The Delphi method, also called a Delphi study, is an intensive questionnaire used to bring a group of experts to a consensus. It is widely used in investment forecasting but has also become commonplace in such diverse fields as medicine, psychology, and political science.
The term “expert” is open to debate but generally means anyone with relevant and substantial experience. Care should be taken to reduce bias as much as possible. For example, convenience sampling, where the researcher simply asks colleagues or other class acquaintances, should be avoided if at all possible.
Using the Delphi Method
The first step in using the Delphi method is recruiting a team of experts in the relevant field; these called panelists. It has been suggested that 7 is the minimum number of panelist for effective use of a Delphi study, but in practice anywhere from 4 to several thousand have been used.
The minimum amount of rounds (iterations) used in a Delphi method study is two.
In round 1, a set of identical questions is provided to each panelist. These questions may be open ended or not, and are typically drawn from relevant literature. The answers of the panelists are collected and kept anonymous, and before the next round, panelists are provided with a summary of the results.
In a two-round Delphi study, the second round is used to clarify the issues addressed in round 1 as well as provide opportunities to the panelists to revise their opinions based on the (anonymous and aggregated) responses to round 1.
If there are more than two rounds, intermediate rounds are used to delve deeper into the issues and further clarify them.
Example of a Delphi Study
Problem: Expected core competencies for incoming college freshmen.
Competencies for freshmen usually include English reading and writing, basic algebra, science, communications and social sciences.
First Iteration
Please state as many suggestions as possible for the following open-ended question.
Q 1. What core competencies should be expected for incoming college freshmen?
Second Iteration
For this example, we’ll assume that the experts identified 10 difference competencies. You may want to read the article on Likert Scales to learn how to set up this iteration.
Q 2. A list of 10 competencies were identified in round 1. Please rate each competency from a scale of 1 to 10, where 1 is not important and 10 is very important.
Third Iteration
Mean scores and standard deviations from round two are provided to the experts, along with their individual rating, They are given an opportunity to change their ratings if they want to,
Q 3. Given the following mean scores, do you want to revise your original rating? This is completely optional.
History and Current Use of the Delphi Method
The Delphi study was first conceived in 1950s by Helmer and Dalkey of the RAND Corporation, an American policy think tank. It was at the beginning of the Cold War, and the technique was meant to be used to obtain ‘the most reliable’ consensus in situations where actual hard data couldn’t be obtained.
The Delphi method is named after the ancient Greek oracle, to which kings went for prophetic forecasts.
Today, the Delphi method is used in science and technology forecasting, in policy-making, and in many other fields. It’s efficacy has been questioned, but, with no viable replacement, it still is widely used.
References
Dalkey NC, Helmer O. An experimental a-application of the Delphi method to use of experts. Manage Sci 1963-9:458-67
Iqbal & Pipon-Young. The Delphi Method. The Psychologist, Vol 22 (pp 598-601). Retrieved from https://thepsychologist.bps.org.uk/volume-22/edition-7/delphi-method on June 10, 2018
Linstone, H.. (1975). The Delphi Method: Techniques and Applications. Addison-Wesley.
Thangaratinam & Redman. The Delphi Technique. The Obstetrician & Gynaecologist
2005;7:120–125. Retrieved from https://obgyn.onlinelibrary.wiley.com/doi/pdf/10.1576/toag.7.2.120.27071 on June 10, 2018.