Article Research Questions -2pages

Topnotch4

GreenspaceandCrime.pdf

Home >Social Science homework help >Article Research Questions -2pages

Article

Greenspace and Crime: An Analysis of Greenspace Types, Neighboring Composition, and the Temporal Dimensions of Crime

Anthony Kimpton 1 , Jonathan Corcoran

1 ,

and Rebecca Wickes 1

Abstract Objectives: There is a growing interest in the relationship between green- space and crime, yet how particular greenspace types encourage or inhibit the timing and types of greenspace crime remains largely unexplored. Drawing upon recent advances in environmental criminology, we introduce an integrated suite of methods to examine the spatial, temporal, and neigh- borhood dynamics of greenspace crime. Methods: We collate administra- tive, census, and crime incident data and employ cluster analysis, circular statistics, and negative binomial regression to examine violent, public nui- sance, property, and drug crimes within 4,265 greenspaces across Brisbane, Australia. Results: We find that greenspace amenities, neighborhood social composition, and the presence of proximate crime generators influence the

1 The University of Queensland, Brisbane, Queensland, Australia

Corresponding Author:

Anthony John Kimpton, The University of Queensland, Brisbane, Queensland 4072, Australia.

Email: [email protected]

Journal of Research in Crime and Delinquency

1-35 ª The Author(s) 2016

Reprints and permission: sagepub.com/journalsPermissions.nav

DOI: 10.1177/0022427816666309 jrc.sagepub.com

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://www.sagepub.com/journalsPermissions.nav

http://jrc.sagepub.com

http://jrc.sagepub.com/

frequency and timing of greenspace crime. Conclusions: Our analyses reveal that particular types of greenspaces are more crime prone than others. We argue that this is largely due to the presence of amenities within green- spaces allied with the sociodemographic context of surrounding neighbor- hoods. We conclude that understanding how these factors influence the behaviors of potential offenders, victims, and guardians is necessary to better understand the spatial distribution of greenspace crime and provide an evidence base for crime prevention initiatives.

Keywords routine activity theory, criminological theory, urban crime, statistical methods, quantitative research, research methods, parks, amenities

Introduction

Greenspaces refer to a range of different public spaces including parks,

gardens, greened thoroughfares, sporting fields, and ovals. They are an

important urban design feature, as they provide unique health benefits for

local residents that include filtering and sequestering airborne and water-

borne toxins (Yang et al. 2005), counter the urban heat island effect (Bowler

et al. 2010; Feyisa, Dons, and Meilby 2014; Kong et al. 2014; Li et al.

2012), and assist in the development of immunity responses against aller-

gens (Hanski et al. 2012). Further, greenspaces offer social benefits for local

residents by strengthening place attachment (Hur, Nasar, and Chun 2010;

Kim and Kaplan 2004; McCunn and Gifford 2014) and social cohesion

(Mason 2010).

While the health and social benefits of greenspace are well supported,

emerging research indicates that greenspace can also generate crime. Scho-

lars suggest that a greenspace can function as a ‘‘social hole,’’ which dis-

rupts community processes necessary for preventing crime (Hipp et al.

2014). However, as greenspaces are morphologically distinct, their ability

to generate crime is arguably also distinct. Greenspace types that include

hidden areas can create opportunities for consensual crimes such as drug

use (Felson and Boba 2010; Hope 1982; Knutsson 1997). Further, green-

space types that attract legitimate greenspace users with playgrounds or

sporting features can also create opportunities for predatory crimes such

as panhandling or pickpocketing, and more violent predatory offenses such

as robbery, assault, or rape (Ceccato 2014; Ellickson 1996; Groff and

McCord 2012). Last, greenspace can provide opportunities for young

2 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

people to engage in public nuisance away from adult handlers and guardians

(Dinkes et al. 2009; Felson and Boba 2010; Snyder and Sickmund 2006).

Since greenspaces can be both places of legitimate public recreation and

places of crime, legitimate users may avoid these areas if problems arise.

This avoidance has consequences for individuals, communities, and moti-

vated offenders. Individuals avoiding greenspaces for safety also forgo the

previously described associated health and social benefits (Hanski et al.

2012; McCunn and Gifford 2014). In areas where many residents avoid

greenspaces, social networks may diminish, which can have a negative

influence on positive social processes (Bairner and Shirlow 2003; Haber-

mas 1991; Jorgensen, Ellis, and Ruddell 2013; Palmer et al. 2005). Last,

motivated offenders may become more inclined to offend when there are

fewer residents to intervene. In time, these greenspaces develop into gang

‘‘set spaces’’ that further increase crime throughout the surrounding areas

(Stodolska, Acevedo, and Shinew 2009; Tita, Cohen, and Engberg 2005).

Despite an emerging interest in crime within greenspace (herein referred

to as ‘‘greenspace crime’’), significant gaps in our understanding of this

association remain. First, as Groff and McCord (2012) argue, the literature

on greenspace crime is mostly theoretical, and the few empirical studies

tend to examine case studies rather than citywide variation in greenspace

crime (for exceptions, see Anderson and West 2006; Crewe 2001; McCord

and Houser 2015). Second, no study compares crime across greenspace

types despite the functional and morphological variety of greenspaces.

According to crime pattern theory and routine activity theory, greenspaces

are ‘‘behavior settings’’ and particular greenspace amenities can be condu-

cive to particular types of crime (Brantingham and Brantingham 1995;

Felson and Boba 2010). Thus, the microplace features ultimately determine

when and where opportunities for crime exist (Clarke 2012). Further, the

temporal dynamics of greenspace crime between greenspace types are not

well understood. Routine activity theory argues that the lifestyle routines of

offenders, victims, and guardians determine whether, where, and when

offending occurs (see Felson and Boba 2010; Haberman and Ratcliffe

2015), but the extent to which greenspace crime corresponds to daily or

weekly lifestyle routines is unclear. Last, there is insufficient scholarship to

reject that greenspace crime may be simply a function of nearby social

context and crime generators. Certainly, we know that crime spatially con-

centrates within poorer neighborhoods (Hirschfield, Bowers, and Brown

1995; Lockwood 2007; Storr, Chen, and Anthony 2004), but we also know

that poorer neighborhoods generally feature fewer greenspaces with limited

amenity variety (Astell-Burt et al. 2014; Crawford et al. 2008; Macintyre,

Kimpton et al. 3

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Macdonald, and Ellaway 2008; Timperio et al. 2007). Thus, prior studies

could be observing the neighborhood externalities of greenspace crime.

This study addresses these four gaps by spatially integrating crime inci-

dent data, census, land use, and local council asset registers into a single

data set. We conduct a cluster analysis to classify greenspaces into a set of

mutually exclusive types and then compare the concentrations of violent,

drug, property damage, theft, or public nuisance crimes across each type.

Next, we employ circular statistics to compare the temporal dynamics of

crime across the greenspace types and then by crime type. Here we examine

whether the greenspace type is associated with particular timings and types

of greenspace crime. The final component of our analysis computes a set of

negative binomial regression models to establish the extent to which the

adjacent social context and the presence of proximal crime generators help

explain variations in greenspace crime.

Background

Greenspace as a Behavior Setting

The emergence of environmental criminology marked an important transi-

tion from the traditional focus on ‘‘whodunit’’ toward one that examined

‘‘wheredunit.’’ Just as epidemiology labored to protect public health by

identifying disease hot spots, environmental criminology labored to protect

public safety by identifying crime hot spots (see Sherman, Gartin, and

Buerger 1989). For example, Sherman, Gartin, and Buerger (1989) revealed

the nonrandom occurrence of crime across a city with 50 percent of crimes

occurring within 3 percent of places. Sherman (1995) offers three hypoth-

eses to explain why crime clusters within places: (1) the patron hypothesis

that offenders tend to congregate at particular places, (2) the management

hypothesis that place managers determine who is present and how they

behave, and (3) the behavior settings hypothesis that places have particular

informal rules, configurations of people and objects, and lifestyle roles that

influence behavior. Their research finds the greatest support for the beha-

vior settings hypothesis, as there is a greater clustering of crime within

places than within people (Sherman, Gartin, and Buerger 1989).

According to the theory of behavior settings, places exhibit ‘‘standing

waves of behavior’’ that are shaped by both social customs and physical

design (Barker 1968). In theory, once these standing waves of behavior

coalesce with the lifestyle routines of place visitors, certain behaviors may

pass the threshold of public acceptability. For example, bars as behavior

4 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

settings have extended trading hours and on-site alcohol consumption that

lower behavioral inhibitions and thus shape on-site standing waves of beha-

vior. Simultaneously, the majority of the workforce commence their recrea-

tional weekend each Friday afternoon thus they share a lifestyle routine.

Combining these two factors may explain why Friday evenings are associ-

ated with behavioral excesses that often culminate in violent assaults (Gru-

besic et al. 2013). However, blaming individual-level alcohol consumption

for behavioral extremes is overly reductionist, given that restaurants are

also behavioral settings that share extended trading hours and on-site alco-

hol consumption yet lack the same association with violence. Perhaps the

distinction is the social customs of restaurants curtail excessive alcohol

consumption.

Similarly, routine activity theory includes lifestyle routines but refer-

ences social encounters rather than places as the focus (Cohen and Felson

1979). From this theoretical perspective, when lifestyle routines determine

chance encounters between motivated offenders and suitable targets, offen-

ders will be more likely to act when guardians are absent to minimize risk

(Felson and Boba 2010). Lifestyle routines are often cyclic by revolving

around diurnal cycles, weekly work schedules, and yearly seasonal changes.

As Ratcliffe (2010:15) notes, ‘‘As the relevant actors—victims, offenders,

guardians, and place managers—adjust their relative densities over time

and around specific places, the opportunities for crime shift and coagulate.’’

Greenspace Crime

In criminology, scholarship has seriously engaged with the association

between particular behavior settings and crime. Schools, shopping centers,

and licensed venues and their link to crime feature strongly in this literature

(Brantingham and Brantingham 1993a; Felson and Boba 2010; Nelson,

Bromley, and Thomas 2001; Snyder and Sickmund 2006). More recently,

studies have considered greenspaces as unique settings that may lead to

particular types of crime (Crewe 2001; Groff and McCord 2012; McCord

and Houser 2015). For the most part, research has focused on the link

between the presence of greenspace and individuals’ perceived fear of

victimization (McCord and Houser 2015) rather than the link between

greenspace and actual crime events. From this literature, we know that

greenspace is associated with fear of crime (Jorgensen et al. 2013; Kuo and

Sullivan 2001; Nasar and Fisher 1993; Sreetheran and Van Den Bosch

2014). Yet greenspace can also enhance feelings of safety and lead to higher

levels of place attachment (Arnberger and Eder 2012; Bonaiuto et al. 1999;

Kimpton et al. 5

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Fried 1982; Hur et al. 2010; Lee and Shen 2013). For example, greenspaces

generally contain more vegetation than other public settings, which Nasar

and Fisher (1993) argue generate fear by potentially concealing predatory

offenders that are prospecting for isolated victims as well as impeding

escapes from victimization. In contrast, Kuo and Sullivan’s (2001) findings

suggest that vegetation reduces fear, perceived incivilities, and perceived

aggression. They attribute these reductions to vegetation’s ability to alle-

viate mental fatigue, and thus the ‘‘psychological precursor to violence’’

(Kuo and Sullivan 2001:346).

