Abstract
The author argues that criminal justice professionals have failed in most cases to critically evaluate programs introduced into the criminal justice system. Other reviews have shown criminal justice research to be largely nonexperimental. This problem is discussed in the context of pretrial diversion, an increasingly popular innovation in criminal justice. A recent project in Illinois illustrates the problems and pitfalls that can be expected when advocating more sophisticated methods of evaluation. In reviewing the research on evaluation of diversion projects, the author concludes that we know little about the effectiveness of this form of pretrial interven tion-projects typically have not been evaluated at all, or they have used weak experimental designs that do not allow for definitive conclusions. The author discusses possible reasons for this nonexperimental attitude and proposes a research methodology for future evaluations of diversion effectiveness. Evaluation of pretrial diversion is only part of a more general problem in criminal justice research, and it leads to a consideration of wide social policy implications.

This publication has 0 references indexed in Scilit: