Ethics and practical limits are part of the design

By UK GCSE Sociology · independent study · myAudience

Original OpenAI imagegen conceptual teaching illustration, generated 4 October 2026; not observational evidence. · AI-generated original asset for this independent guide; no third-party artwork intentionally reproduced.

Ethics and practical limits are part of the design

Ethical evaluation asks what the research could do to participants and others, how obligations conflict and whether the design is justified. It is not a final signature added after choosing a method. The British Sociological Association's professional guidance treats responsibility as continuing throughout research, rather than a recipe that resolves every dilemma. This lesson applies that principle to fictional proposals. Consent, confidentiality and harm are central requirements. A design can be technically informative but ethically unacceptable; conversely, a well-intentioned project can still collect misleading evidence or expose people through careless reporting.

Informed consent requires an understandable explanation of purpose, participation, use of evidence and relevant limits, with a meaningful voluntary choice. A gatekeeper may allow access to an institution but cannot simply consent for everyone within it. Pressure from a teacher, employer or researcher can weaken voluntariness even when a form is signed. Consent can need revisiting when a project changes. A withdrawal plan should explain when withdrawal is possible and any limits after data are irreversibly anonymised or combined. Avoid absolute promises that the design cannot honour. Silence or compliance is not sufficient evidence of freely chosen participation.

Confidentiality concerns controlling disclosure of information entrusted to the researcher. Anonymity means identity is not known or cannot be connected to the data in the relevant process; replacing names with codes is usually pseudonymisation if a linking key remains. A rare combination of details can identify a person even without a name. In a small school, a quotation about a unique incident may be recognisable. Collect only necessary information, restrict access, set a retention plan and consider how publication could reveal identities. These design principles do not replace current institutional or legal requirements, and this lesson makes no promise of absolute secrecy.

Harm can be social or psychological as well as physical. Questions about family conflict can create distress, expose relationships or cause consequences if answers reach a powerful person. The researcher should assess foreseeable risks, offer omission and stopping options, avoid unnecessary detail and plan support and appropriate escalation within competent oversight. Sensitive disclosures can create duties that conflict with confidentiality. Explain limits in advance rather than promise secrecy and later reverse it. Deception and covert access require especially strong scrutiny; educational curiosity is not a justification for involving others without appropriate consent.

Practical issues shape who is included and what is possible. Time and cost may limit interviews, translation, travel or analysis. Access can depend on gatekeepers who prefer favourable participants, introducing selection bias. A brief online questionnaire may be feasible but exclude some users or offer poor privacy. The answer is not simply to choose the cheapest method. Revise the aim, restrict the inference honestly or use an alternative design that avoids unnecessary personal data. A feasible study should have resources to protect participants and analyse evidence competently; collecting more material than can be handled responsibly is not a success.

The illustration depicts an invitation, an open route to leave and restricted storage. It is a conceptual prompt, not proof that consent is informed, refusal is safe or a cabinet guarantees confidentiality. Assess those conditions through the design and its actual operation. For all activities in this batch, evaluate fictional studies and synthetic records; do not collect classmates' family histories or covertly record people. Ethical reasoning is assessed through how a safeguard changes a specific risk and what remains unresolved. A generic list of consent, confidentiality and harm does not explain whether a proposal is acceptable.

Worked example: Original fictional proposal: a teacher asks every learner to describe family arguments on a named questionnaire, says the work will count toward participation, and promises complete secrecy while keeping forms in an open tray. Redesign or reject the study, explaining ethical and practical reasons.

Identify coercion and sensitivity: assessment pressure undermines voluntariness, and family conflict can cause distress or expose others who have not consented.

Challenge the confidentiality promise: names and open storage permit disclosure; rare details may remain identifiable even if names are removed.

Prefer a fictional-case exercise for this teaching aim. If genuine research were necessary, competent oversight would have to establish necessity, consent, safeguards and realistic disclosure limits.

Explain that safeguards cost time and resources. If these cannot be provided, narrow or abandon the project rather than collect unsafe evidence.

Answer: The proposal should not proceed as a classroom participation task. The teacher's assessment power makes refusal costly, the topic invites sensitive disclosures, and the storage contradicts the secrecy promise. For learning research design, fictional cases achieve the aim with much less risk. A genuine study would need an independently justified purpose, appropriate oversight, voluntary participation outside assessment pressure, minimised collection and clear disclosure and withdrawal limits. Removing names alone would not resolve recognisable details or relational harm. Practical feasibility includes resources for protection and analysis, not merely the ability to distribute a form.

Check your understanding: Extended transfer: a fictional youth service wants candid feedback from users, but its manager offers to select only 'reliable' participants and sit in on interviews. Evaluate access, sampling, consent and confidentiality, and propose a defensible alternative. Indicative rubric: 12 points: two connected ethical risks 4, sampling/access analysis 3, alternative 3, reasoned go/no-go judgment 2.

Answer: Manager selection may exclude critical users and shape whose accounts appear. Manager presence can discourage criticism and make refusal unsafe. A competent independent process could offer accessible voluntary invitations to a broader eligible group, private interview arrangements, clear use/disclosure limits and reporting that avoids identification. If independence or safeguards cannot be established, the feedback study should be narrowed or postponed. Permission from the manager alone is insufficient.

Why: Full credit links power relationships to likely response and disclosure risks. 'Get consent' without explaining freedom to decline is incomplete. The alternative must address selection as well as interview privacy and should acknowledge that quotations from a small service can be recognisable. A justified decision to reject the design can earn full credit; accumulating data is not the overriding goal.