The Role Of Place in Outdoor Serious Violence Research Paper

This sample The Role Of Place in Outdoor Serious Violence Research Paper is published for educational and informational purposes only. If you need help writing your assignment, please use our research paper writing service and buy a paper on any topic at affordable price. Also check our tips on how to write a research paper, see the lists of criminal justice research paper topics, and browse research paper examples.

Like many other crime types, outdoor serious violence is not randomly distributed in space. This research paper critically considers why this may be the case from the perspective of environmental criminology. Special consideration is given to routine activity theory, which argues that for a crime to occur three elements need to coincide in space and time: a motivated offender, a suitable target, and the absence of a capable guardian. In the case of outdoor serious violence, both targets and guardians are often mobile. Therefore, to understand the geographic distribution of this type of crime, the movement patterns of offenders, victims, and guardians all need to be considered. But there are different types of outdoor serious violence, each with its own “optimal” place and set of circumstances. For example, predatory violence (e.g., sexualor robbery-related) tends to take place in secluded locations such as quiet streets, parking lots, and parks. In contrast, unplanned or expressive attacks, such as those resulting from alcohol related altercations, are more common in crowded spaces, often near entertainment venues. A third type is gang-related violence, which also has a strong geographical element due to the territorial nature of gang activity. While considering these scenarios from a routine activity perspective, a number of theoretical and methodological issues arise, such as the role and significance of guardianship and the difficulties associated with measuring the ambient population. The article concludes with the implications these issues may have for research and policing.

Introduction

This research paper examines the influence of “place” on the geographical distribution of outdoor serious violence (hereafter, OSV) by considering two separate but interrelated questions. The first considers whether OSV concentrates spatially, either through the formation of clusters (i.e., hot spots) or through common characteristics across disperse crime locations (i.e., the offenses are not clustered geographically but the types of places where they occur are similar in some way). The second question considers the level of specialization of OSV locations; in other words, does OSV occur in places that are different from where other types of crime – or no crime – occur?

This research paper is structured as follows. The first section sets the context by providing a working definition of OSV, a rationale for its study, and an overview of the theoretical framework employed. Three main types of OSV are identified and these are considered separately in the second section, which summarizes the state of the art. The third part of the essay highlights areas of controversy and suggests further research.

Background Description

For the purposes of this research paper, OSV is defined as the unlawful killing, or near killing, of another person which takes place in an open public space (usually a street). As such, it incorporates unlawful homicides and “near misses” where an individual sustains near-lethal injuries (e.g., attempted murder, aggravated assault).

Although researchers often provide summaries of the type of location where homicides occur, it is not common for the type of offense location (such as “outdoor”) to be used as a classification variable per se. Instead, homicides tend to be defined in terms of motive (e.g., sex murder), the offender victim relationship (e.g., stranger homicide), or the victim characteristics (e.g., child homicide). While these variables will no doubt be informative, knowledge about the immediate social and physical setting should also be considered when generating typologies, particularly if the focus is on understanding the crime event itself – as opposed to, for example, an individual’s propensity to engage in criminal activity.

This is precisely the focus of environmental criminology. Within this approach, it is not just the offender’s predisposition to engage in criminal activity that is considered but also the roles of victims and bystanders, as well as a range of other situational variables that may influence the likelihood of a crime occurring. Theories of environmental criminology include routine activity theory (Cohen and Felson 1979), which argues that for a crime to occur three elements need to coincide in space and time: a motivated offender, a suitable target, and the absence of a capable guardian.

Open public spaces are particularly interesting from a routine activity perspective. As the definition implies, they are usually accessible to anyone at any time, which means potential victims and offenders can come together without any access restrictions. Because individuals tend to move through these spaces or remain there only for small time periods, the nature of the environment is in a constant state of flux. Public spaces are also seen as singular in that they belong to everyone and no one at the same time, which results in space users being less inclined to regulate them (e.g., than they would be to regulate their own yard or store; Newman 1972). As Campbell (1986: 120–121) put it,

There is no startingand finishing-time for street life. It is an ongoing party that changes its character constantly as individuals join and leave in the course of the day. The streets are above all public places. Unlike in someone’s home, those present are not beholden to their host, nor are they required to constrain their behaviour out of respect for someone else’s property or territorial rights. The street therefore represents an area that is “up for grabs” in terms of control and territorial rights.

In this research paper, environmental criminology and, more specifically, routine activity theory provide the theoretical framework for discussing why OSV may occur in some places (under certain circumstances) but not in others. The next section summarizes what we already know in this respect.

State Of The Art

When considering whether OSV happens in some places but not in others, one question should be whether OSV clusters geographically. While the identification of geographical clusters for homicide and other forms of serious violence is neither uncommon nor new (e.g., Bullock 1955), only a very small number of authors have reported disaggregated figures based on the type of place where the crime occurred. One such author is Per-Olof Wikstro¨ m, who uses a contextual classification of violence – originally developed by McClintock in 1963 – based on both victim-offender relationship and crime location type.

In a study published in 1994 (cited in Wikstro¨ m 1995), he reported that outdoor homicides were highly concentrated in space and that these hot spots were fairly stable over a fourdecade period (1951–1991). These outdoor homicide hot spots highly overlapped with hot spots of street violence identified in a previous study (Wikstro¨ m 1991). This seems to indicate that either these two offense types are similar in nature or that they are enabled – or facilitated – by the same kind of environment(s).

Both outdoor homicides and outdoor violence were concentrated in the central business district (CBD). From a routine activity perspective, this makes sense. CBDs are places where vast amounts of people congregate, hence making it more likely that a motivated offender and a suitable target will come together in time and space. At the same time, most of these people are strangers to each other, thus creating an environment low in social order, where individuals are less inclined to act as guardians (see Wikstro¨ m 1995).

A Need To Disaggregate

At a more micro level, however, different types of OSV are likely to be associated with different types of places. At least two main types are worth considering: (1) planned or predatory violence and (2) unplanned or expressive attacks. Where an offense is planned (e.g., robbery-related, sexual murders, retaliatory attacks), the offender is likely to have chosen the attack location a priori based on specific criteria, such as the area being isolated and thus lacking a capable guardian. In many cases, predatory offenders would not have planned to go out and commit an offense but rather take advantage of an opportunity as it presents itself. However, these cases as still regarded as “planned” in that thought would have been given in advance as to how to commit the crime and which circumstances and/or locations would be appropriate. Because isolated areas will not only have fewer potential witnesses but also a reduced victim pool, predatory offenders often need to identify their victims at a busier, nearby location (i.e., the encounter site) and follow them to a location where they feel comfortable assaulting the victim (i.e., the attack site). As a result, isolated street segments in close proximity of busy areas may be particularly risky locations (Rossmo 2000).

Unplanned violence such as dispute-related attacks, on the other hand, may occur in busy places where friction is more likely. These offenses are more expressive in nature which means offenders may not be dissuaded by the presence of potential guardians. Outdoor dispute-related violence often occurs near alcohol outlets (e.g., Nelson et al. 2001). This is especially the case if these facilities are in close proximity of each other and have similar closing times, as this results in what has been termed a “potentiation effect” (see Rossmo 1995). In such circumstances, crowds of intoxicated individuals try to navigate the street network and compete for available resources (e.g., taxis, fast food), often leading to disorder, crime, and violence. Additional “crime generators” for outdoor serious violence, albeit to a lesser extent, include retail stores (Nelson et al. 2001) and neighborhood parks (Groff and McCord 2011).

Offender Awareness Spaces

Whatever their motives and levels of planning, offenders “consistently commit crime in neighbourhoods they personally know well or that are very similar in physical, social and economic characteristics to their home neighbourhoods” (Brantingham and Brantingham 1995: 13). In predatory violence, these are areas where they feel comfortable as familiarity provides them with a degree of control (e.g., they know how frequented the street is, what escape routes are available). Predatory offenders are familiar with these areas because they have come to know them while engaging in noncriminal activities or, less commonly, through targeted spatial exploration (i.e., they have familiarized themselves with these areas with the intention of committing crimes there). In unplanned violence, the crime usually occurs while the offender is engaged in everyday routine activities (Rossmo 2000).

