How to turn a broad topic into a focused research question

A topic tells a reader what you are interested in. A research question tells them what you intend to find out. Moving between the two often takes more work than expected, especially when the subject seems to offer dozens of possible directions.

You do not need to settle every methodological detail immediately. You do need a question that is specific enough to guide your reading and realistic enough to investigate with the time and access available to you.

Name the problem before choosing the method

The power-station example below is illustrative; it is not a report of a client project or a completed study.

Start with a short explanation of what is unclear. For example, “I want to study quality management” describes an area. “I want to understand how staff perceive the relationship between quality practices and operational performance in one power station” identifies a more specific question.

The second version still needs refinement. However, it already gives you something to discuss with a supervisor: the practices, the outcome, the participants, and the setting.

Avoid choosing a complex analytical technique just because it sounds impressive. A method should help you answer the question you have, using evidence you can obtain.

Set three boundaries

Write down three decisions in ordinary language:

1. Who or what will you study? This might be a defined group of employees, a set of organisations, or a collection of documents.
2. Where will the study apply? One workplace, one city, or one industry may be a workable starting point. A wider setting usually requires wider access.
3. What relationship or experience matters? Select the main issue you want to understand, rather than collecting every possible factor.

These decisions should narrow the work without removing the reason for doing it. If your topic becomes so narrow that the answer would have little meaning, return to the problem and reconsider which boundary is most useful.

Read enough to check the problem

Before treating your question as a research gap, read a small, relevant set of studies. Note what they examined, which evidence they used, and what their findings leave unresolved. A study set in a different country or organisation does not, by itself, establish that your version makes a useful contribution. Explain why the new setting or perspective could matter.

Keep a short record of the searches and sources you used. If you cannot find a particular study, say that it was not found in your search; avoid claiming that no research exists. George Mason University’s guide to developing a research question also places preliminary reading before committing to a final question.

Check the evidence you can actually access

A promising question may still be impractical. Before committing to it, identify what evidence would be needed and how you would obtain it.

If you plan a survey, think about who can answer the questions, whether they have the relevant knowledge, and whether you can reach enough suitable participants for the intended analysis. If you plan interviews, consider access and the time needed to arrange, conduct, and analyse them. If you plan document analysis, check that the necessary material is available.

Access is not the same as permission. Institutional requirements, organisational approval, and research ethics procedures need to be considered before data collection begins. Discuss the requirements for your particular project with your supervisor.

Write a question that matches the evidence

Compare these two versions:

• “How does quality management improve power station performance?”
• “How are employees’ perceptions of quality management practices associated with their perceptions of operational performance in the selected power station?”

The first version suggests a causal conclusion. The second is more closely aligned with a study based on employee perceptions collected at one point in time. It is longer, but it is also more precise about what the evidence could show.

This does not mean every research question should use the same wording. The useful principle is to avoid promising more than your design can establish.

Check whether the question is ready to discuss

Use this short checklist before taking the question to your supervisor:

• Can you explain what is uncertain and why answering it would be useful?
• Are the main concepts clear enough to define through the literature?
• Have you identified who or what could provide the evidence, and how you would obtain permission?
• Does the wording fit what that evidence could establish?
• Could the work fit the time, word limit, and resources available?

For the illustrative power-station question, “quality practices” and “operational performance” still need definitions. If the intended evidence is employees’ perceptions, the conclusions must remain about those reported perceptions. Access to one workplace also does not automatically make the findings representative of other power stations.

If access falls through, revisit the question and its evidence requirements together. Replacing a survey with interviews changes what you can investigate; it is not simply a smaller version of the same design.

Use the question to organise your next steps

Once you have a working question, use it to decide what to read, what concepts to define, and what information to collect. Your objectives should describe the tasks needed to answer it. Sections that do not contribute to those tasks may be outside the current scope.

Take the question, a short explanation of the problem, and a realistic access plan to your supervisor. Their feedback may change the wording or the design. That is a normal part of developing the study.

For related help, explore our research support and academic writing support pages.