Delphi method

Delphi method
Name	Delphi method
Uses	Structured communication technique for forecasting and consensus building
Related	RAND Corporation, Norman Dalkey, Olaf Helmer

Delphi method. The Delphi method is a structured communication technique, originally developed as a systematic, interactive forecasting method which relies on a panel of experts. The technique involves multiple rounds of questionnaires sent to a panel of experts, with aggregated and anonymized feedback provided after each round to converge towards a consensus. It is widely used for forecasting, policy-making, and project planning where precise data is unavailable, aiming to mitigate the effects of dominant individuals and groupthink.

Overview

The core principle involves eliciting and refining opinions from a selected group through iterative, anonymous questionnaires with controlled feedback. This approach was pioneered by researchers at the RAND Corporation during the Cold War to forecast the impact of technology on warfare. It is distinct from other group decision-making methods like the Nominal Group Technique or Brainstorming due to its emphasis on anonymity and statistical aggregation of responses. The method is often applied in fields such as Public Health, Information Technology, and Environmental Science where expert judgment is paramount.

Process

A typical process begins with the formation of a facilitator team and the selection of a panel of experts, often from diverse disciplines or institutions like the World Health Organization or Massachusetts Institute of Technology. The first round involves an open-ended questionnaire to generate ideas, with responses synthesized into a series of statements or scenarios. In subsequent rounds, panelists rate or rank these items, often using a Likert scale, and receive a statistical summary of the group's response, such as the Interquartile range. This iterative process continues, usually for two to four rounds, until a pre-defined stop criterion, such as stability in responses or consensus measured by metrics like Kendall's coefficient of concordance, is reached.

Applications

Its applications are broad, having been used by entities like the United States Department of Defense for long-range strategic planning and by the European Commission for research priority setting. In healthcare, it has been instrumental for developing clinical guidelines, such as those for the Centers for Disease Control and Prevention, and for identifying key indicators in Oncology or Psychiatry. Within technology, it has aided in forecasting developments in Artificial Intelligence and Cybersecurity. Other notable uses include urban planning for cities like Singapore and resource management for agencies such as the Environmental Protection Agency.

Variations

Several modifications have been developed to address specific needs. The Policy Delphi, for instance, is designed to generate opposing viewpoints on contentious issues, often used in debates surrounding the Kyoto Protocol or Brexit. The Real-Time Delphi, facilitated by software from groups like the Institute for the Future, allows for continuous feedback. The Disaggregative Delphi employs techniques like Cluster analysis to identify distinct opinion groups within the panel. Hybrid approaches combining it with the Analytic Hierarchy Process or Scenario planning are also common in complex studies for organizations like the World Bank.

Criticisms and limitations

Criticisms often focus on methodological weaknesses, such as the potential for low reliability and the influence of the facilitator's design, as noted in journals like the Journal of Marketing Research. The selection of experts, which may favor well-known figures from institutions like Harvard University, can introduce bias and exclude dissenting voices. The statistical convergence towards a median may create an illusion of consensus without genuine agreement, a problem highlighted in studies of Climate change policy. Furthermore, the process can be time-consuming and resource-intensive, limiting its use in rapidly evolving fields like Cryptocurrency markets.

History and development

The method was developed in the 1950s by Norman Dalkey and Olaf Helmer at the RAND Corporation under a project funded by the United States Air Force, initially named "Project Delphi". Its first major application was a study on the use of expert opinion to estimate the number of atomic bombs needed to cripple Soviet Union industry. The methodology was formally published in the 1960s and gained prominence after its use in a report for the Commission on the Year 2000. Its adoption spread through the work of the Futures Group and publications in periodicals like Foreign Affairs, influencing long-range planning in both corporate and governmental spheres globally.

Category:Research methods Category:Forecasting Category:Consensus decision-making