Because the behavior of offenders – just like everyone else’s – is influenced by the least effort principle (Zipf 1949), the areas offenders frequent during the course of their routine activities tend to be close to their residence (or another significant anchor point such as work or entertainment). This translates into crime sites being close to the offender’s home, a finding that has been used to explain spatial distributions of crime and to prioritize suspects during crime investigations (i.e., geographic profiling; Rossmo 2000).

When targets are stationary, examining the distances between the crime sites and the offender’s home (referred to as journey-tocrime; JTC, hereafter) seems appropriate. But when targets are mobile, a greater understanding can be achieved by also considering the location of the victim’s home, as this may be different from that of the crime site. Tita and Griffiths (2005) have devised a mobility-based spatial typology of homicide, based on how these three locations (i.e., the crime site, the offender’s home, and the victim’s home) relate to one another. Five separate types of homicide can be identified in this way, as follows:

internal, where all three locations lie within the same census tract;
predatory, where the crime site and the victim’s home are in the same tract while the offender lives elsewhere;
intrusion, where the crime site and the offender’s home are in the same tract while the victim lives elsewhere;
offense mobility, where the offender and the victim live in the same tract but the crime occurs elsewhere; and
total mobility, where each location is located in a separate census tract.

Outdoor homicides are less likely to be “internal” or “predatory” and, more likely, to be “intrusions” or have “total mobility,” which led the authors to conclude that “events occurring outside are predicated primarily on the mobility of the victim” (Tita and Griffiths 2005: 303).

Gang-Related Violence: A Special Case

Although OSV may be loosely divided into planned (predatory) and unplanned (expressive, altercation) attacks, gang-related violence is worth special attention. This is because gangrelated violence is often turf-related in nature, and, as such, it has a strong spatial component. Gangs tend to claim territories within which they engage in drug-selling and other criminal activities or simply hang out. Although the territories claimed can be as big as a whole neighborhood, gang members tend to congregate in smaller areas defined by Tita et al. (2005: 280) as the gang’s “set space”:

….a gang may “claim” an entire neighborhood as its domain or “turf,” but set space is the actual area within the neighborhood where gang members come together as a gang. Thus, just as sets are part of a larger gang, set space is a subset of a larger gang turf or territory.

Set spaces are usually located in open public spaces, leading to most incidents of gang-related violence taking place outdoors. Where rival gangs occupy territories that are in close proximity of each other, this can lead to an increased prevalence of violence. As described by Valdez et al. (2009: 303) in their study of Mexican American street gangs,

Geographic proximity of gang territories emerged as an important characteristic in explaining these gang homicides. The fact that most of the homicides were between members of gangs bordering the same neighborhood or area may provide an understanding of the escalation of violent acts between these groups of youth in these neighborhoods. The spatial proximity of two gangs results in frequent contact and high visibility that create a volatile environment susceptible to aggression and violence. These findings are similar to the findings of others that gang homicide was more often turf related than drug related.

George Tita and colleagues have carried out extensive work on how the geographic distribution of gang-related homicide – and violence in general – can be explained in relation to the location of the “set space” of different gangs in relation to one another (e.g., rivals vs. neutral vs. friendly; see Tita and Radil 2011 for a recent example).

Gang-related attacks may be planned or unplanned. Planned attacks are usually retaliatory (as part of ongoing violence between rival gangs or as a punishment for not showing “respect”) or used as a means of solving disputes arising during the course of illegal activities suchas drug-selling. Unplanned or spontaneous attacks tend to occur in unclaimed territories or within gang turf if infiltrated by a rival gang member. An example of the latter scenario is described by Valdez et al. (2009: 295–6):

Me and RadioMan were walking over by the wall and didn’t notice we went into their neighborhood. It was around Christmas time. At first I thought they were firecrackers, but then I saw they were shooting at us from the alley. I got shot in the stomach. When I turned around, RadioMan got shot too. All I remember was seeing him lying there, and he died in the ambulance.

