One of my main concerns is how many expert opinions I need as a research methodology to consider a conceptual framework in education. Any ideas is appreciated.
Why would Delphi be the best method? 5-6 experts may be enough. Check if the authors are here on RG:
Information & Management
Volume 42, Issue 1, December 2004, Pages 15-29📷
The Delphi method as a research tool: an example, design considerations and applications
ChituOkoli Suzanne D.Pawlowski https://doi.org/10.1016/j.im.2003.11.002
It is unlikely that you are going to find any statistical basis for setting the sample size, so I think your best bet would be to do a search of recent Delphi studies and determine the typical size range for the number of participants.
Hi Ahmad. As David says, there are no strict guidelines for sample size, (true of many qualitative research methods). In practice, I have run Delphi Panels with the number of experts ranging from about 8 to 15. The important thing is to recruit people with expert knowledge and an interest in the subject. A slightly larger panel size is better as you may get some experts who drop out between rounds of the Delphi, purely due to availability.
A larger panel size will also add robustness to the average or mean rating scores (and standard deviations) if you are asking experts to, for example, rate interventions for effectiveness on a scale of 1 to 5. I have come across a couple of papers describing the use of Delphi Technique in Healthcare Education, for example this one which used a panel size of 13:
Article Using the Delphi technique to develop standards for neonatal...
https://www.omicsonline.org/open-access/utilizing-the-delphi-survey-approach-a-review--2167-1168-1000246.php?aid=51677
http://theijes.com/papers/v3-i4/Version-4/B03404008013.pdf
Thank you Jenny, Shahnaz, David and Beatrice. Great info.
One of my concerns is that the participants may be from different backgrounds. However, I believe in Delphi we want the "questionnaire" to be the same for all participants. Is that correct? Are we able to customize based on the participants background or is it best to stick to the model I am trying to get info and ask the same set of questions to all participants specific to that model? Can we also have qualitative interviews?
Yes it is important that everyone is asked the same questions to provide the mean ratings across all respondents.
Qualittive interviews would be separate from the Delphi Panel exercise.
Thanks Jenny. The participants are anonymous to each other during the study. Can their names be public after the study?
This link is interesting as it considers Delphi from a Qualitative approach with an example: https://nsuworks.nova.edu/cgi/viewcontent.cgi?article=2974&context=tqr
Just to confirm, the anonymity part is for the participants to participants only. I assume the participants would know the researcher, correct?
Also, I assume the researcher can explain the model in a face-to-face interaction with individual participants (separate sessions for each participant), and then start the Delphi rounds. Would that be considered OK with the Delphi approach?
Yes, the participants and researcher would be known to each other through email communications. Explanation of the Delphi model and process is normally done during the first round in the email correspondence and questionnaire document.
In response to your earlier query, if possible I think it would be best to maintain anonymity even after the study is completed, particularly if you have promised anonymity when recruiting participants.
Thanks Jenny. As I am asking the experts to consider a conceptual framework I have designed, I want to have a one on one face to face w each of them to explain the framework first. And then ask for their opinions based on Delphi. Would that work?
I guess it depends whether your conceptual framework is too complicated to explain in an email and summary document. If it is, then doing a series of interviews or running a focus group would probably be a better way of collecting opinion.
It won't be easy to explain just n emails or documents. That is why I am thinking of Delphi mixed with interviews.
I am thinking of Delphi as it puts a more rigorous scientific feel to it, more formal.
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology