This sample Prediction and Crime Clusters 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.
Predicting crime is a necessary condition for its prevention, and crime is most predictable along those dimensions in which it is concentrated. The most established forms of crime’s tendency to concentrate or cluster are repeat offending, repeat victimization, and geographical hot spots, with complementary concepts including supertargets, hot products, hot places, hot targets, risky facilities, risky routes, and crime sprees and spates. This research paper charts the relationship between such clusters, observing how a broad conception of “near repeats,” incorporating crimes with similar situations and characteristics, is a useful unifying concept. Metrics of nearness, or conversely the difference between crime events, may inform efforts to understand and disperse crime clusters. The theories of repeat victimization and hot spots are shown to be overlapping and compatible and suggest a unified theory of clusters should result from greater conceptual integration in this field. The overall aim of such integration should be to inform cluster-busting crime prevention efforts.
Definitions And Concepts
In what follows the term “crime clusters” is used to refer to crime’s tendency to concentrate in time, space, and other dimensions along which it is measured. A range of related concepts and terms has emerged. Repeat victimization can be thought of as crime against the same target, however defined. That can be the same person, household, business, vehicle, place, or other target. Crime follows crime such that a small proportion of targets and places experience a vastly disproportionate amount. One study found that 1 % of people experience 59 % of personal crime including violence and that 2 % of households experience 41 % of property crime (Pease 1998). This is largely responsible for spatial clustering wherein a small percentage of hot spot locations accounted for a significant proportion of all calls for police service (Sherman et al. 1989; Andresen and Malleson 2011). The most chronically victimized targets are termed supertargets, with many studies finding two or so percent of potential targets and locations accounting for around half or more of crime. Hot spots may or may not be synonymous with high-crime areas, depending on geographical scales, which are also disproportionately composed of repeats.
A common feature of retail crime and personal theft is the hot product. These are typically high-value portable electronic goods including – at the time of writing – smartphones, laptops, and SatNav-GPS systems, or low-value items easy to dispose of to unscrupulous retail outlets, such as razor blades or cosmetics. Locations which host crimes of the same type – for example, if pharmacies suffer more robberies than other stores – have been termed hot targets (Velasco and Boba 2000), and those hosting multiple crimes of the same and different types have been explored as hot places (Block and Block 1994), crime generators and crime attractors (Brantingham and Brantingham 1995), and risky facilities (Eck et al. 2007). This can include bars, retail establishments, and land use of different types. Banks are risky facilities with, crucially for the present discussion, some chains and some branches experiencing many repeat robberies, and a few schools are victimized far more than others (Lindstrom 1997).
Crime is common for passengers on some bus routes (Tompson et al. 2009). The term “hot routes” is used in the urban planning literature to refer to roads with high traffic flow (Li et al. 2007), so perhaps risky routes distinguishes those with disproportionate crime. The risky professions tend to be those that provide care or services to the public such as nurses, customer services staff, and emergency services staff. Some taxi drivers are unusually likely to be robbed (Smith 2005), while firefighters, police, and paramedics often find themselves in the firing line. Within those professions, crime is concentrated on a small percentage of oft victimized staff who experience a lot of the crime. One study of police officers found that around half were assaulted but 3.5 % experienced a quarter of all assaults – likely a conservative estimate. But such statements can be frustratingly unnuanced, in that one needs to know, within the general observation, the times, circumstances, and personalities around which events cluster.
Evolving from work on repeat victimization, the term near repeats (Morgan 2001) is used to refer to similar crimes occurring a short time and distance from an earlier crime. The phrase near repeats has mostly been used as synonymous with spatiotemporal proximity. Neighboring and nearby households are more likely to be victimized after a break-in (Townsley et al. 2003; Johnson et al. 2007a, b). Ratcliffe and Rengert (2008) found repeat shootings in Philadelphia were more likely within a city block and two weeks. To return to an earlier issue, repeats and near repeats suggest why crime clusters spatially. Levy and Tartaro note:
In 1996, repeat locations became known as hot dots (Pease & Laycock, 1996; Townsley, Homel, & Chaseling, 2000). “Hot dots” are locations within a hot spot that are known to have a high incidence of victimization – a repeat victimization location. (Levy and Tartaro 2010, p. 300)
Studying car theft, Levy and Tartaro use the concept of repeat victimization locations as a useful way of examining a combination of repeats and near repeats, hot dots and hot spots:
Repeat victimization locations are defined as any street segment on which more than one auto theft occurred during the study period. In Atlantic City the average length of a city block is approximately 150 ft. (Levy and Tartaro 2010, pp. 304–305)
The same offenders are more likely to be those committing repeats, and such returners are more likely to be prolific (Everson 2003). This implies that a crime spree (from the offender’s perspective) or crime spate (from the victim’s perspective), typically defined as more than two similar crimes in a short time period, is usually composed of quick and near repeats. The conceptualization of near repeats is developed further later in this research paper.
