Crime Prevention

David P Farrington. 21st Century Criminology: A Reference Handbook. Editor: J Mitchell Miller. 2009. Sage Publications.

As described by Michael Tonry and David Farrington (1995), criminal justice prevention refers to traditional deterrent, incapacitative, and rehabilitative strategies operated by law enforcement and criminal justice system agencies. Community prevention refers to interventions designed to change the social conditions and institutions (e.g., families, peers, social norms, clubs, organizations) that influence offending in residential communities. These interventions target community risk factors and social conditions such as cohesiveness or disorganization. Situational prevention refers to interventions designed to prevent the occurrence of crimes by reducing opportunities and increasing the risk and difficulty of offending. Developmental crime prevention refers to interventions designed to prevent the development of criminal potential in individuals, especially those targeting risk and protective factors discovered in studies of human development. The focus in this article is on developmental or riskfocused prevention.

The main aim is to summarize briefly some of the most effective programs for preventing delinquency and antisocial behavior whose effectiveness has been demonstrated in high-quality evaluation research. The focus is especially on programs evaluated in randomized experiments with reasonably large samples, since the effect of any intervention on delinquency can be demonstrated most convincingly in these studies (see Farrington & Welsh, 2006).

Risk-Focused Prevention

The basic idea of developmental or risk-focused prevention is very simple: Identify the key risk factors for offending and implement prevention techniques designed to counteract them. There is often a related attempt to identify key protective factors against offending and to implement prevention techniques designed to enhance or strengthen them. Longitudinal surveys are used to advance knowledge about risk and protective factors, and experimental and quasi-experimental methods are used to evaluate the impact of prevention and intervention programs.

Risk-focused prevention was imported into criminology from medicine and public health by pioneers such as David Hawkins and Richard Catalano (1992). This approach has been used successfully for many years to tackle illnesses such as cancer and heart disease. For example, the identified risk factors for heart disease include smoking, a fatty diet, and lack of exercise. These can be tackled by encouraging people to stop smoking; to have a more healthy, lowfat diet; and to exercise more.

Risk-focused prevention links explanation and prevention; links fundamental and applied research; and links scholars, policymakers, and practitioners. The book Saving Children From a Life of Crime: Early Risk Factors and Effective Interventions, by Farrington and Brandon Welsh (2007), contains a detailed exposition of this approach Importantly, risk-focused prevention is easy to understand and to communicate, and it is readily accepted by policymakers, practitioners, and the general public. Both risk factors and interventions are based on empirical research rather than on theories. This approach avoids difficult theoretical questions about which risk factors have causal effects.

What is a Risk Factor?

By definition, a risk factor predicts an increased probability of later offending. For example, children who experience poor parental supervision have an increased risk of committing criminal acts later on. In the Cambridge Study in Delinquent Development, which is a prospective longitudinal survey of 400 London males from age 8 to age 50, a total of 61% of those experiencing poor parental supervision at age 8 were convicted by age 50, compared with 36% of the remainder—a significant difference. Since risk factors are defined by their ability to predict later offending, it follows that longitudinal studies are needed to establish them.

The most important risk factors for delinquency are well-known (Farrington, 2007). They include individual factors such as high impulsiveness and low intelligence, family factors such as poor parental supervision and harsh or erratic parental discipline, peer factors such as hanging around with delinquent friends, school factors such as attending a high-delinquency-rate school, socioeconomic factors such as low income and poor housing, and neighborhood or community factors such as living in a highcrime neighborhood. The focus is on risk factors that can be changed by interventions. There is also a focus on protective factors that predict a low probability of offending, but less is known about them.

Risk factors tend to be similar for many different outcomes, including delinquency, violence, drug use, school failure, and unemployment. This is good news, because a program that is successful in reducing one of these outcomes is likely to be successful in reducing the others as well. This chapter reviews programs that target family, school, peer, and community risk factors.

Family-Based Prevention

The behavioral parent management training developed by Gerald Patterson (1982) in Oregon is one of the most influential family-based prevention approaches. His careful observations of parent-child interaction showed that parents of antisocial children were deficient in their methods of child rearing. These parents failed to tell their children how they were expected to behave, failed to use punishment consistently or monitor their behavior to ensure that it was desirable, and failed to enforce rules promptly and unambiguously with appropriate rewards and penalties. The parents of antisocial children used more punishment (such as scolding, shouting, or threatening), but failed to use it consistently or make it contingent on the child’s behavior.

Patterson’s (1982) method involved linking antecedents, behaviors, and consequences. He attempted to train parents in effective child-rearing methods, namely, noticing what a child is doing, monitoring the child’s behavior over long periods, clearly stating house rules, making rewards and punishments consistent and contingent on the child’s behavior, and negotiating disagreements so that conflicts and crises did not escalate. His treatment was shown to be effective in reducing child stealing and antisocial behavior over short periods in small-scale studies. However, the treatment worked best with children aged 3 to 10 and less well with adolescents. Also, there were problems achieving cooperation from the families experiencing the worst problems. In particular, single mothers on welfare were experiencing so many different stresses that they found it difficult to use consistent and contingent child-rearing methods.

The most important types of family-based programs that have been evaluated will now be reviewed. These are home visiting programs (and especially the work of Olds and rthur Reynolds), parent training programs (especially those used by Carolyn Webster-Stratton, Stephen Scott, and Matthew Sanders), home or community programs with older children (especially those implemented by James Alexander and Patricia Chamberlain), and Multisystemic Therapy or MST (used by Scott Henggeler and Alison Cunningham).

Home Visiting Programs

In the most famous intensive home visiting program, Olds and his colleagues (Olds, Hill, & Rumsey, 1998) in Elmira (New York) randomly allocated 400 mothers to receive home visits from nurses during pregnancy, to receive visits both during pregnancy and during the first 2 years of the child’s life, or to be part of a control group that received no visits. Each visit lasted about 1.25 hours, and the mothers were visited on average every 2 weeks. The home visitors gave advice about prenatal and postnatal care of the child, about infant development, and about the importance of proper nutrition and avoiding smoking and drinking during pregnancy. Thus, this was a general parent education program.

The results of this experiment showed that the postnatal home visits caused a decrease in recorded child physical abuse and neglect during the first 2 years of life, especially by poor unmarried teenage mothers; 4% of visited versus 19% of nonvisited mothers of this type were guilty of child abuse or neglect. This last result is important because children who are physically abused or neglected tend to become violent offenders later in life. In a 15-year follow-up, the main focus was on lower-class unmarried mothers. Among these mothers, those who received prenatal and postnatal home visits had fewer arrests than those who received prenatal visits or no visits. Also, children of these mothers who received prenatal and/or postnatal home visits had less than half as many arrests as children of mothers who received no visits. According to Steve Aos and his colleagues (Aos, Phipps, Barnoski, & Lieb, 2001) from the Washington State Institute for Public Policy, $3 were saved for every $1 expended on high-risk mothers in this program.

Like the Perry project, described later in this chapter, the Child Parent Center (CPC) in Chicago provided disadvantaged children with a high-quality, active learning preschool supplemented with family support. However, unlike Perry, CPC continued to provide the children with the educational enrichment component into elementary school, up to age 9. Focusing on the effect of the preschool intervention, Reynolds and his colleagues (Reynolds, Temple, Robertson, & Mann, 2001) found that compared to a control group, those who received the program were less likely to be arrested for either nonviolent or violent offenses by the time they were 18. The CPC program also produced other benefits for those in the experimental compared to the control group, such as a high rate of high school completion.

Parent Management Training

One of the most famous parent training programs was developed by Carolyn Webster-Stratton (2000) in Seattle. She evaluated its success by randomly assigning 426 4year-old children (most with single mothers on welfare) either to an experimental group that received parent training or to a control group that did not. The experimental mothers met in groups every week for 8 or 9 weeks, watched videotapes demonstrating parenting skills, and then took part in focused group discussions. The topics included how to play with your child, helping your child learn, using praise and encouragement to bring out the best in your child, effective setting of limits, handling misbehavior, how to teach your child to solve problems, and how to give and get support. Observations in the home showed that the children of mothers in the experimental group behaved better than those of the control group mothers.

Webster-Stratton and Mary Hammond (1997) also evaluated the effectiveness of parent training and child skills training with about 100 Seattle children (average age, 5) referred to a clinic because of conduct problems. The children and their parents were randomly assigned to (a) receive parent training, (b) receive child skills training, (c) receive both parent and child training, or (d) be in a control group. The skills training aimed to foster prosocial behavior and interpersonal skills using video modeling, while the parent training involved weekly meetings between parents and therapists for 22 to 24 weeks. Parent reports and home observations showed that children in all three experimental conditions had fewer behavior problems than control children, in both an immediate and a one-year followup. There was little difference in results among the three experimental conditions, although the combined parent and child training condition produced the most significant improvements in child behavior at the 1-year follow-up. It is generally true that combined parent and child interventions are more effective than either one alone.

