A lecturer wants to know how students spend their independent study time, so she posts a survey link in the group chat of the class she teaches, with the line “would really appreciate your help!”. Ninety per cent respond. The rate is suspiciously good, and that is exactly the problem: when the person asking is also the person grading, voluntary is no longer quite voluntary.
Human participant research is not just drug trials and clinical interventions. A questionnaire, an interview, a recording of a seminar all raise ethical questions, and many journals now ask for an approval reference at the submission stage.
When you need ethics review
Rules differ by country and institution, and the committee may be called an IRB, a research ethics board or a research ethics committee. The broad pattern is widely shared:
| Type of study | Usually requires |
|---|---|
| Biomedical interventions, biological samples, clinical trials | Full review |
| Interviews on sensitive topics (mental health, violence, illegal behaviour) | Full review |
| Surveys of minors or people in dependent relationships | Review, often with guardian consent |
| Anonymous adult surveys on non-sensitive topics | Often expedited review or exemption, but the committee decides |
| Analysis of fully anonymised public data | Usually exempt |
The most common misunderstanding: you cannot declare your own study exempt. The committee does that. When a journal asks for ethics approval, “not needed, it was only a survey” rarely suffices; a formal exemption letter does.
Consent forms written for real readers
A consent form should be readable in a few minutes by the participants themselves. At minimum it covers:
- The purpose of the study in plain language.
- What participants will do and how long it takes.
- Possible risks and inconveniences, including small ones such as discomfort with certain questions.
- The right to decline or withdraw at any time without consequences.
- Where data will be stored, for how long, who can see it and whether it will be shared publicly.
- A contact for questions, plus an independent contact, usually the ethics committee.
The classic failure is a five-page form full of legal language. People sign without reading, which means the consent is not truly informed. One clear page beats five exhaustive ones.
Surveying your own students
The teacher–student relationship is a dependent one. Committees typically want to see steps that reduce pressure:
- Someone else, such as a teaching assistant or colleague, distributes and collects the survey rather than the person grading the course.
- Data collection after grades are released, or a commitment not to open responses until the module ends.
- No course credit for participation; if there is an incentive, an equivalent alternative for those who opt out.
- Genuinely anonymous collection, without student IDs or emails unless essential.
The friendly group-chat request should become a neutral invitation from someone else, stating clearly that not taking part has no effect on anything.
Anonymous, pseudonymised, confidential
Three terms that get used interchangeably:
- Anonymous: nobody, including the research team, can link a response to a person.
- Pseudonymised: names are replaced with codes, but a key exists and is stored separately. Under data protection frameworks such as the GDPR this is still personal data.
- Confidential: the team knows who said what but promises not to disclose it.
A survey without names can still identify people. “Female, 45, head of department, at a small college” may describe exactly one person. In qualitative reports, generalise background details (“a faculty-level manager”) and strip identifying specifics from interview quotes.
Storing and sharing data
- Keep identifiable raw data in password-protected institutional storage, not on a personal phone or in a chat app.
- Store the name-to-code key separately from the analysis file.
- If you might share data openly later, say so in the consent form from the start. Without that, sharing later is difficult to justify.
- Delete audio recordings after transcription and checking, if the consent form promised it.
What journals ask, and how to answer
An ethics statement should give the committee name, approval reference, approval date, how consent was obtained (written, electronic, recorded verbal) and how minors were handled if relevant. For example: “The study was approved by the Research Ethics Committee of [institution] (ref. …, date …). Participants gave electronic consent before starting the survey.”
A two-week preparation plan
- Week one: finalise the protocol, questionnaire or interview guide, information sheet and consent form, and a data management plan.
- Ask someone outside your field to read the consent form and tell you what they understood.
- Week two: submit and respond to queries. Do not start collecting data before written approval; journals do not accept retrospective approval.
Your next step: find your institution's ethics committee page, download its consent template and its meeting calendar. Knowing the calendar early can save you a month of delay.
Câu hỏi thường gặp
Do anonymous surveys need IRB approval?
It depends on your institution, but the decision to exempt a study belongs to the committee, not the researcher. Many journals ask for a formal exemption letter.
Can I get ethics approval after collecting data?
Almost never. Committees and journals generally reject retrospective approval, so secure written approval before any data collection.
What do I need to research students under 18?
Usually parental or guardian consent, the young person's own assent, and permission from the school. Committees tend to review these applications more closely.
What is the difference between anonymous and pseudonymised data?
Anonymous data cannot be linked to individuals by anyone. Pseudonymised data replaces names with codes but a key exists, so it is still treated as personal data.
Can I give course credit for taking part in my study?
It is best avoided. If there is any reward, offer an equivalent alternative to students who do not wish to participate so that participation stays voluntary.
