Effectiveness Of Situational Crime Prevention Research Paper ⋆ Research Paper Examples ⋆ EssayEmpire

This sample Effectiveness Of Situational Crime Prevention 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.

Overview

This research paper summarizes the evidence on the effectiveness of situational crime prevention measures. It uses existing reviews of the evaluation literature to draw broad conclusions about the types of situational measures that show promise and identifies some of the mediating factors in determining levels of effectiveness. It also defines and summarizes evidence on other outcomes that evaluations sometimes considered such as cost-effectiveness, sustainability, anticipatory benefit, and displacement/diffusion of benefits. The principles behind process evaluation, which examines how a scheme was implemented, are also outlined. The second part of this section discusses two different approaches to synthesizing evidence on what works – the approach taken by those who advocate detailed consideration of the validity of the evaluation research design and synthesis through metaanalysis (e.g., the Campbell collaboration) and those advocating a realist approach, which uses all relevant evidence and considers how distinct contexts and mechanisms impact on outcome (e.g., Pawson 2006). The paper concludes with an account of what still remains to be done in evaluating SCP, including gaps in our current knowledge and future methodological developments.

Introduction

Situational crime prevention (SCP) entails efforts which seek to alter the situational environment in an effort to disrupt the frequency and harmful effects of reoccurring crime behavior. Unlike other approaches which seek to reduce crime by focusing solely on offenders, the situational prevention approach recognizes the importance which the environment plays in facilitating and, in some cases, causing crime events. Environments where prevention measures can be implemented include not only geographical places but also include adaptations to manufactured products and electronic systems which readily present opportunities for offending. Among geographical environments, 11 general types have been identified (Clarke and Eck 2003), though the most common of these include residential, public ways, recreational, retail, and transport.

Several challenges exist when evaluating situational crime prevention efforts. Some of these are common in applied research settings generally, while others are unique to the evaluation of situational prevention efforts. In part, these challenges stem from the environmental orientation of situational crime prevention.

Conventional research methodology to a large extent originated from research which focused on the individual level. The application of these same research methods to the evaluation of place, product, and system-based interventions is not always compatible. At best, conventional methods require adaptation when used in environmental settings. At worst, they are simply not applicable.

One challenge facing situational crime prevention evaluation is that SCP is a process rather than an intervention(s) in and of itself. This process relies on an action research methodology which calls for (i) an identification of problems, (ii) in-depth analysis of the factors that contribute to and facilitate the problem behavior, (iii) identification of suitable responses and a determination of those which are most feasible, (iv) implementation of those responses, and (v) an assessment of their impact. Inherent in this process is that the implementation of situational measures will be specifically tailored to the unique circumstances of the problem at hand. This means that problem profiles will likely differ across time, geographical settings, products, and systems. What is devised to work in one location or in response to one problem will likely not work at other places or in response to other problem types. Accordingly, random assignment of interventions, as required in classical experiments, runs counter to the deliberative process of situational crime prevention since it calls for blindly applying interventions at random.

A second challenge is that many situational prevention schemes rely on intervention “packages” consisting of multiple prevention measures. For instance, an initiative to reduce residential burglary may include an assortment of tactics such as a media campaign, target hardening measures including the use of security alarms and the fortification of dwelling entry points, and improved lighting, among others. This makes it difficult to identify the role of any single situational intervention as being responsible for any observed reductions in crime. At best, what can be revealed by any evaluation is that the assortment of situational alterations resulted in some change. To compound matters, a third challenge is that most evaluations of situational crime prevention initiatives are conducted retrospectively after the intervention has already taken place, thereby making random assignment impossible.

A fourth challenge is that despite situational crime prevention having a formal theoretical foundation and action framework for understanding crime problems and formulating situational interventions (Clarke 1992), many evaluations of situational prevention efforts include those which were not informed by the framework. While the underlying mechanisms for how and why those situational interventions should work can be understood by the SCP approach despite their ex parte implementation, the remaining caveat is that the situational interventions may not be accurately suited to the specific crime problem that they were intended to address. The intervention may be incomplete or simply irrelevant to the nature of the problem. This presents the possibility of producing false negatives (erroneous conclusions of ineffectiveness) when trying to determine the impact of the situational interventions that were implemented.

