Crime Prevention






Strengthen the law


Family law act (1996) – automatic arrest for breach of non-molestation orders

Child abuse suspects have to leave the family home rather than the children


Different strategies are needed for different crimes.


Date rape


American students issued with pamphlet from the American College Health Association.  Warns of alcohol, miscommunication, challenges stereotypes, gives examples of how to handle particular scenarios.

Literature has been inconsistent – be passive / aggressive, Cry / do not show weakness, Scream / do not scream. (Morgan 1986).

Modern approaches emphasise the situation (e.g. ensure there is an escape route) or society (e.g. exploding myths and emphasising the trauma that ensues for the rape victim).  Can you think what the effects would be in terms of the victim being able to form future relationships?

How could you help to prevent date rape in your institution?


Four main approaches to crime prevention


1.      Primary prevention – reduce the chances of an individual becoming a criminal

2.      Secondary prevention – Prevent the first time offender from becoming a hardened criminal. As in the James Bulger case where the boys have been released before having to experience an adult prison where they were likely to become confirmed criminals.

3.      Tertiary prevention – Trying to prevent chronic offenders from committing further offences.

4.      Situational crime prevention – reducing opportunities for crime to take place. Locking your bike with a D lock rather than a chain that can easily be cut through.


Primary prevention


Within this approach energies are directed towards preventing the onset of criminality, and known risk factors associated with childhood anti-social behaviour have been identified. These include poverty and poor housing; poor parenting; association with delinquent peers, siblings and partners; low intelligence; poor school performance and truancy; high levels of impulsiveness and hyperactivity; and being brought up by a criminal parent (Farrington, 1996). Children who are exposed to multiple risks are disproportionately likely to become serious or persistent offenders.

Longitudinal study of boys from vulnerable backgrounds.  In the 1940s experimental group of ‘difficult’ boys given personal and social counselling from social workers plus additional academic tutoring for 5 years.  Compared to a control group 30 years later more in the experimental group had committed at least two offences (McCord 1978).  The intervention did not work in other words.  Perhaps the intervention started too late for the boys who had already been identified as ‘difficult’.  Or perhaps the study had led to the boys identifying with their label and thus the results reflect this self-fulfilling prophecy.  Perhaps the intervention had been inadequate.


Other ineffective strategies include individual casework, counselling, corporal punishment, school suspension, information campaigns and fear arousal (Oottfredson, 1997). Those strategies which do seem to produce success are:


·         Early home visits and preschool education programmes

·         Family—school initiatives

·         Anti-bullying strategies in schools.

·        Parental training for school children

·        Information packs for practitioners on how to spot child abuse

·        Support groups for new and unskilled mothers. 



The relationship between the ability to defer gratification and social competence has been demonstrated by the use of the ‘marshmallow test’ (developed by Mischel et al in the 1960’s).  The test is very simple.  The young child is told that they can either eat one marshmallow now or wait until the experimenter returns a short while later in order to be rewarded with two marshmallows.  Some children could not wait and ate the one marshmallow as soon as the experimenter left the room.  Adolescents were traced who had taken the ‘marshmallow test’ at the age of four.  Those who had immediately eaten the one marshmallow (unable to defer gratification) were found to be more troubled socially and academically, unlike those who had resisted temptation who were more socially competent, confident and successful at school (Shoda et al., 1990).  Notice however that the marshmallow test confounds the ability to delay gratification with the ability to follow rules.  If the child at four was not able to follow a simple rule then it should not be that surprising that they were still unable to follow rules later on in life.

Accepting that the inability to delay gratification at four is in someway responsible for later social problems then it would seem that training in being able to delay gratification might have some benefits.


Utting (1996) provides accounts of 30 such programmes operating in the UK, such as ‘Sure Start’ and ‘On Track’.

There is certainly evidence that early intervention at the preschool or initial school level seems to be successful, whether the focus is on the children (providing training in social-cognitive skills in order to reduce impulsivity) or on parental training to encourage more effective attachment (Hawkins et al. 1987). The best known preschool intervention is ‘Operation Headstart which was introduced in America in the 1960s and attempted to accelerate cognitive development in children from high-risk families before their entry to school. Although the evidence for sustained cognitive gains is limited, what these children did seem to gain was enhanced social competence, including reduced aggression and the ability to defer gratification (Zigler and Hall, 1987).

Farrington (2001) suggests that:


Logically, it must be better to prevent offending by intervening early in life than to wait until someone has committed many offences and then intervene many victims will be spared. And yet most crime reduction resources are devoted to the police, court, prison and probation services and very few to prevention. (p.182)


What is recommended is the setting up of a national agency with a primary mandate for the early prevention of offending, funding and co-ordinating the activities of local agencies while establishing an ongoing cost—benefit analysis of research and practice in the prevention of crime. Farrington suggests that prevention programmes should aim not only to tackle risk factors but also to strengthen protective factors.

Problems include the cost of covering such a large number of people with potential problems and the danger of labelling such people and thus encouraging them to identify with the label and in turn to take up crime.



Here the intervention occurs when an individual is showing some signs of antisocial behaviour and has already committed a single or minor offence. The aim is to minimise legal intervention in recognition of the fact that being ‘processed’ and labelled within the criminal justice system is likely to increase the development of a criminal identity and the probability of further offending. In these cases an informal warning or caution from the police will be given together with a possible referral to the local Juvenile Liaison Bureau, where an interdisciplinary team will work with the young person and their family to identify and resolve recurring problems.

There is also the concept of restorative justice which emphasises the value of ‘reintegrative shaming’ of early offenders focusing on the offence rather than the offender but requiring the offender to apologise for their actions and make reparation (Braithwaite, 1989). This involves diverting offenders from court and involving them in a process whereby they quite literally make reparation to their victims by apologising and offering some financial compensation or even repairing the damage done. The rationale behind such schemes is that offenders have to reflect on their actions and the consequences of those actions, accepting responsibility and facing their victim while trying to make amends. While it may sound exceedingly simple it is clear that not only is such a scheme inexpensive, but it can be very effective in terms of reducing recidivism and also in helping the victims of crime. Evaluation of such a scheme operating in Northamptonshire has been very favourable, with an estimated cost of £720 per case, in comparison to the cost of £2500 if offenders were taken to court, and a 35% re-offending rate by comparison with 67% (Audit Commission, 1996).

Intermediate treatment programmes began in the 1970s and are designed to take the young offender away from his environment and to show them how to manage in society and to broaden their experiences.  In the Bulger case Thomson and Venables were allowed out of their secure homes and taken on shopping trips and to football matches.  Unfortunately, the public find it difficult to understand the purpose behind these trips and argue that criminals are being rewarded.  Intermediate programmes have not been totally successful because of lack of planning, support and resources.  A review found that schemes lacked consistent goals and procedures, and were poorly managed despite the enthusiasm of staff (Bottoms and McWilliams, 1990).



