The extreme variability of the 50 projects under the NIE Program to Increase the Participation of Minorities and Women in Educational Research and Development necessitates an evaluation design that is flexible enough to accomodate significant differences in project characteristics, yet comprehensive enough to allow meaningful aggregation of cross-project data for evaluation at the program level. This report of a program evaluation method that meets these two objectives is divided into two parts. Part I includes descriptive overview data on major project dimensions that clearly illustrate the wide diversity of the characteristics among the 50 projects. Two key variables, project orientation and targeted participant population, emerged as a basis for logical sub-grouping of projects into a classification matrix. In Part II, possible methods and timing of data collection approaches are outlined in depth and potential data sources are reviewed. The section also outlines a conceptual framework that focuses on the context and process factors influencing project activities and potential outcomes. A final section details a series of recommendations directed to NIE policy makers, program-related personnel, and others interested in the problem of increasing the participation of women and minorities in educational research and development. (CJ)