Jorgensen et al. (2013) argue these contradictory findings are attributable

to the omission of physical and social cues—such as hedges that could

potentially conceal offenders and uninhabited areas that may ultimately

influence research findings. To control for social cues, they provided parti-

cipants with a series of photos capturing different greenspace microsettings

and required participants to rate the level of fear evoked from each photo.

Interestingly, their findings suggest that both sexes are sensitive to similar

environmental cues in the photographs but only female participants are

sensitive to the social cues, for example, the individuals featured in the

greenspace image. While their experiment provides interesting findings,

others argue that the relationship between social cues and fear of crime is

more complex (Hunter and Baumer 1982). Hunter and Baumer found that

unless the person is familiar with someone they encounter or feel connected

to the neighborhood, then ‘‘each additional person represents another poten-

tial offender’’ (1982:127). As such, the experiment may be both heightening

and abstracting fear, given that the social cues are out of neighborhood

context and the people unfamiliar.

These studies are useful in the development of fear of crime reduction

strategies to ‘‘design out fear,’’ yet they do not provide an understanding

of why greenspace crime might be higher in some places and not others.

Few studies of greenspace crime examine greenspaces as crime genera-

tors across a city’s landscape and fewer still consider the features of

greenspaces that might explain why crime concentrates in some green-

spaces and not others. Indeed, we are only aware of the three studies that

examine the greenspace—crime association at the macro level—that is,

across a metropolitan area. Using calls to police in Boston, Crewe (2001)

find that crime reports increase with proximity to the Boston Southwest

Corridor parkland. Groff and McCord (2012) examine greenspace crime

specifically within ‘‘neighborhood parks’’ throughout Philadelphia and

find that violent, property, and disorder crime incidents cluster within

and around this greenspace type, and that particular greenspace amenities

6 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

reduce these associations, for example, sporting facilities and lighting

within the greenspace and public transport stops by the perimeter. The

authors suggest that sporting amenities may formalize place guardianship

by accommodating sporting clubs—for example, club members and coa-

ches—while amenities such as lighting and public transport stops may

extend periods of legitimate visitation thus extending periods of place

guardianship. In a third study, McCord and Houser (2015) examine crime

within the ‘‘park environs’’—the streets adjacent to the greenspace—in

Philadelphia and Louisville. Their results reveal that violent, property,

and disorder crime clusters in the areas surrounding greenspace and that

this clustering occurs for the majority of greenspaces in the two cities.

Further, the presence of greenspace lighting and the provision of park

benches, drinking fountains, and parking lots reduced crime within the

park environs (McCord and Houser 2015). This provides evidence that

particular greenspace amenities may promote guardianship in both green-

spaces and their adjoining streets.

By considering greenspace amenities, the latter two studies lend sup-

port for Jacob’s (1964) influential ‘‘eyes on the street’’ theory. This thesis

suggests (a) that amenity users provide ‘‘natural surveillance’’ within

their field of view and (b) ‘‘multiuse’’ settings can attract visitors across

multiple periods therefore extending the period of natural surveillance

(Jacobs 1964). For example, a greenspace with a playground and a sport-

ing field may attract parents accompanying toddlers in the early afternoon

and sporting teams in the late afternoon, thus providing an extended

period of natural surveillance through multiple visitor types. In contrast,

amenity-poor greenspaces may lack the appeal to attract capable guar-

dians but sufficient tree cover to ‘‘camouflage’’ particular criminal activ-

ities (Felson and Boba 2010:31). Thus, greenspace amenities can be

crime limiting or crime generating depending upon the public behaviors

they promote (Brantingham and Brantingham 1995; Groff and McCord

2012).

Greenspace crime case studies shed further light on this. For example,

Rhodes and colleagues (2007) find that injecting drug users choose places

that afford concealment from interruption, identification, and shaming

from friends, family, and colleagues when injecting. Interestingly, reduc-

ing hedge heights reduces places of concealment and subsequently

reduces drug use within greenspaces (Knutsson 1997). Installing new

amenities that increase the ‘‘awareness space’’ of place guardians also

reduces drug use (Knutsson 1997; also see Brantingham and Brantingham

1993b).

Kimpton et al. 7

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

The Temporal and Neighborhood Contexts of Greenspace Crime

Behavior setting theory and routine activity theory highlight the importance

of time in understanding the dynamics of crime (Felson and Boba 2010;

Haberman and Ratcliffe 2015). By observing the timing of crime, researchers

attempt to capture lifestyle routine activities—both diurnal and weekly—that

account for the temporary movements of individuals between the places

where they may become victimized, protect as guardians, or seize the oppor-

tunity to offend (see Felson and Boba 2010). These routines are important for

generating site-specific social norms (Barker 1968) that constrain particular

behaviors within specific periods of the day or week. An example of this is

the consumption of alcohol to specific periods of the day (Ratcliffe 2006).

Ratcliffe (2001) argues that offenders choose particular times of the day to

offend. Looking specifically at burglary, he finds that offenders will engage

in burglary when they expect to encounter fewer capable guardians. In other

words, potential offenders are aware of lifestyle routines and choose when

they offend accordingly. Thus, burglars generally target homes during busi-

ness hours and workplaces during nonbusiness hours. This is consistent with

routine activity theory that states that offenders are more likely to offend

where there are fewer capable guardians. Brunsdon and Corcoran (2006) also

find that temporary population shifts can leave opportunities for property

damage and public disorder offenses, given that these crimes cluster within

the inner city between the hours 11 p.m. and 4 a.m. This is the same general

period that the inner-city workforce is more likely to be located elsewhere,

and thus absent as place guardians. Similarly, Herrmann’s (2013) study

reveals that when the intended target is a person rather than place, crime will

occur within ‘‘hot streets’’—or points of intersection between offender and

victim and where guardians are momentarily absent. We are unaware of any

studies that examine the temporal patterning of greenspace crime, but it is

plausible that greenspaces have multiple offending patterns, with each depen-

dent upon whether the intended target is person or property.

In addition to the timing of greenspace crime, we argue that social and

physical context surrounding greenspaces will influence crime. When crime

generators are adjacent to other behavior settings, the opportunity for crime

is heightened. For example, Groff and Lockwood (2014) observed higher

levels of property damage and disorder crime around licensed venues.

Similarly, Pridemore and Grubesic (2011) observed higher levels of violent

crime around licensed venues. Parker (1993) argues that licensed venues

combine two contextual conditions that may influence violence. The first

contextual condition is that licensed venues serve alcohol, which can

8 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

disinhibit individuals who usually refrain from using violence to settle

disputes. The second condition is that licensed venues can attract people

who enjoy watching violence, and thus actively goad the intoxicated to

settle their disputes with violence. Collins (2009) argues that such settings

can provide a ‘‘moral holiday’’ for such people from their usual moral

constraints, since they just need to claim they were intoxicated to absolve

their amoral behaviors. Since people leaving licensed venues do not leave

their intoxication at the door, it may explain the higher levels of violence

toward people and property surrounding these licensed venues. Roncek and

Lobosco (1983) also observed higher levels of property damage, violence,

and burglary near high schools. Felson and Boba (2010) argue that high

school attendance spatially concentrates adolescents who have relatively

poor inhibitions, and their school day routine provides them with a routine

window of opportunity to offend while they return from school and their

parents are still within the workplace. Conversely, Groff and McCord

(2012) found that greenspaces adjacent to schools had fewer disorder

crimes. It is therefore plausible that the proximity of licensed venues and

high schools may influence crime in neighboring greenspaces.

In summary, Groff and McCord (2012) argue that while the greenspace

crime literature is theoretically rich, it remains evidence poor. This provides

opportunities to examine whether greenspace types are distinct behavior settings

that influence the types and timings of crime, and whether greenspace crime is,

at least in part, a function of the immediate surrounding social and physical

context. These opportunities lead us to the following four research questions:

Research question 1: Is greenspace type associated with greenspace

crime?

Research question 2: Is greenspace type associated with the timing of

greenspace crime?

Research question 3: Is the neighborhood social composition asso-

ciated with greenspace crime?

Research question 4: Is the presence of neighborhood crime genera-

tors associated with greenspace crime?

Data and Methods

The Research Site

We examine greenspace crime in the Brisbane Statistical Division (BSD). The

BSD is located in the southeast of the state of Queensland, Australia. Five local

councils comprise this region. Collectively, these councils provide and

Kimpton et al. 9

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

maintain 4,265 greenspaces across the region. Greenspace comprises 3 percent

of the BSD and contains 2 percent of the BSD total crime. Further, a small

proportion of these greenspaces account for all observed greenspace crime:

violent crime concentrates in 9 percent of all greenspace, theft within 17

percent, drug within 6 percent, public nuisance within 7 percent, and property

damage within 15 percent of greenspaces (Figure 1). The risky facilities theory

points toward the complex and dynamic interactions among offenders, vic-

tims, and place guardians to explain this unequal distribution of crime between

seemingly homogenous settings (Eck, Clarke, and Guerette 2007). We suggest

that the variability of greenspace amenities and spatial context surrounding

greenspaces can explain this unequal distribution of greenspace crime.

Data Sources, Spatial Integration, and Preparation

We combine five data sets to capture greenspace amenities, crime, and

neighborhood context. The combined asset registers from all five local

Figure 1. Percentage of total greenspace crime � crime type and greenspace.

10 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

councils provide the park amenities for all BSD greenspace. We procure all

information on public transport stops (including bus, train, and ferry stops)

located within or immediately adjacent to greenspaces from the Translink

(2015) website. Combined, these data provide a complete account of green-

space amenities associated greenspace crime, which we employ for classi-

fying greenspace into general types.

Our crime data come from the Queensland Police Service (QPS 2015).

The QPS administer an application programming interface for the public to

extract crime data. Using this interface, we extracted the timing and loca-

tion of all available crime incidents from January 1, 2007 until December

31, 2011 across the BSD. These data provided information on crime types

and the timing of crime for all incidents occurring within and adjacent to

greenspaces.

The joined Digital Cadaster DataBase (Department of Natural Resources

and Mines 2012) and the Queensland Valuations and Sales data set (QVAS;

Australian Business Research 2011) provided information at the level of the

land parcel. Collectively, these data sources provide the spatial boundaries

and their associated land use for each land parcel in the BSD. We use these

data to identify greenspace. As noted previously, greenspace operationali-

zations generally comprise public parks, gardens, greened walkways, nature

reserves, and sporting fields and ovals. Given that the QVAS database

describes 100 different land use types, we restrict our selection to land use

types characterized by ‘‘natural ecology’’ and ‘‘public access’’ (see Bur-

gess, Harrison, and Limb 1988; Comber, Brunsdon, and Green 2008;

Coolen and Meesters 2012; Dinnie, Brown, and Morris 2013; Feyisa

et al. 2014; Lachowycz and Jones 2013). Within QVAS, these particular

land use are labeled ‘‘sports clubs/facilities;’’ ‘‘sports ground, racecourse,

and airfield;’’ and ‘‘parks and gardens.’’ Further, we exclude all national

and state nature reserves from our selection, given that (1) nature reserve

placement is generally unintentional, with the urban form built to accom-

modate and preserve them; (2) local councils rarely administer nature

reserves, thus they are less likely to be spatially distributed according min-

imum policy standards; and (3) visitors generally seek nature reserves to

encounter novelty rather than the familiarity sought from neighborhood

greenspace. For these reasons, we consider national and state parks con-

ceptually distinct from urban greenspaces (Kemperman, Borgers, and Tim-

mermans 2002; Maat and De Vries 2006).

As we argue that crime generators near greenspaces, in particular,

licensed venues and schools, may influence crime within greenspaces, we

use an online registry of Queensland liquor licenses to locate BSD venues

Kimpton et al. 11

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

that dispense alcohol (Office of Liquor and Gaming Regulation 2015) and

the 2012 StreetPro data set to locate all BSD high schools (Pitney Bowes

Insight 2015). Multiple studies suggest that alcohol dispensing venues

lower adult inhibitions (Knutsson 1997; Stelzig 2012) and schools spatially

concentrate unsupervised and relatively uninhibited adolescents (Felson

and Boba 2010; Roncek and Lobosco 1983), which can increase crime with

nearby places that include greenspace.

Last, to capture the social composition of neighborhoods surrounding the

greenspaces, we employ 2011 census data collected and distributed by the

Australian Bureau of Statistics (ABS 2011). We operationalize the sur-

rounding neighborhood context as all statistical area level 1 (SA1) units

that are contiguous to each greenspace. In Australia, the SA1 is the smallest

available spatial unit provided by the ABS for demographic composition.

Each SA1 unit has an average of 162 households and 402 residents, thus it is

roughly the equivalent of a U.S. census block group or a U.K. output area.

In our analyses, we spatially overlay the resultant 4,265 greenspace land

parcels, 49,180 greenspace amenities, 4,093 SA1 units, and 1,219,377

crime incidents to examine the association between greenspace types and

crime. We also calculate the Moran’s I of SA1 crime counts to test for the

presence of spatial autocorrelation and find that violent (I ¼ .3648, p < .001), theft (I ¼ .2198, p < .001), drug (I ¼ .4138, p < .001), public nuisance (I ¼ .2866, p < .001), and property damage (I ¼ .3739, p < .001) crime is positively spatial autocorrelated (across the study area), which suggests that

offending is geographically concentrated in particular locales of the BSD.

Greenspace Types

Drawing on the literature (Gehl 2010; Groff and McCord 2012; Henriksen

and Tjora 2014), we classify the 25,952 greenspace assets registered across

the 4,265 greenspaces in our study site into one of the nine theoretically

relevant greenspace amenity types (see Appendix Table A1). We then

associate these greenspace amenities types with nearby public transport

stops, greenspace size, and greenspace roundness with a cluster analysis

that employs Gower’s dissimilarity matrix (1971) and Ward’s linkage

method (1963) to simultaneously compare the binary and continuous green-

space characteristics. Further, we compare the Caliński and Harabasz’

pseudo-F (1974) and Duda and Hart’s (1973) pseudo-T index scores of all

cluster sets generated with fewer than 10 clusters (see Appendix Table A2),

and examine the amenities that characterize the clusters (see Table 1) before

concluding that the four-cluster set provides a qualitatively robust

12 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

greenspace typology within the BSD context. We also label the clusters

‘‘amenity rich,’’ ‘‘sit or play,’’ ‘‘transport,’’ and ‘‘amenity poor,’’ to reflect

these qualitative distinctions and employ discriminant analysis to evaluate

the reliability of our cluster analysis regarding misclassification (p < .05).

Units of Analysis

We employ two units of analysis throughout this study. Our w2 and circular t-tests employ greenspace type as the unit of analysis for examining (1) the

association between greenspace type and crime type, (2) the daily, and (3)

weekly timing of greenspace crime. Since daily and weekly timing has a

circular distribution, we recode crime timing as angular units where 0�

(functionally equivalent to 360�) is midnight in a daily distribution and midnight on Sunday in a weekly distribution. Our negative binomial regres-

sion analyses employ greenspace as the unit of analysis for examining the

social and physical context of greenspace crime.

Dependent Variables

While the QPS crime incident data distinguish multiple crime types, we

examine specifically violent, theft, drug, public nuisance, and property

damage since each of these crime types has a theoretical association with

Table 1. Greenspace Types and Amenity Presence.

Cluster Number 1 2 3 4 TotalCluster Name Amenity Rich Sit or Play Transport Amenity Poor

Barbecue and tables (%) 77 9 0 0 12 Buildings (%) 31 4 0 0 5 Dog enclosure (%) 6 2 0 0 1 Managers (%) 3 0 0 0 0 Formal sports (%) 51 9 0 0 9 Informal sports (%) 49 32 0 0 13 Lights (%) 58 27 0 0 13 Playground (%) 87 49 0 0 21 Public transport (%) 26 18 100 0 13 Seating (%) 99 66 0 0 26 Roundness (mean) 0.50 0.49 0.45 0.46 0.47 Hectares (mean) 5.46 2.66 4.00 3.26 3.49 Total 578 789 257 2,641 4,265

Kimpton et al. 13

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

greenspaces. Our violent crime includes the QPS homicide, assault, and

robbery crime incidents; theft includes both theft and the handling of stolen

goods; public nuisance includes good order offenses and illegal alcohol

consumption; drug crimes include consumption and dealing; and property

damage includes unlawful entry and arson.

Explanatory Variables

When examining the social and physical context of greenspace crime, we

employ several explanatory variables coded as follows:

Greenspace type is coded as a categorical variable where the ‘‘amenity-

poor’’ greenspace type serves as the reference category. We expect

this category to stand apart from other greenspace types by lacking the

amenities necessary for attracting place guardians (Groff and McCord

2012).

Crime timing is coded as the proportion of greenspace crime occurring

between 5p.m. and 5 a.m. to examine the nighttime association (see

Felson and Poulsen 2003). To observe the weekend association with

greenspace crime, we use the proportion of greenspace crime occur-

ring between 6 p.m. on Friday and 6 a.m. on Monday (see Uittenbo-

gaard and Ceccato 2012). We employ these variables in our negative

binomial regression analyses, which we describe below.

Neighborhood crime rate is the per capita rate of crime incidents within

the surrounding neighborhoods of a greenspace (e.g., the SA1s con-

tiguous to the greenspace, which represents our neighborhood context

as previously discussed). Since we detected the presence of positive

spatial autocorrelation at the SA1 within the BSD, there is the need to

account for the geographic concentration of crime in our analytic

approach by observing the neighborhood crime rate.

Neighborhood social composition includes several variables that are asso-

ciated with crime in the broader literature, namely, population density,

the proportion of adolescents, residential instability, ethnic diversity,

and income inequality (see Boggess and Hipp 2010; Friedson and

Sharkey 2015; Peterson and Krivo 2009; Hipp, Tita, and Boggess

2009; Mason et al. 2009; Morenoff, Sampson, and Raudenbush 2001;

Sampson, Raudenbush, and Earls 1997). All these neighborhood mea-

sures are aggregated and coded from the 2011 ABS census data as

follows: (1) population density divides the neighborhood population

by the area of neighborhood (hectares), (2) adolescents is the

14 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

percentage of adolescents (aged 15–20) within the total population, and

(3) residential instability is the percentage of residents who lived at

their current address for less than five years. Further, (4) ranked ethnic

diversity employs the Blau’s diversity index (Blau 1977; also Sampson

1984) using the country of birth as the distinguishing characteristic and

partitions these index scores into deciles to interpret findings. Last, (5)

ranked economic disadvantage employs principle component factor

analysis to calculate an index score from median household income,

adult resident proportion unemployed, household proportion living

under the poverty line (AU$800, that is, half the national median), and

household proportion headed by a single mother, which is again parti-

tioned into deciles for interpreting findings.

Neighborhood crime generators again employs all SA1 contiguous to the

greenspace as the neighborhood context and counts all neighborhood

land parcels identified as either (1) high schools or (2) licensed venues

to observe two types of greenspace crime generators.

Analytic Approach

There are three stages to our analytical approach. First, we employ a w2

analysis to determine whether there is an association between greenspace

type and crime type. Second, we employ circular t-tests to determine if

crime timing varies significantly between greenspace types or crime types

(see Mardia 1972; Wheeler and Watson 1964) and provide circular plots

(Salgado-Ugarte and Pérez-Hernández 2014) to visually observe the daily

and weekly temporal dynamics of greenspace crime (see Brunsdon and

Corcoran 2006). Last, we compute a negative binomial regression model

for each crime type to examine the association between neighborhood

context and greenspace crime and calculate the Moran’s I of using the

residuals and the inverse distance between every greenspace to determine

whether the externalities of our models are spatially autocorrelated.

Results

Our first research questions asked whether there is an association between

greenspace type and crime type, thus, our w2 analysis employs the green- space type as the unit of analysis. Our results suggest that there is an

association (p < .001) with public nuisance crime occurring disproportio-

nately within the sit or play and transport greenspace types and property

Kimpton et al. 15

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

damage crime disproportionately within the amenity-rich and amenity-poor

greenspace types (Table 2).

Our second research question asked whether the timing of greenspace

crime varies by greenspace type. To compare timings of crime, greenspace

type was again the unit of analysis, and we started by employing tests to

detect whether each distribution was significantly distinct from a uniform

distribution, and followed by testing whether these distributions signifi-

cantly differed by greenspace type. All daily and weekly timings of crime

were significantly different from a uniform distribution (Appendix Table

A3). Except for violent crime, the daily timing of crime significantly varied

by greenspace type (p < .001). Likewise, except for property damage crime,

the weekly timing of crime significantly varied by greenspace type (p <

.01). We plotted distributions in separate matrices (Figures 2 and 3, respec-

tively) to visualize when this variability occurred. For example, daily

around 1 p.m., there was a drug crime increase within the transport green-

space type (Figure 2, C3). This is in contrast to a decrease within the sit or

play greenspace at this same time (Figure 2, B3). This finding suggests that

these greenspace types are distinct behavior settings for drug crime. In

contrast, the greatest increase in property damage crime was around mid-

night in all greenspace types, but particularly the amenity-rich and sit or

play greenspace types (Figure 2, A5 and B5). This indicates that these

greenspace types are similar behavior settings for property damage inci-

dents. Similarly, drug crime increases across the weekend within both the

sit or play and transport greenspace types (Figure 3, B3 and C3) but not

within the amenity-rich greenspace type (Figure 3, A3). We argue this

demonstrates that sit or play and transport greenspace types are similar

behavior settings for drug crime on the weekend, yet they are dissimilar

Table 2. Crime Incidence � Crime Type and Greenspace Type.