Most of the clustering concepts referred to so far relate to crime recurring over relatively short periods of time. Yet crime is also common against certain targets over a longer period. Averdijk (2010) explored the area of victimization over the life course, or “victim careers,” identifying a clear need for further research. Similar terminology had previously been developed in relation to the criminal careers of places (see chapters in Eck and Weisburd 1995). In a similar vein, the onset, frequency, duration, and decline in the criminal careers of stolen products have been explored (Mailley et al. 2008). The trajectory of a product theft career depends on the extent to which it is craved, that is, the extent to which it is concealable, removable, available, valuable, enjoyable, and disposable (Clarke 1999; Wellsmith and Burrell 2005). Heavy valueless items have little if any theft career.
Hence, the study of crime clusters, concentration, or repeats has evolved on a piecemeal basis. One result is the diverse lexicon of sometimes overlapping terms and concepts described so far. Naming specific clusters is useful because it provides the handles with which to grasp how crime against certain targets, times, and places can be predicted and thereby prevented. Yet it is clear that there are many common threads emerging that are naturally leading the field toward increased conceptual and theoretical integration, and the bulk of the remainder of this research paper outlines how this is occurring.
Near Repeats As A Broad Inclusive Concept
The concept of near repeats has been discussed so far as referring primarily to spatial and temporal proximity. However, the concept of nearness can be applied more widely. Spatial proximity is perhaps the easiest form of nearness of which to conceive, but a crime may also be a near repeat in terms of any quantifiable characteristic or combination of quantifiable characteristics. Quantification here includes classification. For example, hot product theft is a form of near repeat along the dimension of product similarity. Drive offs from service stations may reveal a form of near repeat in terms of forecourt layout or oil company concerns. A few professions are disproportionately targeted, and within those a few individuals with common characteristics (such as the particular population or area they serve) may be disproportionately victimized. Linking profession and time of victimization yields, for example, insights about the victimization of emergency room nurses in the late evening. Near repeats may be inferred from similarities between types of recreational establishment or location, staffing levels, and client surveillability. Route characteristics in public transport will reveal forms of near repeat either directly or in interaction with time of day and day of week.
Two further points about the near repeats terminology are warranted. First, near repeat provides a useful and distinct term because it embodies the mechanism by which crime is patterned, namely, their nearness along at least one dimension. Nearness may vary, and there are degrees of nearness because some crimes are more similar than others, an issue returned to later. However, a broad concept of near repeats includes a range of types of crime cluster where similarity between event, participants, and contexts is key. Second, it could be suggested that terms such as “hot” or “risky” are equally good as near repeats. They are catchy terms and this can be a surprisingly important component of academic concepts and theories – think of Buckyballs or of chaos theory, the latter of which was frequently contested (because the name inaccurately represents the field) until its popularity brought credibility. However, unlike the term near repeats, they do not embody the notion of why repetition occurs. Near repeats seems a preferable general term because it embodies the notion of similarity and difference between crimes across any dimension or combination of dimensions. This is critical because it is similarity that defines all forms of crime cluster. Defining how concentration is distinguished from a non-concentration of crime is avoided here, but a simple definition as a rule of thumb would be two standard deviations from the mean.
One way of thinking about the distinctiveness of near repeats is to consider it as a reflection of offender targeting practices, summed across offenders. Target selection will, in the aggregate, be represented in the pattern of near repeats. Locally, departures from the aggregate pattern will reflect the signature of prolific individuals.
