The stated goals of World Health Organization-supported behavioral research are applied: to contribute to planning and policy decision-making, and to improve health care delivery methods. The evidence suggests the organization is getting less for its behavioral research dollar than it ought to: much research has been of poor quality, and researchers often appear more concerned with research design elegance than with practical application of results. Professional and structural factors explain this picture. Physician-dominated research committees that evaluate research applications assume that quantitative hypothesis-testing investigation is the only acceptable research model. Hence, in their grant applications behavioral scientists conform to the expectations of research committees, stressing quantitative methods and slighting the qualitative approaches that often are more productive in providing operational information. Research committee members also often fail to understand the scope of behavioral research possibilities: hoped-for results are limited to information on how to change community behavior more nearly to conform to the needs of health care delivery programs. Research on organization policies and programs is viewed as irrelevant and perhaps even threatening. To improve the quality and utility of WHO-sponsored behavioral research, the author suggests a workshop of organization personnel and short-term consultants and temporary advisors with prior experience, to evaluate past research, to identify its strengths and weaknesses, and to recommend modified research support procedures to produce higher quality, operationally-useful results.