Agreeing Consent and Attribution Before a Research Interview
Discuss participation, recording, attribution and later use before an interview, keeping those permissions separate from the project invitation.
Consent and attribution should be settled in writing before a research interview begins, and they should be settled as two separate permissions. The first covers what participation permits: the purpose, audience, outputs, retention and any later re-use of the contribution. The second covers identity: whether the interview is recorded, whether the participant's words may be quoted, and whether those quotations carry a name, a role, a general description or no identifier at all. Treating these as one signature is the most common source of disagreement after publication, because a participant who is content to inform a study may still object to being named in it, or to being named in a different document three years later.
What participation permits
A consent conversation that stays at the level of "a research interview about skills policy" leaves the participant unable to judge what they are agreeing to. Useful scope statements are specific about four things.
Purpose and output. Name the study, the commissioning organisation and the form of the output: a report, a briefing, an article, an internal memorandum, a dataset with accompanying analysis. If the output form is not yet fixed, say what the realistic options are rather than describing the project in general terms.
Audience and publication route. A contribution to a closed internal review and a contribution to a public website carry different exposure. State which applies, and whether a press summary, oral briefing or public event is contemplated.
Later use. Consent given for one report does not automatically extend to teaching materials, a subsequent book, an archive deposit or a proposal. If later use is plausible, describe the categories now and agree whether fresh permission will be sought each time. A narrow permission that is respected is worth more than a broad one that the participant did not understand.
Choice within the interview. Participants should know that they can decline an individual question, ask for a remark to be treated as background, or stop the interview, without withdrawing from the study altogether. These are operational choices rather than new consent, and they are easier to exercise when they have been described in advance.
Recording, transcription and quotation are separate decisions
Recording, quotation and naming carry different consequences and should be documented separately, even when they are agreed in the same conversation.
| Permission | What it authorises | What it does not settle |
|---|---|---|
| Audio or video recording | Capture and transcription for analysis | Publication of clips; how long the file is kept; who may listen to it |
| Anonymised quotation | Use of the participant's words without a name | Whether a role, sector or location may be given; re-identification in a small field |
| Named quotation | Quotation attributed by name and, typically, role | Use in media work, marketing or later publications outside the agreed scope |
| Paraphrase without quotation | Use of the substance of a view in findings | Whether the source of the argument may be described, even without words |
Two practical consequences follow. First, a participant may reasonably accept recording while refusing named quotation; the agreement should make that combination easy to record. Second, anonymity is a spectrum rather than a setting. In a specialised field, a job title and a policy area can identify someone as effectively as a name, so the conversation should cover what contextual detail may accompany an unattributed quotation.
Withdrawal limits that need explaining
Withdrawal is the point at which consent documents most often overpromise. A participant who is told they may withdraw "at any time" will reasonably expect their contribution to disappear from the finished report, which is rarely possible once analysis has begun. Explain the boundary before the interview rather than after it.
Points worth stating plainly:
- The date or stage after which withdrawal of the contribution from the analysis is no longer practicable, for example once a draft has been circulated or a synthesis completed.
- What remains possible after that point: correcting factual errors, asking that a specific quotation not be used, or noting a change of position.
- What happens to recordings and transcripts, including deletion or retention periods and who holds the files.
- Whether withdrawal ends participation only, or also any earlier permission to be named.
- That withdrawal does not apply to material already published, though a correction or clarification may be sought.
Where a participant asks for a stricter arrangement than the standard one, it is better to record the exception than to give a verbal assurance that the project cannot honour.
Confidentiality limits
Confidentiality commitments should describe their own edges. Research teams may be subject to institutional ethics review, funder reporting or legal obligations that limit what can be kept back, and the participant is entitled to know that those obligations exist. Confidentiality of content should also be distinguished from confidentiality of identity. An interview can be quoted anonymously while the fact of participation is public, or both can be restricted. The researcher should describe the applicable obligations as they understand them and point the participant to the relevant institutional or legal source rather than offering a personal guarantee.
The UK Research Integrity Office's Code of Practice for Research identifies research governance, ethics and data management as areas of good practice and is a starting point for checking the appropriate institutional process in the United Kingdom; it is general guidance rather than a certification of any particular project. Researchers working under other frameworks should check the consent requirements that apply to their institution, funder and jurisdiction, including data protection obligations, before the interview is scheduled.
A discussion checklist
The following worksheet is a suggested working format for the pre-interview conversation. It is not a validated instrument, not an institutional requirement and not a substitute for the consent documentation the researcher's own organisation requires. Its purpose is to make sure each decision is taken explicitly and recorded in the same place.
| Domain | Question to settle | Recorded as |
|---|---|---|
| Purpose | Which study and which output is this contribution for? | Named study and output type |
| Choice | Which questions may be declined, and what are the withdrawal terms and cut-off? | Ground rules and the withdrawal statement |
| Recording | Is the interview recorded, who transcribes it, how long is it kept? | Recording decision and retention period |
| Audience | Where will the output appear, and is media or public presentation contemplated? | Publication route |
| Quotation | Anonymous, named, role only, or no quotation at all? | Quotation permission with identifier level |
| Confidentiality limits | What must be excluded, and what obligations limit confidentiality? | Exclusions and stated limits |
A short written summary of the answers, sent to the participant before the interview, gives them a chance to correct a misunderstanding while correction is still cheap. Where an invitation comes from Policy Expert Network, the invitation is described as setting out the question, intended use, deliverables, preparation, timetable, attribution and confidentiality limits, and an invited expert can accept, request changes or decline without losing future eligibility. The same principle applies to any research engagement: the project agreement, rather than a general description of the programme, should specify the actual scope.
Worked example: a fictional checklist in use
The following example is fictional. It illustrates how the worksheet might be completed; the periods and arrangements shown are particular to this invented case and imply no standard.
A researcher at a fictional institute is interviewing a former local authority analyst about the design of a devolved skills programme. The output is a report for a fictional regional board, to be published on the board's website and summarised in a four-page briefing.
| Domain | Completed entry |
|---|---|
| Purpose | One report on programme design and one summary briefing for the regional board; no use in teaching or commercial work without a new agreement |
| Choice | Individual questions may be declined; withdrawal of the contribution is possible up to 14 days after the interview and before a draft is circulated |
| Recording | Audio recorded; transcribed by the researcher; recording deleted 12 months after publication |
| Audience | Board website and printed briefing; no press release quotations without separate agreement |
| Quotation | Named quotation with job title approved; no quotation describing the participant's former employer's internal process without a further check |
| Confidentiality limits | No confidential material to be discussed; where a third party could be identified, the quotation is held or generalised |
Two entries do the most work. The withdrawal entry sets a date that can actually be honoured, and the quotation entry distinguishes a general attribution from a specific and more sensitive one. If the board later wants the interview recorded for a public podcast, that is a new use and requires a fresh conversation rather than an extension of the original consent.
Reopening the agreement
Consent and attribution are not settled permanently by a single exchange. A new audience, a move from anonymous to named quotation, an archive deposit, a change of commissioner or a request to reuse material in a bid all warrant a return to the participant. Recording the original terms in a form that can be found again (attached to the interview file rather than held in correspondence) makes that return straightforward and keeps the published output consistent with what was agreed.
Before accepting an interview date, read the attribution and confidentiality terms in the invitation, note any point that is narrower or broader than you would accept, and put your proposed permissions in writing so that the researcher can confirm or amend them in reply. The guide on what to check before accepting an expert project sets out the surrounding questions of scope and fee, and the organisations page describes how research engagements are structured.