Interviews focusing on homicide, sexual assault, aggravated assault, robbery, and burglary provided data for this study. They were conducted in seven areas: Salem, Mass.; Baltimore County, Md.; Greenville, S.C.; New Orleans, LA.; Kansas City, Mo.; Minneapolis, MN.; and San Jose, Calif. Practitioners explained their actions in recently closed cases in two sites and simulated decisionmaking processes using scenario scenes in six sites. Most victims were interviewed by mail. The sample consisted of approximately 389 victims, 112 police officers, 101 prosecutors, and 48 judges from the six scenario case sites and 21 practitioners from the real case sites. Analysis showed that evidence rather than victim harm is the most important factor in the decision to accept the case for prosecution, and type of conviction and defendant's prior record are the primary considerations in sentencing. Physical injury and victim's age were significant in screening sexual assault cases, but appeared important because they contributed additional evidence. Victims in sites with active, prosecutor-based victim/witness programs reported higher levels of satisfaction with the criminal justice system than those in sites without such programs. When asked what the system could do to increase their satisfaction, victims indicated they wanted to be better informed about the case, the offender punished more harshly, and improved social services. Police believed strongly in greater victim involvement, while prosecutors and judges saw no need for change. The report discusses policy implications of these findings and supplies tables and footnotes. The appendix contains the interview instruments, discussion of the analysis, and an explanation of the mail survey techniques.