Scott and his colleagues (Scott, Spender, Doolan, Jacobs, & Aspland, 2001) evaluated the Webster-Stratton parent training program in London and Chichester, U.K. About 140 mainly poor, disadvantaged children aged 3 to 8 who were referred for antisocial behavior were randomly assigned to receive parent training or to be in a control group. The parent training program, based on videotapes, covered praise and rewards, setting limits, and handling misbehavior. Follow-up parent interviews and observations showed that the antisocial behavior of the experimental children decreased significantly compared to that of the controls. Furthermore, after the intervention, experimental parents gave their children more praise to encourage desirable behavior and used more effective commands to obtain compliance.

Sanders and his colleagues (Sanders, Markie-Dadds, Tully, & Bor, 2000) in Brisbane, Australia, developed the Triple-P Parenting program. This program either can be delivered to the whole community in primary prevention using the mass media or can be used in secondary prevention with high-risk or clinic samples. Sanders et al. evaluated the success of Triple-P with high-risk children aged 3 by randomly assigning them either to receive Triple-P or to be in a control group. The Triple-P program involves teaching parents 17 child management strategies, including talking with children, giving physical affection, praising, giving attention, setting a good example, setting rules, giving clear instructions, and using appropriate penalties for misbehavior (a “time-out,” or sending the child to his or her room). The evaluation showed that the Triple-P program was successful in reducing children’s antisocial behavior.

Other Parenting Interventions

Another parenting intervention, termed functional family therapy, was developed by Alexander in Utah (see Sexton & Alexander, 2000). This aimed to modify patterns of family interaction by modeling, prompting, and reinforcement; to encourage clear communication of requests and solutions between family members; and to minimize conflict. Essentially, all family members were trained to negotiate effectively, to set clear rules about privileges and responsibilities, and to use techniques of reciprocal reinforcement with each other. The program was evaluated by randomly assigning 86 delinquent youths to experimental or control conditions. The results showed that this technique halved the recidivism rate of minor delinquents in comparison with other approaches (client-centered or psychodynamic therapy). Its effectiveness with more serious offenders was confirmed in a replication study using matched groups.

Chamberlain (1998) in Oregon evaluated treatment foster care (TFC), which was used as an alternative to custody for delinquent youths. Custodial sentences for delinquents were thought to have undesirable effects, especially because of the bad influence of delinquent peers. In treatment foster care, families in the community were recruited and trained to provide a placement for delinquent youths. The TFC youths were closely supervised at home, in the community, and in the school, and their contacts with delinquent peers were minimized. The foster parents provided a structured daily living environment with clear rules and limits, consistent discipline for rule violations, and one-to-one monitoring. The youths were encouraged to develop academic skills and desirable work habits.

In the evaluation, 79 chronic male delinquents were randomly assigned to treatment foster care or to regular group homes where they lived with other delinquent youths. A 1-year follow-up showed that the TFC boys had fewer criminal referrals and lower self-reported delinquency. Hence, this program seemed to be an effective treatment for delinquency.

Multisystemic Therapy

Multisystemic therapy (MST) is an important multiplecomponent family preservation program that was developed by Henggeler and his colleagues (Henggeler, Schoenwald, Borduin, Rowland, & Cunningham, 1998) in South Carolina. The particular type of treatment is chosen according to the particular needs of the youth. Therefore, the nature of the treatment is different for each person. MST is delivered in the youth’s home, school, and community settings. The treatment typically includes family intervention to promote the parent’s ability to monitor and discipline the adolescent, peer intervention to encourage the choice of prosocial friends, and school intervention to enhance competence and school achievement.

In an evaluation by Henggeler et al. (1998), 84 serious delinquents (with an average age of 15) were randomly assigned either to receive MST or the usual treatment (which mostly involved placing the juvenile in a setting outside the home). The results showed that the MST group had fewer arrests and fewer self-reported crimes in a 1-year follow-up. In another evaluation, in Missouri, Charles Borduin and his colleagues randomly assigned 176 juvenile offenders (with an average age of 14) either to MST or to individual therapy focusing on personal, family, and academic issues. Four years later, only 29% of the MST offenders had been rearrested, compared with 74% of the individual therapy group (cited in Aos et al., 2001). According to Steve Aos et al. (2001), MST had one of the highest cost-benefit ratios of any program. For every $1 spent on it, $13 were saved in victim and criminal justice costs.

Unfortunately, disappointing results were obtained in a large-scale independent evaluation of MST in Ontario, Canada, by Alan Leschied and Alison Cunningham (1998). Over 400 youths who were either offenders or at risk of offending were randomly assigned to receive either MST or the usual services (typically probation supervision). Six months after treatment, 28% of the MST group had been reconvicted, compared with 31% of the control group, a nonsignificant difference. Therefore, it is unclear how effective MST is when it is implemented independently.

Is Family-Based Intervention Effective?

Evaluations of the effectiveness of family-based intervention programs have produced both encouraging and discouraging results. In order to assess effectiveness according to a large number of evaluations, Farrington and Welsh (2003) reviewed 40 evaluations of family-based programs, each involving at least 50 persons in experimental and control groups combined. All of these had outcome measures of delinquency or antisocial child behavior. Of the 19 studies with outcome measures of delinquency, 10 found significantly beneficial effects of the intervention and 9 found no significant effect. Happily, no study found a significantly harmful effect of family-based treatment.

Over all 19 studies, the average effect size (d, the standardized mean difference) was .32. This was significantly greater than zero. When it was converted into the percentage reconvicted, a d value of .32 corresponded to a decrease in the percentage reconvicted from 50% to 34%. Therefore, it was concluded that, taking all 19 studies together, they showed that family-based intervention had substantial desirable effects. Also, there was evidence that some programs (e.g., home visiting) had financial benefits that greatly exceeded program costs.

School-Based Prevention

The next section reviews school-based prevention programs, most of which also had a family-based component. The Perry preschool program is reviewed first. This is perhaps the most influential early prevention project, because it concluded that $7 were saved for every $1 expended. Then the famous programs combining child skills training and parent training, implemented in Montreal by Richard Tremblay and in Seattle by David Hawkins, are reviewed, and also anti-bullying programs by Dan Olweus in Norway and Peter Smith in England.

Preschool Programs

The most famous preschool intellectual enrichment program is the Perry project, carried out in Ypsilanti (Michigan) by Lawrence Schweinhart and David Weikart (see Schweinhart et al., 2005). This was essentially a “Head Start” program targeted at disadvantaged African American children. Members of a small sample of 123 children were assigned (approximately at random) to experimental and control groups. The experimental children attended a daily preschool program—backed up by weekly home visits—usually lasting 2 years (covering ages 3-4). The aim of the “plan-do-review” program was to provide intellectual stimulation, to increase thinking and reasoning abilities, and to increase later school achievement.

This program had long-term benefits. John BerruetaClement (1984) showed that at age 19, members of the experimental group were more likely to be employed, more likely to have graduated from high school, more likely to have received college or vocational training, and less likely to have been arrested. By age 27, the experimental group had accumulated only half as many arrests on average as the controls. Also, they had significantly higher earnings and were more likely to be homeowners. Regarding the women in the experimental group, more were married, and fewer of their children were born to unmarried mothers.

The most recent follow-up of this program, evaluating the participants at age 40, found that it continued to make an important difference in their lives. Compared to the control group, those who received the program had significantly fewer lifetime arrests for violent crimes (32% vs. 48%), property crimes (36% vs. 56%), and drug crimes (14% vs. 34%), and they were significantly less likely to be arrested five or more times (36% vs. 55%). Improvements were also recorded in many other important life course outcomes. For example, significantly higher levels of schooling (77% vs. 60% graduating from high school), better records of employment (76% vs. 62%), and higher annual incomes were reported by the program group compared to the controls.

Several economic analyses show that the financial benefits of this program outweighed its costs. The Perry project’s own calculation included crime and non-crime benefits, intangible costs to victims, and even projected benefits beyond age 27. This generated the famous costbenefit ratio of 7 to 1. Most of the benefits (65%) were derived from savings to crime victims. The most recent cost-benefit analysis of participants at age 40 found that the program produced $17 in benefits per $1 of cost.