Mapping The Relationships
This section describes the matrix of crime clusters shown as Table 1 which builds on Farrell (2005) and Farrell and Pease (2008). Following routine activity theory, the columns relate to targets, locations, or offenders. The rows denote four dimensions along which individual crimes cluster, described further below: spatial unit, time, crime type, and modus operandi. The matrix is a heuristic device that allows different types of crime cluster to be viewed simultaneously in terms of their key characteristics,
Prediction and Crime Clusters, Table 1 Matrix of crime clusters
though of course it is a simplification of reality because in practice many of the cells overlap and some of the possible characterizations of near repeat clusters are excluded.
The first column of Table 1 identifies crime against the same or similar (including perceived similar) targets. Such crime may be in the same or a nearby place, but not all repeat targets are spatially alike: Consider a victim who is robbed, harassed, and assaulted at different places, perhaps on the basis of ethnicity, sexual orientation, ostentatious wealth, or perceived vulnerability. The cell in the second row of the first column identifies the role of time. Many repeats and near repeats occur quickly, particularly when committed by the same offenders. Continuing down the first column, many repeats are of the same crime type. The row for tactic or modus operandi identifies the fact that crime is often repeated using the same method, particular quick repeats of the same crime type by the same offender, as when a burglar reenters a home through the same rear patio door.
In practice, crime is usually common along many dimensions. The limiting case here is crime against the same target in the same place shortly after the previous crime and committed by the same offender using the same tools and tactics. In many instances, however, nearness is greater in some dimensions than others. Crimes concentrated upon the same victims and product types but not necessarily temporally concentrated have been noted already in the context of the criminal careers of places, victims, and products.
Just as the definition of target was broad, the second column of Table 1 defines repeats at the same location, one nearby or with similar characteristics. The term location or place is scalable. It could cover a region, a country, city, neighborhood, building, apartment, street corner or inter-section, or other precise geographical or virtual coordinate. Consequently, the intersection with spatial repeats (the top cell in the second column) reflects the importance of possible variation in spatial unit. A street or neighborhood which hosts a cluster of burglaries, or a shopping mall suffering a cluster of robberies, may be the same location for many crime reduction purposes. Within those locations, however, there are specific targets where crime is repeated – particular houses, retail outlets, street corners, or alleyways. Levy and Tartaro are worth quoting at length in relation to unit of analysis:
Other aspects of the event should be studied to better understand the repeat victimization phenomenon. One such concept is the “unit of analysis” for the initial criminal event. As the unit of analysis changes, the way in which police and researchers view the crime may also change. Using auto theft as an example, if the unit of analysis is the car, and the car is stolen and dismantled, there is no chance of that car being stolen again. However, if the unit of analysis is the location from which the car is stolen, the fate of the first car has little to do with whether another car could be stolen from that lot. In this situation, there is a chance for repeat victimization, another car could be stolen from that lot and the lot would then be victimized twice. Similarly, the location may be repeatedly victimized, but the owner may not be. If the location is a parking lot, many cars can be stolen from the lot, but they may not belong to the same owner. An owner can be yet another type of repeat victim. The owner could have a different car stolen at different points in time, from the same or different locations. Though the same car has not been victimized again, the owner has still been victimized repeatedly. (Levy and Tartaro 2010, p. 301)
It may be necessary to consider other unit types for analysis. The recommended practice for police seeking to prevent repeat victimization has long been to use that unit of analysis which worked best to facilitate prevention, although that may be easier with hindsight.
It is also necessary to consider the composition and nature of exact and near repeats, as these may inform how crime prevention resources are best allocated. Hence, Table 1 includes concepts relating to attractors and generators, hot spots, and risky facilities. Spatially defined repeats may also vary in the time to repetition, in the type of crime and the modus operandi, or some combination of these relating to the criminal career of the place.
Column three of Table 1 refers to repeats by the same offenders. Progressing down the column, these may be at the same or a similar location, may recur quickly, and may be the same type of offence committed by the same tactic. The same offender committing further crimes shortly afterward could result in a spree or spate, usually defined as more than two crimes in quick succession, and which may or may not involve the same crime types and tactics. However, crimes of the same type may be indicative of offence specialization. It should be noted that using offender characteristics to identify commonality is only possible after detection. The more usual causal direction involves taking event characteristics to suggest likely offenders.
The matrix provides an overview of a few of the key dimensions of the various clusters and helps clarify their relationships. Since patterning is a necessary condition of nonrandom prediction, the matrix may well provide information and insight that assists in the development of cluster-busting preventive responses.
