Social Prevention and Community Safety

The three-level framework adapted from public health medicine provides a conceptual map for organising crime prevention activity. Its value is not just definitional. It forces practitioners to ask, before commissioning any programme, which population is being targeted and at what stage of the causal chain the intervention is aimed.

Primary prevention addresses the general population without identifying any individual or group as at risk. Public awareness campaigns about domestic violence, universal school programmes in conflict resolution, and urban design principles that create naturally defensible public spaces all operate at this level. The reach is wide; the effect on any one person is usually modest.

Secondary prevention focuses on individuals or communities where risk factors are already elevated. A youth worker assigned to a school in a disadvantaged area, a targeted home-visiting programme for families under stress, or a police diversion scheme for first-time young offenders before prosecution all operate here. The intervention is more intensive because the need is more specific.

Tertiary prevention addresses people who have already offended, aiming to reduce reoffending. Rehabilitative programmes within prison, post-release mentoring, drug treatment courts, and restorative justice conferencing all belong to this category. Tertiary prevention overlaps substantially with criminal justice, and the distinction between treatment and punishment is often contested.

Developmental crime prevention draws on longitudinal studies of offending to identify the risk and protective factors present at different life stages. Key risk factors identified in studies such as the Cambridge Study in Delinquent Development (Farrington, UK) and the Dunedin Multidisciplinary Health and Development Study (New Zealand) include: low birth weight and prenatal exposure to toxins, harsh or inconsistent parenting, low school attainment, association with delinquent peers, and residence in disadvantaged neighbourhoods. Developmental programmes seek to reduce these risks or strengthen the protective factors that buffer against them.

The Perry Preschool Project, run in Ypsilanti, Michigan from 1962 to 1967, provided high-quality preschool education and weekly home visits to 58 African-American children from low-income families at risk of school failure. A matched control group received no programme. Follow-up studies through age 40 found that programme participants had significantly lower arrest rates, higher earnings, and greater educational attainment than controls. A cost-benefit analysis published by the RAND Corporation estimated returns of approximately seven dollars for every dollar invested, primarily through reductions in crime costs.

The Nurse-Family Partnership (NFP), developed by David Olds in the United States, provides home visits by trained nurses to low-income, first-time mothers from pregnancy through the child's second year. Randomised controlled trials in Elmira (New York), Memphis (Tennessee), and Denver (Colorado) showed significant reductions in child maltreatment, maternal welfare dependence, and, in follow-up studies, criminal records among the children in their teens and twenties. The NFP model has been replicated in the UK as the Family Nurse Partnership and piloted in Australia and the Netherlands.

School-based programmes target the adolescent transition, when delinquency rates peak. Anti-bullying programmes, social skills training, mentoring, and school-based restorative practices have evidence bases of varying strength. The strongest evidence, from systematic reviews by the Campbell Collaboration, supports structured social skills programmes that teach emotion regulation, problem-solving, and resistance to peer pressure. Mentoring programmes show positive effects on school attachment but weaker effects on offending directly.

The multi-agency community safety partnership model rests on three premises: crime has determinants that cross institutional boundaries, no single agency has either the information or the authority to address all of them, and joint governance produces better outcomes than parallel action by separate agencies. These premises are widely accepted in the policy literature, though the evidence for the third is more mixed than the first two.

In England and Wales, the Crime and Disorder Act 1998 created statutory Community Safety Partnerships (then called Crime and Disorder Reduction Partnerships) in every local authority area. Each partnership is required to conduct a regular strategic assessment of crime and disorder in its area, produce a three-year strategy, and report on outcomes. Responsible authorities include the police, the local council, the fire and rescue service, and the local health board. Co-opted partners include schools, housing associations, and voluntary organisations. The Home Office issues guidance and statutory performance indicators, but partnerships have discretion in how they prioritise and deliver.

Comparable structures exist elsewhere. In Australia, state governments operate community safety plans that require police, local councils, health departments, and community organisations to coordinate. South Africa's Community Policing Forums, established under the South African Police Service Act 1995, bring community members into formal oversight of local policing. In India, the Integrated Child Development Services scheme, though primarily a health and nutrition programme, incorporates protection from violence and crime as an outcome, and local self-government bodies under the Panchayati Raj system are formally charged with public safety coordination in rural areas. These varied forms share the multi-agency logic even when the institutional architecture differs.

Crime is not evenly distributed across cities. Research from the United States, United Kingdom, Australia, and India consistently finds that a small proportion of street segments or postal areas accounts for a disproportionate share of recorded crime. This concentration has two implications. First, place-based prevention can be highly efficient: intensive work in a handful of hot spots can produce large aggregate reductions. Second, the concentration of crime in specific areas is not accidental but reflects cumulative disadvantage, poor housing stock, lack of economic opportunity, and degraded public space.

