A critical analysis and evaluation of the Delphi technique, a systematic method for eliciting expert opinion on a variety of topics. The author examines the methodological principles and key assumptions of Delphi in the light of current standards for social experimentation, test design, sampling, use of experts, and interpretation of findings. He concludes that conventional Delphi is an unreliable and scientifically unvalidated technique in principle and probably in practice. Except for its possible value as an informal exercise for heuristic purposes, Delphi should be replaced by demonstrably superior, scientifically rigorous questionnaire techniques and associated experimental procedures using human subjects. Users are urged to work with psychometrically trained social scientists who can apply rigorous techniques tailored to specific needs. Delphi should not be used until its principles, methods, and fundamental applications can be established experimentally as scientifically tenable.