- The text on this page is taken from an equivalent page of the IEHIAS-project.
The Delphi method was one of the first formal expert elicitation methods. It involves multiple rounds of expert judgements. Experts respond initially to a specific question or series of questions, and their responses are collated. In subsequent rounds, respondents are then provided with (anonymous) feedback on their results and the results of others, and invited to further justify, expand or amend their opinions.
The ultimate aim of the Delphi method is to achieve consensus among the experts. It thus deliberately operates by exerting pressures on participants to move towards the common view, or to justify their insistence on maintaining a view that is different from others. As such it has several dangers, one of which is that it will produce a spurious convergence - one achieved for the sake of the exercise, rather than any real change in opinion. It can also result in dilution of the well-founded knowledge of individuals by the weight of voice of other, less well-informed participants.
In various forms, the method is still used for expert elicitation. However, for elicitations for which consensus among experts is not required or not justifiable, other methods are more suitable. Useful information on alternative methods are provided on the FOR-LEARN website; (see also Expert elicitation methodology).
- Amara, R. and Lipinski, A. 1972 Some views on the use of expert judgment. Technological Forecasting and Social Change 3, 279-289.*
- Brown, B.B., Cochran, S.W. and Dalkey, N.C. 1969 The Delphi method. II: structure of experiments. Rand Rep. RM-5957-PR.
- Dalkey, N.C. 1969 The Delphi method: an experimental study of group opinion. Rand Rep. RM-5888-PR. 1969.
- Rowe, G. and Wright, G. 1999 The Delphi technique as a forecasting tool: issues and analysis. International Journal of Forecasting 15, 353-375.
- Steinert, M. 2009 A dissensus based online Delphi approach: an explorative research tool. Technological Forecasting and Social Change 76, 291-300.