Measuring The Relationships
In offender profiling, behavioral linkage analysis (BHA) seeks to determine whether, in the absence of physical evidence, unsolved crimes were committed by the same serial offender. For example, with a burglary where there is no DNA or other evidence, the behavioral characteristics of the crime (e.g., type of property targeted, means of access and egress, type of items stolen, the presence of gratuitous mess) are used to develop a behavioral profile. The characteristics of different crimes can then be compared to determine, within parameters, whether they may plausibly have been committed by the same offender. There are multiple comparison techniques which use metrics of the goodness of a match between case characteristics. Some crimes are more similar than others. Two of the more prominent measures are Jaccard’s similarity coefficient and the taxonomic similarity index. Thus, behavioral linkage analysis has been used to determine whether, in the absence of physical evidence, unsolved crimes were committed by the same offender, based on across-crime similarity coefficients (Melnyk et al. 2010).
BHA is predicated on two assumptions (Canter 2004). The first is behavioral stability – the assumption that offenders are reasonably consistent in their offending so that a series of their crimes has common characteristics. The second is behavioral distinctiveness – the notion that each offender’s crime series embodies particular characteristics that effectively serve as their behavioral signature.
The application of linkage analysis and similarity coefficients to a crime event profiling, rather than offender profiling, sits well with evolving approaches to crime cluster analysis. Crime clusters can be linked by factors other than the offender. They have other similar characteristics and similar underlying causal mechanisms relating to the opportunity structure (e.g., the characteristics of frequently stolen products). Crime event linkage analysis is thus predicated on the supposition that similarity across crimes is a key factor underpinning clusters of crime irrespective of the dimensions across which common features are held. The orientation of such analysis should be to inform crime understanding and consequent prevention efforts. That could involve the detection of serial offenders but seems more likely to involve designing out crime.
Most concepts relating to crime clusters hinge on the similarity between criminal events. This can be gauged via an index of similarity. Consider how crimes differ. One crime is only perfectly identical to itself. Thus, “identical” is the theoretic maximum level of commonality. Two entirely unrelated crimes of, say, different types in different countries and eras are in effect completely dissimilar (other than that they both involve human behavior and breach of a criminal code). This theoretically maximum difference defines the theoretically minimum level of similarity. All other criminal events lie on a spectrum somewhere in-between.
Quick repeat burglaries by the same offender against the same household are very similar crimes, as is repeated domestic violence involving the same partners, household, and situation. They are sufficiently similar to have been termed “exact” repeats (Summers 2010), differing primarily in time of occurrence. Likewise, simultaneous attacks on two computer networks by the same hacker using the same modus operandi differ primarily in the geographical location of the target. Hence, they are also exact repeats insofar as they differ primarily in one dimension (of course, the violence may result in different injuries, and different goods may be stolen in repeat burglaries, but they are assumed similar for present purposes).
One step further away is a spatially near repeat burglary of a neighboring household shortly after a burglary. This differs in two elements – time and target – from the prior burglary, although only by a relatively small amount in each case. Other forms of crime commonality are typically very similar or “near” in relation to other characteristics. Frequently stolen hot products such as some types of smartphones, laptops, and GPS-SatNavs may be stolen by the same offenders using the same modus operandi (such as sneak theft) but differ in terms of time and place. Racist attacks against members of the same ethnic group by the same perpetrator have different specific targets. However such targets are similar in key characteristics whether that is portability and high resale value or perceived ethnicity. For present purposes, the target is the same and they are forms of near repeat victimization. Whereas the near repeat burglary was defined by spatial proximity, the near repeat theft or assault is defined by the nearness of interchangeability of the target. That is, as noted earlier, the concept of “nearness” is not confined to the spatial and temporal variables. Here the broad inclusive definition of near repeats is important because it demonstrates that it is the degree of similarity between crimes that is the key to crime clusters. For present purposes, spatially near repeat burglaries might be considered equally similar as the targeting of hot products which are near repeat thefts. A hot spot has crimes that are similar in spatial terms but perhaps in little else. This means a hot spot warrants closer scrutiny to determine if it is composed of repeat victimization of the same target or victimization of different targets and crime types. The important point, however, is the following: Similarity is the key to prediction and informed crime prevention. The greater the commonality across crimes, the potentially more informed the response and the greater the preventive scope. The predictability of a repeat crime depends on the type of crime and the context. Hence, quick repeats of the same crime type against the same target, by the same offenders using the same tactic, present the greatest potential for crime prevention. Such precise repeats should be easiest to prevent because the maximum amount of information is available. Other forms of crime commonality are derivatives with less similarity or, conversely, a higher index of dissimilarity.
