This paper presents the methodology for evaluating the Monroe-Livingston demonstration project's capitation payment system (CPS), based in Rochester, New York, for chronic mentally ill patients. To allow for both patient and provider choice within the experimental design, 1,587 CPS-eligible patients were randomly assigned at the start of the study to control or experimental conditions, with the intent of capturing in the experimental group a significant number of patients who would later be enrolled in the CPS. Protocols, which included measures of symptomatology, functioning, and resource utilization, were completed at baseline for 422 of the 605 patients contacted for inclusion in the study. The baseline control group included 143 patients; the experimental group included 279 patients, 153 of whom were eventually enrolled in the CPS.