Delphi and its potential impact on information systems

Abstract

The Delphi method is basically defined as a method for the systematic solicitation and collation of informed judgments on a particular topic. The concept of "informed" here could mean poor people, if the subject were poverty, as well as the usual interpretation of "experts." The method has two important characteristics which distinguish it considerably from a polling procedure. The first is feedback, where the judgments of the individuals are collected, possibly formulated as a group response and fed back. Thus, each individual may view the results and consider whether he wishes to contribute more to the information and/or reconsider his earlier views. This round or phase structure may go through three to five iterations in the usual paper and pencil exercise. The second characteristic is that all responses are anonymous. The reasons for anonymity are much discussed in the literature and will not be reviewed here. However, there are circumstances where complete anonymity could be relaxed. In some cases it may be useful for the respondents to know who is participating in order to insure awareness that a peer group is involved in the discussion. Also, when a highly specialized subtopic enters the discussion it may be appropriate to permit an expert to endorse an item.