School Programs

The Montreal longitudinal-experimental study combined child skills training and parent training (see McCord & Tremblay, 1992). Tremblay and his colleagues identified disruptive (aggressive or hyperactive) boys at age 6, and randomly allocated over 300 of them to experimental or control conditions. Between ages 7 and 9, the experimental group received training designed to foster social skills and self-control. Coaching, peer modeling, roleplaying, and reinforcement contingencies were used in smallgroup sessions on such topics as “how to help,” “what to do when you are angry,” and “how to react to teasing.” Also, their parents were trained using the parent management training techniques developed by Gerald Patterson (1982).

This prevention program was successful. By age 12, the experimental boys committed less burglary and theft, were less likely to get drunk, and were less likely to be involved in fights than the controls (according to self-reports). Also, the experimental boys had higher school achievement. At every age from 10 to 15, the experimental boys had lower self-reported delinquency scores than the control boys. Interestingly, the differences in antisocial behavior between experimental and control boys increased as the follow-up progressed. A later follow-up showed that fewer experimental boys had a criminal record by age 24 (Boisjoli, Vitaro, Lacourse, Barker, & Tremblay, 2007).

One of the most important school-based prevention experiments was carried out in Seattle by Hawkins and his colleagues (Hawkins, Catalano, Kosterman, Abbott, & Hill, 1999). They implemented a multiple-component program combining parent training, teacher training, and child skills training. About 500 first-grade children (aged 6) in 21 classes in 8 schools were randomly assigned to be in experimental or control classes. The children in the experimental classes received special treatment at home and school that was designed to increase their attachment to their parents and their bonding to the school. Also, they were trained in interpersonal cognitive problem solving. Their parents were trained to notice and reinforce socially desirable behavior in a program called “Catch Them Being Good.” Their teachers were trained in classroom management—for example, to provide clear instructions and expectations to children, to reward children for participation in desired behavior, and to teach children prosocial (socially desirable) methods of solving problems.

This program had long-term benefits. By the sixth grade (age 12), experimental boys were less likely to have initiated delinquency, while experimental girls were less likely to have initiated drug use. In a later follow-up, Hawkins and his colleagues (Hawkins, Catalano, Kosterman, Abbott, & Hill, 1999) found that, at age 18, the full intervention group (those who received the intervention from Grades 1-6) admitted less violence, less alcohol abuse, and fewer sexual partners than the late intervention group (Grades 5-6 only) or the control group. According to Steve Aos and his colleagues (2001), over $4 were saved for every $1 spent on this program.

In Baltimore, Hanno Petras, Sheppard Kellam, and their colleagues (2008) evaluated the “Good Behavior Game” (GBG), which aimed to reduce aggressive and disruptive child behavior through contingent reinforcement of interdependent team behavior. First-grade classrooms and teachers were randomly assigned either to the GBG condition or to a control condition, and the GBG was played repeatedly over 2 years. In trajectory analyses, the researchers found that the GBG decreased aggressive/disruptive behavior (according to teacher reports) up to Grade 7 among the most aggressive boys, and also caused a decrease in antisocial personality disorder at ages 19-21. However, effects on girls and on a second cohort of children were less marked.

There have been a number of comprehensive, evidencebased reviews of the effectiveness of school-based programs by Denise Gottfredson, David Wilson, and their colleagues (see Sherman, Farrington, Welsh, & MacKenzie, 2006). Meta-analyses identified four types of school-based programs that were effective in preventing delinquency: school and discipline management, classroom or instructional management, reorganization of grades or classes, and increasing self-control or social competency using cognitive-behavioral instruction methods. Reorganization of grades or classes had the largest average effect size (d = .34), corresponding to a significant 17% reduction in delinquency.

After-school programs (e.g., recreation-based, dropin clubs, dance groups, and tutoring services) are based on the belief that providing prosocial opportunities for young people in the after-school hours can reduce their involvement in delinquent behavior in the community. After-school programs target a range of risk factors for delinquency, including association with delinquent peers. Welsh and Akemi Hoshi identified three highquality after-school programs with an evaluated impact on delinquency (see Sherman et al., 2006). Each had desirable effects on delinquency, and one program also reported lower rates of drug use for participants compared to controls.

