Ethics sections are often written last and treated as a formality. Examiners and ethics committees read them as evidence of whether you understood your own design: who could be affected, how, and what you did about it.
This post covers the core principles, what each means in practice for surveys, interviews and secondary data, the approval process, and how to write the ethics section of your methodology.
The Core Principles
| Principle | What it means | In practice |
|---|---|---|
| Informed consent | Participants agree freely, knowing what is involved | Information sheet, consent form, right to withdraw |
| Voluntary participation | No pressure or undue incentive | No recruiting your own students or direct reports without safeguards |
| Confidentiality and anonymity | Participants cannot be identified without agreement | Pseudonyms, removing identifying details, aggregated reporting |
| Data protection | Personal data handled lawfully and securely | Minimal collection, encrypted storage, a retention period, GDPR compliance in the EU and UK |
| Avoiding harm | No physical, psychological, social or financial harm | Risk assessment, support contacts, the option to skip questions |
| Research integrity | Honest conduct and reporting | No fabrication, falsification or plagiarism; transparent AI use |
Informed Consent
Consent is informed only if participants understand the purpose of the study, what they will do, how long it takes, the risks and benefits, how their data will be stored and used, and that they can withdraw. Write the information sheet in plain language at the reading level of your participants.
- Surveys: a consent statement on the first page and a tick box before the first question is usually enough.
- Interviews: a signed or recorded consent, plus separate consent for audio recording and for quotations.
- Withdrawal: state until when data can be withdrawn. Once anonymised and aggregated, it often cannot be.
- Deception: only where scientifically essential, approved in advance, and followed by a full debrief.
Confidentiality and Data Protection
- Collect only the personal data your analysis needs.
- Store data on university-approved, encrypted storage, not a personal laptop or a free cloud account.
- Keep consent forms and the key linking pseudonyms to names separately from the data.
- Watch for deductive disclosure: in a small organisation, "a female senior manager in finance" may identify one person.
- State how long data is kept and how it is destroyed.
Vulnerable Groups and Sensitive Topics
Children, patients, people in care, prisoners, refugees and people in dependent relationships with the researcher need extra safeguards: guardian consent, gatekeeper approval, and closer scrutiny of risk. The same applies to sensitive topics such as health, sexuality, crime or trauma. Provide support contacts and plan what you will do if a participant becomes distressed or discloses a risk.
A methodology chapter that covers ethics properly
ThesisAI drafts full academic documents with methodology chapters that address ethics, and checks every citation against the paper it came from.
LoslegenSecondary Data and Online Research
Using existing datasets does not remove ethical obligations. Check the licence and terms of use, and whether the data were collected with consent for reuse. Public social media posts raise their own questions: quoting a post verbatim can identify its author through search, so paraphrase or aggregate where possible, and follow platform terms on data collection.
Research Integrity and AI
Integrity covers how you report, not only how you treat participants: no fabricated or altered data, honest reporting of non-significant results, proper citation and no plagiarism. If you used AI tools in your research or writing, follow your university's policy and declare the use. Our guide on using AI ethically in academic writing explains how.
Getting Ethics Approval
- Check whether your study needs review. Most studies involving people or personal data do.
- Complete your institution's application form with your supervisor.
- Attach the information sheet, consent form, instruments and data management plan.
- Wait for approval before contacting any participant. Data collected before approval usually cannot be used.
- Submit an amendment if your design changes.
Writing the Ethics Section
Weak: "All ethical guidelines were followed. Participants gave consent and their data was kept confidential."
Strong: "The study received approval from the Faculty Research Ethics Committee (ref. FREC-2026-114). Participants received an information sheet and gave written consent, with separate consent for audio recording. Interviews were transcribed and pseudonymised within 48 hours, and job titles were generalised to prevent identification within the organisation. Participants could withdraw until 14 days after their interview, before transcripts were combined for analysis. Data were stored on the university's encrypted server and will be deleted five years after the award of the degree."
Place the section at the end of your methodology, and put the information sheet and consent form in an appendix.
FAQs About Research Ethics
What are the main ethical considerations in research?
Informed consent, voluntary participation, confidentiality and anonymity, data protection, avoiding harm, care for vulnerable groups, and research integrity.
Does a literature review need ethics approval?
Usually not, because it involves no participants or personal data. Check your institution's rules, especially for systematic reviews of sensitive records.
What happens if I collect data before ethics approval?
The data generally cannot be used in your thesis, and it may be treated as misconduct. Always wait for written approval.
How long does ethics approval take?
From a few days for low-risk studies to several months for studies involving patients or vulnerable groups. Plan for it in your timeline.
Ethics is part of the design, not a paragraph added at the end. Identify who could be affected by your study, apply each principle to your specific methods, get approval before you start, and report exactly what you did.