Chapter 5

CONCLUSIONS

Clearly, we are past the era in which some observers believed that "nothing works" to prevent youth violence. Numerous programs have demonstrated their effectiveness in reducing risk factors for serious violence. At the same time, there is a pressing need to evaluate more youth violence prevention programs. Of the hundreds of programs currently in use throughout the United States, only six met the criteria for a Model program, and 21 met the criteria for a Promising program. Of the 266 school-based program modules reviewed by Gottfredson et al. (in press), all of which were formally evaluated against a control or comparison group, only 10 percent received the highest score for scientific rigor (the experimental design standard used here). For most violence, crime, and drug prevention programs now being implemented, there is simply no evidence regarding effectiveness. Although well-designed program evaluations are expensive and time-consuming, they are the only way to determine the effectiveness of existing youth violence prevention programs.

Nearly half of the most thoroughly evaluated strategies for preventing youth violence are ineffective, however, and a few are even harmful. It is in society’s best interest to evaluate programs before exposing children and adolescents to them—otherwise we run the risk of harming young people rather than helping them.

The most effective youth violence prevention programs are targeted appropriately, address several age-appropriate risk and protective factors in different contexts, and include several program components that have been shown to be effective. This finding is consistent with research showing that youth violence is affected by numerous risk and protective factors that span several environmental contexts (individual, family, school, peer group, community) and several stages of a youth’s life (see Chapter 4).

GOING TO SCALE

While identifying best practices in youth violence prevention is critical to reducing the number of young people involved in and affected by violence, it is not the last step. The manner in which a program is implemented can have an enormous impact on its effectiveness—even the best programs are effective only when implemented with high quality and fidelity to the program’s design. In other words, using an effective strategy is only part of what is required to achieve effective results. Details of program delivery, including characteristics of the youths receiving the intervention, the setting in which they are treated, and the intensity or duration of the intervention, play important roles in determining effectiveness. Programs must be delivered with design fidelity, to a specific population of youths, within a specific context, and for a specific period of time.

Unfortunately, very little is known about how to preserve a prevention program’s positive effects when it is implemented on a wide-scale or national level. What research has been conducted indicates that effective implementation is at least as important to a program’s success as the characteristics and content of the program itself (Petersilia, 1990; Lipsey, 1992a, 1992b). Studies of program implementation consistently find that effectiveness depends on the following principles, according to a review by Petersilia (1990, p. 130):

• The project addresses a pressing local problem.
• The project has clearly articulated goals that reflect the needs and desires of the "customer."
• The project has a receptive environment in both the parent organization and the larger system.
• The organization has a leader who is committed to the objectives, values, and implications of the project and who can devise practical strategies to motivate and effect change.
• The project has a director who shares the leader’s ideas and values and uses them to guide the implementation process and ongoing operation of the project.
• Practitioners make the project their own rather than being coerced into it; that is, they buy into it, participate in its development, and have incentives to maintain its integrity during the process of change.
• The project has clear lines of authority: There is no ambiguity as to who is in charge.
• The change and its implementation are not complex and sweeping.
• The organization has secure administrators, low staff turnover, and plentiful resources.

Gendreau et al. (1999) organize these same principles into four categories: general organizational factors, program factors, change agent factors, and staffing activities. While they acknowledge the importance of a program’s characteristics, such as its theoretical basis, they also stress that positive change and success are dependent on much more than the specific characteristics of a prevention program or intervention. Characteristics of the implementer, the environment in which the program is implemented, and even the target population have a significant influence on overall program effects.

Both the Petersilia and Gendreau et al. studies discuss characteristics of effective implementation within a correctional setting. The Centers for Disease Control and Prevention’s (CDC) Best Practices of Youth Violence Prevention (Thornton et al., 2000) and a recent review by Gottfredson et al. (2000) suggest that many of the same characteristics help determine the success of violence and delinquency prevention programs. In particular, the CDC study highlights the importance of training, monitoring, and supporting the staff who implement a program on the local level. An appropriate match between staff and the target population can also contribute to program success, particularly in parent- and family-based programs. Staff must be committed to the program, experienced with the general strategy being used, knowledgeable about the target community, and capable of managing group dynamics and overcoming resistance. Likewise, as noted by Petersilia, maintaining community involvement is a key element of program success. Finally, linking a youth violence prevention program to existing strategies and support agencies in the community or school can contribute to success (Thornton et al., 2000).

A similar group of implementation characteristics affects the success of school-based delinquency prevention programs, according to Gottfredson and colleagues (2000). In a study of more than 1,200 schools throughout the United States, they found that extensive, high-quality training and supervision, as well as support for the program from the principal of the school, are key elements of success. Schools also appear to have greater success with standardized materials and methods, as well as programs that can be incorporated into the regular school program. Consistent with Petersilia’s principles, local buy-in and initiation of school-based delinquency prevention are important predictors of program success. Multiple sources of information, including the use of an expert to assist with training and implementation, also help to ensure positive results. Improvements in any or all of these factors should improve the quality of the overall prevention program—and its effects on youths.

The CDC recommends monitoring the progress and quality of program implementation on a local level. This step can be particularly important when implementing Model programs. The proven effectiveness of these programs in multiple, long-term studies makes them suitable for implementation on a wide, or even national, scale, but even Model programs are successful only when implemented with fidelity. While it is not always necessary to conduct expensive outcome evaluations of Model programs, given their demonstrated positive effects and ongoing national evaluations, it is critical to monitor the quality of implementation on the local level.

Scientific research has established the effectiveness of a number of prevention programs, and evaluation studies are sure to identify more in the near future. Although the studies cited above offer valuable guidance, more research is needed on how to implement youth violence prevention programs with fidelity on a national scale, how to monitor program fidelity on this scale, and how to increase community and agency capacity for implementing these programs. In addition, large-scale program dissemination will affect the overall benefits of individual youth violence prevention programs. Addressing these issues will require a major investment of time and resources, but it is the essential next step in the continuing effort to find effective solutions to the problem of youth violence.