See the web pages ‘types and effectiveness of punishments’ and ‘offender treatment programmes’.


Providing appropriate treatment programmes for chronic and serious offenders falls into this category. However, harsher forms of tertiary prevention include selective incarceration keeping high-risk offenders out of circulation altogether, as has been suggested in relation to serious sex offenders and repeat offenders; or severe intervention for young offenders in the form of military-style training, such as ‘boot camps’ and ‘secure training centres’. While there may be instinctive public appeal in some of these options there are no guarantees of success, and serious reservations in terms of human rights. Not all high-risk offenders are correctly classified, and the removal of criminally active individuals from the street may well simply open up criminal opportunities for others eager to step into their shoes.

Government proposals to incarcerate indefinitely those deemed to be Dangerous Severe Personality Disordered (DSPN) may also give cause for concern. DPSN is not a clinical diagnosis but is intended to apply to those individuals who pose a risk to the public, even though they may not have committed a criminal offence. While the protection of the public is important there must also be serious consideration of civil liberties.


Hardening the target

This involves making the target physically difficult to break into, like the many changes in car design. Mayhew et al (1976) showed that the compulsory fitting of steering column locks to cars in West Ger­many in the 1960s led to a 60% drop in car theft.

Practically individual houses can be hardened against burglars with good locks, door chains, high gates, alarms and security lights.  Target hardening, however, can be expen­sive.


Removing the target

This is the removal of the potential targets of crime. For example, the payment of salaries by bank account reduces the number of security vans carry­ing large sums of cash.

During the 1960s and 1970s, British Gas was detoxified (that is, it is no longer lethal) and conse­quently the suicide rate has declined. So if people could not ‘gas themselves’, then there was no displacement to other methods of suicide.


Formal surveillance

This involves ‘authority figures’ in different ways: police patrols reduce general levels of crime, bus conductors reduce vandalism of the bus, shop assis­tants reduce shoplifting, door attendants in apart­ment blocks reduce burglary.

One strategy has been to focus police surveillance on particular situations in which crime is likely to occur. In this light, Ekblom (1980) discusses the operation of police truancy patrols. These patrols, working in conjunction with an education welfare officer (EWO), pick up children playing truant found on the streets. The justification for this is twofold: to prevent offending, given the link between truancy and delinquency (Belson 1975); and to return the child to school and allow the EWO to examine the reasons for the truancy. The effectiveness of the patrols has proved difficult to measure but, as Ekblom notes, the potential is certainly apparent for those children detected.

In the absence of formal police surveillance there are other possibilities for crime prevention. Sturman (1980) reports a study of vandalism on buses which has clear policy implications for reducing damage through formal surveillance. In comparing damage to different types of bus, Sturman found that vandalism was much greater on buses without a con­ductor; further, on all buses, with or without a conductor, damage was greatest at the rear on the upper deck. As Sturman suggests, bus com­panies must balance the costs of running defaced vehicles against the costs of repairs and of conductors to supervise passengers. Other studies have also pointed to the importance of an ‘authority figure’ in reducing crime. Walsh (1978) reported that the presence of a shop assistant reduces levels of shoplifting. Waller and Okihiro (1978) suggested that apart­ment blocks supervised by doormen are at less risk for burglary. R.V.G. Clarke (1982) notes that supervision of public telephones in public houses and launderettes is associated with low levels of vandalism; car parks with attendants have lower rates of damage to cars; while the presence of football club stewards reduces hooliganism on trains by supporters.

While the visible appearance of employees appears to be advantageous in reducing offending, the formality of the surveillance can be increased by the use of closed circuit television (CCTV); indeed, CCTV has become a feature in many shops, banks, and building societies. Burrows (1980) describes the use of CCTV to attempt to reduce robbery and theft on the London Underground. CCTV was installed at four Underground stations, positioned to give a view of the parts of the stations plat­forms, ticket halls, escalators which are high-risk areas for crime. In the stations with CCTV there were 252 reported offences in the twelve months prior to the installation of the CCTV: this fell to seventy-five offences in the twelve months after installation, which is 70 per cent fewer crimes. Over the same period, the comparable figure was a fall of just over 25 per cent in fifteen Underground stations geographically close to the four stations with CCTV. On 238 other stations over the whole of the Underground system the fall was 38 per cent. This pattern of figures clearly illustrates that while reported crime was falling throughout the system, the fall was greatest where the CCTV was in operation. However, the figures from the fifteen stations close to the CCTV-protected stations give some cause for concern. The significantly lower fall in offending in these stations suggests that displacement, par­ticularly of theft, was occurring. As Burrows notes, if displacement was occurring then ‘it may have nullified up to 85 per cent of the savings in theft offences apparently produced by CCTV installations’ (1980: 81).

The introduction of CCTV in town centres has reduced crime for example; there was a reduc­tion of 27% in Northampton after the introduction of 120 cameras in 1993 (Harrower, 1998).


Informal surveillance

Another effective form of situational crime prevention involves the community taking action to protect itself. The introduction of neighbourhood watch schemes has proved popular, and Mayhew, Elliott and Dowds (1989) describe such schemes as:


[having] made more of an impact, in terms of visibility if nothing else, than any other community crime prevention effort in Britain (p.51).


Members of these schemes are encouraged and supported by the police to keep an eye out for any suspicious activity in their community, and to inform the authorities if they do see anything untoward. This activity is very much seen as a partnership with the police, as evidenced by the advertising slogan ‘Crime together we’ll crack it’. The effectiveness of neighbourhood watch schemes, however, seems to lie more in the fostering of community spirit and reducing fear of crime rather than in reducing crime itself (Brantingham and Brantingham, 1990). There is a danger of an over-zealous commitment to community protection which can lead to vigilantism, particularly true in areas where communities have wanted to get rid of prostitution, or known sex offenders.

While the results of these types of intervention targeting specific communities can look quite impressive, there have been criticisms. They can be costly, and reducing criminal opportunities in one area may result in displacement, whereby criminals turn their attention to other types of crime or other areas. Pease (1994), however, feels this view may be a little pessimistic, especially since introducing crime prevention techniques in specific communities may also provide much-valued protection in those areas particularly beset by crime. The pattern of crime ensures that some areas are significantly more at risk than others, and it is there that attention needs to be focused in an attempt to discover what factors make an area more vulnerable.


Crime prevention programmes

Welsh and Farrington (1999) reviewed the studies on crime prevention programmes. Two examples proved to be particularly successful. First was the Seattle Community Crime Prevention Programme launched in 1973. Advice was given to household­ers to avoid burglary as well as public information about crime. It included property marking and neighbourhood watch schemes. There was a 61% drop in crime after one year, with no evidence of ter­ritorial displacement.

