Consent requirements are set by the ethics committees governing your study and by the law of the country where the research takes place — often both. This article is about making consent meaningful in practice; it does not replace the requirements of your approvals.
Your ethics application is approved. The consent form has been reviewed, translated, and signed by every participant. On paper, consent is complete.
Whether the people who signed it understood what they were agreeing to is a separate question, and across a language and context difference the answer is often no.
Where approved forms fail as consent
Four gaps, all common:
The form is written for the committee, not the participant. Ethics committees want completeness, which produces long documents in formal register. Completeness and comprehension work against each other past a certain length.
Translation is literal. A technically accurate translation of a sentence written in academic English produces a sentence nobody speaks. Words like "anonymised", "data controller", "withdraw at any time without penalty" often have no everyday equivalent.
Signing is habitual. In many contexts, a form presented by someone who appears official is signed without reading — not from carelessness, but because that is what one does with forms.
Refusing feels impossible. Where the researcher arrives with local authority, or the participant is a patient, student, or employee in a hierarchy, "you may decline" is formally true and practically hard.
The last gap is the one that most needs deliberate work, because no amount of rewriting the form addresses it.
Translation: do it properly
Three steps, and the second is the one usually skipped:
Translate for comprehension, not word-for-word. Brief the translator that the goal is that a person with basic schooling understands what is being asked — not that the English is faithfully rendered.
Back-translate. A second translator, who has not seen the original, translates back into English. Comparing the two catches meaning that shifted — and it catches it reliably.
Test with two or three people from the actual participant group. Have them read it and then explain in their own words what the study wants from them. This takes an afternoon and finds problems no review process catches.
Budget and schedule for all three. A translation done the week before fieldwork starts will be literal, untested, and probably wrong in places that matter.
When literacy is limited
Written forms are not the only valid route, and in some settings they are the worst one.
Practical approaches, subject to what your approvals permit:
Read the information aloud and discuss, rather than handing over a document.
Use a witnessed process where an independent person confirms the information was given and understood — common in ethics frameworks for exactly this situation.
Recorded verbal consent where written signatures are inappropriate or risky for participants.
Visual materials explaining the study.
Whichever route, build in a comprehension check: ask the person to say back what will happen and what they can do if they change their mind. If they cannot, consent has not happened yet regardless of what has been signed.
Working through interpreters and local assistants
This is where much of the real consent process happens, and it is usually under-prepared.
Four things:
Brief them on the science, not just the words. Someone who understands why the study is being done can answer the questions participants actually ask.
Agree how to handle questions they cannot answer — bring them to you rather than improvise.
Be explicit that they must not encourage participation. A local assistant who wants to be helpful can turn an invitation into a request that is hard to refuse, without intending to.
Consider who they are in the community. An assistant with authority over participants — a teacher, a health worker, a local official — changes the voluntariness of the whole interaction. Sometimes the right choice is someone with less standing.
Consent as a process, not a moment
Three practices that matter more than the initial signature:
Re-confirm at later contacts. In longitudinal work, a brief check at each round — still willing to continue? — is both more ethical and more accurate than one signature at the start.
Make withdrawal genuinely available. Give a name and a contact that works locally, in the local language. A university email address in another country is not an available route.
Be explicit about what withdrawal means for data already collected. Participants often assume it can all be deleted, and if that is not the case, they need to know before they agree.
Five things to check before the first interview
Does the form say plainly what will happen to the person, in what order, and for how long?
Has it been back-translated and tested with real participants?
Can a participant refuse without any consequence they can perceive — and is there anyone present whose presence makes refusal harder?
Is there a working local contact for questions and withdrawal?
Do the local assistants know what they must not say?
Working through these five takes a day and changes consent from an administrative artefact into the protection it is meant to be.
Why do approved consent forms often fail?
They are written for the committee rather than the participant, translated literally, signed out of habit, and presented in situations where refusing feels impossible.
What translation step is usually skipped?
Back-translation by a second translator who has not seen the original — comparing the two reliably catches meaning that shifted.
What should be done when literacy is limited?
Read the information aloud and discuss it, use a witnessed process or recorded verbal consent where permitted, and always include a comprehension check.
What must local assistants be told?
That they must not encourage participation — someone wanting to be helpful can turn an invitation into a request that is hard to refuse.
