‘More of an art than a science’? The development, design and mechanics of the Delphi Technique

• Published in: Research in social and administrative pharmacy Vol. 18; no. 1; pp. 2230 - 2236
• Main Authors: Drumm, Sarah, Bradley, Catriona, Moriarty, Frank
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.01.2022
• Subjects: Consensus; Delphi; Delphi Technique; Expert; Group judgement; Humans; Reproducibility of Results; Research Design; Study design; Surveys and Questionnaires; Study design; Group judgement; Delphi; Consensus; Expert
• ISSN: 1551-7411; 1934-8150; 1934-8150
• DOI: 10.1016/j.sapharm.2021.06.027

The Delphi Technique is a group judgement method which is typically used to reach agreement from a group of people with expertise in a particular area. It is an iterative process where panel members complete questionnaires over several rounds, often rating their agreement/disagreement against a statement, with changes made in later rounds based on the feedback received. It has been used widely in pharmacy-related studies relevant to education, research and practice. This paper provides a critical analysis of the various design choices which researchers may consider when planning a Delphi namely the panel of participants, the use of the Likert scale, the effect of feedback, what constitutes consensus and the number of rounds. It also gives an overview of the development and origins of the Delphi, and discusses the advantages and disadvantages of the technique. Advantages include that the Delphi can be conducted with panel members in different geographical locations in their own time, however the technique can therefore take longer to conduct and lacks face-to-face discussion. Patient experts may be less comfortable participating in a relatively complex survey, however the anonymous nature of the process can be more inclusive in allowing participants to feedback candidly. This paper shows the importance of careful planning of the design choices to ensure the reliability and validity of the Delphi.