In the UK, there was the Kirkholt Burglary Pre­vention Project (Rochdale) between March 1987 and February 1990. This included the marking of property, ‘target hardening’ (the reduction of vul­nerable points of entrance to the house like win­dows without locks), neighbourhood watch, and ‘target removal’ (for example, the removal of coin gas and electricity meters). Over the three years of the project, reported burglaries fell from 526 per annum to 132 (a reduction of 75%).


Personal skills

Another approach to preventing crime is to change the potential victim. In other words, potential vic­tims are made more aware of the risks, and given strategies to deal with those risks. This approach has been applied in the UK in, for example, government booklets like ‘Practical Ways to Crack Crime’. Rec­ommendations include avoiding isolated bus stops at night, not stopping for hitch-hikers when driving alone, and not parking in a dimly-lit area.

Most of the advice is aimed at women and what they can do to protect themselves from crime. But Stanko (1990) argues from a feminist perspective that this type of advice raises fears rather than lessens them. In fact, by placing the burden of safety on women’s shoulders, it is ‘formulating a new ver­sion of blaming women for their victimization’ (p. 179). Furthermore, Stanko argues that this type of advice ignores the danger of violence from known men. Statistically, rape is more likely to be commit­ted by someone known to the victim rather than a complete stranger.

Personal skills can also be applied to children in relation to sexual abuse. These include encouraging children to be assertive when being touched in an unpleasant way and the concept of body ownership.


Environmental Crime Prevention


Situational crime prevention

Substituting natural gas for toxic gas in British homes was followed by a reduction of the number of suicides.  People have a rational choice to commit a crime or not; if the opportunity is not there then people generally should choose not to commit crime.  This explanation of criminal behaviour was called the rational choice model (Cornish and Clarke, 1986).  The compulsory fitting of steering column locks to all cars in West Germany during the 1960s lead to a 60 per cent reduction in car theft.  There was also a reduction in motorbike deaths following the introduction of compulsory crash helmets probably because they felt they would be noticeable riding a motorbike without a helmet (Mayhew et al. 1989).


Text Box: Situational crime prevention takes the following forms:

•	target hardening
e.g strengthening coin boxes in telephone kiosks, fitting steering column locks, installing anti-robbery screens in banks.
•	controlling access to crime targets
e.g installing entry phones, appointing caretakers, using fencing.
•	surveillance
e.g installing Closed Circuit Television (COT’!) and burglar alarms, providing better street lighting.
•	target removal
e.g installing removable car radios, setting up Women’s Refuges, accepting credit cards rather than cash, organising secure late night transport.
•	property identification
e.g marking property, vehicle licensing.
•	reducing temptation
e.g introducing gender-neutral telephone listings, off-street parking.
•	alerting consciences
e.g installing roadside cameras, posting notices alerting people to crime and the role they can play in reducing it.
•	controlling factors which undermine constraint
e.g banning alcohol consumption from public places, imposing age checks in premises selling alcohol.
•	making compliance easier
e.g introducing more litter bins.
(see Home Office Research Study 187, 1998)

The most common form of environmental intervention is target hardening.  If the targets of crime are difficult this alters the balance of costs and benefits that the offender ways up.  Examples include strengthening: boxes in public telephone kiosks or vending machines, making such machines usable by credit card only, making burglar alarms visible, security lighting, car immobilisation, lockable wheel nuts, etc.  Many of these techniques are used by the general public and are recommended by the police through the use of crime prevention leaflets.  Target removal includes the use of credit cards instead of cash, paying wages by direct transfer to bank accounts and the provision of late-night public transport.


The problem with these types of intervention is that crime can be shifted elsewhere for example once old cars became difficult to steal older cars were targeted.  Pease (1994) felt that this attitude was rather pessimistic and suggested that crime prevention in areas where a crime is prevalent would be seen as a welcome strategy in such areas.  However recently the case of Tony Martin, the Norfolk farmer who shot an intruder, highlights how criminals are turning to the less protected rural areas as a softer option to urban areas.


Another strategy would be to increase the risk of detection and this can be done by using cameras to monitor activity in specific area such as car parks in town centres.  A controlled study on the London Underground demonstrated a 70 per cent reduction in crimes of robbery and theft in the year after the CC TV was installed (Burrows, 1980).  The first British town to adopt a CC TV was Kings Lynn and this was funded partly by raising car-parking fees; the results were so successful that the general public were happy to pay the higher cost.  In 1991 there were 200 cases of car theft or damage in Kings Lynn but in 1992 after CC TV was installed the number of cases was down to 10.  In Northampton after the introduction of CC TV in 1993, using 120 cameras, police claimed that there has been a 27% decrease in crime; however house burglaries, not covered by CC TV, have increased.  Evidence of displacement?


Electronic tagging has been introduced in the UK and has helped to relieve prisoner overcrowding.  The offender wears a small transmitter, or tag, on his or her ankle and the monitoring centre picks up the signal.  In this way any violation of curfews will become apparent.  Nellis (1991) has noted several problems with electronic tagging:

1.      The reliability of the equipment

2.      The effectiveness of monitoring

3.      Infringement of civil liberties.


Neighbourhood watch schemes have proved popular although Mayhew et al. (1989) has described such schemes as:

“…(having) made more of an impact, in terms of visibility if nothing else, than any other community crime prevention effort in Britain.”


The effectiveness of neighbourhood watch schemes lies more in fostering a community spirit that reduces the fear of crime rather than reducing crime itself (Brantingham and Brantingham, 1990) and there is the danger of vigilantism.  This is particularly true were communities have wanted to get rid of prostitution.


Another way of reducing crime is to pay attention to architectural design and the environment.  Oscar Newman (1972) drawing on the work of Jane Jacobs (1961) suggests that designing built environments in such a way as to increase defensible space can reduce crime.  Newman suggests the following:


Defensible space is areas where residents feel some sort of ownership.  In many built up environments there are larger areas, which no one owns or controls.  These areas become magnets for criminal activity and the rest of the community therefore avoid these areas, which leads to their deterioration.  Newman advanced several recommendations to increase defensible space and thereby reduce crime. These were

      the use of boundary markers (real or symbolic barriers, such as pathways, small walls or fences) to reduce the likelihood of intrusion so that residents had control over the outside areas adjacent to their homes and over hallways and staircases

      planning which enabled surveillance of these semi-public areas, for example by the strategic placement of windows

      a design which encouraged positive social attitudes and a sense of community such that residents felt that the area was theirs and worth looking after; such housing should not be stigmatised and should not be designed to set it off from other types of housing such that it was easily identified as housing for the poor

      projects should be sited in low-crime areas.


