Part 5 of a seven-volume report made by Graduate Medical Educational Advisory Committee (GMENAC) on the relationship between the education environment and specialty and subspecialty career choices is presented. After background information, Section 2 examines such conceptual and methodological issues as boundaries of the educational environment, specialty preferences vs. specialty choices vs. specialty attainment, timing of specialty preference formation and choice, aggregating or grouping specialties, input-output analyses, and the interrelationships among specialty and other career choices. In a third section on physician specialty choice literature (does it indicate possible mechanisms for modifying specialty distribution via the educational environment?) admission/student characteristics and institutional environment influences are discussed. The final section offers several recommendations such as: the applicant pool must be broadened with regard to students' individual characteristics and to reduce the financial barriers to medical education which limit diversity; undergraduate medical education must be broad-based; the first year of graduate medical education should be a broad-based clinical experience to serve as the foundation for further specialty training; and programs of loans and scholarships should be expanded. Extensive references are provided. (LC)