The feedback of information to assistant district attorneys (ADA's) concerning civilian witnesses (notifications and expected court attendance) is examined. The findings indicate that civilian witness attendance in court facilitates substantive outcomes. For examples, guilty plea doubled from 12 to 25 percent and dismissals declined from 23 to 14 percent. The report also shows that the project in fact increases attendance by civilian witnesses in court: attendance at the first postarraignment appearnace rose from 45 percent and 39 percent, for baseline and control samples respectively, to 55 percent for a project sample. The report also suggests, on the basis of very limited data, that the project's feedback of information to ADA's increased placement of cases on second call when witnesses who have not yet arrived are expected to appear and that offering of pleas increased when witnesses have not been contacted or are reported to be unwilling to appear. The report concludes by recommending that the project direct more resources toward reducing a high level of nonappearances (45 percent) among civilian witnesses. Tables and footnotes are included. Three appendixes give methodology of the study, characteristics of civilian witnesses, and reliability of information on the Court Part Information Sheets (CPIS). (Author abstract modified.)