A fifth challenge is that there are a variety of developing concepts such as displacement (the movement of crime to other places, times, tactics, targets, or offenses), diffusion of benefits (a reduction in crime surrounding intervention areas), anticipatory benefits (the premature reduction in crime prior to the implementation of an intervention), and questions of costeffectiveness (an assessment of the gains achieved by an intervention in relation to the expended cost) which also have to be accommodated in any comprehensive evaluation of situational crime prevention schemes. Each of these requires different research designs if they are to be adequately assessed. To study spatial displacement and diffusion effects, for instance, requires the use of a buffer zone surrounding treatment areas in addition to conventional comparison areas. For an assessment of anticipatory benefits, a time series design must be utilized in order to reveal any premature reductions in crime prior to the intervention having been put in place. Lastly, assessments of cost-effectiveness require determinations of the expenditures involved in the implementation of the prevention scheme as well as some of accounting of crimes prevented and their associated cost savings.

These challenges explain, to some degree, the distribution of research methods that have been commonly used in the evaluation of situational crime prevention to date. A review of SCP evaluations (see Guerette 2009) used over the last nearly four decades found that the majority of studies used some variety of a quasi-experimental design with many using a combination of a time series and before-after design with comparisons. Only 3 % (n ¼ 6) of the evaluations relied upon randomized experimentation.

Despite these challenges, considerable progress has been made in understanding the effects of situationally focused crime reduction schemes. The following synthesizes the current state of situational crime prevention research:

On the aggregate, the majority of SCP efforts are reportedly effective according to study authors. Roughly three out of four evaluations indicate beneficial reductions in crime (Guerette 2009).
The effectiveness of SCP holds true even after adjusting for displacement and diffusion effects (Guerette and Bowers 2009; Hesseling 1994; Eck 1993).
Diffusion of benefits is just as likely to occur as displacement (Guerette and Bowers 2009; Guerette 2009).
SCP effects are often compounded by anticipatory benefits. Smith et al. (2002) found that in some cases crime reduction can be observed before the official start date of a situational prevention scheme.
Situational prevention measures appear to have more enduring reduction effects compared to focused policing operations (Braga and Bond 2008).
The large majority of SCP efforts evaluated comprised techniques which either increase the effort necessary to commit crime and/or increase the risk involved (Guerette 2009).
Most evaluations of SCP rely on quasiexperimental designs. Randomized experiments remain rare (Guerette 2009).
SCP measures have most commonly been implemented and evaluated in four place types: residential, retail, public ways, and transport (Eck and Guerette 2012; Guerette 2009).
On the aggregate cost assessments of SCP measures are less common (less than half of evaluations compute or allow for such determinations (Guerette 2009)), though some focused reviews have found them to be cost efficient (e.g., Welsh and Farrington 2008a).
Little is known about the sustained impact of SCP measures beyond 1 or 2 years. Most evaluations (nearly two-thirds) of SCP have post-evaluation follow-up periods of 24 months or less (Guerette 2009).

The following sections present in more detail the prevailing findings regarding the evaluation of situational crime prevention initiatives. The sections draw heavily from existing systematic reviews of situational crime prevention schemes and report on the nature of SCP evaluation, some emerging findings of effectiveness, and discuss what is known about the unique concepts involved in SCP evaluations, namely, those of displacement and diffusion, anticipatory benefits, sustainability, and cost-effectiveness.

The Current Evidence On What Works

A small body of systematic reviews of situational crime prevention evaluations has consistently found them to be effective. These reviews have examined SCP evaluations in a variety of ways. Some have assessed the collective conclusions reported by study authors, while others have sought to examine findings disaggregated by place and intervention types. Still others have examined more narrowly specific intervention types, such as CCTV and street lighting. Reviews of evaluations are always hampered by the underlying methodological rigor of the evaluation studies which are reviewed. Generally, those with stronger research designs are viewed as more trustworthy, and many reviews place thresholds on the levels of design rigor that will be included in the review. Despite the variability in the types of research designs and the focus of the review, on the aggregate, evaluations of SCP suggest that situational alterations at problem places can effectively reduce crime.

In a descriptive review of conclusions drawn by study authors of 206 SCP evaluations (Guerette 2009), it was found that three out of four (75 %, n = 154) concluded that the intervention was effective overall. Twelve percent (n = 24) concluded that the situational intervention was not effective, while 6 % (n = 12) reported mixed findings, and 8 % (n = 16) of the study outcomes were inconclusive. Relying on this same sample of evaluations, another review examined study findings across five common place types (Eck and Guerette 2012) resulting in a subsample of 149 evaluations. In that review variation in effectiveness across place types was found; however, for all place types, effectiveness was reported by at least 60 % of the evaluations (see Table 1). Though all recreational interventions appeared successful, the small number makes any meaningful conclusion unreliable, since more evaluations will likely find negative effects. Considered differently, situational efforts in public outdoor settings and in residential places had the most failures, but even in those settings, ineffective interventions were comparatively small compared to those found to be effective and those with mixed results.

One interesting finding in this review was that situational interventions at public places compared to other settings appeared less successful. There are many possible reasons for this, including (i) public places are more open and less contained; (ii) they may host a variety of crime behaviors some of which are not expected to be prevented by the situational measure; (iv) their implementation may be flawed; and (v) the techniques used were “off the shelf” and not informed by careful analysis of the targeted crime problem.

An examination of the success rate of commonly used interventions also found them to be mostly effective, although with some variation across intervention type. Table 2 presents interventions that were used in decreasing order of frequency (totals column) and shows thepercent of each intervention type demonstrating different outcomes (effective, not effective, mixed, inconclusive). Seventy-nine percent of the 149 studies reviewed entailed the use of just seven situational intervention techniques. Interestingly, though most were found effective, the most frequently used interventions showed the lowest rate of effectiveness. Similar to before, this may be the consequence of the tendency to use “off-the-shelf” techniques rather than those tailored to the specific problem at hand.

Effectiveness of Situational Crime Prevention, Table 1 Effectiveness of place-based intervention evaluations by common place types

Effectiveness of Situational Crime Prevention, Table 2 Effectiveness of the most used interventions

It is important to note that these findings are drawn from the study authors’ conclusions and must be taken with consideration of the types of evaluation methods that were used. Stronger evaluation designs allow for more confident determinations that it was in fact the situational intervention that was responsible for any observed crime reduction, while weaker evaluation designs leave open the possibility that something else caused the change. Because of this only gross determination of situational crime prevention effectiveness can be concluded.

In separate meta-analytic reviews of the two most common SCP techniques, closed circuit television (CCTV), and street lighting, similar positive findings were found (Welsh and Farrington 2008a, b). Using meta-analytical techniques, a review of 41 evaluations of CCTV found that overall the technique produced moderate significant reductions in crime though this too varied by place type. CCTV was found to be most effective when used in parking lots to reduce vehicle-related crime. The findings also revealed that they tended to be more effectively applied in the United Kingdom than in the United States. Using similar methods, a meta-analysis of 13 evaluations of the effects of improved street lighting found that they too achieved significant reductions in crime. This reduction was not restricted to nighttime crimes (which would be predicted to be most influenced by improved street lighting). Thus, a form of temporal diffusion of benefits was observed.

State Of The Art: Measuring Outcomes Other Than Crime Reduction

As mentioned in the introduction, there are outcomes other than whether or not crime has been reduced in the area of intervention that are of interest to crime prevention practitioners. These are designed to answer the following questions: Is the intervention good value for money? How long does the crime reductive effect of an intervention last? Does the intervention shift the crime problem elsewhere? Particularly comprehensive evaluations attempt to answer these questions, but evidence here is limited because the data and the methods required in undertaking such research are complex. This section will summarize some of the issues and evidence where it exists under each of these headings, respectively.

Is The Intervention Good Value For Money?

In order to answer this question, a cost-benefit analysis is necessary (Roman and Farrell 2002). This involves estimating the cost of the intervention in monetary terms. This is no simple undertaking as there are many costs associated with setting up and running a scheme. These costs include personnel, transport, equipment, premises, training, and advertising. Ideally the costs should include everything from employment of full-time personnel and substantial hardware (such as CCTV cameras or alleygates) to the cost of fuel and office supplies. A further complication is how to cost in-kind contributions or voluntary personnel. For example, a scheme might be (in fact is nearly always) introduced in an environment where personnel are already working on crime prevention tasks, so they might purely absorb the further activity without the need for new staff; however, there is still an opportunity cost associated with the time that they devote to the new project. There is also the matter of approximating local costs for personnel and staff; there is likely to be considerable variation in costs by region and to be accurate analysis should use local rather than national cost estimates. In order to reliably collect cost information, evaluators need to gather information from the very beginning – it is virtually impossible to get a reliable estimate of cost retrospectively.

Once the costs of the scheme itself have been collected, these costs need to be balanced against the financial benefit of the crimes that have been averted as a consequence of the preventative activity. This involves obtaining two estimates: one quantifying the number of crimes of each type that have been prevented and the other quantifying the costs that would result if those crimes had occurred. Once more, acquiring this information is no mean feat. The outcome effect size (e.g., the percentage reduction in crime) needs converting to a count of crimes prevented (see, e.g., Johnson et al. 2004). Following this, these counts need to be multiplied by the associated financial cost to society that would have been incurred as a result of the crime and summed to get an estimate of overall financial benefit. Data on the costs of crime are available for some countries. For example, the UK Home Office have estimates available for 2005 (Dubourg et al. 2005) for a range of crime types. The costs are national ones and include costs in anticipation of crime (e.g., security and insurance expenditure), as a consequence of crime (e.g., property stolen, lost output, and emotional/physical impact), and in response to crime (e.g., the police costs and those of the criminal justice system). For reference, this source estimates the cost of a common assault at £1,440, a dwelling burglary at £3,268, and a homicide at just over £1.4 million. There are obviously questions about these data in terms of how the estimates are made (particularly the more intangible costs such as emotional impact), how the costs change over time, and how useful national costs are at the local level.

There have been surprisingly few (published) studies that undertake cost-benefit analysis of situational crime prevention measures. Of note are the street lighting evaluation by Painter and Farrington (2001), the Kirkholt project evaluation (Forrester et al. 1990), an evaluation of an anti-robbery target hardening project (Clarke and McGrath 1990), two CCTV evaluations (Skinns 1998 and Gill and Spriggs 2005), and an evaluation of alley-gating (Bowers et al. 2004a). What is evident from these studies is that in the majority of cases, such measures were not only effective at reducing crime but also cost-effective in that the financial benefit of the intervention outweighed the cost. Of course, this particular set of findings is not likely to be a truly representative sample, but the results can be seen as giving some cause for optimism. See Table 3 for more details of the cost-benefit analysis estimates for these examples.

Effectiveness of Situational Crime Prevention, Table 3 Cost-benefit analysis of situational measures

How Long Does The Crime Reductive Effect Last?

Answering this question involves dealing with a number of related concepts, some of which were mentioned in the introduction. These include anticipatory benefit (the potential reduction in crime before the official start of the scheme), sustainability (the length of time over which a reductive effect is observed), and residual deterrence (a reduction in crime which temporarily continues when the intervention has come to an end). As is the case for cost-benefit analysis, there are few SCP evaluations that systematically test for these possibilities.

Smith et al. (2002) outline a number of potential causes of anticipatory reductions in levels of crime and search for empirical examples. From a conceptual point of view, they highlight a number of different mechanisms: training that staff have received before the official start of the scheme being immediately utilized, pre-scheme fact finding activity (such as surveys), changing pre-implementation surveillance levels, or publicity causing changes in offenders perception of risks associated with offending in certain areas. Empirical evidence found a possible anticipatory effect in approximately half of series of situational prevention schemes (22 of 52) where enough information was available for such detailed pre-scheme assessment. Looking for anticipatory benefit needs to be done with caution to ensure that it is not a dummy effect being detected. In order to rule out the possibility of an “anticipatory effect” in fact being regression to the mean, a seasonal fluctuation, or a change in recording practice, suitable controls need to be used in the analysis.

Sherman (1990) reported that reductions in crime as a consequence of a police crackdown were realized during the schemes intervention period but that these also endured for some time after the implementation had ceased. This is attributed to a temporary change in offenders’ perception of risk involved in committing crime that extends beyond the lifetime of the operation. A related point is the issue of sustainability. Unlike those that follow individual life courses of offending behavior, situational prevention evaluations rarely extend beyond 1 or 2 years following the original implementation of an initiative. This means that there is little information concerning the length of the period for which such measures remain effective. Situational measures which use long-lasting hardware or change the physical layout of the environment are both examples where there is scope for crime reductive effects to last over long periods of time. Using an evaluation method that stops monitoring outcome after 1 or 2 years runs the risk of significantly underestimating the crime reductive effect of these measures. It also limits the availability of information concerning the point at which hardware should be updated or replaced. A piece of equipment might have a 10-year warranty – but is it cost-effective to replace it at this point?

The Bowers et al. (2004a) alley-gating evaluation illustrates this point. The evaluation assessed outcomes at two different stages. The analysis separated alley-gates that had been in place for less than a year from those that had been in place a year or more. This demonstrated that in fact the gates only became cost-effective after they had been in for some time; the benefitcost ratio was £0.96 for those in for less than a year and £1.86 for those in a year or more.

Does The Intervention Shift The Crime Problem Elsewhere?

One of the most common criticisms leveled at situational interventions is that crime will simply relocate to other times and places since they do not address what are seen as the root causes of crime. There are a number of different forms of potential displacement (e.g., Hesseling 1994; Barr and Pease 1990), of which spatial displacement (movement of crime to nearby areas) is the form which tends to be most commonly analyzed in evaluation studies. Other commonly listed possibilities include temporal displacement (crime moves to another time), target displacement (crime is directed away from one target to another), crime type displacement (one kind of crime is substituted for another), tactical displacement (one method of committing crime is replaced by another), and offender replacement (one offender is arrested but replaced by another). There is of course another possibility, which is that the positive effect of the scheme is larger than originally anticipated; this is known as “diffusion of crime control benefits” (Clarke and Weisburd 1994). For example, diffusion takes place when reductions in crime occur in untreated areas or targets that are close to those that have received treatment or in types of crime similar to those targeted by the scheme.

A number of reviews of displacement and diffusion have been conducted. Three of these were undertaken in the 1990s (Barr and Pease 1990; Eck 1993; and Hesseling 1994). These early displacement reviews were fairly consistent in finding that displacement was often not observed, and in cases where it was, it tended to be smaller in scale than the reduction shown in the intervention areas. Of the 33 studies reviewed by Eck (1993), only three (9 %) appeared to suggest a substantial amount of displacement (where the negative outcome associated with displacement outweighed the positive gain as a result of the intervention). Similarly, reviewing 55 studies, Hesseling (1994) found that in 40 % of cases, no displacement at all was reported, and six studies appeared to demonstrate a diffusion of benefits. Finally, Barr and Pease (1990) took a more conceptual approach in their review and considered the equity of crime distribution, arguing that even in the minority event of total displacement, a redistribution of crime can still achieve a desirable social gain.

A more recent review of displacement and diffusion effects among situational crime prevention measures sought to update these earlier findings (Guerette and Bowers 2009). This review reported evidence from 102 evaluations of situational focused crime prevention projects and involved analysis of 574 separate observations relating to potential displacement or diffusion of benefit. Of these observations, the study authors reported displacement in 26 % and diffusion of benefit in a further 27 %. Interestingly these figures varied slightly depending on the type of displacement considered. For example, temporal and target displacement were slightly more commonly observed, as was a spatial diffusion of benefit. However, the reliability of these variations is restricted by the low number of evaluations in some of the displacement categories. For 13 of the 102 studies, the study design and the information available allowed a more detailed independent quantitative assessment. Here analyses suggested that when spatial displacement did occur, it tended to be less than the treatment effect, suggesting that on balance the intervention was still beneficial. Moreover, a diffusion of benefit appeared to be the more likely outcome.

A further systematic review of the degree to which SCP measures displaced crime or lead to diffusion of benefit was conducted by Johnson et al. (2011) using procedures closely in line with those recommended by the Campbell Collaboration (see http://www.campbellcollaboration.org/). This found that for the 13 studies in which sufficient information was available for the quantitative calculation of effect sizes, geographic displacement was by no means inevitable for this type of intervention. It was rarely the case that crime increased in the potential displacement areas following intervention. In fact, crime was just as likely to decrease in the areas that surround an intervention area, demonstrating diffusion of benefit. This review was limited in scope by low numbers of suitable studies, and those that were included tended to involve SCP measures that aimed to increase surveillance. That said it is still encouraging that displacement does not appear to be the inevitable outcome of SCP that it has sometimes been made out to be.

Possible Controversies: Levels And Types Of Evidence

One debated issue in SCP evaluation is the type of evidence that should be used in the assessment of overall levels of effectiveness of particular interventions. There are two schools of thought: those that consider that only high-quality evaluation research should be included in such an assessment and those that, while conscious of validity issues, believe it is more important to have an assessment which understands and accounts for differences in local circumstances which will ultimately provide more evidence on the context in which the measures will work and the mechanisms by which the schemes operate.

Those that advocate firm quality control on the evaluation research design often refer to the Maryland Scale of Scientific Evidence (Sherman et al. 1997). This grades research designs according to their internal validity and deems randomized control trials the highest standard, quasi-experiments with multiple control groups as the next and evaluations which have no control or no temporal aspect the lowest standard. Under this paradigm restrictions are put in place in terms of the evidence that should be used in reviewing effectiveness. Commonly this means discounting evidence that is not level 3 or above on the 5-point Maryland Scale. Some SCP reviews that adhere to this guidance include the CCTV and street lighting reviews conducted by Welsh and Farrington (2008a, b), the review of Neighborhood Watch conducted by Bennett et al. (2008), and the POP review conducted by Weisburd et al. (2008). All of these reviews were undertaken using the procedures outlined by the Campbell Collaboration.

The approach which is more focused on understanding the process by which situational measures work and how this is moderated by the particular context of an intervention is generally referred to as the realistic evaluation approach (see Pawson 2006 and Pawson and Tilley 1997). A realist review starts with a model based on program theory (a statement of how an intervention works) and adds layers of evidence onto this from, for example, empirical evaluations. As the layers of evidence are added, it is likely that the initial model will need refining in able to sufficiently absorb it. Hence, a review might find that an intervention works differently in a newly revealed context. A significant feature of this approach is that the evidence is not just limited to evaluations. Other bits of evidence that might be included are theoretical pieces on process, agency reports on implementation, or initial bids for funding to name a few. The focus is on taking the most relevant bits of evidence from a range of different sources. The realist approach does not rule out using evidence from a study because of a poor research design but looks instead for what can be seen as “gold nuggets” even if they are in otherwise problematic documents. A further consideration often raised under the realist approach is the degree to which implementation failure occurs. That is when an intervention does not operate as original planned or parts of the intended activity fail to materialize (see Knutsson and Clarke (2006) for a detailed account).

A third option of course is to take some sort of hybrid approach. A review by Van der Knaap et al. (2008) on prevention of violence in public domains planned to do just that. This review classified each of 48 publications describing 36 violence reduction interventions into one of the five levels on the Scientific Methods Scale. Most were level 3 or above and many were level 4 or above. The authors choose to use only those that were level 4 or above in a vote counting review. The results found that of those where sufficient evidence was available, nine appeared to be effective and six potentially effective, with the remaining approaches looking less promising. As well as undertaking this systematic enquiry, the authors also searched the documents for details of program theory. They found three overarching mechanisms for schemes deemed as effective: ones that were cognitive and focused on learning, teaching, and training; ones that used the social environment to reward or punish behavior; and ones that focused on risk reduction to victims by promoting protective factors. This example of an integrated approach is not specifically related to SCP intervention but demonstrates that there are strengths to reviewing in this way as it does not neglect either of the philosophies outlined above. However this approach has the distinct drawback that it is likely to be very time consuming in nature.

Open Questions

There are some remaining gaps in our knowledge concerning evaluation of situational prevention measures. It is obvious from the discussion above that these include a more extensive evidence based on the cost-effectiveness of different approaches and further research on the sustainability of interventions in terms of estimating the lifetime over which they are effective. Also useful would be further research into the extent of displacement of types other than the more traditionally measured spatial variety. This undertaking presents a number of data collection and methodological challenges. For example, more detailed information recording precisely what times an intervention was active between or precisely which individuals or properties within a scheme area were treated would be necessary to enable a reliable assessment of temporal or target switch displacement. A future methodological challenge would involve examining the interactions between potential types of displacement. For example, it is conceivable that an intervention caused a crime switch that involved both a different target and occurred at a different time?