Zones of territorial influence

Based on ethological theories of territoriality is the notion that if people perceive certain areas as their own space they will take a pride in and defend them. Additionally, a community, which has a shared sense of territoriality, is more likely to repel intruders. Architectural design can take account of this by establishing real or symbolic barriers, for example fencing, which should encourage territoriality.

Text Box: S. Wilson (1980) investigated vandalism and defensible space on a housing estate in an inner London borough. As might be predicted, vandalism occurred mainly on the ground floor, particularly in the areas where children played. Public areas such as entrances and underground garages were particularly heavily vandalized, with glazing the most common target. Wilson reported little overall variation in the levels of vandalism in housing blocks of different design. However, there were some variations of note: if entrances also acted as routes to other locations, thereby affording access to outsiders, vandalism was unduly high; alternatively, if the entrance design was such that it was exclusively for the use of residents then vandalism was lower. The child density in a given block was one of the best predictors of vandalism: when too great numbers of children were concentrated on a single estate, vandalism was at its peak. Wilson suggests that future design of housing estates might look to a number of points to increase surveillance and defensible space and so prevent vandalism:

Dwellings should overlook outside areas so that children at play can be seen; impersonal space which is not part of residents’ territory should be reduced; and entrances should be made discrete for residents’ use only. Reducing child densities through dispersal of families with children is problematic but may have a significant part to play in prevention. Where possible, families with children should be housed in buildings small enough not to need lifts or much semi-public access space..
More robust materials should be used in construction and repair. . . and where possible the use of glazing reduced. (S. Wilson 1980: 61—2)

Opportunities for surveillance

If buildings are designed in such a way as to allow residents to naturally observe areas, and to recognise outsiders, it is more probable that any offences will be noticed early and reported. Lack of surveillance was caused by several factors

      Buildings housed large numbers of people with many entrances, all of which gave access to the whole building. With such frequent pedes­trian traffic, residents could not be distinguished from outsiders who could, once in the building, move freely within it using the undefended staircases.  Wilson (1980) (see box) compared housing blocks in inner London and, in particular, entrance design. If the entrance acted as routes to other locations there was more vandalism than if the entrance was solely for the residents.


      The areas between the buildings were desolate open spaces with no territorial markings.


      The height of the buildings meant that there were ‘corridors’ of space between them that could not be overlooked from the windows.


Image and milieu

The design of buildings conveys a visual image and an identity, as does the setting. For instance, high-rise blocks all look the same whereas signs of indi­viduality tend to signal a private area. Similarly, a vandalised area can convey disorder and apathy.


Support for Newman’s theory was drawn from a study in which he looked at crime rates in 100 estates in New York and found that the greater the amount of defensible space, the lower the incidence of crime. Further support was offered by a comparison he made between two adjacent housing estates in New York City, that of Van Dyke

Van Dyke

Entrance to Van Dyke Houses

Central grounds of Van Dyke Houses

Brownsville Houses

 and Brownsville. Although Brownsville was eight years older than Van Dyke, it had a far lower crime rate. The design of this development meant that it had a considerable amount of defensible space. It consisted of six X-shaped buildings with some three-storey wings and entrances which, being used by a relatively small number of families, were easy to keep an eye on. Anyone approaching an entrance could be seen from any one of a large number of windows. Within the buildings the hallways and stairwells were easily monitored; children played on them and residents often left their doors ajar. This greater defensibility meant that there were strong bonds between the neighbours and a positive attitude towards the location and the police, as well as lower maintenance and crime rates. Van Dyke, in contrast, consisted of mostly large 14-storey buildings separated by open spaces that provided little or no defensible space. A similar project, that of Pruitt-Igoe in St Louis, which was designed to reduce crime and vandalism by incorporating open areas between buildings and vandal-proof fixtures, had become such a ruined wasteland that it had to be demolished.

Further support for Newman’s theory comes from a project in Ohio (reported in Bell et al, 1996) in which a neighbourhood called Five Oaks was split from one large residential area into several small ones by the blocking of certain streets and alleys and the use of speed bumps. Once these ‘mini-neighbourhoods’ were established, crime rates decreased by 26 per cent. However, the use of barriers to separate areas into smaller units and restrict access has been heavily criticised by some groups who see this as a means of restricting the access of poorer people to higher-income areas.

In Britain

Like America, Britain has had considerable problems with housing projects that were at one time considered to be ‘Utopian’ (Colenian, 1990). Jephcott & Robinson (1971) conducted a large-scale study of high-rise developments in Britain, interviewing nearly 1,000 residents of 168 multistorey blocks. There tended be an equal number of likes and dislikes, with likes being the interiors of the flats and features independent of design such as a good bus service and local amenities. The dislikes, on the other hand, were very much rooted in the design features inherent in high-rise blocks: the lifts, the loneliness and isolation, the entrances, vandalism, noise, poor maintenance and problems of refuse disposal.

Alice Coleman, a defender of Newman’s theory, has extended the work on defensible space and points to various design features (namely number of storeys, number of dwellings per entrance, number of dwellings per block, overhead walkways and spatial organisation) which she believes encourage all manner of antisocial behaviour. She maintains that some design features promote child crime by undermining the normal child­rearing practices that operate when residences are separate and do not incorporate shared space (Coleman, 1990). She compared housing estates with blocks of flats (apartments) in terms of litter, graffiti, vandal damage, children in care, urine pollution and faecal pollution. The 3,893 houses surveyed showed far less sign of social breakdown than purpose­-built flats. Litter was less common, graffiti extremely rare and excrement virtually unheard of. Like those before her, Coleman acknowledges that design is not the only factor in the promotion or prevention of social breakdown but thinks that criminology would do well not to ignore its probably considerable influence.


Rubenstein et al. (1980) have identified the major design approaches, which can be used to make space more defensible:


   improving external lighting;

   reducing opportunities for offender concealment;

   reducing unassigned open spaces;

   locating outdoor activities in sight of windows;

   increasing the number of designated walkways or paths;

   increasing pedestrian activity


Whilst Poyner (1991) adds the elements of fencing, house improvements, road closure and landscaping. All of these variables can contribute to a reduction in crime in terms of increased commitment to the community, increased likelihood of reporting and a higher risk of detection acting as a deterrent. Moreover, they can add to the quality of life experienced by residents.


In relation to house burglaries, Brown and Altman (1983) suggest the use of three features, which can help deter burglars in addition to the normal security measures — barriers, markers and traces. Barriers can be fences, hedges, walls and gates, whilst markers and traces include indications of occupation, for example house nameplates and garden furniture. Garages also help prevent burglaries because they make it hard for potential burglars to know whether anyone is home.


