Progress and prospects in youth violence-prevention evaluation

The evaluation of youth violence-prevention projects using sound methods is very important. Up to now, the evaluation literature has (1) inadequately heeded known epidemiologic patterns of violence, (2) failed to differentiate types of violence, and (3) failed to differentiate the levels of risk-influence addressed by the intervention. The reports in this supplement have considered these past deficiencies. These reports also make some notable advances, such as illustrating the role that efficacy trials can and should play in violence prevention, demonstrating that programs can and should be evaluated and that design characteristics such as random assignment are plausible, balancing the practical and theoretical aspects of violence prevention, and reflecting the importance of the setting on the style and scope of the intervention. The reports and projects also have remaining limitations. Length of planned follow-up is generally quite brief. In some cases the rationale for the intervention is not adequately explained. Some reports do not clarify whether the intervention is intended for all youths or selected high-risk youth. The next steps for these and similar projects are to determine the program impact and implementation, strive for longer follow-up, document conditions that interact with proximal impact on distal outcomes, and further broaden evaluation efforts into modifying situations instead of modifying only individuals. Finally, we must recognize that the goal of evaluation is not to declare all earnest efforts effective, but to determine which efforts merit further consideration.