The appropriate terminology is likely to be that of difference rather than similarity. Difference might be preferred when metrics are developed because it allows lower values to refer to more similar crimes – that is, it is more appropriate to refer to a low difference score. Similar indices exist in various fields including the Hamming distance, and the Sorensen Index and various metrics for DNA fingerprint matches. Critical elements of crime event similarity are the target, the location, the time, the crime type, the modus operandi, and the offender. Measures of the extent of difference within each variable require more sophisticated indices. However, there is a likely benefit to simplicity in a crime event difference index as it would be useful primarily to the extent it served as a heuristic device for crime prevention.
Toward A Crime Cluster Theory
Theories of repeat victimization and hot spots emerged around the same time and from a similar epidemiological tradition. As key areas of crime commonality where theory is established, if they prove compatible, then this may provide a platform for further integration of both theory and crime control practice. Hence, while recognizing the need for a wider repertoire of variables to be considered in understanding crime patterns, this section of the paper reverts to the more familiar ones because they are the most established and readily comprehended.
Cohen and Felson, in the landmark study that developed routine activity theory, anticipated more recent theories of crime clustering when they noted
[T]he effects of the convergence in time and space of these elements [suitable targets, likely offenders, and the absence of capable guardianship] may be multiplicative rather than additive. That is, their convergence by a fixed percentage may produce increases in crime rates far greater than that fixed percentage. (Cohen and Felson 1979, p. 604)
There are two main theories of hot spots. These are embodied in the concepts of crime attractors and generators. Generators are places with high flows of people which yield spatial concentrations of crime, even though one crime may be no more related to another than elsewhere. Attractors are places that gain a reputation for crime and thereby attract likely offenders. The same place can be a generator and an attractor, because a generator is likely to attract would-be offenders due to the rich supply of possible targets. Brantingham and Brantingham (1995, pp. 7–8) observe that
Crime generators are particular areas to which large numbers of people are attracted for reasons unrelated to any particular level of criminal motivation they might have or to any particular crime they might end up committing. Typical examples might include shopping precincts; entertainment districts; office concentrations; or sports stadiums. … Crime generators produce crime by creating particular times and places that provide appropriate concentrations of people and other targets (Angel, 1968) in settings that are conducive to particular types of criminal acts. Mixed into the people gathered at generator locations are some potential offenders with sufficient general levels of criminal motivation that although they did not come to the area with the explicit intent of doing a crime, they notice and exploit criminal opportunities as presented (either immediately or on a subsequent occasion).
There is much in common here with Cohen and Felson’s notion of multiplicative interaction effects. Further, the Brantinghams state that
Crime attractors are particular places, areas, neighbourhoods, districts which create wellknown criminal opportunities to which strongly motivated, intending criminal offenders are attracted because of the known opportunities for particular types of crime. Examples might include bar districts; prostitution areas; drug markets; large shopping malls, particularly those near major public transit exchanges; large, insecure parking lots in business or commercial areas. The intending offender goes to rough bars looking for fights or other kinds of ‘action.’
These definitions are consistent with those in the more recent work of Kinney et al. (2008). With respect to repeat victimization, Farrell and Pease (1993, p. 13) offered three theories. The first is that
It is possible that a first victimisation does not increase the probability of repeat victimisation, but merely flags the high prior probability of victimisation, which is attested again by its suffering a repeat burglary.
which has become known as the risk heterogeneity or flag theory. The second is that
The same offenders may return to take things they had forgotten the first time, or for which they now have fencing opportunities. Since they know house layout and exit points, problems are less numerous than for the first offence.
which has become known as the event dependency or boost theory. A successful crime boosts the chances of repetition, so that, for example, bank robbers who escape with a good haul are more likely to return to the same bank branch. The third theory is that
The offenders on the first occasion may tell others of the remaining goods, and those whom they tell return to commit offences.
which could be termed the buddy theory. Drawing upon this work, a theory of the relationship between repeat victimization and high-crime areas was proposed independently to the work on attractors and generators but which also drew on Cohen and Felson’s (1979) notion of multiplicative interactions. The theoretical model draws on routine activity and has three principal steps. The first observes that linear increases in interactions between suitable targets, potential offenders, and conducive environments (the absence of guardianship) produce nonlinear increases in crime, the effect on repeat victimization depending upon the specifics of the respective changes. Thus, this interactor is similar to a crime generator. The second step adds a contagion or boost component wherein victimized targets have a heightened risk of experiencing further crime. Such risk is known to increase with each victimization so that a small proportion of targets progress to become chronically victimized supertargets. This explains why repeat victimization accounts disproportionately for crime in high-crime areas as found elsewhere (Johnson et al. 1997). The interactor-contagion model was developed in Farrell et al. (1994, 2005).
