%202.png)
Ethics Clearance Application
Builds an ethics submission that passes first time: proportionate risk, matching participant materials, honest anonymity claims and a real timeline.
Free. Works on Claude, ChatGPT, Gemini or any assistant that accepts a skill file.
What this skill does
The method, encoded.
Most ethics applications are not returned because the research is unethical. They are returned because they are incomplete or inconsistent. The consent form promises anonymity the design cannot deliver. The information sheet is written at postgraduate reading level for participants who are not postgraduates. The retention period is blank. The poster says thirty minutes and the information sheet says an hour. Each of those triggers a revision cycle, and a cycle is usually a month.
This skill treats the submission as a set of documents that must describe the same study. It builds the timeline backwards from committee meeting dates, because ethics approval is the most commonly underestimated dependency in a dissertation. It calibrates risk in both directions, since inflating trivial risk invites scrutiny and dismissing real risk invites referral. It writes information sheets for participants rather than for markers, itemises consent so people can decline recording and still take part, and insists on a truthful withdrawal cut-off.
It also handles the parts students find hardest: re-identification in small or specialist samples, the additional protections each vulnerable category requires, distress protocols with a named service and a named escalation route, whether original consent covers a secondary use, and how to disclose AI processing of participant data.
Best used for
- Assembling a first ethics submission from a finished methodology
- Writing information sheets and consent forms that match the study
- Judging what review level a study needs
- Building distress protocols with a real escalation route
- Assessing re-identification risk in small or specialist samples
- Responding to a committee's requested revisions
Typical inputs
What you give it.
The institution's own ethics form, guidance notes and meeting dates, A settled methodology, procedure, instrument and data plan, The supervisor's involvement and co-signature, Institutional data protection and retention policy, Documentation of original consent, for secondary data, A recently approved application from the same department
Typical outputs
What you get back.
A backward-planned submission timeline against committee meeting dates, A plain-language study description in the committee's terms, A four-direction risk table with mitigations and residual risk, Consistent recruitment materials with the incentive stated identically, A participant information sheet at the participants' reading level, An itemised consent form with a truthful withdrawal statement, A data management table with one row per artefact, A re-identification assessment with reporting mitigations, Named protections per applicable vulnerable category, A threshold-based distress and disclosure protocol with escalation route, An AI and third-party processing statement, A cross-document consistency matrix and submission checklist, A point-by-point response document for requested revisions
Method coverage
What the skill works through.
- Why ethics applications get returned
- Finding the route, the review level and the meeting dates
- Building the timeline backwards from fieldwork
- Assessing risk in four directions without inflating or dismissing it
- Recruitment materials and the undue inducement question
- Writing an information sheet participants can actually use
- A consent form that matches what will really happen
- Data management across the whole lifecycle
- Anonymised, pseudonymised, identifiable: using the right word
- Re-identification risk in small and specialist samples
- Additional protections for vulnerable participants
- Distress protocols with a named escalation route
- Does the original consent cover your secondary use
- Disclosing AI processing of participant data
- The cross-document consistency check
- Responding to requested revisions
Download
Free skill. One file.
Enter your email once. Every skill you download after that takes a single click.
How to install
Add the skill file and the five kernel protocols to a Claude Project, a ChatGPT Project, a Gemini Gem, or paste them at the top of any assistant conversation. Then give it your real research material, not a description of it.
Download skillQuestions
Common questions.
Why do ethics applications get rejected?
Usually for completeness and consistency rather than for ethical problems. Blank fields, missing appendices, a poster and an information sheet that disagree, a consent form promising unlimited withdrawal when the protocol anonymises at transcription, or an anonymity promise a small-sample design cannot deliver. Reading the protocol, information sheet and consent form side by side finds most of these before submission.
How long does ethics approval take?
Longer than students plan for, and the delay is structural rather than variable: committees meet on fixed dates with submission deadlines before each. Build the timeline backwards from your fieldwork start, allowing one cycle for the first submission and one for revisions. In most institutions that puts the submission four to ten weeks before fieldwork. Note also that piloting and recruitment advertising usually require approval too.
What reading level should a participant information sheet be?
Around general newspaper level unless your participants are a specialist group, and lower where literacy or language varies. Short sentences, no methodological vocabulary, no acronyms, active voice, questions as headings. A useful test is whether someone outside your field can read it once and explain what would happen to them.
Can I promise participants anonymity?
Only if the design can deliver it. In small or specialist samples, colleagues routinely identify each other from a role, a tenure and a distinctive account, even without names. The honest approach is to assess re-identification for your specific sample and reporting plan, name the mitigations you will use, and tell participants plainly that although you will not identify them, others might recognise an account, so that they can decide.
What is the difference between anonymised and pseudonymised data?
Pseudonymised data has identifiers replaced with codes while a linking key still exists, which means the people can still be re-identified and the data remains personal data. Anonymised data has no key and cannot reasonably be re-identified by anyone. Most student research is pseudonymised and describes itself as anonymised, which is inaccurate and carries different obligations.
Do I need to tell participants that AI will process their data?
Yes, if any external or automated service touches their data, including transcription, translation, coding assistance or analysis support. State in the application which tools, at which step, whether data leaves the institution's controlled environment, and whether the provider retains or trains on it. Say the same thing in plain language in the information sheet, and describe how a human verifies any AI-produced transcription or coding that feeds a finding.
Does my ethics application cover a change I made after approval?
No. A material change to the design, population, instrument or procedure requires an amendment, which is a short submission against the existing approval. Amendments are cheap; a study conducted outside its approval is a governance problem that can surface at examination and affect publication.
The skill chain
Works well with.
Research where people already are.
Analyse it where you already work.
Yazi helps researchers conduct surveys, AI interviews and longitudinal research directly through WhatsApp.
