A Delphi Survey is a series of questionnaires that allow experts to develop ideas about potential future developments around an issue. The questionnaires are developed throughout the process in relation to the responses given by participants.
Delphi Surveys are used to gather collective forecasts through questionnaires about likely or possible developments in particular areas. Delphi Surveys can be carried out face to face, online or by post. In online versions, participants are given their own login and password to access the site. This is useful when the expert participants are very busy people. The technique aims to derive the benefit of the opinions of a group of experts, while avoiding the disadvantages of 'group-think' and group dynamics where certain individuals dominate the discussion. The process takes place in a number of stages:
• The first questionnaire either asks the participants to individually identify issues and generate as many ideas as possible or to answer more close ended questions such as the likely dates for specific developments.
• The second questionnaire anonymously feeds back all the ideas and forecasts sent in the first round to all participants. This questionnaire also provides space for participants to refine each idea, comment on their strengths or weaknesses and to suggest new ideas.
• An additional questionnaire then summarises the input from the second questionnaire and asks for further clarification, strengths, weaknesses, and new ideas. This stage can be repeated as many times as necessary until consensus on key points is reached.
• The end product is either a consensus amongst the participants on likely and possible future developments, or a wide range of possible developments and their relative strengths and weaknesses.
• People are chosen to participate in a Delphi Survey because of their expertise or specific knowledge of the topic area. This expertise does not have to be academic and could instead be the knowledge of an experienced service user.
• It is important that the participants reflect a wide range of views and perspectives so that there will be a meaningful exchange of ideas.
• A Delphi survey may involve 10 to 50 participants.
• As the Delphi survey is often run as a remote method, it does not have costs for venues or transport. However, the cost of sending surveys by post or setting up the survey online remains.
• It can sometimes be suitable to provide an incentive for the participants, such as a small payment.
Approximate time expense
• As the participants in a Delphi Survey are often very busy, it is sometimes necessary to allow them plenty of time (weeks, if not a month) to complete each round of surveys.
• It is quite time consuming to remind and encourage participants to complete the surveys.
• It is important that participants are aware of the level of time commitment required in advance of joining as the many rounds can be demanding.
• Participants can complete the surveys in their own time.
• People can freely express their opinions and critique ideas as anonymity is maintained throughout the process.
• Discussions are not dominated by one participant's authority or personality.
• The series of questionnaires generates a much wider set of ideas than each individual participants would generate by completing them independently.
• Results will not be skewed towards influential individuals.
• It may be hard to coordinate and motivate the group.
• Results may be influenced by the collective bias of the participants.
• The surveys are limited to forecasting future developments and are unable to recommend action.
• Will not build relationships or generate a dialogue between participants.
The Delphi Method was developed by the Rand Corporation for the American military as way of forecasting future technological developments. Since then, it has been widely used in the field of science and technology forecasting to combine expert opinion on the likelihood and impact of technological developments, particularly on warfare.
Image by Thomas Kunsch©
Gather individual pre-existing opinions
Gather informed and considered opinions (deliberation)
Generate new ideas (innovation)
Create a shared vision amongst participants
Number of participants
A Group which broadly reflects the Demographic make up of a certain community or population
Statistically representative sample of a population
Representatives of wider interest groups (stakeholders)
Level of awareness and interest
participants know about some aspects/can roughly articulate some interest
Crime and justice
Culture and arts
Environment and climate change
Health and well-being
Housing and Planning
Science and technology
Limit search to...
... online processes
Level of involvement
Children and young people
Ethnic minority groups
Groups with low levels of literacy/confidence
People with learning difficulties