Delphi_Technique - Predictive Analysis: The Delphi Method...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Predictive Analysis: The Delphi Method The Delphi Method is one of several analytic techniques used to make estimates or predict future human behavior or human conditions. The Delphi Method takes its name from the legend of the Oracle at Delphi in ancient Greece. According to the legend, those wishing to gain a glimpse into the future would gather in a special chamber deep within a temple dedicated to the god Apollo located in the coastal town of Delphi. The crowd would listen to “the Pythia,” a woman, obviously in an entranced state, perched on a three-legged stool, who channeled prophetic messages from the deity through the temple’s priests. The priests translated the Pythia’s otherwise unintelligible utterances. Thus, the ancient priests were the “experts” in interpreting the Pythia’s messages about the future. (Note: Recent scientific investigation reveals that the Pythia’s entranced state could have been due to gaseous emissions along tectonic fault lines in the Delphi region—whether the oracle could see the future is another matter.) The Delphi Method of estimation or prediction is also based on asking the “experts” about the future. Its central assumption offers that the collective judgment and wisdom of several experts is better than the estimates and predictions of any one expert. Developed originally by the RAND Corporation, the Delphi Method is a technique for eliciting and combining judgments systematically from a group of experts. The basic Delphi Method begins with a series of first round questions asked individually of experts on the subject. The experts submit their judgments. The results of the first round judgments are then tabulated and the results sent back to the experts for modification. In essence, the experts are asked in the second round to reevaluate their original judgments in light of the average estimates calculated in the first round. This procedure of reevaluation is continued for several rounds until a fairly high degree of consensus is reached, or until the experts no longer modify their previous estimates. The Delphi Method questions are commonly transmitted as part of a mail survey. Following normal procedures for constructing survey questions, the researcher must ensure that questions provide good operational measures of the concepts they are trying to estimate or predict. If possible, it is best to use interval or ordinal measured questions to facilitate returning measures of central tendency (mode, median, mean) to the experts in subsequent rounds. The use of Likert, Gutmann, or other scales for question answers is common. In the second round of the survey, the experts are normally given only the measures of central tendency for the answers to the questions in the first round, and asked to explain in detail when their second round judgments (answers) differ substantially from the first round’s measures of central tendency. This same procedure, asking for details of why subsequent judgments differ widely from the previous round’s measures of
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 08/08/2010 for the course HLS 401 taught by Professor Dr.collier during the Fall '09 term at E. Kentucky.

Page1 / 9

Delphi_Technique - Predictive Analysis: The Delphi Method...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online