Entrance to Van Dyke Houses


Criticism of Newman’s ideas

Newman’s work has provided a commonsense and practical approach to crime prevention but it is not without its critics. The relationship between crime rates and defensible space is only correlational and therefore does not demonstrate cause and effect. Furthermore, as Repetto (1976) pointed out, we should be cautious about any theory heavily dependent on a single case study (the comparison between Brownsville and Van Dyke), especially since there were many other housing projects that could have been studied but were not. Other factors may also have been neglected in the analysis of crime rates. Mawby (1977), in a study based in Sheffield, found that while business premises based in the lower floors of residential high-rise buildings did indeed suffer from higher crime rates, this was because of greater reporting of crime rather than design features.  In a study of vandalism to telephone kiosks, Mayhew et al. (1980b) tested the ‘defensible space’  hypothesis that kiosks would be less of a target for vandalism if they were overlooked by people’s homes. Analysis of the frequency of vandalism showed, as predicted, that there was less van­dalism when the kiosk could be seen from a home. The effect was not however of high magnitude and, as R.V.G. Clarke notes, this ‘perhaps emphasises that defensible space is only part of the solution to crime in public housing settings’ (1982: 224). It is probably more than coincidence that, as in Wilson’s (1980) study, the major determinant of vandalism was the number of children living in the vicinity. With children in mind, Hope (1986) discusses the prevention of school burglary through design of the school buildings.


Taylor et al. (1980) suggest that Newman’s model is inadequate because it does not take sufficient account of other social factors such as the number of families on welfare which could have contributed to the different crime rates. They contend that sociocultural variables and social conditions, as well as design, determine the level of crime in any neigh­bourhood. Merry (1981) suggests that defensible space is necessary for crime prevention but it is not enough on its own. Even when the archi­tecture of a site lends itself to defensible spaces these may not be defended if there is a heterogeneous ethnic mix of residents which results in them not intervening in each other’s affairs.


Several studies indicate that there are certain social factors which may, at least in some cases, have a greater impact on crime than defensible space does. Wilson (1980) found that rates of vandalism are most closely corre­lated with the number of children living in an area: the more children, the more vandalism.


Newman himself now acknowledges that social variables, such as the percentage of families on welfare benefits, the percentage of single-parent families, the income level and the ratio of teenagers to adults, are more closely related to crime rates than are the design features of the immediate environment. However, even though defensible space may not be the best explanation for crime, it offers a very practical way of improving things.


Other research has shown that design changes by them­selves could not produce long-term reductions in crime. For example, design changes (like removing overhead walkways) on the Mozart Estate (London) cut burglaries, assaults and street robberies for five months, but then they rose again (quoted in Hall, 1995).



Most of the environmental approaches to crime prevention rely very much on the principle of informal social controls operating in cohesive communi­ties and the idea that environmental design can help to encourage a sense of cohesion. As Murray (1995) points out, however, improved environmental and building design can enhance the quality of life in a community, but if there are elements within that community, which continue to contribute to a pattern of disorder, or there are serious rifts within a community, then no amount of environmental intervention will produce a reduction in crime. Thus, those areas which could most benefit from crime prevention strategies are probably the least likely to show an improvement. Skogan (1990) feels that these areas are characterised by disorder and he distinguishes between physical disorder (abandoned or neglected buildings, broken streetlights, litter, etc.) and social disorder (public drinking, prostitution, sexual innuendo, etc.). Whilst this might sound rather judgemental in the sense that different people presumably perceive disorder differently, Skogan found that the area in which people lived was a more significant indicator of the number of disorders they reported than any other personal characteristic. Skogan went on to identify disorders as being most numerous in areas with low neigh­bourhood stability, poverty and a high ethnic minority population. He suggested that the significant consequences of disorder included less willing­ness by residents to help one another, higher crime rates and a desire to leave the neighbourhood.


Situational measures of crime prevention ignore the ‘person’ side of the person and environment interaction. Farrington, with particular emphasis on delinquency prevention through schools, is firmly of the opinion that ‘it is just as plausible to locate the causes of delinquency in the individual as in the environment’ (1985: 40), a point which is recognized by pro­ponents of environmental measures, who point to social factors alongside architectural design (0. Newman 1980). It must be emphasized that the theoretical stance in situational prevention, which has a great deal in com­mon with social learning theory (R.V.G. Clarke 1982), postulates an interaction between the person and the environment. To prevent or reduce crime both sides of the interaction must be considered. In a critique of situational crime control, Trasler works towards just such a position:

Policies of crime reduction really demand two strategies: deterring occasional or low-rate offenders from committing crimes . . . and identifying and incapacitating high-rate, persistent offenders. Situa­tional crime control offers effective measures for the first, but is likely to have little impact on the second group. (Trasler 1986: 24)


Another objection lies in the domain of civil liberties; recordings on videotape, photographs on credit cards, and so on. (Although, as Laycock (1986) describes, the marking of personal property can reduce the incidence of domestic burglary.) This is clearly an area in which moral and political decisions have to be made, as R.V.G. Clarke notes: ‘A certain level of crime may be the inevitable consequence of practices and institutions we cherish or find convenient and the costs of reducing crime below this level may be unacceptable’ (1982: 227). A similar argu­ment applies to objections concerning the aesthetics of an environment at worst bristling with guard dogs, barbed wire, and security cameras. Situational measures can have an effect but their implementation requires caution, planning, debate and consultation.

Another objection centres on the notion of ‘displacement’, that the criminal, frustrated by situational measures, will change either the time, place, method, or form of the crime (Reppetto 1976). Displace­ment is difficult to measure accurately, but there is little doubt that the studies of car steering locks (Mayhew et al. 1980a) and CCTV (Burrows 1980) did show evidence of this phenomenon. The nature and extent of the displacement will certainly be determined by a number of factors. The motivation of the offender may be important: the ‘opportunistic offender’ may be deterred by situational measures, whereas the ‘profes­sional criminal’ is forced to devise other strategies and more sophisticated techniques. The level of implementation is important, so that only partial prevention (as with steering locks) influences displacement. In addition, removing a target may lead to opportunities for other offences: the shift to credit cards may reduce robbery or pilfering but opens the door to computer crime, accounting frauds, and so on. In the final analysis it is a question of balance: if the gains in crime prevention, in both finan­cial and human terms, outweigh the effects of displacement, in a manner acceptable to the majority, then situational measures may be worthwhile. The great problem is the difficulty in measuring the exact effects of situa­tional interventions, so that in the final analysis their implementation becomes something of a social gamble.