Neighbourhood regeneration programmes address the physical and social fabric of high-crime areas. The Housing Action Trusts in the UK during the 1990s, the Hope VI HOPE VI programme in the US that replaced decayed public housing with mixed-income developments, and the Neighbourhood Renewal Fund in England all combined physical renovation with employment and community capacity building. Evaluations of these programmes show crime reductions, but disentangling the effect of regeneration from the displacement of vulnerable residents, a form of spatial displacement of the problem rather than its solution, is methodologically difficult.

Crime Prevention Through Environmental Design (CPTED), derived from Jane Jacobs's observations about urban vitality and Oscar Newman's concept of defensible space, provides design principles for reducing crime through the physical environment. Natural surveillance (ensuring that spaces are overlooked by residents and passers-by), territorial reinforcement (clear delineation between public and private space), and access management (controlling movement through an area) are the three core principles. CPTED has been adopted in building regulations and urban design guidance in the UK (Secured by Design), Australia (SafeGrowth), and many other jurisdictions.

The shift toward evidence-based crime prevention is one of the defining features of the field since the 1990s. Lawrence Sherman and colleagues' 1997 report to the US Congress, 'Preventing Crime: What Works, What Doesn't, What's Promising', applied a five-point scientific methods scale to hundreds of evaluations and identified the conditions under which prevention programmes succeed or fail. The Maryland Report, as it became known, established the evidence hierarchy that the UK What Works Centre for Crime Reduction, the Campbell Collaboration Crime and Justice Group, and equivalent bodies in Australia and Canada have subsequently built on.

What the evidence shows is roughly this. Early childhood programmes with strong randomised trial designs (Perry, NFP, Chicago Child-Parent Centers) show genuine long-term effects on offending, though effects are concentrated among the highest-risk participants and fade without sustained support. School-based social skills programmes have consistent modest positive effects. Mentoring, hot-spots policing (which straddles social and situational prevention), and drug courts have reasonable evidence bases. Boot camps, Scared Straight programmes (taking at-risk youth to meet prisoners), and untargeted information campaigns have either null or negative effects in rigorous trials.

Systematic reviews repeatedly find that effect sizes are larger in studies with weaker designs, meaning that poorly controlled evaluations tend to overestimate how well programmes work. This is not a reason for cynicism about prevention; it is a reason to fund rigorous evaluation alongside programme delivery. The UK Home Office and the US National Institute of Justice now require evaluation plans as a condition of funding for major prevention programmes.

Cross-national transfer of programmes raises its own questions. A programme designed for low-income African-American families in 1960s Michigan does not automatically translate to a Tamil Nadu village, a Stockholm suburb, or a Cape Town township. Cultural adaptation, local institutional capacity, and the specific risk-factor profile of the target population all affect whether a programme's effects replicate. The evidence base for adapted versions of proven programmes is thinner than for the original settings.

Social crime prevention attracts criticism from several directions simultaneously, which reflects the breadth of interests it engages.

From the left, critics argue that social prevention individualises what are structural problems. Poverty, labour market exclusion, racial discrimination, and housing insecurity generate crime not because individuals lack social skills or adequate supervision, but because the social structure creates unequal opportunities and unequal exposure to criminalisation. Programmes that target parenting or peer groups leave the structure intact and may stigmatise the communities they serve. This critique is associated with critical criminology and with writers such as Pat Carlen, who argues that social crime prevention often functions as governance of the poor rather than improvement of their conditions.

From a conservative direction, concerns focus on cost, efficacy, and the displacement of personal responsibility onto programmes. Long-term developmental programmes require sustained public investment across electoral cycles, and their benefits are diffuse and delayed. Politicians face incentives to fund visible short-term enforcement rather than prevention whose payoff is measured in reduced arrest rates fifteen years later.

A third set of tensions arises within the community safety partnership model itself. Research on partnerships in England, Wales, Australia, and South Africa consistently finds that police partners tend to dominate agendas, that health and social care priorities are marginalised, and that accountability for outcomes is diffuse. When a partnership fails to reduce domestic violence or youth violence in an area, it is unclear which partner bears responsibility. This accountability gap is a structural feature of multi-agency governance, not a management failure in any particular partnership.

There is also a tension between community voice and professional planning. Programmes designed with communities tend to have better local legitimacy, but community preferences are not always well aligned with evidence of what works. A community may demand more visible policing when the evidence favours a family support programme. Managing this tension without ignoring either evidence or legitimacy is one of the persistent practical challenges of the field. Links between criminological research and community safety practice have been studied under the label of 'evidence-based policing', which extends naturally to prevention. See also Social Bond and Self-Control Theories for the theoretical grounding of many developmental prevention approaches.