How to Use the Modified Delphi Protocol Prompt to Plan a Consensus Study Reviewers Can Trust

A Delphi study looks simple from the outside: send experts a survey, repeat it, and report what they agree on. Reviewers know where these studies go wrong. The consensus threshold appears after the results, one stakeholder group dominates the panel, attrition between rounds goes unreported, and the feedback panelists saw is never described. The Modified Delphi Consensus Study Protocol Builder: Panel Eligibility and Recruitment, Round 1 Item Generation, 9 Point Rating Rounds, A Priori Consensus and Stopping Rules, Controlled Feedback Reports, and a Reporting Checklist prompt helps you write the protocol before Round 1 opens, so every decision is fixed and defensible.

What the prompt produces

1. A protocol summary stating the aim, why a modified Delphi fits your situation, and what the steering group does.
2. A panel section with checkable eligibility criteria per stakeholder group, recruitment routes, and how group balance is protected.
3. A Round 1 design using a 9 point scale with three bands, an unable to score option, and free text for new items, coded by two researchers.
4. A consensus definition for consensus in, consensus out, and no consensus, labeled as a team choice made before data collection.
5. Rules for later rounds, including who is invited and when the study stops.
6. A controlled feedback report layout per item: median, interquartile range, band percentages, the panelist's own previous score, and results by group.
7. Attrition and bias controls covering response rates, reminders, anonymity, and missing ratings.
8. A build note for your survey platform.
9. A reporting checklist mapping each decision to your methods section.

How to fill the inputs

ResearchQuestion should say what the panel decides and what the result will be used for. A list for a curriculum and a list for a clinical guideline need different panels.

PanelPlan names stakeholder groups, eligibility criteria, and a target per group. Write criteria you can verify on a screening form.

SeedItems says where your starting items come from and how many there are. This is what makes the Delphi modified.

RatingPlatform is REDCap, Qualtrics, or another tool. The build note adapts to it.

Timeline and EthicsContext shape how many rounds are realistic and what consent wording you need.

Reading the example output

The example plans a study on competencies for new graduate nurses in rural emergency departments:

• The modified design is justified. Forty two statements already exist from a scoping review, so Round 1 rates them instead of asking an open question.
• Eligibility is concrete, such as at least two years in an educator or charge role at a rural ED.
• Panel size is handled honestly. The protocol states a target of 15 per group, notes there is no single correct size, and reports results by group.
• The consensus table is dated. Thresholds are written as the steering group's choice, with a rationale paragraph to be fixed before launch.
• The feedback mock leaves blanks for values that do not exist yet, instead of inventing numbers.
• The REDCap note is specific: one instrument per round, a survey queue, and previous scores piped into the next round.

Tips for better results

• Ask for the screening form questions as a follow up, built directly from the eligibility criteria.
• Pilot the Round 1 survey with two or three people outside the panel to check wording and time.
• Decide in advance whether Round 1 non responders are invited later, and write it down.
• Keep a dated decision log. It turns into your methods section.

Mistakes to avoid

• Do not choose or adjust thresholds after you see the ratings.
• Do not report only the final list. Show how many items entered and left each round.
• Do not let panelists see who said what. Anonymity is part of the method.
• Do not ignore the reporting guideline your target journal expects.

Who it is for

Graduate students planning a thesis Delphi, clinician researchers building a consensus list for a guideline or curriculum, education researchers designing competency frameworks, and research coordinators setting up multi round surveys.

Related PromptDig links

Open the Modified Delphi Consensus Study Protocol Builder: Panel Eligibility and Recruitment, Round 1 Item Generation, 9 Point Rating Rounds, A Priori Consensus and Stopping Rules, Controlled Feedback Reports, and a Reporting Checklist prompt and describe your question and panel. For more research design prompts, Browse more prompts. If you use a methods prompt that keeps your studies tidy, Share a prompt.