Preventing Crime & Creating Safer Communities
Home | Distance learning | Information & news | Discussions | Join the network | Contact | Useful links

Crime Prevention Information & News

What research can tell us about how to prevent crime (and what it cannot tell us)

Henry Shaftoe (revised 23/01/01).

An enormous amount of academic research has been undertaken into the "causes" of crime and criminality. Criminologists in Europe and North America have, for many years, managed to carve entire careers out of ever more esoteric research into both the criminal mind and the social context of crime. The trouble is, that very little of this research can be used directly by practitioners on the "front line" of crime control. There are a number of reasons for this:

• Most criminologists write for the benefit of other criminologists, rather than for practitioners. There is a pressure on academics to "get published" and the specialist journals are attracted to erudite and esoteric submissions;
• Many research reports conclude with recommendations for further research, rather than with practical proposals for modified practice;
• Most significantly, social research into "slippery" subjects such as crime are notoriously difficult to isolate from other influences, so that researchers can hardly ever say that a particular change in outcome (say a reduction in a crime level) was conclusively the result of a particular input (say more police on the beat).

Specifically in relation to crime prevention, there are very few research studies that have had a major influence on practice. This is partly to do with the problem of isolating the "cause" and "effect", described above, and partly because durable improvements to community safety may only become clear after many years. However, a handful of research evaluations have become highly influential in terms of preventative practice and three of these are briefly described below:

Oscar Newman and "Defensible Space"
Oscar Newman is an American architect who, in the late 1960s, was able to carry out an experiment in improving the design of a social housing estate whilst leaving a similar adjacent estate untouched. His modifications aimed at giving people more control and surveillance over the communal spaces around their homes. Crime fell in the improved area, compared to the neighbouring estate. This improvement has been maintained to the present day, according to Newman, although he also stresses the importance of what he calls "community of interest" (ie: community development and cohesion).

The High/Scope pre-school programme
This is another project started in the 60s in the USA. A control group of children were offered high quality pre-school education and activities which were designed not only to be of educational value but also to give the children control over what they chose to do. 25 years later those who were part of the project have benefitted in numerous ways, compared to their "untreated" peers: they are more literate, better employed, in more stable relationships, and much less likely to have been involved in delinquency and crime.

The Cambridge Study in Delinquent Development
This longitudinal research study has followed the fortunes of a large cohort of randomly selected boys who were attending primary school in 1961/2. The study has found that those who have ended up involved in crime, emerged from circumstances characterised by: low income, poor housing, living in run-down areas, parental conflict, family breakdown, poor parental supervision, harsh and erratic discipline, high degrees of impulsiveness and hyperactivity and low school attainment.

Other pieces of research, less famous than the above, have nevertheless some useful implications for policy and practice. A selection of these are listed below:

"Burglars on Burglary" was a small study carried out by Bennett and Wright in 1984. They interviewed a number of convicted burglars, asking them how they chose a target and what would put them off committing a crime. They generally chose premises that looked unoccupied and were particularly put off by dogs, but not necessarily by burglar alarms (which indicated there might be something worth stealing!). They either didn't think about the consequences of their possible apprehension (some were too drunk anyway!) or they deliberately put such thoughts out of their minds, believed they wouldn't get caught, or thought they would be caught and punished eventually so might as well go on offending in the meantime.

"Repeat Victimisation" based on a study of burglary patterns on an estate in Rochdale, but already implicit in the practice of giving security advice to burglary victims, was the observation that people who had been burgled once were, statistically, much more likely to get burgled again. This suggests that extra precautions and security measures should be targetted on properties that have already been burgled.

"The Kansas City Police Patrolling Experiment". In the 1970s, as part of a controlled experiment, the Kansas City Police divided the city into three comparable areas. In one area, patrols were doubled and sometimes tripled. In another, patrols were eliminated altogether, although the police continued to respond to 999 calls. In the third area no changes were made. After a year, the police checked to see how the crime rates differed in the three areas. They had stayed the same in all three areas and the public had barely noticed the difference. Adding or removing patrols had no effect on crime.

"Single Shot Solutions". A number of evaluations of crime prevention activities have indicated that, although single measures aimed at crime prevention (such as lock fitting programmes, lighting improvements or entryphones) can reduce crime occurances, they have a much greater durable impact if they are combined with other measures (such as community development or area management changes).

"Broken Windows".  On a public car park in the USA, a car was deliberately abandoned but kept under invisible surveillance. For many days the car lay untouched, so the researchers removed the number plates and left the bonnet propped open. This triggered an almost immediate and escalating response from the "community". The following day thieves removed the battery and then in rapid sequence over the next few days, all removable parts were stolen. Finally, local youths descended on the car to smash the windows and dent the body work. This modest experiment in "auto abuse" led to the concept of neighbourhood deterioration known as "broken windows" and police preventative approaches aimed at nipping this deterioration in the bud have recently been described as "zero tolerance" strategies.

"Social Efficacy and Community Cohesion". A number of studies in the 1990s discovered that neighbourhoods with high levels of "social capital" (defined as trust, mutual aid, cohesiveness and acquaintance support networks) had lower levels of crime and violence. This suggests that design, management and development activities that encourage "bonding" and stability within neighbourhoods are likely to improve community safety. So, for example, community development, local co-operatives, neighbourhood management and the urban villages movement, although initiated for different reasons, are also likely to reduce crime and insecurity.

Survey Research
Apart from academic research into crime and criminality, there is another type of research which is very valuable for crime prevention and community safety practice - the survey. This will usually be based on a particular locality or crime problem and simply asks a representative sample of the local or target population about their experience of crime, their fears and perceptions about safety and their thoughts about improving the situation. Not only can such surveys provide an excellent reference point for preventative action, but they can act as the "before" stage for the subsequent evaluation of the effectiveness of any practical work undertaken. For this "longitudinal research study" to be significant and helpful for future practice, the implementation or work phase must be carefully monitored to check whether what has been proposed is actually undertaken as specified.