Start with the people behind the data
A research question sets the boundaries of a project. It influences whom you approach, what you ask them to share, and which conclusions you may draw. That makes privacy and ethics part of question design, not just matters to address after choosing a method. Before settling on a question, identify whose experiences or records it would involve and how the work might affect them.
Suppose a student wants to know why classmates miss lectures. A question such as “Which students skip class because of personal problems?” invites speculation about identifiable people and potentially sensitive circumstances. “What barriers to attending lectures do students describe?” leaves room for useful answers without assuming a cause or requiring the researcher to label individuals. The second question does not eliminate ethical concerns, but it offers a more respectful starting point.
Ask only for information the question requires
A focused question can reduce unnecessary data collection. Separate the information needed to answer it from details that might be interesting but are not essential. If the aim is to understand whether a new timetable makes a service easier to use, do you need participants’ names, exact addresses, or personal histories? A question about overall experiences may be answerable through broad descriptions rather than detailed individual profiles.
Be careful with questions that join several data sources. Combining survey answers with attendance logs, for example, may reveal more about a person than either source alone. Likewise, removing names does not always make information anonymous: a distinctive role, location, or set of experiences may still point to someone. When drafting the question, consider whether you can use less detailed information, a narrower comparison, or an approach that does not require linking records.
Consider consent, power, and possible harm
Who can freely choose whether to take part? A request from a teacher, manager, or service provider may feel difficult to decline, even when participation is described as optional. A question that depends on obtaining candid accounts from people who rely on the researcher may need a different setting, a different recruitment approach, or a different question. Think about this before deciding that interviews or surveys are the obvious way forward.
Also ask what could happen if an answer were repeated, misunderstood, or made identifiable. Questions about health, finances, workplace conflict, or experiences of discrimination may carry risks beyond ordinary inconvenience. Do not assume that a topic is safe merely because it is interesting or because some information is publicly accessible. If a project involves people or their records, check the applicable institutional requirements and seek appropriate guidance before collecting information.
Build fairness into the wording
A question can create an ethical problem by embedding an unsupported judgment. “Why are older employees resistant to new tools?” treats a broad group as though it shares one attitude. “How do employees with different levels of experience describe the challenges of adopting a new tool?” makes the assumption visible and open to testing. Neutral wording does not mean avoiding difficult topics; it means allowing evidence to challenge your expectations.
Consider, too, whose perspective the question excludes. If you ask only supervisors whether a workplace policy helps staff, you may miss the experiences of people affected by it. On the other hand, including more participants is not automatically better if it exposes people to needless demands or risks. Aim for a question whose scope supports a fair answer without asking anyone to disclose more than the purpose warrants.
Test the question before committing to it
Write a short plan alongside your draft question: what information would answer it, how would you obtain that information, who might be identifiable, and how would you present the result? This exercise can reveal that an apparently simple question requires intrusive details. Revise the question if you cannot explain why particular information is needed or how you would avoid exposing participants in the final account.
A practical final check is to imagine describing the project to someone whose information it would use. Could you explain the question plainly, including what you would not collect and what the findings could reasonably show? If not, narrow the scope or reconsider the method. Ethical framing is not a promise that every risk has disappeared. It is a way to make the project’s purpose, limits, and responsibilities clearer from the start.