Amenity Rich Sit or Play Transport Amenity Poor

Crime Type n % n % n % n %

Violent 340 10 129 7 167 8 291 11 Theft 1,376 42 616 35 775 39 1,237 49 Drug 319 10 291 17 307 15 166 7 Public nuisance 336 10 460 26 506 25 294 12 Property damage 880 27 261 15 234 12 560 22 Total 3,251 100 1,757 100 1,989 100 2,548 100

Note. Pearson w2(12) ¼ 683.3457, p < .001.

16 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

behavior settings compared to amenity-rich greenspace during the same

period. These findings provide evidence that greenspace types are distinct

behavior settings that influence the timings and types of offending beha-

viors and are thus likely to influence the social norms that influence the

reporting of offending (Barker 1968; Sherman, Gartin, and Buerger 1989).

Table 3. Negative Binomial Regression of Crime Counts by Crime Type.

Violent Theft Drug Nuisance Damage Crime Type b (SE) b (SE) b (SE) b (SE) b (SE)

Amenity poor type as the reference category

G re

e n sp

ac e Amenity-rich

type a

1.502 *** 0.755*** 1.461*** 1.498*** 1.454*** (.41) (.17) (.25) (.26) (.21)

Sit or play type 0.176 0.190 0.650* 0.801* 0.599** (.24) (.16) (.31) (.32) (.19)

Transport type 1.130*** 0.830*** 0.956* 2.322*** 0.738*** (.31) (.22) (.40) (.57) (.20)

T im

Night 0.028*** 0.039*** 0.035*** 0.030*** 0.029*** (.00) (.00) (.00) (.00) (.00)

Weekend 0.026*** 0.044*** 0.037*** 0.042*** 0.027*** (.00) (.00) (.00) (.00) (.00)

N e ig

h b o rh

o o d

Crime rate per 1,000

�0.001 0.000** 0.000*** �0.000 0.002*** (.00) (.00) (.00) (.00) (.00)

Population density

�0.001 0.006 �0.006 0.009 0.011 (.01) (.01) (.01) (.01) (.01)

Adolescents �0.028 �0.030 �0.010 �0.114*** �0.019 (.02) (.02) (.03) (.02) (.02)

Residential instability

�0.003 0.002 �0.014 �0.003 �0.006 (.01) (.01) (.01) (.01) (.01)

Ethnic diversity �0.026 �0.004 0.017 0.020 �0.105** (.03) (.02) (.03) (.03) (.03)

Economic disadvantage

0.232*** 0.114*** 0.028 0.124*** 0.172*** (.03) (.02) (.03) (.04) (.03)

Schools 0.484*** 0.275*** 0.238 0.614** 0.275** (.12) (.07) (.12) (.20) (.10)

Licensed venues 0.014* �0.004 0.003 0.020 0.011** (.01) (.01) (.01) (.01) (.00)

Constant �4.500*** �3.356*** �3.465*** �4.291*** �3.270*** (.89) (.36) (.69) (.61) (.49)

Observations 4,233 4,233 4,233 4,233 4,233 Moran’s I of

residuals .027*** .020*** .023*** .019*** .021***

*p < .05. **p < .01. ***p < .001.

Kimpton et al. 17

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Using the same analytic approach, we next compared greenspace crime

timing by crime type. Our circular t-test results suggested type of crime

significantly varied throughout day and week within all greenspace types

(p < .001). For example, violent crime generally increased around 3 p.m.

and around 9 p.m. within all greenspace types, while property damage crime

generally increased later in the evening around midnight (Figure 2). This

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

22 2

Midday

20 4

18 6

16 8

14 10

Midnight

A. Amenity Rich B. Sit or Play C. Transport D. Amenity Poor

1. V

io le

2. T

he �

D ru

g 4.

P ub

lic N

ui sa

nc e

5. P

ro pe

rt y

D am

ag e

Figure 2. Greenspace type � crime type matrix containing circular plots across a daily period (each segment represents a two-hour period).

18 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

crime type variation throughout the day and week provides evidence for

routine activities theory, which posits that offending occurs during the

spatial and temporal intersection of offender and target away from guar-

dians (Felson and Boba 2010). For example, aggressors prospecting for

low-risk targets may encounter children without adult supervision traveling

home from school around 3 p.m. or people alone after dark around 9 p.m.,

S S

F M

T T

WA1

S S

F M

T T

WB1

S S

F M

T T

WC1

S S

F M

T T

WD1

S S

F M

T T

WA2

S S

F M

T T

WB2

S S

F M

T T

WC2

S S

F M

T T

WD2

S S

F M

T T

WA3

S S

F M

T T

WB3

S S

F M

T T

WC3

S S

F M

T T

WD3

S S

F M

T T

WA4

S S

F M

T T

WB4

S S

F M

T T

WC4

S S

F M

T T

WD4

S S

F M

T T

WA5

S S

F M

T T

WB5

S S

F M

T T

WC5

S S

F M

T T

WD5

A. Amenity Rich B. Sit or Play C. Transport D. Amenity Poor

1. V

io le

2. T

he �

D ru

g 4.

P ub

lic N

ui sa

nc e

5. P

ro pe

rt y

D am

ag e

Figure 3. Greenspace type � crime type matrix containing circular plots across a weekly period (each segment represents one day in the week).

Kimpton et al. 19

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

while vandals prospecting for low-risk targets may encounter unguarded

greenspace amenities around midnight.

Our third and final research questions asked whether neighborhood

social composition and crime generators are associated with greenspace

crime. In these analyses, greenspace is our unit of analysis. As our previous

analyses demonstrated a relationship between crime timing and greenspace

crime, we control for the proportion of crimes that occur at night and the

proportion of crimes that occur during the weekend in our analyses. With

the amenity-poor greenspace type as the reference category, we found that

the amenity-rich greenspaces type that accommodates the broadest range of

visitor activities was associated with a 349 percent increase in violent crime

(Incidence Rate Ratio [IRR] ¼ 4.493, p < .001; see Table 3 and Appendix Table A4), 113 percent increase in theft (IRR ¼ 2.127, p < .001), 331 percent increase in drug crime (IRR ¼ 4.309, p < .001), 347 percent increase in public nuisance crime (IRR ¼ 4.472, p < .001), and 328 percent increase in property damage crime (IRR ¼ 4.280, p < .001). The sit or play greenspace type that accommodates fewer visitor activities was also asso-

ciated with increased crime but to a lesser degree. For example, this green-

space type was associated with a 92 percent increase in drug crime (IRR ¼ 1.916, p < .05), 123 percent increase in public nuisance crime (IRR ¼ 2.277, p < .05), and 82 percent increase in property damage crime (IRR ¼ 1.820, p < .01). Last, the transport greenspace type that lacks amenities other than a

public transport stop was once more associated with increased crime. For

example, this greenspace type was associated with a 210 percent increase in

violent crime (IRR ¼ 3.096, p < .001), 129 percent increase in theft (IRR ¼ 2.294, p < .001), 160 percent increase in drug crime (IRR ¼ 2.600, p < .05), 920 percent increase in public nuisance crime (IRR ¼ 10.197, p < .001), and 109 percent increase in property damage (IRR ¼ 2.092, p < .001). The timing of crime also influences levels of crime in greenspaces. For example,

nighttime was associated with 3 percent increase in violent crime (IRR ¼ 1.028, p < .001), 4 percent increase in theft (IRR ¼ 1.039, p < .001), 4 percent increase in drug crime (IRR ¼ 1.036, p < .001), 3 percent increase in public nuisance crime (IRR ¼ 1.030, p < .001), and 3 percent increase in property damage crime (IRR ¼ 1.029, p < .001). Likewise, the weekend was associated with 3 percent increase in violent crime (IRR ¼ 1.027, p < .001), 4 percent increase in theft (IRR ¼ 1.045, p < .001), 4 percent increase in drug crime (IRR ¼ 1.038, p < .001), 4 percent increase in public nuisance crime (IRR ¼ 1.043, p < .001), and 3 percent increase in property damage crime (IRR ¼ 1.027, p < .001). These findings also suggest that greenspace types are distinct behavior settings.

20 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Next, we examined the social context of greenspace. Neighborhood theft

(IRR ¼ 1.000, p < .01), drug (IRR ¼ 1.000, p < .001), and property damage crime (IRR ¼ 1.002, p < .001) were each associated with greenspace crime but with small effect sizes. Interestingly, each 1 percent increase in neighbor-

hood adolescents was associated with an 11 percent reduction in greenspace

public nuisance crime (IRR ¼ 0.892, p < .001). Further, a decile increase in ranked diversity was associated with a 10 percent reduction in property theft

(IRR ¼ 0.901, p < .01). In contrast, a decile increase in ranked disadvantage was associated with a 26 percent increase in violence (IRR ¼ 1.262, p < .001), 12 percent increase in theft (IRR ¼ 1.121, p < .001), 13 percent increase in public nuisance (IRR ¼ 1.133, p < .011), and 29 percent increase in property damage (IRR ¼ 1.188, p < .001). Possible explanations of why disadvantaged neighborhoods have more crime in public places include that

they can struggle to attract and maintain the community institutions that often

reduce crime (Peterson, Krivo, and Harris 2000), their residents more fre-

quently belong to offending subcultures (Lockwood 2007), or that residents

often have greater reliance on public facilities which increases their exposure

to victimization (Graif, Gladfelter, and Matthews 2014).

Last, we examined the nearby crime generators. The presence of each

school was associated with a 62 percent increase in violence (IRR ¼ 1.622, p < .001), 32 percent increase in theft (IRR ¼ 1.317, p < .001), 85 percent increase in public nuisance (IRR ¼ 1.847, p < .01), and 32 percent increase in property damage (IRR ¼ 1.316, p < .01). The presence of each licensed venue was associated with a 1 percent increase in violent (IRR ¼ 1.014, p < .05) and property damage crime (IRR ¼ 1.011, p < .01). While multiple studies suggest that schools (Brantingham and Brantingham 1993a; Felson

and Boba 2010; Nelson et al. 2001; Snyder and Sickmund 2006) and

licensed venues (Groff and Lockwood 2014; Grubesic et al. 2013) generate

neighborhood crime, our findings specifically examine how they may be

off-site crime generators for greenspace crime. Our findings indicate that

these specific crime generators spatially concentrate adolescents and intoxi-

cated persons into greenspaces, yet our data do not allow us to determine if

members of these groups are victim or offender (Felson and Boba 2010).

We concluded by calculating the Moran’s I of the model residuals. We

found no evidence of spatial autocorrelation for greenspace violent (I ¼ 0.027, p < .001), theft (I ¼ 0.020, p < .001), drug (I ¼ 0.023, p < .001), public nuisance (I ¼ 0.019, p < .001), or property damage crime (I ¼ 0.021, p < .001). The results suggest that the indicators allied with our analytic

framework adequately accounted for the spatial processes related to the

distribution of greenspace crime.