The interactor-contagion model offers a stepping-stone from repeat victimization to hot spots. Some hot spots are defined solely in terms of repeat victimization. Others include multiple targets that may or may not be repeatedly victimized. The mechanism that underlies generators is common to these theoretical perspectives. The interactor-contagion model incorporates increases in targets and suitable environments plus the same offenders returning, in addition to the increase in likely offenders that is the characteristic of attractors.
This brief review suggests theories of repeat victimization, and hot spots are compatible and to a large extent already integrated, at least implicitly, though differences in terminology and emphasis mean this not always obvious. Interactor-generator notions are common to both. Statistically, these result in a random chance or Poisson distribution of victimization across targets in the absence of a boost or attractor effect, and this has been a mainstay of research on repeat victimization since Sparks et al. (1977). Boosts, buddies, and attractors offer similar but distinct explanatory mechanisms. Boost denotes increased likelihood of the same offenders returning, buddies denotes increased likelihood of criminal associates returning, while attractors denote increased likelihood of other offenders returning.
The links between the theories bodes well for integration of theory and concepts relating to crime clusters. It seems that perhaps crime concentration of all types take place by the same basic mechanisms. Hence, a unified theory of crime clusters would include flags, interactorgenerators, boosters, buddies, and attractors. Such a theory appears to account for changes in the number and interactions of suitable targets, suitable environments, and likely offenders, and thereby for variation in both individual and area-level risk. It might be applied more universally across clusters so that, for example, hot products flag their attractiveness and are found more frequently at interactorgenerators, with their theft producing boost, buddy, and attractor effects.
Crime clusters are central to prediction, and absent prediction there is no prevention. The term clusters was used as a shorthand for, or stepping-stone to, a broad definition of near repeats incorporating the various dimensions and types of crime commonality and concentration.
A diverse set of crime clusters and commonality was reviewed, and crime event difference indices may assist in the exploration of how similarity, or nearness, is key to crime’s clustering in all its forms. A comparison of the theories of key crime clusters showed them to be overlapping and compatible and suggests that the unification of theory in this area may promote cluster-busting crime prevention practice.
- Andresen MA, Malleson N (2011) Testing the stability of crime patterns: implications for theory and practice. J Res Crime Del 48(1):58–82
- Averdijk M (2010) Individuals victimization patterns over time. Haveka, Alblasserdam
- Block RL, Block CR (1994) Space, place and crime: hot spot areas and hot places of liquor-related crime. In: Eck J, Weisburd D (eds) Crime and place. Crime prevention studies, vol 4. Criminal Justice Press, New York
- Brantingham PJ, Brantingham PL (1995) Criminality of place: crime generators and crime attractors. Eur J Crim Pol Res 3(3):1–26
- Canter DV (2004) Offender profiling and investigative psychology. J Investig Psych Offender Profiling 1:1–15.
- Clarke RV (1999) Hot products: understanding, anticipating and reducing demand for stolen goods. Police research series paper 112, Home Office, London
- Cohen L, Felson M (1979) Social change and crime rate trends: A routine activity approach. Am Sociol Rev 44:588–605
- Eck JE, Clarke RV, Guerette RT (2007) Risky facilities: crime concentration in homogenous sets of establishments and facilities. In: Farrell G, Bowers KJ, Johnson SD, Townelsey M (eds) Imagination for crime prevention: essays in honour of Ken Pease. Crime prevention studies, vol 21. Criminal Justice Press, New York
- Eck JE, Weisburd D (1995) Crime and place, volume 4 of crime prevention studies. Criminal Justice Press, Monsey
- Everson S (2003) Repeat victimisation and prolific offending: chance or choice? Int J Police Sci Manag 5:180–194
- Farrell G (2005) Progress and prospects in the prevention of repeat victimisation. In: Tilley N (ed) The handbook of crime prevention and community safety. Willan, Cullompton, pp 145–172
- Farrell G, Pease K (1993) Once bitten, twice bitten: repeat victimisation and its implications for crime prevention. Police Research Group, Crime Prevention Unit Paper 46. Home Office, London
- Farrell G, Pease K (2008) Repeat victimization. In: Wortley R, Mazerolle L (eds) Environmental criminology and crime analysis. Cullompton, Willan
- Farrell G, Clark K, Ellingworth D, Pease K (1994) High crime rates, repeat victimization, and routine activities. Paper presented to the Cropwood conference. University of Cambridge, Institute of Criminology, Cambridge
- Farrell G, Clark K, Ellingworth D, Pease K (2005) Of targets and supertargets: a routine activity theory of high crime areas. Internet J Criminol pp 1–25
- Johnson SD, Bernasco W, Bowers K, Elffers H, Ratcliffe J, Rengert G, Townsley M (2007a) Space-time patterns of risk: a cross national assessment of residential burglary victimization. J Quant Criminol 23:201–219
- Johnson SD, Birks DJ, McLaughlin L, Bowers KJ, Pease K (2007b) Prospective crime mapping in operational context. Home Office, London
- Johnson SD, Bowers KJ, Hirschfield A (1997) New insights into the spatial and temporal distribution of repeat victimisation. Br J Criminol 37(2):224–241
- Kinney JB, Brantingham PL, Wuschke K, Kirk MG, Brantingham PJ (2008) Crime attractors generators and detractors: land use and urban crime opportunities. Built Environ 34(1):62–74
- Levy MP, Tartaro C (2010) Repeat victimization: a study of auto theft in Atlantic city using the WALLS variables to measure environmental indicators. Crim Just Policy Rev 21:296–318
- Li X, Han J, Lee JG, Gonzalez H (2007) Traffic density-based discovery of hot routes in road networks. In: Proceedings of the 10th international symposium on spatial and temporal databases, Boston. Available at http://www.xiaolei.org/research/pubs/sstd07.pdf
- Lindstrom P (1997) Patterns of school crime: a replication and empirical extension. Br J Criminol 37(1):121–130
- Mailley J, Garcia R, Whitehead S, Farrell G (2008) Phone theft index. Secur J 21(3):212–227
- Melnyk T, Bennella C, Gauthiera DJ, Gauthier D (2010) Another look at across-crime similarity coefficients for use in behavioural linkage analysis: an attempt to replicate Woodhams, Grant and Price (2007). Psychol Crime Law 17(4):359–380
- Morgan F (2001) Repeat burglary in a Perth suburb: indicator of short-term or long-term risk? In: Farrell G, Pease K (eds) Repeat victimization. Criminal Justice Press, Monsey, pp 83–118
- Pease K (1998) Repeat victimisation: taking stock. Police Research Group, Home Office, London
- Ratcliffe JH, Rengert GF (2008) Near repeat patterns in Philadelphia shootings. Secur J 21:58–76
- Sherman LW, Gartin PR, Buerger ME (1989) Hot spots of predatory crime: routine activities and the criminology of place. Criminology 27(1):27–55
- Smith MJ (2005) Robbery of taxi drivers. ProblemOriented Guides for Police Problem-Specific Guide Series No. 34. Department of Justice, Washington, DC
- Summers L (2010) Virtual repeats and near repeats. In: Fisher B, Lab S (eds) Encyclopedia of victimology and crime prevention. Sage, Los Angeles, pp 1043–1047
- Tompson L, Partridge H, Shepherd N (2009) Hot routes: developing a new technique for the spatial analysis of crime. Crime Mapp J Res Prac 1(1):77–96
- Townsley M, Homel R, Chaseling J (2003) Infectious burglaries: a test of the near repeat hypothesis. Br J Criminol 43:615–633
- Velasco M, Boba R (2000) Tactical crime analysis and geographic information systems: concepts and examples. Crime Mapping News 2(2):1–4
- Wellsmith M, Burrell A (2005) The influence of purchase price and ownership levels on theft targets: the example of domestic burglary. Br J Criminol 45:741–764
Free research papers are not written to satisfy your specific instructions. You can use our professional writing services to buy a custom research paper on any topic and get your high quality paper at affordable price.