manchester, drawing from research from the UCL Jill Dando Institute of Security and Crime Science. It also describes a refinement to the original processes adopted in Trafford after an evaluative analysis and site visits. The initial focus in Greater Manchester has been on residential burglary. This predictive mapping is now being extended to other types of crimes such as pedal cycle theft, vehicle crime and street robbery. The approach uses standard technology available to police and community safety agencies, and in Trafford’s case the responses have made use of existing resources. The briefing note describes the theory that underpins the predictive mapping approach, analysis that should be conducted, police and local partner response opportunities, results from Trafford, and practical considerations. The final section lists a number of resources for further information.
Thesis
The premise that where crime has occurred previously, informs where crime is likely to occur in the future has long been used for geographically targeting police and public safety services. Hotspot analysis is the most applied technique that is based on this premise – using crime data to identify areas of crime concentration, and in turn predict where crime is likely to occur. However, the extent to which hotspot analysis can accurately predict spatial patterns of crime has not been comprehensively examined. The current research involves an examination of hotspot analysis techniques, measuring the extent to which these techniques accurately predict spatial patterns of crime. The research includes comparing the prediction performance of hotspot analysis techniques that are commonly used in policing and public safety, such as kernel density estimation, to spatial significance mapping techniques such as the Gi* statistic. The research also considers how different retrospective periods of crime data influence the accuracy of the predictions made by spatial analysis techniques, for different periods of the future. In addition to considering the sole use of recorded crime data for informing spatial predictions of crime, the research examines the use of geographically weighted regression for determining variables that statistically correlate with crime, and how these variables can be used to inform spatial crime prediction. The findings from the research result in introducing the crime prediction framework for aiding spatial crime prediction. The crime prediction framework illustrates the importance of aligning predictions for different periods of the future to different police and prevention response activities, with each future time period informed by different spatial analysis techniques and different retrospective crime data, underpinned with different theoretical explanations for predicting where crime is likely to occur.
Book
Book description: The growing potential of GIS for supporting policing and crime reduction is now being recognised by a broader community. GIS can be employed at different levels to support operational policing, tactical crime mapping, detection, and wider-ranging strategic analyses. With the use of GIS for crime mapping increasing, this book provides a definitive reference. GIS and Crime Mapping provides essential information and reference material to support readers in developing and implementing crime mapping. Relevant case studies help demonstrate the key principles, concepts and applications of crime mapping. This book combines the topics of theoretical principles, GIS, analytical techniques, data processing solutions, information sharing, problem-solving approaches, map design, and organisational structures for using crime mapping for policing and crime reduction. Delivered in an accessible style, topics are covered in a manner that underpins crime mapping use in the three broad areas of operations, tactics and strategy. * Provides a complete start-to-finish coverage of crime mapping, including theory, scientific methodologies, analysis techniques and design principles. * Includes a comprehensive presentation of crime mapping applications for operational, tactical and strategic purposes. * Includes global case studies and examples to demonstrate good practice. * Co-authored by Spencer Chainey, a leading researcher and consultant on GIS and crime mapping, and Jerry Ratcliffe, a renowned professor and former police officer. This book is essential reading for crime analysts and other professionals working in intelligence roles in law enforcement or crime reduction, at the local, regional and national government levels. It is also an excellent reference for undergraduate and Masters students taking courses in GIS, Geomatics, Crime Mapping, Crime Science, Criminal Justice and Criminology.
Article
In this chapter, we endeavour to predict the trends in predictive microbiology that are going to shape its development as a multidisciplinary field. We integrate the most commonly used, population and single-cell models of bacterial kinetics into a top-down framework. We predict that modellers will need to face various complexities induced by interactions at different levels: between food and microorganism, between cells and species and between molecular elements at the intracellular level. Hence, advances in the area of complex systems (e.g. network science, systems biology and stochastic modelling) will have a significant effect on the future development of predictive microbiology.
Article
Predicting when and where crimes are likely to occur is crucial for prioritizing police resources. In a companion paper (Johnson and Bowers 2004), we showed that burglaries clustered within 1-2 months and up to 300-400 metres of a prior burglary, i.e. a domestic burglary could profitably trigger time- and space-limited resource deployment around the burglary. Research is here presented which refines the priorities that should be given in such deployment. It demonstrates that (a) whereas repeat victimization proper tended to occur in more deprived areas, space-time clustering was more evident in affluent areas, (b) houses next to a burgled home were at a substantially heightened risk relative to those located further away, particularly within one week of an initial burglary, (c) properties located on the same side of the street as a burgled house were at significantly greater risk compared with those opposite, even when corrections are made for differences in linear distances between homes, (d) houses with probably identical layouts (e.g. houses two, four, six, etc., doors away from a burgled property) were slightly more at risk than those with the reverse layout, but these differences are too slight to inform crime reduction practice. Taken together, these patterns may be used to prioritize attention within the two months after, and up to 400 metres from, a prior burglary identified by Johnson and Bowers (2004) as encompassing homes at elevated burglary risk.
Article
This paper explores one aspect of spatial dependence for the offence of burglary, utilising epidemiological methods for the study of infectious diseases to investigate the phenomenon of near repeat victimization. The near repeat burglary hypothesis states that proximity to a burgled dwelling increases burglary risk for those areas that have a high degree of housing homogeneity and that this risk is similar in nature to the temporarily heightened risk of becoming a repeat victim after an initial victimization. The near repeat hypothesis was tested on 34 months of police recorded burglary data across a high crime area of Brisbane, Australia. Near repeats were shown to exist in the study area, mainly in suburbs containing homogeneous housing. Little or no housing diversity, in terms of the type of physical construction and general appearance of dwellings, serves to restrict the extent of repeat victimization. Housing diversity allows offenders a choice of targets, and favoured targets will be 'revisited' by burglars. Near identical targets usually present no motive for an offender to favour one property over another. Thus in areas with low housing diversity, victim prevalence should be higher than in areas with heterogeneous housing.