Kimpton et al. 21

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Discussion and Conclusion

In this article, we examined the association between greenspace type, crime

timing, crime type, and neighborhood context. We find that greenspaces are

heterogeneous settings, and this heterogeneity influences both the type and

the timings of greenspace crime. Further, our results support several broader

theoretical frameworks. For example, Barker’s (1968) behavior settings

theory proposes that amenities and social norms within a setting influence

visitor behavior. At least in the Brisbane context, greenspace types repre-

sent qualitatively distinct behavior settings, which in turn influence whether

and when greenspace crime occurs. In line with routine activities theory

(Felson and Boba 2010), we find that workday and school-day lifestyle

routines also influence greenspace crime. In further support of environmen-

tal criminology, we also find that the neighboring area is particularly con-

sequential for greenspace crime.

Our results contribute to the extant literature in two important ways.

First, we demonstrate that greenspaces are heterogeneous and greenspace

types influence both the timing and the frequency of crime. From our

analyses, greenspace types represent distinct behavior settings that are capa-

ble of influencing offender and guardian behaviors. Moreover, all green-

space crime was strongly linked to the presence of greenspace amenities.

This suggests that amenities attract both offenders and guardians alike, and

as a consequence, greenspaces become contested spaces. Second, few stud-

ies of greenspace crime have considered the influence of surrounding neigh-

borhoods on greenspace crime (see Groff and McCord 2012 or McCord and

Houser 2015, for exceptions). We find that greenspace crime is strongly

associated with the neighborhood setting and the routine activity patterns of

key organizations. Equally importantly, the presence of schools in sur-

rounding areas is also associated with higher crime for four of the five

crime types we examined. In our sample, crime generally concentrates after

3 p.m. when school children are likely to be least supervised by adults with

workday lifestyle routines and again during the night when most potential

guardians with general lifestyle routines are asleep. By demonstrating the

spatial externalities of greenspace crime and the link between greenspace

crime and routine activities, we are able to explain why similar greenspace

types might experience different levels of crime.

Given that greenspace type and location have important implications for

greenspace crime, we highlight three important policy implications. First, the

‘‘park-standards’’ approach that is widely adopted by local councils to ensure

that all residents live within a maximum range of a greenspace and that there is

22 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

a minimum amount of greenspace per resident (Byrne, Sipe, and Searle 2010)

generally excludes specifications concerning the types of greenspace that are

required. As we found that greenspace type influences greenspace crime, we

conclude that there is generally no regulatory mechanism to determine which

urban community receives a relatively safe greenspace type and which

receives a relatively criminogenic greenspace type. Further, we find that the

high-amenity greenspace type is associated with more crime, thus retrofitting

criminogenic greenspaces with further amenities appears to be an ineffective

crime reduction strategy unless the amenities increase guardianship and report-

ing behaviors. Second, we have demonstrated a limitation of deeming partic-

ular greenspace types as ‘‘criminogenic’’ since we discovered that the strength

of these associations varied throughout the day and week. For this reason, we

urge planners to consider the daily and weekly lifestyle roles that their green-

space designs will fulfill since this may determine when they are criminogenic.

Last, greenspace crime appears sensitive to neighborhood crime rates, neigh-

borhood social composition, and other neighborhood crime generators. This

suggests that standardizing greenspace provision across socially distinct

neighborhoods is unlikely to produce uniform crime outcomes. For example,

our findings suggest that residents of disadvantaged neighborhoods are more

likely to encounter violence, thieves, public nuisances, and damaged public

property by local councils increasing greenspace provision. Further, position-

ing greenspace near high schools and licensed venues may create low-risk

spaces for motivated offenders to encounter potential victims away from

guardians as both go about their daily and weekly lifestyle time–space routines

(Brantingham and Brantingham 1993a). A more tailored approach is therefore

needed when designing safe neighborhood greenspaces that is sensitive to both

the physical and social characteristics of the neighborhood.

While this study advances the greenspace crime literature, there are

limitations. We could not directly attribute the greenspace crime reporting

to neighborhood residents, which limits how we interpret neighborhood

social processes of place guardianship. Increasingly, criminologists are

modeling ambient populations (see Andresen 2011; Felson and Boivin

2015), but suitable data were unavailable at the time of our study. Likewise,

pervasive computing studies increasingly capture actual time–space flows

of ambient populations using cellular network data (see Barabási, González,

and Hidalgo 2008; Isaacman et al. 2011), but these data were also unavail-

able within our study frame. If the availability of these data broadens to the

Australian context, and more specifically Brisbane, then it will become

possible to revisit this topic. Alternatively, there may be study frames where

these data are already available, which affords researchers with new

Kimpton et al. 23

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

opportunities to greatly expand understandings of how neighborhood guar-

dianship influences greenspace crime.

The environmental burden of greenspace crime was the focus of our

study, and we found the association to be more complex than is generally

assumed. We found that whether and when crime occurs depends upon (1)

greenspace type, (2) greenspace location, and (3) greenspace social context.

Urban researchers, policymakers, and planners must be mindful of these

three burdening factors when advocating, provisioning, and designing

greenspace. This is necessary to continue to counteract many of the nega-

tive impacts of urbanism without introducing further problems for residents

to contend with. This approach will ultimately assist in actualizing

Howard’s ([1898] 1965) vision of urban settings as places that combine the

best aspects of both town and country lifestyles for every urban resident.

Appendix

Table A1. Key Words Associated with Amenity Types Employed by the Parsing Program.

Amenity Types Key Words within the Amenity Description

Playground ‘‘playground,’’ ‘‘swing,’’ ‘‘rocker (rota roca),’’ ‘‘spinner (supa nova),’’ ‘‘softfall,’’ ‘‘pedal power,’’ ‘‘play,’’ ‘‘jungle gym,’’ ‘‘giant revolving disk type e,’’ ‘‘maze,’’ ‘‘slide,’’ ‘‘seesaw,’’ ‘‘spring rocker,’’ ‘‘digger,’’ and ‘‘monorail’’

Eating ‘‘table,’’ ‘‘barbecue,’’ and ‘‘firewood’’ Seating ‘‘furniture,’’ ‘‘bench,’’ and ‘‘seat’’ Dog off-leash area ‘‘dog’’ Managers ‘‘museum/resource center,’’ ‘‘pcyc,’’ ‘‘library,’’ ‘‘visitor center,’’

‘‘information booth,’’ and ‘‘information centers’’ Formal sports ‘‘shot put,’’ ‘‘hammer throw,’’ ‘‘equestrian,’’ ‘‘horse,’’ ‘‘Aussie

rules,’’ ‘‘afl,’’ ‘‘sporting field,’’ ‘‘stadium,’’ ‘‘goal post,’’ ‘‘goal,’’ ‘‘club,’’ ‘‘sporting clubhouse,’’ ‘‘stand,’’ ‘‘golf,’’ ‘‘baseball,’’ ‘‘cricket,’’ ‘‘hockey,’’ ‘‘rugby,’’ ‘‘soccer,’’ ‘‘basketball,’’ ‘‘basketball_netball,’’ ‘‘handball,’’ ‘‘netball,’’ ‘‘tennis,’’ ‘‘volleyball,’’ ‘‘sporting court,’’ ‘‘basketball/netball,’’ ‘‘boules court,’’ and ‘‘lawn bowls/croquet green’’

Informal sports ‘‘fitness exercise equipment,’’ ‘‘upper body equipment,’’ ‘‘exercise station,’’ ‘‘bike,’’ ‘‘bmx,’’ ‘‘skate,’’ ‘‘fitness exercise equipment,’’ ‘‘upper body equipment,’’ and ‘‘exercise station’’

Enclosed spaces ‘‘shower,’’ ‘‘toilet,’’ and ‘‘change room’’ Lights ‘‘light’’

24 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Table A2. Pseudo-F and Pseudo-T Scores Calculated for the Final Nine Cluster Sets.

Clusters F T

2 7.05 30.27 3 9.07 1.02 4 6.88 11.89 5 7.30 34.04 6 5.93 2.98 7 5.02 15.38 8 5.49 13.94 9 4.90 0.78 10 4.49 4.83

Table A3. Rayleigh, Kuiper, and Rao’s Tests of Uniformity.

Greenspace Types Uniformity Test

Crime Type

Violent Theft Drug Nuisance Damage

Amenity rich Rayleigh .00 .00 .00 .00 .00 Kuiper .00 .00 .00 .00 .00 Rao .00 .00 .00 .00 .00

Sit or play Rayleigh .00 .00 .00 .00 .00 Kuiper .00 .00 .00 .00 .00 Rao .00 .00 .00 .00 .00

Transport Rayleigh .00 .00 .00 .00 .00 Kuiper .00 .00 .00 .00 .00 Rao .00 .00 .00 .00 .00

Amenity poor Rayleigh .00 .00 .00 .00 .00 Kuiper .00 .00 .00 .00 .00 Rao .00 .00 .00 .00 .00

Kimpton et al. 25

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research,

authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research,

authorship, and/or publication of this article: This research was supported by the

Australian Research Council (DP150101293).

Table A4. Negative Binomial Models Version Reported Using Incident Rate Ratios.

Crime Types Violent Theft Drug Nuisance Damage IRR (SE) IRR (SE) IRR (SE) IRR (SE) IRR (SE)

Amenity poor type as tde reference category

G re

e n sp

ac e

Amenity-rich typea 4.493*** 2.127*** 4.309*** 4.472*** 4.280*** (1.83) (0.35) (1.08) (1.16) (0.88)

Sit or play type 1.192 1.209 1.916* 2.227* 1.820** (0.28) (0.20) (0.60) (0.71) (0.35)

Transport type 3.096*** 2.294*** 2.600* 10.197*** 2.092*** (0.97) (0.50) (1.03) (5.79) (0.41)

T im

e Night 1.028*** 1.039*** 1.036*** 1.030*** 1.029*** (0.00) (0.00) (0.00) (0.00) (0.00)

Weekend 1.027*** 1.045*** 1.038*** 1.043*** 1.027*** (0.00) (0.00) (0.00) (0.00) (0.00)

N e ig

h b o rh

o o d

Crime rate per 1,000

0.999 1.000** 1.000*** 1.000 1.002*** (0.00) (0.00) (0.00) (0.00) (0.00)

Population density 0.999 1.006 0.994 1.009 1.011 (0.01) (0.01) (0.01) (0.01) (0.01)

Adolescents 0.973 0.971 0.990 0.892*** 0.982 (0.02) (0.01) (0.03) (0.02) (0.02)

Residential instability

0.997 1.002 0.986 0.997 0.994 (0.01) (0.01) (0.01) (0.01) (0.01)

Ethnic diversity 0.975 0.996 1.017 1.021 0.901** (0.02) (0.02) (0.03) (0.03) (0.03)

Economic disadvantage

1.262*** 1.121*** 1.028 1.133*** 1.188*** (0.04) (0.02) (0.03) (0.04) (0.03)

Schools 1.622*** 1.317*** 1.269 1.847** 1.316** (0.19) (0.09) (0.15) (0.37) (0.13)

Licensed venues 1.014* 0.996 1.003 1.020 1.011** (0.01) (0.01) (0.01) (0.01) (0.00)

Observations 4,233 4,233 4,233 4,233 4,233 Moran’s I of

residuals .027*** .020*** .023*** .019*** .021***

*p < .05. **p < .01. ***p < .001.

26 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

References

ABS (Australian Bureau of Statistics). 2011. Census of Population and Housing.

Canberra, Australia: ABS. Retrieved September 23, 2015 (http://www.abs.gov.

au/websitedbs/censushome.nsf/home/datapacks).

Anderson, Soren T. and Sarah E. West. 2006. ‘‘Open Space, Residential Property

Values, and Spatial Context.’’ Regional Science and Urban Economics 36:

773-89.

Andresen, Martin A. 2011. ‘‘The Ambient Population and Crime Analysis.’’ The

Professional Geographer 63:193-212. doi: 10.1080/00330124.2010.547151.

Arnberger, Arne and Renate Eder. 2012. ‘‘The Influence of Green Space on Com-

munity Attachment of Urban and Suburban Residents.’’ Urban Forestry &

Urban Greening 11:41-49.

Astell-Burt, Thomas, Xiaoqi Feng, Suzanne Mavoa, Hannah M. Badland, and Billie

Giles-Corti. 2014. ‘‘Do Low-income Neighbourhoods Have the Least Green

Space? A Cross-sectional Study of Australia’s Most Populous Cities.’’ BMC

Public Health 14:1-11. doi: 10.1186/1471-2458-14-292.

Australian Business Research. 2011. ‘‘Queensland Valuations and Sales Dataset.’’

Retrieved November 8, 2011 (http://www.abr.com.au/site/live/index.php).

Bairner, Alan and Peter Shirlow. 2003. ‘‘When Leisure Turns to Fear: Fear, Mobi-

lity, and Ethno-sectarianism in Belfast.’’ Leisure Studies 22:203-21.

Barabási, Albert-László, Marta C. González, and César A. Hidalgo. 2008. ‘‘Under-

standing Individual Human Mobility Patterns.’’ Nature 453:779-82. doi: 10.

1038/nature06958.

Barker, Roger G. 1968. Ecological Psychology: Concepts and Methods for Studying

the Environment of Human Behavior. Stanford, CA: Stanford University Press.

Blau, Peter M. 1977. ‘‘A Macrosociological Theory of Social Structure.’’ The Amer-

ican Journal of Sociology 83:26-54. doi: 10.1086/226505.

Boggess, Lyndsay N. and John R. Hipp. 2010. ‘‘Violent Crime, Residential Instability

and Mobility: Does the Relationship Differ in Minority Neighborhoods?’’ Journal

of Quantitative Criminology 26:351-70. doi: 10.1007/s10940-010-9093-7.

Bonaiuto, Marino, Antonia Aiello, Marco Perugini, Mirilia Bonnes, and Anna Paola

Ercolani. 1999. ‘‘Multidimensional Perception of Residential Environment

Quality and Neighbourhood Attachment in the Urban Environment.’’ Journal

of Environmental Psychology 19:331-52.

Bowler, Diana E., Lisette Buyung-Ali, Teri M. Knight, and Andrew S. Pullin. 2010.

‘‘Urban Greening to Cool Towns and Cities: A Systematic Review of the Empiri-

cal Evidence.’’ Landscape and Urban Planning 97:147-55.

Kimpton et al. 27

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://www.abs.gov.au/websitedbs/censushome.nsf/home/datapacks

http://www.abr.com.au/site/live/index.php

http://jrc.sagepub.com/

Brantingham, Patricia L. and Paul J. Brantingham. 1993a. ‘‘Nodes, Paths and Edges:

Considerations on the Complexity of Crime and the Physical Environment.’’

Journal of Environmental Psychology 13:3-28.

Brantingham, Patricia L. and Paul J. Brantingham. 1993b. ‘‘Environment, Routine,

and Situation: Towards a Pattern Theory of Crime.’’ Pp. 259-94 in Routine

Activity and Rational Choice, Vol. 5, edited by R. V. G. Clarke and M. Felson.

New Brunswick, NJ: Transaction.

Brantingham, Patricia L. and Paul J. Brantingham. 1995. ‘‘Criminality of Place.’’

European Journal on Criminal Policy and Research 3(3):5-26.

Brunsdon, Chris and Jon Corcoran. 2006. ‘‘Using Circular Statistics to Analyse

Time Patterns in Crime Incidence.’’ Computers, Environment and Urban Sys-

tems 30:300-19. doi: 10.1016/j.compenvurbsys.2005.11.001.

Burgess, Jacquelin, Carolyn Harrison, and Melanie Limb. 1988. ‘‘People, Parks and

the Urban Green: A Study of Popular Meanings and Values for Open Spaces in

the City.’’ Urban Studies 25:455-73.

Byrne, Jason, Neil Sipe, and Glen Searle. 2010. ‘‘Green around the Gills? The Challenge

of Density for Urban Greenspace Planning in Seq.’’ Australian Planner 47:162-77.

Caliński, Tadeusz and Jerzy Harabasz. 1974. ‘‘A Dendrite Method for Cluster

Analysis.’’ Communications in Statistics 3:1-27.

Ceccato, Vania. 2014. ‘‘The Nature of Rape Places.’’ Journal of Environmental

Psychology 40:97-107. doi: 10.1016/j.jenvp.2014.05.006.

Clarke, Ronald V. 2012. ‘‘Opportunity Makes the Thief. Really? And So What?’’

Crime Science 1:1-9.

Cohen, Lawrence E. and Marcus Felson. 1979. ‘‘Social Change and Crime Rate

Trends: A Routine Activity Approach.’’ American Sociological Review 44:588.

Collins, Randall. 2009. Violence: A Micro-sociological Theory. Oxford, UK: Prin-

ceton University Press.

Comber, Alexis, Chris Brunsdon, and Edmund Green. 2008. ‘‘Using a GIS-based

Network Analysis to Determine Urban Greenspace Accessibility for Different

Ethnic and Religious Groups.’’ Landscape and Urban Planning 86:103-14.

Coolen, Henny and Janine Meesters. 2012. ‘‘Private and Public Green Spaces:

Meaningful but Different Settings.’’ Journal of Housing and the Built Environ-

ment 27:49-67.

Crawford, David, Anna Timperio, Billie Giles-Corti, Kylie Ball, Clare Hume,

Rebecca Roberts, Nick Andrianopoulos, and Jo Salmon. 2008. ‘‘Do Features

of Public Open Spaces Vary According to Neighbourhood Socio-economic

Status?’’ Health & Place 14:889-93.

Crewe, Katherine. 2001. ‘‘Linear Parks and Urban Neighbourhoods: A Study of the

Crime Impact of the Boston South-West Corridor.’’ Journal of Urban Design 6:

245-64. doi: 10.1080/13574800120105779.

28 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Department of Natural Resources and Mines 2012. ‘‘Digital Cadaster Database.’’

Retrieved March 16, 2012 (http://qldspatial.information.qld.gov.au/catalogue/

custom/detail.page?fid¼f4091CAF1-50E6-4BC3-B3D4-229AA318231Ag). Dinkes, Rachel, Jana Kemp, Katrina Baum, and Thomas D. Snyder. 2009. Indicators

of School Crime and Safety: 2008. Washington, DC: U.S. Department of Justice.

Dinnie, Elizabeth, Katrina M. Brown, and Sue Morris. 2013. ‘‘Community, Coop-

eration and Conflict: Negotiating the Social Well-being Benefits of Urban

Greenspace Experiences.’’ Landscape and Urban Planning 112:1-9.

Duda, R. O. and P. E. Hart. 1973. Pattern Classification and Scene Analysis. Vol. 3.

New York: Wiley.

Eck, J. E., R. V. Clarke, and R. T. Guerette. 2007. ‘‘Risky Facilities: Crime

Concentration in Homogeneous Sets of Establishments and Facilities.’’ Crime

Prevention Studies 21:225-64.

Ellickson, Robert C. 1996. ‘‘Controlling Chronic Misconduct in City Spaces: Of

Panhandlers, Skid Rows, and Public-space Zoning.’’ Yale Law Journal 105:

1165-248.

Felson, Marcus and Rachel Boba. 2010. Crime and Everyday Life. Vol. 4. Thousand

Oaks, CA: Sage.

Felson, Marcus and Rémi Boivin. 2015. ‘‘Daily Crime Flows within a City.’’ Crime

Science 4:1-10. doi: 10.1186/s40163-015-0039-0.

Felson, Marcus and Erika Poulsen. 2003. ‘‘Simple Indicators of Crime by Time of

Day.’’ International Journal of Forecasting 19:595-601. doi: 10.1016/S0169-

2070(03)00093 -1.

Feyisa, Gudina Legese, Klaus Dons, and Henrik Meilby. 2014. ‘‘Efficiency of Parks

in Mitigating Urban Heat Island Effect: An Example from Addis Ababa.’’ Land-

scape and Urban Planning 123:87-95. Retrieved August 23, 2016. (http://dx.doi.

org/10.1016/j.landurbplan.2013.12.008).

Fried, Marc. 1982. ‘‘Residential Attachment: Sources of Residential and Commu-

nity Satisfaction.’’ Journal of Social Issues 38:107-19.

Friedson, Michael and Patrick Sharkey. 2015. ‘‘Violence and Neighborhood Dis-

advantage after the Crime Decline.’’ Annals of the American Academy of Polit-

ical and Social Science 660:341-58. doi: 10.1177/0002716215579825.

Gehl, Jan. 2010. Cities for People. Washington, DC: Island Press.

Gower, J. C. 1971. ‘‘A General Coefficient of Similarity and Some of Its

Properties.’’ Biometrics 27:857-71.

Graif, C., A. S. Gladfelter, and S. A. Matthews. 2014. ‘‘Urban Poverty and Neigh-

borhood Effects on Crime: Incorporating Spatial and Network Perspectives.’’

Sociology Compass 8:1140-55.

Groff, Elizabeth and Eric McCord. 2012. ‘‘The Role of Neighborhood Parks as

Crime Generators.’’ Security Journal 25:1-24.

Kimpton et al. 29

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://qldspatial.information.qld.gov.au/catalogue/custom/detail.page?fid={4091CAF1-50E6-4BC3-B3D4-229AA318231A}

http://dx.doi.org/10.1016/j.landurbplan.2013.12.008

http://jrc.sagepub.com/

Groff, Elizabeth R. and Brian Lockwood. 2014. ‘‘Criminogenic Facilities and Crime

across Street Segments in Philadelphia: Uncovering Evidence about the Spatial

Extent of Facility Influence.’’ Journal of Research in Crime and Delinquency 51:

277-314. doi: 10.1177/0022427813512494.