Routine Activity Theory And Outdoor Serious Violence

Routine activity theory provides a useful framework for explaining the spatial distribution of OSV. The routine activities of both offenders and victims dictate where crime occurs, although some predatory offenders may also purposively familiarize themselves with an area not normally frequented with a view to committing crimes there (i.e., active spatial exploration; see Table 1). Guardianship may play a lesser role in OSV as compared to violent incidents taking place in other locations. This is because, as argued earlier, the public nature of the outdoor environment can make bystanders less willing to intervene, particularly when exposed to expressive violence (Felson 1993).

Outdoor Serious Violence: The Role of Place, Table 1 Variables influencing the activity spaces of motivated offenders, suitable targets, and capable guardians in relation to predatory and expressive outdoor serious violence

Most predatory violent offenders look for a particular type of victim (as opposed to a particular individual; Rossmo 2000); hence, the location of predatory attacks is mainly determined by the offender’s routine activities and the absence of a capable guardian which may be a bystander or some form of formal guardianship (e.g., CCTV camera). Where the victim has been predetermined a priori, the victim’s routine activities will be the most decisive factor.

Expressive attacks occurring outdoors tend to be either alcoholor gang-related. As described earlier, the former usually take place in close proximity of alcohol outlets and are overrepresented in downtown areas. Gang-related OSV, on the other hand, is often reported in gang turf or in unclaimed areas nearby.

Controversies And Open Questions

Despite extensive research on the spatial concentration and specialization of serious violence, consensus is still to be reached in relation to some issues. These issues, as well as the impact of recently emerging techniques, are considered in this section.

Should Homicide Be Considered A Separate Offense Or Is It Just An Outcome?

As Gottfredson and Hirschi (1990: 34) put it, “the difference between homicide and assault may simply be the intervention of a bystander, the accuracy of a gun, the weight of a frying pan, the speed of an ambulance or the availability of a trauma centre.” In a similar vein, Block (1977: 10) argues that “most killings are the outcome of either an aggravated assault or a robbery which somehow progressed beyond the degree of harm intended by the offender” and that this should prompt us to regard lethal and near-lethal attacks as similar in nature. In this way, homicide and near-lethal attacks are seen as a single form of crime albeit with outcomes varying in severity and which can be positioned along a continuum.

A contrasting view is offered by Felson and Messner (1996: 520), who state that “a substantial proportion of homicide offenders really do intend to kill their victims and not merely injure them” and claim this is an indication that homicides and near-lethal violence are different entities. The fact of the matter is that intent to kill is difficult – if not impossible – to measure and that, when an assumption is made as to whether intent to kill is present, this is imperfectly associated with the outcome of the offense (i.e., in manslaughter, there is no intent to kill but a fatal outcome, while attempted murder results in a nonfatal outcome despite the intent being lethal).

Although most authors tend to favor the former position, most homicide research still considers just homicide data, as opposed to, for example, also including attempted murder and aggravated assault. A number of studies have included both in an attempt to identify risk factors associated with lethality, which has resulted in some interesting findings (e.g., Weaver et al. 2004). The question to consider here is if lethal and near-lethal violent acts are essentially of the same nature, should future research aim to include both in their analyses? This approach might be advantageous in that it may provide a fuller picture, while also increasing sample sizes, which can be particularly helpful when studying low incidence crimes such as homicide.

Does Alcohol-Related Homicide Really Cluster Around Alcohol Outlets?

While most research findings point to an association between the spatial distribution of alcoholrelated homicide (and violence in general) and that for alcohol outlets, contradictory results have also been reported. For instance, Block and Block (1995) described how hot spots for “liquorinvolved” homicide did not overlap those for alcohol outlets (i.e., taverns and liquor stores) or even hot spots for crime within such premises.

But Block and Block – like many other authors – included in their analysis all instances of homicide where alcohol was thought to have played a role, regardless of where the incident took place. Including residence-based offenses in the analysis has the potential of obscuring any associations that may exist. The situational context in such cases is markedly different, which means different crime facilitators may be at play: while the disinhibiting effects of alcohol may apply in all cases, the environment within and around taverns is thought to present additional risk factors (e.g., crowds, competition for limited resources) which are tied – geographically – to the taverns themselves.

Michael Townsley and colleagues at Griffiths University in Australia have recently carried out an observational study into violence around taverns. What is interesting about this research is that it specifically examines crowding and waiting in line as precipitating factors (see Wortley 1998). One of the findings is that it is not so much overcrowding that leads to aggression, but how pedestrians’ direction of travel (as influenced by the street network and surrounding land use) increases the likelihood of individuals running into each other (Townsley and Grimshaw 2013).

Studies like this where the microenvironments of violence are examined are essential in helping us understand when a potential offender may be motivated enough to physically attack another individual. As stated earlier, guardianship is thought to have a limited role in such cases, and the roles of victims and offenders are often interchangeable. This means that research may be more informative – from a preventative viewpoint – if it focuses on the circumstances that increase an individual’s propensity to violence and his/her ability to inflict serious harm.

Interpreting Journey-To-Crime Distances

As discussed, offenders tend to commit (violent) crime in familiar areas close to their home. This has been reported to be particularly the case for violent offenses, as compared to property crime. But these figures should be interpreted with caution, as analyses often involve offenses that are committed in the offender’s own home. When these cases are excluded, the mean (or median) distances traveled are likely to be on a par with those for property crime.

To illustrate, the JTC mean distance for the 363 homicides in Pizarro et al. (2007) study was 1.8 miles; if the homicides taking place at the offender’s home are excluded (67 in total), the mean JTC increases to 2.2 miles. Unsurprisingly, the biggest differences are in domestic homicide: when considering all domestic cases (71), the mean JTC is 0.6 miles; if zero JTC distances are excluded, the mean JTC increases to 2.9 miles.

Of course it is important to determine which proportion of offenses takes place within the offender’s home, but these should be examined separately when providing summary statistics.

Assessing Victimization Risk

Dangerous places are not necessarily those where many incidents of outdoor serious violence occur. This is because dangerousness relates to the odds of being victimized and, as such, it is mediated by a denominator that represents the population at risk (i.e., the number of potential victims). The number of residents is the most commonly used denominator when calculating violent crime rates. This may be appropriate for some crime types, such as domestic violence, which tend to occur in the home. For crimes that occur outdoors, however, a better denominator would be the number of people on the street, often referred to as the ambient population. This is particularly important when calculating rates for central business districts, which tend to have busy streets but small resident populations.

While resident population statistics can usually be obtained from national census polls, ambient population figures are not readily available, meaning researchers are forced to generate these data themselves. This is usually done through observation, where a fieldworker stations him/ herself at a “gate” (i.e., a specific point within the street network) and counts the number of pedestrians (or vehicles) crossing such a gate over a particular period of time. Data may be collected at the same gate at various times of the day, week, or year, to determine whether the ambient population varies accordingly.

For instance, Chainey and Desyllas (2006) used a pedestrian model to calculate victimization rates of street robbery and “snatch” theft in central London, England. The model was calibrated with relevant key variables (i.e., land use patterns, availability of public transport, sidewalk capacity, and straight-line visibility for easier navigation) and observation data from 231 block segments in 107 different locations. The application of the model to a new area for testing revealed a strong association between estimated and actual pedestrian counts (R2 ¼ 0.74).When the spatial distributions of the two crime types were examined, it emerged that the maps displaying the victimization rates based on estimated pedestrian counts bore little resemblance to those where the resident population was used as a denominator instead. Both types of rates led to distributions that were also different from simple crime density hot spot maps.

But studies of this nature are rare. Collecting data to measure the ambient population can be incredibly resource intensive, especially for larger areas. This has led a number of authors to question whether the changes potentially observed in the new crime rates (i.e., as compared to the rates where the resident population is used as the denominator) are sizeable enough to justify the cost (e.g., Cohen et al. 1985).

In recent years, however, at least two different methods of estimating the ambient population have become available, namely, LandScan and Space Syntax. LandScan is a geographic data set developed by the Oak Ridge National Laboratory which provides ambient population estimates worldwide (see Dobson et al. 2000). These estimates are calculated on the basis of census information, physical geography data, lighting intensity, and other spatial data sets. Despite the level of resolution (approximate one-squarekilometer or 30-arcsecond square cells) still not being fine enough for microlevel analyses, LandScan has already been demonstrated to improve our understanding of the spatial distribution of homicide and other violent crime (e.g., Andresen 2011). At the time of writing, LandScan data at an improved 90m resolution (3-arcsecond) were already available for the US and select OCONUS areas, with plans to extend this further.

A second methodological advance in this area comes from the Space Syntax literature (Hillier and Hanson 1984). Space Syntax provides a means for systematically measuring the configuration of street networks. It does this by representing the street network as an inverted graph where the nodes are the block segments and the links (between them) denote whether the nodes intersect. Using this inverted graph and formulae adapted from graph theory, a number of measures are generated for each segment. Examples of these measures are integration, which indicates how accessible a block segment is within a particular area, and choice, which is the number of times a segment is passed through when generating all shortest paths. These measures have been shown to correlate with pedestrian flows and vehicular flows (Hillier and Iida 2005).

Unlike LandScan, Space Syntax is not normally used to generate pedestrian counts that can then be used as denominators in the calculation of victimization rates (although there is no reason why this should be avoided). Instead, Space Syntax studies tend to examine the relationship between integration (and other graph-derived measures) and crime levels. These tend to indicate that it is more segregated (isolated) segments that are associated with higher crime counts. This applies to residential burglary, street robbery (e.g., Hillier and Sahbaz 2008), and lower level violence (Friedrich et al. 2009). In contrast, Summers and Johnson (2011) found outdoor serious violence – including homicide – to be associated with more integrated (busier) segments.

A problem with Space Syntax is that the proxy measures it generates are atemporal. Values are assigned purely based on the structure of the street network; because this is constant over time, so too are the measures generated. But pedestrian and vehicular flows are known to vary dependent on the time of the day, week, or year. Efforts have been made to address this issue by disaggregating crime data by time of offense (e.g., Hillier and Sahbaz 2008), but more work needs to be done, especially in relation to incorporating land use data into the analyses.

In regard to LandScan, daytime and nighttime ambient population data have been made available for the USA (see LandScan USA). It is expected that these estimates will be further disaggregated in the near future and extended to other geographical areas.

A developing field awash with contradictory findings, combined with a lack of research in relation to OSV, calls for more studies to be performed in this area before a full understanding of the spatial distribution of this crime type can be achieved. Research incorporating various measures of the ambient population, different crime types, and complementary data sets (e.g., land use, census) would be particularly helpful in clarifying inconsistencies.

Situating The Crime Event Within The Neighborhood Context

To achieve a better understanding of the context of the crime event, it is important to establish not just how many people are around but also what these people are like. In recent years, researchers have used multilevel analyses and spatial econometric statistical models to examine the joint effect of microlevel variables (e.g., number of bystanders) and neighborhood characteristics (e.g., deprivation) on the spatial distribution of crime. For instance, Bernasco and Block (2009) employed such an approach with robbery data for the city of Chicago to examine what affected offender target choices. Their results suggested that offenders avoided neighborhoods high in collective efficacy and those where the predominant racial or ethnic group was different from theirs. Collective efficacy is defined as “social cohesion among neighbors combined with their willingness to intervene on behalf of the common good” (Sampson et al. 1997: 918). Proximity measures for a lack of collective efficacy include ethnic heterogeneity and high levels of residential mobility.