A further useful line of enquiry would be in more systematically assessing the degree to which implementation intensity affects the level of reduction seen. Intensity refers to the dosage of intervention received per treated subject. It is usually expressed in financial terms and is represented, for instance, as pounds or dollars spent per head of the treated population. It is unclear whether the relationship between dosage/intensity and crime reduction is a linear one (Bowers et al. 2004b). It might be the case that there is a level of spending per head at which point cost-effectiveness is optimized and after which any further investment gives diminishing returns. Evaluations that look at the relationship between intensity and crime reduction remain few and far between (see, e.g., Ekblom et al. 1996).

There could usefully be more syntheses of evidence on effectiveness conducted from the perspective of both schools of thought outlined above. There are plenty of SCP interventions for which (to the authors’ knowledge) there has not been a review. For example, there might usefully be a synthesis examining the degree to which controlling access to facilities, reducing anonymity, or reducing provocations can reduce crime. There should also be more syntheses that use mixed methods such as those illustrated by Van der Knaap et al. (2008).

Finally, to feed into the body of evidence on what works from a situational perspective, more individual evaluations are necessary. A framework for optimizing these in terms of the production of new research evidence rather than replication of already established effects would perhaps be the most desirable way forward. For example, the Welsh and Farrington (2008a) review of CCTV described results from 44 evaluations in a number of different contexts and across a number of different countries. This provides fairly solid evidence of the extent to which CCTV shows promise as a crime reduction measure and facilitates discussion of the factors that are likely to moderate its effectiveness. However, from a brief scan of the published academic literature, there appear to have been far fewer evaluations of the effectiveness of replacing beer glasses with safer alternatives (Warburton and Shepherd 2000; Anderson et al. 2009) and more evidence on how generalizable the findings of these studies would be useful, both from a qualitative and a quantitative perspective. Having said that more novel evaluations would be useful, it is important to balance this with a sufficient number of replications to increase the external validity of the evidence base.

Bibliography:

Anderson P, Chisholm D, Fuhr D. (2009) Effectiveness and cost-effectiveness of policies and programmes to reduce the harm caused by alcohol. Lancet, 373(9682):2234–2246.
Barr R, Pease K (1990) Crime placement, displacement and deflection. In: Tonry MH, Morris N (eds) Crime and justice: a review of research, vol 12. University of Chicago Press, Chicago
Bennett T, Holloway K, Farrington D (2008) The effectiveness of neighborhood watch. Campbell Syst Rev 2008:18. doi:10.4073/csr.2008.18
Bowers KJ, Johnson SD, Hirschfield AFG (2004a) Closing off opportunities for crime: an evaluation of alleygating. Eur J Crim Pol Res 10(4):283–308
Bowers KJ, Johnson SD, Hirschfield AFG (2004b) The measurement of crime prevention intensity and its impact on levels of crime. Brit J Criminol 44(3):1–22
Braga AA, Bond BJ (2008) Policing crime and disorder hot spots: a randomized, controlled trial. Criminology 46:577–607
Clarke RV (1992) Introduction. In: Clarke RV (ed) Situational crime prevention: successful case studies. Harrow and Heston, Albany
Clarke RV, Weisburd D (1994) Diffusion of Crime Control Benefits: Observations on the Reverse of Displacement. In RV Clarke (ed.) Crime prevention studies, Vol 2. NY: Criminal Justice, Monsey, pp. 165–182
Clarke RV, Eck J (2003) Become a problem solving crime analyst in 55 small steps. Jill Dando Institute of Crime Science, London. http://www.popcenter.org/library/ reading/pdfs/55stepsUK.pdf
Clarke RV, McGrath G (1990) Cash reduction and robbery prevention in Australian betting shops. Secur J 1:160–163
Dubourg R, Hamed J, Thorns J (2005) The economic and social costs of crime against individuals and households 2003/04 Home office on-line report 30/05
Eck JE (1993) The threat of crime displacement. Crim Justice Abstr 253:527–546
Eck JE, Guerette RT (2012) “Own the place, own the crime” prevention: how evidence about place-based crime shifts the burden of prevention. In: Loeber R, Welsh BC (eds) The future of criminology. Oxford University Press, Oxford
Ekblom P, Law H, Sutton M, Crisp P, Wiggins R (1996) Safer cities and domestic Burglary. Home office research study no. 164. Home Office, London
Forrester D, Frenz S, O’Connell M, Pease K (1990) The Kirkholt Burglary prevention project: phase 2, crime prevention unit paper no. 23. Home Office, London
Gill M, Spriggs A (2005) Assessing the impact of CCTV. Home office research study, no. 292. Home Office, London
Guerette RT (2009) The pull, push and expansion of situational crime prevention evaluation: an appraisal of thirty-seven years of research. In: Tilley N, Knutsson ZJ (eds) Evaluating crime reduction initiatives crime prevention studies, vol 24. Criminal Justice, Monsey, pp 29–58
Guerette R, Bowers K (2009) Assessing the extent of crime displacement and diffusion of benefit: a systematic review of situational crime prevention evaluations. Criminology 47(4):1331–1368
Hesseling R (1994) Displacement: a review of the empirical literature. In: Clarke RV (ed) Crime prevention studies, vol 3. Criminal Justice, Monsey
Johnson SD, Bowers KJ, Jordan P, Mallender J, Davidson N, Hirschfield AFG (2004) Estimating crime reduction outcomes: how many crimes were prevented? Evaluation 10(3):327–348
Johnson SD, Bowers KJ, Guerette R (2011) Crime displacement and diffusion of benefit: a review of situational crime prevention measures. In: Welsh B, Farrington D (eds) The oxford handbook of crime prevention. Oxford University Press, Oxford
Knutsson J, Clarke RV (eds) (2006) Putting theory to work: Implementing situational prevention and problem-oriented policing. Crime Prevention Studies Vol 20. NY: Criminal Justice, Monsey
Knutsson J, Tilley N (eds) (2009) Evaluating crime reduction initiatives, vol 24, Crime prevention studies. Criminal Justice, Monsey
Painter KA, Farrington D (2001) The financial benefits of improved street lighting, based on crime reduction. Light Res Technol 33:3–12
Pawson R (2006) Evidence-based policy: a realist perspective. Sage, London
Pawson R, Tilley N (1997) Realistic evaluation. Sage, London Roman J, Farrell G (2002) Cost-benefit analysis for crime prevention: opportunity costs, routing savings and crime externalities. Crime Prev Stud 14:53–92
Sherman L (1990) Police crackdowns: initial and residual deterrence. In: Tonry M, Morris N (eds) Crime and justice: an annual review of research. University of Chicago Press, Chicago, pp 1–48
Sherman L, Gottfredson D, MacKenzie D, Eck J, Reuter P, Bushway S (1997) Preventing crime: what works, what doesn’t, what’s promising. National Institute of Justice, U.S. Department of Justice, Washington, DC
Skinns D (1998) Doncaster CCTV surveillance system: second annual report of the independent evaluation. Faculty of Business and Professional Studies, Doncaster College, Doncaster
Smith MJ, Clarke RV, Pease K (2002) Anticipatory benefit in crime prevention. In: Tilley N (ed) Crime prevention studies, vol 1. Criminal Justice, Monsey, pp 71–88
Van der Knaap L, Leeuw F, Bogaerts S, Nijssen L (2008) Combining Campbell standards and the realist evaluation approach: the best of two worlds? Am J Eval March 2008(29):48–57
Warburton AL, Shepherd J (2000) Effectiveness of toughened glassware in terms of reducing injury in bars: a randomised controlled trial. Inj Prev 6(1):36–40
Weisburd D, Telep CW, Hinkle JC, Eck JE (2008) The effects of problem-oriented policing on crime and disorder. Campbell Syst Rev 2008:14. doi:10.4073/csr.2008.14
Welsh BC, Farrington DP (2008a) Effects of closed circuit television surveillance on crime. Campbell Syst Rev 2008:17. doi:10.4073/csr.2008.17
Welsh BC, Farrington DP (2008b) Effects of improved street lighting on crime. Campbell Syst Rev 2008:13. doi:10.4073/csr.2008.13