Grubesic, Tony H., William Alex Pridemore, Dominique A. Williams, and Loni

Philip-Tabb. 2013. ‘‘Alcohol Outlet Density and Violence: The Role of Risky

Retailers and Alcohol-related Expenditures.’’ Alcohol and Alcoholism 48:

613-19. doi: 10.1093/alcalc/agt055.

Haberman, Cory P. and Jerry H. Ratcliffe. 2015. ‘‘Testing for Temporally Differ-

entiated Relationships among Potentially Criminogenic Places and Census Block

Street Robbery Counts.’’ Criminology 53:457-83.

Habermas, Jürgen. 1991. The Structural Transformation of the Public Sphere: An

Inquiry into a Category of Bourgeois Society. Cambridge, MA: MIT press.

Hanski, Ilkka, Leena Von Hertzen, Nanna Fyhrquist, Kaisa Koskinen, Kaisa Torppa,

Tiina Laatikainen, Piia Karisola, Petri Auvinen, Lars Paulin, Mika J. Mäkelä,

Erkki Vartiainen, Timo U. Kosunen, Harri Alenius, and Tari Haahtela. 2012.

‘‘Environmental Biodiversity, Human Microbiota, and Allergy Are Interrelated.’’

Proceedings of the National Academy of Sciences of the United States of America

109:8334-39. doi: 10.1073/pnas.1205624109.

Henriksen, Ida Marie and Aksel Tjora. 2014. ‘‘Interaction Pretext: Experiences of

Community in the Urban Neighbourhood.’’ Urban Studies 51:2111-24.

Herrmann, Christopher. 2013. ‘‘Street-level Spatiotemporal Crime Analysis: Exam-

ples from Bronx County, NY (2006–2010).’’ Pp. 73-104 in Crime Modeling and

Mapping Using Geospatial Technologies, Geotechnologies and the Environ-

ment, Vol. 8, edited by M. Leitner. Baton Rouge, LA: Springer.

Hipp, John R., Jonathan Corcoran, Rebecca Wickes, and Tiebei Li. 2014. ‘‘Exam-

ining the Social Porosity of Environmental Features on Neighborhood Sociabil-

ity and Attachment.’’ PloS One 9:1-13. doi: 10.1371/journal.pone.0084544.

Hipp, John R., George E. Tita, and Robert T. Greenbaum. 2009. ‘‘Drive-bys and

Trade-ups: Examining the Directionality of the Crime and Residential Instability

Relationship.’’ Social Forces 87:1778-812. doi: 10.1353/sof.0.0184.

Hirschfield, A., K. J. Bowers, and P. J. B. Brown. 1995. ‘‘Exploring Relations

between Crime and Disadvantage on Merseyside.’’ European Journal on Crim-

inal Policy and Research 3:93-112. doi: 10.1007/bf02242931.

Hope, Tim. 1982. Burglary in Schools: The Prospects for Prevention. London, UK:

Home Office.

Howard, Ebenezer. 1965 [1898]. Garden Cities of To-morrow. Cambridge, MA:

M.I.T. Press.

Hunter, Albert and Terry L. Baumer. 1982. ‘‘Street Traffic, Social Integration, and Fear of

Crime.’’ Sociological Inquiry 52:122-31. doi: 10.1111/j.1475-682x.1982.tb01244.x.

30 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

Hur, Misun, Jack L. Nasar, and Bumseok Chun. 2010. ‘‘Neighborhood Satisfaction,

Physical and Perceived Naturalness and Openness.’’ Journal of Environmental

Psychology 30:52-59.

Isaacman, Sibren, Richard Becker, Ramón Cáceres, Stephen Kobourov, Margaret

Martonosi, James Rowland, and Alexander Varshavsky. 2011. ‘‘Identifying

Important Places in People’s Lives from Cellular Network Data.’’ Pp. 133-51 in

Pervasive Computing, Lecture Notes in Computer Science, Vol. 6696, edited by

K. Lyons, J. Hightower, and E. Huang. Berlin, Heidelberg, Germany: Springer.

Jacobs, Jane. 1964. The Death and Life of Great American Cities. Ringwood, Aus-

tralia: Penguin.

Jorgensen, Lisa J., Gary D. Ellis, and Edward Ruddell. 2013. ‘‘Fear Perceptions in

Public Parks: Interactions of Environmental Concealment, the Presence of Peo-

ple Recreating, and Gender.’’ Environment and Behavior 45:803-20.

Kemperman, Astrid, Aloys Borgers, and Harry Timmermans. 2002. ‘‘Incorporating

Variety Seeking and Seasonality in Stated Preference Modeling of Leisure Trip

Destination Choice: Test of External Validity.’’ Transportation Research Record

1807:67-76.

Kim, Joongsub and Rachel Kaplan. 2004. ‘‘Physical and Psychological Factors in

Sense of Community: New Urbanist Kentlands and Nearby Orchard Village.’’

Environment and Behavior 36:313-40.

Knutsson, Johannes. 1997. ‘‘Restoring Public Order in a City Park.’’ Pp. 133-51 in

Policing for Prevention: Reducing Crime, Public Intoxication and Injury, Vol. 7,

edited by Homel, Ross. New York: Willow Tree Press.

Kong, Fanhua, Haiwei Yin, Philip James, Lucy R. Hutyra, and Hong S. He. 2014.

‘‘Effects of Spatial Pattern of Greenspace on Urban Cooling in a Large Metropolitan

Area of Eastern China.’’ Landscape and Urban Planning 128:35-47. Retrieved

August 23, 2016 (http://dx.doi.org/10.1016/j.landurbplan.2014.04.018).

Kuo, Frances E. and William C. Sullivan. 2001. ‘‘Environment and Crime in the Inner

City: Does Vegetation Reduce Crime?’’ Environment and Behavior 33:343-67.

Lachowycz, Kate and Andy P. Jones. 2013. ‘‘Towards a Better Understanding of the

Relationship between Greenspace and Health: Development of a Theoretical

Framework.’’ Landscape and Urban Planning 118:62-69.

Lee, Tsung Hung and Yen Ling Shen. 2013. ‘‘The Influence of Leisure Involvement

and Place Attachment on Destination Loyalty: Evidence from Recreationists

Walking Their Dogs in Urban Parks.’’ Journal of Environmental Psychology

33:76-85.

Li, Xiaoma, Weiqi Zhou, Zhiyun Ouyang, Weihua Xu, and Hua Zheng. 2012.

‘‘Spatial Pattern of Greenspace Affects Land Surface Temperature: Evidence

from the Heavily Urbanized Beijing Metropolitan Area, China.’’ Landscape

Ecology 27:887-98.

Kimpton et al. 31

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://dx.doi.org/10.1016/j.landurbplan.2014.04.018

http://jrc.sagepub.com/

Lockwood, Daniel. 2007. ‘‘Mapping Crime in Savannah: Social Disadvantage, Land

Use, and Violent Crimes Reported to the Police.’’ Social Science Computer

Review 25:194-209.

Maat, Kees and Paul De Vries. 2006. ‘‘The Influence of the Residential Environ-

ment on Green-Space Travel: Testing the Compensation Hypothesis.’’ Environ-

ment and Planning A 38:2111-27.

Macintyre, Sally, Laura Macdonald, and Anne Ellaway. 2008. ‘‘Do Poorer People

Have Poorer Access to Local Resources and Facilities? The Distribution of Local

Resources by Area Deprivation in Glasgow, Scotland.’’ Social Science & Med-

icine 67:900-14. doi: 10.1016/j.socscimed.2008.05.029.

Mardia, Kanti V. 1972. ‘‘A Multi-sample Uniform Scores Test on a Circle and Its

Parametric Competitor.’’ Journal of the Royal Statistical Society 34:102-13.

Mason, Michael J., Jeremy Mennis, J. Douglas Coatsworth, Thomas Valente, Frank

Lawrence, and Patricia Pate. 2009. ‘‘The Relationship of Place to Substance Use

and Perceptions of Risk and Safety in Urban Adolescents.’’ Journal of Environ-

mental Psychology 29:485-92. doi: 10.1016/j.jenvp.2009.08.004.

Mason, Susan G. 2010. ‘‘Can Community Design Build Trust? A Comparative

Study of Design Factors in Boise, Idaho Neighborhoods.’’ Cities 27:456-65. doi:

10.1016/j.cities.2010.07.003.

McCord, Eric S. and Kimberly A. Houser. 2015. ‘‘Neighborhood Parks, Evidence of

Guardianship, and Crime in Two Diverse Us Cities.’’ Security Journal 28:1-18.

McCunn, Lindsay J. and Robert Gifford. 2014. ‘‘Interrelations between Sense of

Place, Organizational Commitment, and Green Neighborhoods.’’ Cities 41:20-29.

Morenoff, Jeffrey D., Robert J. Sampson, and Stephen W. Raudenbush. 2001.

‘‘Neighborhood Inequality, Collective Efficacy, and the Spatial Dynamics of

Urban Violence.’’ Criminology 39:517-60.

Nasar, J. L. and B. Fisher. 1993. ‘‘Hot-spots of Fear and Crime—A Multimethod

Investigation.’’ Journal of Environmental Psychology 13:187-206.

Nelson, A. L., R. D. F. Bromley, and C. J. Thomas. 2001. ‘‘Identifying Micro-

Spatial and Temporal Patterns of Violent Crime and Disorder in the British City

Centre.’’ Applied Geography 21:249-74. doi: 10.1016/S0143-6228(01)00008-X.

Office of Liquor and Gaming Regulation. 2015. ‘‘Online Licence Search.’’

Retrieved May 28, 2015 (https://www.business.qld.gov.au/industry/liquor-

gaming).

Palmer, Catherine, Anna Ziersch, Kathy Arthurson, and Fran Baum. 2005. ‘‘Danger

Lurks around Every Corner: Fear of Crime and Its Impact on Opportunities for

Social Interaction in Stigmatised Australian Suburbs.’’ Urban Policy and

Research 23:393-411.

Parker, Robert Nash. 1993. ‘‘Alcohol and Theories of Homicide.’’ Advances in

Criminological Theories 4:113-42.

32 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

https://www.business.qld.gov.au/industry/liquor-gaming

http://jrc.sagepub.com/

Peterson, Ruth D. and Lauren J. Krivo. 2009. ‘‘Segregated Spatial Locations, Race-

ethnic Composition, and Neighborhood Violent Crime.’’ The Annals of the

American Academy of Political and Social Science 623:93-107. doi: 10.1177/

0002716208330490.

Peterson, Ruth D., Lauren J. Krivo, and Mark A. Harris. 2000. ‘‘Disadvantage and

Neighborhood Violent Crime: Do Local Institutions Matter?’’ Journal of

Research in Crime and Delinquency 37:31-63. doi: 10.1177/0022427800

037001002.

Pitney Bowes Insights. 2015. ‘‘Streetpro 2012.’’ Retrieved April 1, 2015 (http://

www.pitneybowes.com/us/location-intelligence/location-business-data/street

pro-australia.html).

Pridemore, William Alex and Tony H. Grubesic. 2011. ‘‘Alcohol Outlets and Com-

munity Levels of Interpersonal Violence: Spatial Density, Outlet Type, and

Seriousness of Assault.’’ Journal of Research in Crime and Delinquency 50:

132-59. doi: 10.1177/0022427810397952.

QPS (Queensland Police Service). 2015. ‘‘Crime Locations 2000—Present.’’

Retrieved July 31, 2015 (https://data.police.qld.gov.au/api).

Ratcliffe, Jerry. 2001. ‘‘Policing Urban Burglary.’’ Pp. 1-6 in Trends and Issues in

Crime and Criminal Justice, edited by A. Graycar. Canberra: Australian Institute

of Criminology.

Ratcliffe, Jerry H. 2006. ‘‘A Temporal Constraint Theory to Explain Opportunity-

based Spatial Offending Patterns.’’ Journal of Research in Crime and Delin-

quency 43:261-91. doi: 10.1177/0022427806286566.

Ratcliffe, Jerry H. 2010. ‘‘The Spatial Dependency of Crime Increase Dispersion.’’

Security Journal 23:18-36. doi: 10.1057/sj.2009.16.

Rhodes, Tim, Louise Watts, Sarah Davies, Anthea Martin, Josie Smith, David Clark,

Noel Craine, and Marion Lyons. 2007. ‘‘Risk, Shame and the Public Injector: A

Qualitative Study of Drug Injecting in South Wales.’’ Social Science & Medicine

65:572-85. doi: 10.1016/j.socscimed.2007.03.033.

Roncek, Dennis W. and Antoinette Lobosco. 1983. ‘‘The Effect of High Schools on

Crime in Their Neighborhoods.’’ Social Science Quarterly 64:598-613.

Salgado-Ugarte, Isaı́as H. and Marco A. Pérez-Hernández. 2014. ‘‘Kernel Density

Estimation for Circular Data.’’ Presented at the 2014 Mexican STATA User

Group meeting, Juárez, Mexico. Retrieved August 23, 2016 (http://www.stata.

com/meeting/mexico14/).

Sampson, Robert J. 1984. ‘‘Group Size, Heterogeneity, and Intergroup Con-

flict: A Test of Blau’s Inequality and Heterogeneity.’’ Social Forces 62:

618-39.

Sampson, Robert J. 1997. ‘‘Neighborhoods and Violent Crime: A Multilevel Study

of Collective Efficacy.’’ Science 277(5328):918-24.

Kimpton et al. 33

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://www.pitneybowes.com/us/location-intelligence/location-business-data/streetpro-australia.html

https://data.police.qld.gov.au/api

http://www.stata.com/meeting/mexico14/

http://jrc.sagepub.com/

Sherman, Lawrence W., Patrick R. Gartin, and Michael E. Buerger. 1989. ‘‘Hot

Spots of Predatory Crime: Routine Activities and the Criminology of Place.’’

Criminology 27:27-55. doi: 10.1111/j.1745-9125.1989.tb00862.x.

Sherman, Lawrence W. 1995. ‘‘Hot Spots of Crime and Criminal Careers of Places.’’

Crime and Place 4:35-52.

Snyder, Howard N. and Melissa Sickmund. 2006. Juvenile Offenders and Victims:

2006 National Report. Washington, DC: Department of Justice.

Sreetheran, Maruthaveeran and Cecil C. Konijnendijk Van Den Bosch. 2014. ‘‘A

Socio-ecological Exploration of Fear of Crime in Urban Green Spaces—A Sys-

tematic Review.’’ Urban Forestry & Urban Greening 13:1-18. doi: http://dx.doi.

org/10.1016/j.ufug.2013.11.006.

Stelzig, William Cameron. 2012. ‘‘Examining Crime Symbiosis: The Impact of

Alcohol Outlets on Violence in Nearby Neighborhood Park Environs.’’ Master’s

Thesis, College of Arts and Sciences, University of Louisville, Louisville, KY.

Stodolska, Monika, Juan Carlos Acevedo, and Kimberly J. Shinew. 2009. ‘‘Gangs of

Chicago: Perceptions of Crime and Its Effect on the Recreation Behavior of

Latino Residents in Urban Communities.’’ Leisure Sciences 31:466-82.

Storr, C. L., C.-Y. Chen, and J. C. Anthony. 2004. ‘‘‘‘Unequal Opportunity’’:

Neighbourhood Disadvantage and the Chance to Buy Illegal Drugs.’’ Journal

of Epidemiology & Community Health 58:231-37. doi: 10.1136/jech.2003.

007575.

Timperio, A., K. Ball, J. Salmon, R. Roberts, and D. Crawford. 2007. ‘‘Is Avail-

ability of Public Open Space Equitable across Areas?’’ Health & Place 13:

335-40.

Tita, George, Jacqueline Cohen, and John Engberg. 2005. ‘‘An Ecological Study of

the Location of Gang ‘‘Set Space.’’’’ Social Problems 52:272-99.

Translink. 2015. ‘‘Translink Data.’’ Retrieved April 23, 2015 (http://translink.com.

au/about-translink/open-data).

Uittenbogaard, Adriaan and Vania Ceccato. 2012. ‘‘Space-time Clusters of Crime in

Stockholm, Sweden.’’ Review of European Studies 4:148-56. doi: 10.5539/res.

v4n5p148.

Ward, Joe H. 1963. ‘‘Hierarchical Grouping to Optimize an Objective Function.’’

Journal of the American Statistical Association 58:236-44.

Wheeler, Stanley and Geoffrey S. Watson. 1964. ‘‘A Distribution-free Two-sample

Test on a Circle.’’ Biometrika 51:256-57. doi: 10.2307/2334214.

Yang, Jun, Joe Mcbride, Jinxing Zhou, and Zhenyuan Sun. 2005. ‘‘The Urban Forest

in Beijing and Its Role in Air Pollution Reduction.’’ Urban Forestry & Urban

Greening 3:65-78. doi: 10.1016/j.ufug.2004.09.001.

34 Journal of Research in Crime and Delinquency

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://dx.doi.org/10.1016/j.ufug.2013.11.006

http://translink.com.au/about-translink/open-data

http://jrc.sagepub.com/

Author Biographies

Anthony Kimpton is a human Geographer within the School of Geography, Plan-

ning and Environmental Management at The University of Queensland, Australia.

His research interests are at the interface of Geography, Sociology, Criminology,

and Urban Planning with a particular focus on urban greenspace, social sustainabil-

ity, and walkability.

Jonathan Corcoran is a professor in Human Geography within the School of

Geography, Planning and Environmental Management at The University of Queens-

land, Australia. His research interests lie in the fields of Human Geography, Demo-

graphy, Spatial Science and Regional Science. His publications cover a broad suite

of topics, including human mobility and migration, human capital, and urban fires,

each of which has a focus on quantitative methods. He has worked closely with a

range of government agencies both to inform and evaluate operational and strategic

planning through the development of geographic-based tools. He is co-editor of

Papers in Regional Science.

Rebecca Wickes is Senior Lecturer in Criminology in the School of Social Science

at the University of Queensland in Brisbane, Australia. Dr. Wickes is the lead

investigator of the Australian Community Capacity Study, a multisite, longitudinal

study of place. Her research focuses on demographic changes in urban communities

and their influence on community regulation, crime and disorder. She has published

in journals such as Criminology, Journal of Research in Crime and Delinquency,

Journal of Quantitative Criminology, American Journal of Community Psychology,

The Journal of Urban Affairs, among others.

Kimpton et al. 35

at Auraria Library on September 12, 2016jrc.sagepub.comDownloaded from

http://jrc.sagepub.com/

<< /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /None /Binding /Left /CalGrayProfile (Gray Gamma 2.2) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Warning /CompatibilityLevel 1.3 /CompressObjects /Off /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages false /CreateJDFFile false /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.1000 /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams true /MaxSubsetPct 100 /Optimize true /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo true /PreserveFlatness false /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Remove /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true ] /AntiAliasColorImages false /CropColorImages false /ColorImageMinResolution 266 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Average /ColorImageResolution 175 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.50286 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.40 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.76 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /CropGrayImages false /GrayImageMinResolution 266 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Average /GrayImageResolution 175 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50286 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.40 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.76 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /CropMonoImages false /MonoImageMinResolution 900 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Average /MonoImageResolution 175 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50286 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox false /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (U.S. Web Coated \050SWOP\051 v2) /PDFXOutputConditionIdentifier (CGATS TR 001) /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown /Description << /ENU <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> >> /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ << /AsReaderSpreads false /CropImagesToFrames true /ErrorControl /WarnAndContinue /FlattenerIgnoreSpreadOverrides false /IncludeGuidesGrids false /IncludeNonPrinting false /IncludeSlug false /Namespace [ (Adobe) (InDesign) (4.0) ] /OmitPlacedBitmaps false /OmitPlacedEPS false /OmitPlacedPDF false /SimulateOverprint /Legacy >> << /AllowImageBreaks true /AllowTableBreaks true /ExpandPage false /HonorBaseURL true /HonorRolloverEffect false /IgnoreHTMLPageBreaks false /IncludeHeaderFooter false /MarginOffset [ 0 0 0 0 ] /MetadataAuthor () /MetadataKeywords () /MetadataSubject () /MetadataTitle () /MetricPageSize [ 0 0 ] /MetricUnit /inch /MobileCompatible 0 /Namespace [ (Adobe) (GoLive) (8.0) ] /OpenZoomToHTMLFontSize false /PageOrientation /Portrait /RemoveBackground false /ShrinkContent true /TreatColorsAs /MainMonitorColors /UseEmbeddedProfiles false /UseHTMLTitleAsMetadata true >> << /AddBleedMarks false /AddColorBars false /AddCropMarks false /AddPageInfo false /AddRegMarks false /BleedOffset [ 9 9 9 9 ] /ConvertColors /ConvertToRGB /DestinationProfileName (sRGB IEC61966-2.1) /DestinationProfileSelector /UseName /Downsample16BitImages true /FlattenerPreset << /ClipComplexRegions true /ConvertStrokesToOutlines false /ConvertTextToOutlines false /GradientResolution 300 /LineArtTextResolution 1200 /PresetName ([High Resolution]) /PresetSelector /HighResolution /RasterVectorBalance 1 >> /FormElements true /GenerateStructure false /IncludeBookmarks false /IncludeHyperlinks false /IncludeInteractive false /IncludeLayers false /IncludeProfiles true /MarksOffset 9 /MarksWeight 0.125000 /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /DocumentCMYK /PageMarksFile /RomanDefault /PreserveEditing true /UntaggedCMYKHandling /UseDocumentProfile /UntaggedRGBHandling /UseDocumentProfile /UseDocumentBleed false >> ] /SyntheticBoldness 1.000000 >> setdistillerparams << /HWResolution [288 288] /PageSize [612.000 792.000] >> setpagedevice