Anti-Bullying Programs

School bullying is a risk factor for later offending, and several school-based programs have been effective in reducing bullying. The most famous of these was implemented by Olweus (1993) in Norway. The general principles of the program were to create an environment characterized by adult warmth, interest in children, and involvement with children; to use authoritative child rearing, including warmth, firm guidance, and close supervision, since authoritarian child rearing is related to child bullying; to set firm limits on what is unacceptable bullying; to consistently apply nonphysical sanctions for rule violations; to improve monitoring and surveillance of child behavior, especially on the playground; and to decrease opportunities and rewards for bullying.

The Olweus (1993) program aimed to increase awareness and knowledge of teachers, parents, and children about bullying and to dispel myths about it. A 30-page booklet was distributed to all schools in Norway describing what was known about bullying and recommending what steps schools and teachers could take to reduce it. Also, a 25-minute video about bullying was made available to schools. Simultaneously, the schools distributed to all parents a four-page folder containing information and advice about bullying. In addition, anonymous selfreport questionnaires about bullying were completed by all children.

Each school received feedback information from the questionnaire, about the prevalence of bullies and victims, on a specially arranged school conference day. Also, teachers were encouraged to develop explicit rules about bullying (e.g., do not bully, tell someone when bullying happens, bullying will not be tolerated, try to help victims, try to include children who are being left out) and to discuss bullying in class, using the video and role-playing exercises. Also, teachers were encouraged to improve monitoring and supervision of children, especially on the playground.

The effects of this anti-bullying program were evaluated in 42 Bergen schools. Olweus (1993) measured the prevalence of bullying before and after the program using selfreport questionnaires completed by the children. Since all schools received the program, there were no control schools. However, Olweus compared children of a certain age (e.g., 13) before the program with different children of the same age after the program. Overall, the program was very successful because bullying decreased by half.

A similar program was implemented in 23 schools in Sheffield (U.K.) by Peter Smith and Sonia Sharp (1994). The core program involved establishing a “whole school” anti-bullying policy, raising awareness of bullying, and clearly defining roles and responsibilities of teachers and students so that everyone knew what bullying was and what they should do about it. In addition, there were optional interventions tailored to particular schools: curriculum work (e.g., reading books, watching videos), direct work with students (e.g., assertiveness training for those who were bullied), and playground work (e.g., training lunchtime supervisors). This program was successful in reducing bullying (by 15%) in primary schools, but had relatively small effects (a 5% reduction) in secondary schools.

Maria Ttofi and her colleagues (Ttofi, Farrington, & Baldry, 2008) completed a systematic review of the effectiveness of anti-bullying programs in schools. They found 59 high-quality evaluations of 30 different programs. They concluded that, overall, anti-bullying programs were effective. The results showed that bullying and victimization were reduced by about 17% to 23% in experimental schools compared with control schools.

Peer Programs

There are few outstanding examples of effective intervention programs for antisocial behavior targeted at peer risk factors. The most hopeful programs involve using high-status conventional peers to teach children ways of resisting peer pressure. Nancy Tobler and her colleagues (Tobler, Lessard, Marshall, Ochshom, & Roona, 1999) found that these were effective in reducing drug use. Also, in a randomized experiment in St. Louis, Ronald Feldman and his colleagues (Feldman, Caplinger, & Wodarski, 1993) showed that placing antisocial adolescents in activity groups dominated by prosocial adolescents led to a reduction in their antisocial behavior (compared with antisocial adolescents placed in antisocial groups). This suggests that the influence of prosocial peers can be harnessed to reduce antisocial behavior. However, putting antisocial peers together can have harmful effects.

The most important intervention program whose success seems to be based mainly on reducing peer risk factors is the “Children at Risk” program, which targeted high-risk adolescents (average age, 12) in poor neighborhoods of five cities across the United States. Eligible youths were identified in schools and randomly assigned to experimental or control groups. The program was a comprehensive, community-based prevention strategy targeting risk factors for delinquency, including case management and family counseling, family skills training, tutoring, mentoring, after-school activities, and community policing. The program was different in each neighborhood.

The initial results of the program were disappointing, but a 1-year follow-up by Adele Harrell and her colleagues (Harrell, Cavanagh, & Sridharan, 1999) showed that (according to self-reports) youths in the experimental groups were less likely to have committed violent crimes and used or sold drugs. The process evaluation showed that the greatest change was in peer risk factors. Experimental youths associated less often with delinquent peers, felt less peer pressure to engage in delinquency, and had more positive peer support. In contrast, there were few changes in individual, family, or community risk factors, which was possibly linked to the low participation of parents in parent training and of youths in mentoring and tutoring. In other words, there were problems of implementation of the program, linked to the serious and multiple needs and problems of the families.

Mentoring programs usually involve nonprofessional adult volunteers spending time with young people at risk for delinquency, dropping out of school, school failure, or other social problems. Welsh and Hoshi (2002) identified seven mentoring programs (of which six were of high quality) that evaluated the impact on delinquency. Since most programs had desirable effects, Welsh and Hoshi concluded that community-based mentoring was a promising approach in preventing delinquency. Similarly, a metaanalysis by Darrick Jolliffe and David Farrington (2008) concluded that mentoring was often effective in reducing reoffending.

Community Programs

In the interests of maximizing effectiveness, what is needed is a multiple-component, community-based program including several of the successful interventions listed above. Many of the programs reviewed in this article are of this type. However, “Communities That Care” (CTC) is an additional program that has many attractions. Perhaps more than any other program, it is evidence-based and systematic: The choice of interventions depends on empirical evidence about what are the important risk and protective factors in a particular community and on empirical evidence about “what works.” It has been implemented in at least 35 sites in England, Scotland, and Wales and also in the Netherlands and Australia.

CTC was developed as a risk-focused prevention strategy by Hawkins and Catalano (1992), and it is a core component of the U.S. Office of Juvenile Justice and Delinquency Prevention’s Comprehensive Strategy for Serious, Violent, and Chronic Juvenile Offenders. CTC is based on a theory (the social development model) that organizes risk and protective factors. The intervention techniques are tailored to the needs of each particular community. The “community” could be a city, a county, a small town, or even a neighborhood or a housing estate. This program aims to reduce delinquency and drug use by implementing particular prevention strategies that have demonstrated effectiveness in reducing risk factors or enhancing protective factors. It is modeled on large-scale, community-wide public health programs designed to reduce illnesses such as coronary heart disease by tackling key risk factors. There is great emphasis in CTC on enhancing protective factors and building on strengths, partly because this is more attractive to communities than tackling risk factors. However, it is generally true that health promotion is more effective than disease prevention.

CTC programs begin with community mobilization. Key community leaders (e.g., elected representatives, education officials, police chiefs, business leaders) are brought together with the aim of getting them to agree on the goals of the prevention program and to implement CTC. The key leaders then set up a community board that is accountable to them, consisting of neighborhood residents and representatives from various agencies (e.g., school, police, social services, probation, health, parents, youth groups, business, church, media). The community board takes charge of prevention on behalf of the community.

The community board then carries out a risk and protective factor assessment, identifying key risk factors in that particular community that need to be tackled and key protective factors that need enhancing. This risk assessment might involve the use of police, school, social, or census records or local neighborhood or school surveys. After identifying key risk and protective factors, the community board assesses existing resources and develops a plan of intervention strategies. With specialist technical assistance and guidance, they choose programs from a menu of strategies that have been shown to be effective in well-designed evaluation research.

The menu of strategies listed by Hawkins and Catalano (1992) includes prenatal and postnatal home visiting programs, preschool intellectual enrichment programs, parent training, school organization and curriculum development, teacher training, and media campaigns. Other strategies include child skills training, anti-bullying programs in schools, situational prevention, and policing strategies. The choice of prevention strategies is based on empirical evidence about effective methods of tackling each particular risk factor, but it also depends on what are identified as the biggest problems in the community. While this approach is not without its challenges and complexities (e.g., cost, implementation, establishing partnerships among diverse agencies), an evidence-based approach that brings together the most effective prevention programs across multiple domains offers the greatest promise for reducing crime and building safer communities.


High-quality evaluation research shows that many programs are effective in reducing delinquency and antisocial behavior, and that in many cases the financial benefits of these programs outweigh their financial costs. The best programs include general parent education, parent management training, preschool intellectual enrichment programs, child skills training, mentoring, teacher training, anti-bullying programs, and multisystemic therapy.

The time is ripe to mount a large-scale, evidence-based, integrated national strategy for the reduction of crime and associated social problems, including rigorous evaluation requirements. This approach should implement programs to tackle risk factors and strengthen protective factors, and it could be based on “Communities That Care.” Primary prevention has been effective in improving health, and it could be equally effective in reducing delinquency and antisocial behavior in all countries.