Abstract:This article reports the findings of an empirical study of arbitration in Arizona's general jurisdiction civil trial courts. The study found that the arbitration program's primary goals of providing faster and less expensive resolution of cases, reducing the court's workload, and maintaining or enhancing the satisfaction of users, were not entirely being met. Many cases did not meet arbitration deadlines and court case processing time standards. Most cases eligible for arbitration concluded before a hearing was held, and those cases that had a hearing seemed more likely to have been diverted from settlement than from trial. Consequently, the arbitration program was likely to affect the court's workload in a relatively small proportion of cases, was more likely to reduce the use of court pretrial rather than trial resources, and was unlikely to substantially reduce litigants' costs. But arbitration did increase access to a hearing on the merits. Lawyers who represented clients in arbitration had generally favorable assessments of the process and award, but expressed concerns about the adequacy of arbitrators’ knowledge of both substantive issues and arbitration procedures. A majority of lawyers favored retaining compulsory arbitration and some of its basic components but changing policies relating to arbitrator service and assignment. These findings, which were consistent with studies in other jurisdictions, suggest that court-connected arbitration does not have negative consequences, but also does not consistently or substantially improve the effectiveness and efficiency of dispute resolution.