Most studies of this kind have focused on residential burglary and other property crime. However, recent research by Summers and Johnson using homicide and attempted murder data for London, England, seems to indicate that neighborhoods of similar racial profiles are also favored by offenders engaging in OSV (Summers and Johnson 2011). Another factor found to be associated with higher crime rates was proximity to the offender’s home area. Though promising, more research needs to be carried out in this area to advance understanding of how the interaction between offender characteristics and micro-and meso-level situational variables influence the spatial distribution of OSV.

Concluding Thoughts

The aim of this research paper was to provide an overview of what we already know about OSV and to highlight the areas within this field where controversy and gaps in knowledge remain. As previously stated, our understanding of this phenomenon is severely restricted by the fact that research findings are seldom disaggregated by location type. Outdoor crime events are qualitatively different in two main ways. First, because individuals tend to move through these spaces or only remain there for limited time periods, the environment is highly fluid. The second difference between outdoor and indoor violence is that guardianship is severely impaired in the former, as its public nature makes bystanders less willing to intervene.

Still, the presence of bystanders is important for some types of OSV, namely predatory attacks. Offenders committing such offenses are likely to be deterred by potential guardians. As a result, the placement of these offenses will be different to that of other forms of OSV, as predatory attacks will tend to take place in more isolated areas (or at times of the day where it is more isolated, e.g., late at night).

An understanding of where and under what circumstances OSV might occur is important from a preventative perspective. Police in many cities are aware of the issues leading to violence in downtown areas and have in some cases implemented interventions that are informed by such knowledge, including taxi marshals and staggered closing times for taverns and nightclubs. More recent research (e.g., Townsley and Grimshaw 2013) aims to improve understanding and lead to further preventative responses, such as modifying the street layout and land use to minimize pedestrians running into each other. Crime prevention efforts by police – and other agencies – in outdoor settings are likely to be markedly different from what may be required in other environments, which again calls for researchers to treat such offenses as a separate entity.

To sum up, there is much we already know in relation to homicide and other forms of serious violence, but it is important to separate events taking place in different settings. This will aid our understanding of the drivers within such localized environments, which will in turn help us to develop more fitting preventative strategies.

Bibliography:

Andresen MA (2011) The ambient population and crime analysis. Prof Geogr 63(2):193–212
Bernasco W, Block R (2009) Where offenders choose to attack: a discrete choice model of robberies in Chicago. Criminology 47:93–130
Block R (1977) Violent crime. Lexington Books, Lexington
Block RL, Block CR (1995) Space, place and crime: hot spot areas and hot places of liquor-related crime. In: Eck JE, Weisburd D (eds) Crime and place, vol 5, Crime prevention studies. Criminal Justice Press, Monsey, pp 145–183
Brantingham PL, Brantingham PJ (1995) Criminality of place: crime generators and crime attractors. Eur J Criml Policy Res 3(3):1–26
Bullock HA (1955) Urban homicide in theory and fact. J Crim Law Criminol Police Sci 45(5):565–575
Campbell A (1986) The streets and violence. In: Campbell A, Gibbs JJ (eds) Violent transactions. Blackwell, Oxford, pp 115–132
Chainey S, Desyllas J (2006) Modelling pedestrian movement to measure on-street crime risk. In: Liu L, Eck J (eds) Artificial crime analysis systems: using computer simulations and geographic information systems. Idea Group Inc., Hershey
Cohen LE, Felson M (1979) Social change and crime rate trends: a routine activity approach. Am Sociol Rev 44(4):588–608
Cohen LE, Kaufman RL, Gottfredson MR (1985) Riskbased crime statistics: a forecasting comparison for comparison for burglary and auto theft. J Crim Justice
13(5):445–457
Dobson JE, Bright EA, Coleman PR, Durfee RC, Worley BA (2000) LandScan: a global population database for estimating populations at risk. Photogramm Eng Remote Sens 66(7):849–857
Felson R (1993) Predatory and dispute-related violence: a social interactionist approach. In: Clarke RV, Felson M (eds) Routine activity and rational choice, vol 5, Advances in criminological theory. Transaction Press, New Brunswick, pp 103–126
Felson RB, Messner SF (1996) To kill or not to kill? Lethal outcomes in injurious attacks. Criminology 34(4):519–545
Friedrich E, Hillier B, Chiaradia A (2009) Anti-social behavior and urban configuration: using space syntax to understand spatial patterns of socio-environmental disorder. In: Koch D, Marcus L, Steen J (eds) Proceedings of the 7th international space syntax symposium. KTH, Stockholm
Gottfredson MR, Hirschi T (1990) A general theory of crime. Stanford University Press, Stanford
Groff E, McCord ES (2011) The role of neighborhood parks as crime generators. Secur J 25(1):1–24
Hillier B, Hanson J (1984) The social logic of space. Cambridge University Press, Cambridge
Hillier B, Iida S (2005) Network and psychological effects in urban movement. In: Cohn AG, Mark DM (eds) Proceedings of spatial information theory: international conference, COSIT 2005, Ellicottsville, 14–18 Sept 2005. Springer, Berlin
Hillier B, Sahbaz O (2008) An evidence based approach to crime and urban design. In: Cooper R, Boyko C, Evans G, Adams M (eds) Urban sustainability for the 24 hour city. Routledge, London
McClintock FH (1963) Crimes of violence. Macmillan, London
Nelson AL, Bromley RDF, Thomas CJ (2001) Identifying micro-spatial and temporal patterns of violent crime and disorder in the British city centre. Appl Geogr 21:249–274
Newman O (1972) Defensible space. Macmillan, New York
Pizarro JM, Corsaro N, Yu SV (2007) Journey to crime and victimization: An application of routine activities theory and environmental criminology to homicide. Vict Offenders 2:375–394
Rossmo DK (1995) Strategic crime patterning: problem-oriented policing and displacement. In: Block CR, Dabdoub M, Fregly S (eds) Crime analysis through computer mapping. Police Executive Research Forum, Washington, DC, pp 1–14
Rossmo DK (2000) Geographic profiling. CRC Press, Boca Raton
Sampson RJ, Raudenbush SW, Earls F (1997) Neighborhoods and violent crime: a multilevel study of collective efficacy. Science 277(15):918–924
Summers L, Johnson SD (2011) The influence of the street network on the spatial distribution of outdoor serious violence. Paper presented at the 20th international symposium of environmental criminology and crime analysis (ECCA), Durban, 18–21 July 2011
Tita GE, Griffiths E (2005) Traveling to violence: the case for a mobility-based spatial typology of homicide. J Res Crime and Delinq 42:275–308
Tita GE, Radil SM (2011) Spatializing the social networks of gangs to explore patterns of violence. J Quant Criminol 27(4):521–545
Tita GE, Cohen J, Engberg J (2005) An ecological study of the location of gang “set space”. Soc Probl 52(2):272–299
Townsley M, Grimshaw R (2013) The consequences of queueing: Crowding, situational features and aggression in entertainment precincts. Crime Prev and Community Safety 15(1):23–47
Valdez A, Cepedea A, Kaplan C (2009) Homicidal events among Mexican American street gangs: a situational analysis. Homicide Stud 13(3):288–306
Weaver GS, Wittekind JEC, Huff-Corzine L, Corzine J, Petee TA, Jarvis JP (2004) Violent encounters: a criminal event analysis of lethal and nonlethal outcomes. J Contemp Crim Justice 20:348–368
Wikstrom P-OH (1991) Urban crime, criminals and victims: the Swedish experience in an Anglo-American comparative perspective. Springer, New York
Wikstrom P-OH (1995) Preventing city-center street crimes. Crime Justice 19:429–468
Wortley R (1998) A two-stage model of situational crime prevention. Stud Crime Crime Prev 7:173–188
Zipf GK (1949) Human behavior and the principle of least effort. Addison-Wesley Press, Oxford