Whilst situational crime prevention tries to change the environment, another approach is to target potential victims, changing their behaviour in such a way as to reduce the likelihood of their victimisation. Women and children have tended to be the main targets of such intervention because of their vulnerability in relation to interpersonal violence and campaigns providing information and advice have become very popular. Perhaps not surprisingly, there have been criticisms from feminists because of the assumption behind these campaigns that those who need to change are the likely victims rather than the perpetrators, but more informed campaigns also call on men to change their attitudes and behaviour. The most well known of these is the impressive zero tolerance campaign against domestic violence which was introduced in Edinburgh during 1992 by the council’s women’s unit. The campaign was unusual because, although hard-hitting, it was both educational and empathic, appealing to the whole community to do something about violence in the family. What the numerous posters and leaflets signalled was that the time to tolerate such abuse was over and that without demonising perpetrators or sensationalising victims, the power to end it lay in the community.


Rational choice theory


See rational choice theory notes, as this theory has some relevance to crime prevention.


Routine activity theory

Sharing some similarities with rational choice theory is routine activity theory (Cohen and Felson, 1979). According to this view, if a crime is to occur there must be a combination of three elements, i.e. a motivated offender, a suitable victim, and the absence of a capable guardian. The theory was originally developed as an explanation of personal-contact predatory crimes such as muggings, but was later extended to cover a wider range of offences.

According to this view, the reason why cities harbour so much crime is not to do with the architectural environment (see above) but rather the fact that cities bring together the three elements, i.e. motivated offend­ers, vulnerable targets and a lack of suitable guardians. The theory does not consider the motivations for offending, but rather simply assumes that there are within society a number of likely offenders. Similarly, ‘suit­able victims’ are not defined but might include anyone whom the offender may choose to attack. The reader may however wish to note that some writers (e.g. Hindelang et at, 1978) have suggested that a vic­tim’s lifestyle may contribute significantly to their chances of victimisation. For example, the people with whom an individual associ­ates, the number of times they choose to go out at night alone and the area in which they live may all contribute to a person’s chances of victim­isation.

The term ‘guardian’ is used in the traditional sense of anyone whose presence deters crime. However, Felson (1995) also talks about others who may influence an offender. These include ‘handlers’ who supervise potential offenders and ‘managers’ who monitor the places where crime might occur.

Felson (1993) suggests that it is elements of our modern Western lifestyle which contribute to the high level of crime in society. For exam­ple, most homes will possess fairly high value and desirable electrical goods such as video recorders and, increasingly, computers. Many homes are also left unoccupied for much of the day while owners are out at work. Furthermore, many of today’s young, mobile householders are unlikely to spend time building up good relations with their neighbours or potential guardians. These three elements in combination may make victimisation of certain targets more likely.

Some writers have recently sought to bring together the ideas incorpo­rated in rational choice theory and routine activity theory (for example Clarke and Felson, 1993). When the two views are combined this tends to lead to an increased focus on the relationship between certain places and crime. The fundamental question is whether certain locations actually generate crime, or whether they simply provide a suitable environment in which an almost inevitable crime just happens to occur (Sherman et al., 1989). Writers such as Newman (see above) have assumed that the former is the case, while others have argued that the environment does not so much cause crime as merely provide an appropriate setting for its occur­rence. Perhaps the truth lies somewhere in between these two positions with certain environments more likely to encourage or harbour criminal activity and certain environments less likely to do so.

Before moving on we should perhaps note the shift in emphasis which both rational choice theory and routine activity theory suggest. Unlike many early sociological theories of crime, both views see crime not so much as a pathological activity requiring explanation, but rather as an almost normal everyday activity in today’s society (Garland, 1996). As such, crime is seen as a risk which can be calculated and, hopefully, avoided or at least managed.

Although there have been some failures in initiatives which utilised sit­uational prevention methods, many more schemes have met with success (see Pease, 1997: 970 for a brief review). Many of these initiatives are of the type which might be expected. For example, a great deal of the British Home Office’s crime prevention budget is now taken up with the installation of CCTV in public places (Koch, 1996). While schemes utilis­ing CCTV have often met with success (Home, 1996) a number of writers have expressed concern about CCTV’s widespread adoption. One reservation is that the quality of images generated, especially by the less expensive type of equipment, means that the ‘evidence’ provided by the cameras is not particularly useful. Research currently being carried out by Vicki Bruce in Edinburgh suggests that mistaken identifications can occur even when apparently good CCTV images are obtained.

There are other concerns over the expertise and training of the staff who monitor the images provided by the cameras. Personnel may be untrained or inefficient, or may choose to use the equipment inappro­priately. In addition, some have argued that CCTV is an invasion of privacy, a concern heightened when some tapes are sold to television companies for the purposes of entertainment.

CCTV is an obvious example of primary prevention, but other mea­sures are perhaps less obvious. For example the introduction of technology which enabled telephones to display the calling number on a potential victim’s phone served to reduce the number of obscene phone calls (Clarke, 1991). Smith and Burrows (1986) showed how simple changes in a hospital’s management procedures could reduce the amount of fraud. Another study established that certain bus seats are more prone to vandalism than others, and this enabled the bus company to use vandal-resistant materials in the most vulnerable seats (Sturman, 1980). The latter study is an example of risk analysis, which allows levels of risk to be calculated and crime prevention initiatives to be targeted appropriately (Pease, 1997: 972).


Crime displacement

One potential problem in all the discussions of situational prevention is the thorny issue of displacement. The concern is that by introducing measures which make some targets less vulnerable, the determined crim­inal will not be deterred but will simply switch their attention to a slightly easier target. Thus, if all new BMWs are fitted with sophisticated alarm and immobilisation systems, the thief may simply target older cars with less sophisticated protection. The bank robber who is deterred by the installation of a state-of-the-art security and surveillance system in the local branch of Barclays may switch his attentions to the local building society instead. The mugger who is deterred by a large police presence may simply wait until the officers are occupied elsewhere before striking. A rapist may wait until his victim is out of range of the CCTV cameras before attacking. Perhaps even more worrying are cases of displacement in which an offender switches not so much the place or time of an attack, but rather the type of offence. A burglar who is deterred by the increas­ing number of domestic alarms may instead satisfy his need for money by switching to muggings.

As Heal and Laycock (1986) have noted, if offending is merely dis­placed, then situational crime prevention measures are worth little in the long term. The problem is that it is often difficult to prove or disprove whether displacement has occurred. If a crime prevention initiative in one small geographical area proves to be successful, it might be thought a simple matter to measure whether the reduction is mirrored by a cor­responding increase in adjoining areas. However, this is far too simplistic, and, one might argue, naïve. For example, the burglar may shift attention not to the adjoining area, but to one in a different police district or county. In this case, the increase may go unrecorded by any researcher trying to prove or disprove the existence of displacement. Similar problems may well be encountered if the offender switches from one type of crime to another. Thus a domestic burglary prevention scheme may lead to a reduction in the number of domestic burglaries, but to an increase in the number of commercial premises targeted. The police may fail to acknowledge this when reporting on the success of their original initiative.

Those who believe that displacement is likely or perhaps even inevitable are basing their beliefs on a number of assumptions. For example, it is assumed that most if not all offenders are highly motivated and will eventually overcome the initial attempts to prevent their crimi­nal behaviour. Such an assumption seems to suggest that criminals are addicted to their criminal lifestyle (Hodge et al., 1997) and will find some way of victimising society no matter what is done. This may however be a rather simplistic assumption.

Supporters of the displacement theory also seem to presume that offenders are almost infinitely adaptable and they can turn their hand to almost anything. Thus, if the acquisition of money is seen as the primary motivating factor, then it may be immaterial whether this is obtained through robbery, burglary or extortion. However, one must question whether this is actually true. Although some offenders do have convic­tions for a wide range of offences, the vast majority tend to concentrate on a rather restricted range of crimes.  In any case, if a criminal does switch the type of crime they commit then this change might be for the better or for worse. Barr and Pease (1992) make an important distinction between malign and benign displacements. The former involves situations in which an offender switches to a more seri­ous type of offence, while the latter refers to a scenario in which an offender changes to a less serious type of crime. Although both would technically be classified as displacement, one would obviously be seen as less worrying than the other.

A final assumption made by advocates of the displacement hypothesis is that there are an almost unlimited number of alternative (and equally attractive) targets available in society. This would appear to be a pre­sumption with little empirical foundation. Not all criminals continue to offend irrespective of what is tried, as the success of some crime preven­tion initiatives confirms. If it were possible to prevent totally the unauthorised taking of cars, what crime would we expect the joy rider to turn to as an ‘equivalent alternative’?

We can thus see that it is all but impossible to prove whether displace­ment will or will not occur in any given situation. A number of the assumptions made by advocates of displacement theory appear to be questionable. While theme is some evidence of displacement following the introduction of certain initiatives, displacement is by no means inevitable (for example Hesseling, 1994). It is surely unrealistic to assume that every crime prevented will be displaced. If half of the crime initially prevented by an intervention is displaced, then it could still be argued that the initiative was worthwhile.

Pease (1997) suggests that a fundamental flaw in the displacement argument is a failure to consider how crime patterns arise.  Rather than talking about crime displacement, Barr and Pease (1990) prefer to use the term crime deflection when discussing prevention ini­tiatives. Such a term simply acknowledges that a crime has been prevented, without an over-concern for its possible displacement. Barr and Pease argue that even if some crime is displaced, crime prevention strategists can address this fact, and the original initiative should not be simply abandoned. Barr and Pease (1990) suggest further that whereas crime displacement is always seen as a failure of the initia­tive, crime deflection is not necessarily so. As was noted above, if one particular crime is deflected, and instead a much less serious crime com­mitted, then this might reasonably be labelled a success.

If, as has been argued above, many primary prevention strategies have been successful, then the cynical reader may well be asking why there is still so much crime in society. We should perhaps note that in terms of policy, intentions and actions are not always equivalent (Crawford, 1998). For example, although the British government sees crime prevention as an important priority, Barclay (1995) notes that only three per cent of crime and criminal justice expenditure is spent on crime prevention. Pease (1997) notes that because over three-quarters of Home Office crime prevention expenditure is devoted to CCTV schemes, local authorities quickly learn that the way to get money is to apply for funding for local CCTV schemes. Thus, although a CCTV scheme may not necessarily be the best solution to a local problem, the authority is more likely to be successful in its bid for funding if it goes down this path. There is a fallacy in this approach, for as Pease notes:


...           good primary prevention requires a clear crime focus, an objective analysis of the presenting problem, and a choice of means from among those avail­able. What is actually happening is that, by its patterns of expenditure, the Home Office demonstrates the marginality of crime prevention; by the way it makes money available, it by-passes the necessary analysis, and once the money hits the maelstrom of local politics, expenditure is a matter more of need and expedience than crime control.

(1997: 980)


Pease goes on to argue that while situational crime prevention is a good way to prevent certain types of crime, it is less adaptable to certain other forms. Thus, such initiatives may well deter a proportion of property crimes, but will probably have less success in deterring crimes such as personal assaults. He also notes that situational crime prevention strategies may be more likely to be adopted by commercial organisations than by public authorities. A company whose financial survival is being threat­ened by a large amount of theft will tend to be highly motivated to take measures to reduce the level of loss. However, local authorities and national governments, while acknowledging the very real concerns of its citizens, may be less motivated to bring in situational crime prevention strategies.


The general public (bystander intervention)


This area of research attracted a great deal of activity after the murder of Kitty Genovese in New York in 1964, when those who heard her cries for help or watched the attack on her from the safety of their apartments failed to help, or even to telephone the police.

Unfortunately, for the practical purposes of crime prevention, these studies do not take us very far. In an earlier review (Feldman 1977, Chapter 4) the author drew the following conclusions from the (mainly simulated) research arrangements which had been used.

1.  In general the acquisition, performance and maintenance of help­ing behaviours, of which bystander intervention to frustrate a crime is an example, are responsive to the same social and cognitive variables as behaviours in general.

2.  Helping is increased by the manifest distress of the victim, the presence of a helping model, and the similarity of the victim to the bystander.

3.   Restraints against helping include potential loss to the helper (whether physical or financial), his lack of self-perceived competence,

and the assumption that others are both available and willing to help (the diffusion of responsibility hypothesis).

In addition, the author suggested that an onlooker is more likely to report a crime if the report is anonymous, police protection is assured, and he or she is not required to appear in court. Sheleff and Schichor (1980) support and amplify these speculations. Writing from legal experience, they argue that bystanders are less willing to intervene the more they are aware of what will happen if they are called as witnesses in court, namely the loss of their time and hence their money, the stress, possibly amounting to humiliation, of the cross-examination, and the possibility of subsequent violence by the accused. They suggest also that the court is a “total institution” in which the convenience of judges and lawyers is the first consideration, that of witnesses the last, and cite evidence that witnesses frequently find the experience annoy­ing and frustrating, and the tasks of recall and presentation very difficult.

They conclude that such experiences could easily become well known and so deter potential witnesses from getting involved. Indeed, a simulation study (Fedler and Pryor 1984) found that the most frequent reason for nonreporting was a desire to avoid subsequent “hassles” (involvement with the police and the courts).


Who is reported? Fedler and Pryor (1984) found that a small, well dressed, person observed shoplifting is more likely to be reported, if witnessed by a bystander, than a large poorly dressed one (the former “doesn’t need it;” the latter “could do me harm”). Mawby (1985) suggests that victims seen as more vulnerable, and individuals as opposed to corporations (corner shops rather than supermarkets), are more likely to be helped, as are neighbours rather than strangers. Other things being equal (such as the potential costs of reporting), the more serious the crime observed the more likely it is to be reported, but when the meaning of an event is unclear, onlookers will tend to categorize it as less serious than it is in fact (Himmelfarb 1981).


Who reports? A simulation study of shoplifting by Gelfand, Hartmann, Walder, and Page (1973) suggested that males, the middle-aged, and those raised in rural settings were more likely to report a witnessed crime. Since then there have many such studies, although most have been in noncriminological contexts. A meta-analysis by Steblay (1987) of 65 studies found a significant amount of support for the hypothesis

that “country people are more helpful than city people.” The over 60’s were much more likely to “report” in the study by Fedler and Pryor (1984) than any other age group. Possible explanations for this include the elderly being more public-spirited and having more time on their hands.

To return from simulation studies to the real world of offences, there is good evidence that many crimes are witnessed, and that witness evidence is related to the clearance rate. Both findings help to explain the persisting interest in this area of research. A national survey in The Netherlands by Steinmetz (1985) found that in 1982 no fewer than 36 per cent of those interviewed had witnessed at least one of 13 types of crime, including threatening or violent behaviour (14 per cent) and shoplifting (12 per cent). Some kind of intervention (usually warning the police) was claimed by just over one-quarter of those who had witnessed an offence. Intervention was related to the personal costs of the particular involvement (warning the police was much easier and hence more frequent than a direct confrontation with the offender).

Mawby (1985) reviewed witness involvement in recorded crime in the Northern English town of Sheffield in 1971. He found that 82 per cent of reported crimes were reported by the victim as against only 7 per cent by witnesses. However, whereas only 33 per cent of the former were cleared, this was true for 57 per cent of the latter, many of which were serious.

It seems that if witnesses to crime are willing and able to observe carefully, to report immediately to the police, and to give evidence in court, the clearance rate may well rise, if only moderately. However, increasing the probability of these actions would require a massive effort in public persuasion; such campaigns in the area of crime are in their infancy, at best.


Who, if anyone, goes beyond witnessing with a view to reporting, and actually intervenes in an attempt to prevent the crime, perhaps even attempts to apprehend the perpetrator? Houston (1980) interviewed 32 people who had intervened directly in dangerous types of crime (muggings, armed robbery, bank hold-ups) in the past 10 years and compared them with ‘noninterveners.” The interveners were more likely to have been a victim of crime or to know victims personally. They were also taller, heavier, and better trained in first-aid and life-saving (though there was no difference in physical combat skills) and were more likely to describe themselves as physically strong, aggressive, emotional and principled. This fits in with a report by London (1970) of those who risked their own lives to help others escape death at the hands of the Nazis in World War II Europe. Moriarty (1975) reported that prior preparation was also relevant those who had been asked to do so previously were more likely to intervene to prevent a theft.




By citizens

Lavrakas and Lewis (1980) analysed data from four American com­munities and concluded that the self-protection methods used by citizens fall in two clusters: one concerned with avoidance, the second with access control (often termed target-hardening, see below). Avoidance behaviours involve people restricting their actions in certain contexts which they would perform in safer ones. Examples include: not going out at night, going out by day only when accompanied; at all times driving rather than walking; avoiding certain areas completely even when in a vehicle; when walking keeping clear of certain types of stranger, and carrying limited amounts of cash on one’s person. Access control included locking doors and windows, and installing extra devices such as burglar alarms.

Those surveyed did not mention surveillance, whether formal, of the Neighbourhood Watch type, or of the informal sort allowed by certain types of environmental arrangements. But, there was much active self-protection. In one of the communities surveyed, Portland, OR, between one-quarter and one-half of those interviewed carried a gun or some other weapon for protection. Riger, Gordon, and Labailly (1982) found that American women, surveyed separately, also used avoidance tactics and access control. The use of the former was predicted by level of fear, perceived physical competence, ethnicity and education. Neither report provides information on the effective­ness of these two clusters of techniques in protecting against victimiza­tion. (To do so would require a longitudinal combination of these studies and a survey of victimization.)


By employers

A number of reports make suggestions, some tested empirically, as to how storeowners, and employers in general, might reduce theft, whether by employees or by shoppers. McNees et al. (1976) measured shoplifting from a department store by checking 25 key items each day. Next, anti-shoplifting signs were introduced (“this store is protected by.. .“); the result was a considerable decrease in stealing. McClaughlin (1976) takes a somewhat different tack, suggesting that stores should reward shoppers and employees for reporting shoplifting, and that prices should be raised or lowered according to the level of shoplifting, thereby making reporting by shoppers more likely.

In a large scale self-report study of employees in a different section of industry, Hollinger and Clark (1983) found that males admitted higher levels of theft from work than females, and that certainty of detection and severity of punishment were both associated with reduced theft, but much less so for younger males than for other groups. This implies that an employer able to choose from a large pool of applicants should select older married employees and, if employing younger people, should prefer females to males (other factors being equal, such as skill and experience). Such a personnel policy combined with high levels of apparent certainty of detection (highly visible closed-circuit TV systems) and of known severity of punishment (“all thieves will be prosecuted”) should sharply reduce employee theft, according to Hollinger and Clark (1983).

Brown and Pardue (1985) describe the introduction of a personnel selection inventory as a component of the hiring procedures of a chain of South Eastern retail drugstores. Over a three-year period, there was a significant reduction in losses, amounting to more than $1 million. But, there are also potential problems in such approaches, for example, illegal discrimination in hiring, and a potentially unpleasant work atmosphere leading to increased absenteeism and staff-turnover.

An insurance-based approach to employers, to improve self-protection against robbery and burglary, is urged by Litton (1982), an executive of an insurance company. He argues that the cost of insurance premiums can be a powerful means of persuading employers to take effective crime prevention measures, such as the use of high quality safes and intruder alarms, and the employment of professional carriers for valuables, as well as adequate initial vetting of new staff. Clients could also be encouraged by their insurers to insist that their architects incorporate crime-prevention devices at the design and planning stage of new buildings. Insurance discounts could be given for all such measures. Both parties would then benefit (fewer losses and fewer claims).


Zero tolerance

Risk assessment

Government site dedicated to the reduction of crime


Peter B. Ainsworth, 2000, Psychology and Crime, Longman, 0-582-41424-5

Kevin Brewer, 2000, Psychology and Crime, Heinemann, 0-435-80653-X

Diana Dwyer, 2001, Angles on Criminal Psychology, nelson thornes, 0-7487-5977-8

Philip Feldman, 1993, The Psychology of Crime, Cambridge, 0-521-33732-1

Julie Harrower, 1998, Applying Psychology to Crime, Hodder & Stoughton, ISBN 0-340-70556-6.

Julie Harrower, 2001, Psychology in Practice – Crime, Hodder & Stoughton, 0-340-84497-3

Clive R. Hollin, 1989, Psychology and Crime, Routledge, 0-415-01807-2




Battle begins in junkies' very personal war zone