Every strong dissertation, thesis or research project has one thing in common: a clear research question at its centre. It decides what literature you read, what data you collect, which methods make sense and how your argument is structured. Yet many students treat the question as a formality to rush through in week one — and then spend months fighting a project that never quite holds together. Learning to write a precise research question is one of the highest-value skills in academic work.
What a Research Question Does
A research question is not a topic. “Social media and teenagers” is a topic; “How does evening social media use affect the sleep quality of secondary school students?” is a question that can actually be investigated. The question converts curiosity into a task with edges: it names who or what is involved, what relationship or experience is being explored, and often where or when. Everything in your project should be traceable back to it. If a chapter, a source or a survey item does not help answer the question, it probably does not belong in the project.
The question also protects you during writing. When you are lost in month three, rereading it tells you what the next paragraph is for. Supervisors and examiners know this, which is why the question features so heavily in proposals and vivas.
From Topic to Question: A Workable Process
Start broad, then narrow in stages. First, name your general area and read enough background literature to see what is already known and where the disagreements or gaps are. Second, draft several candidate questions — aim for five or six, in different shapes: descriptive (what is happening?), comparative (how do two groups differ?), relational (how does X affect Y?), and evaluative (how well does this intervention work?). Third, interrogate each candidate with the tests in the next section. Fourth, pick the survivor and refine its wording until every term is precise.
Notice that reading comes before finalising. Students often lock a question before knowing the literature, then discover their exact study was published two years ago. A scan of existing work — organised the way we describe in our guide to avoiding plagiarism and building on prior scholarship honestly — saves that pain.
The Five Tests of a Good Research Question
Clear. Someone outside your field should understand what is being asked. Define any term that could be read two ways. Focused. One question, one main relationship or phenomenon. If your draft contains “and” joining two separate investigations, split it. Answerable. You must be able to gather evidence — data, texts, observations — that genuinely answers it, within your access and ethics limits. Significant. The answer should matter to someone: a gap in knowledge, a practical problem, a policy debate. Feasible. It fits your timeframe, skills and word count. A question needing a three-year national survey is a fine ambition and a poor undergraduate dissertation.
Apply the tests ruthlessly. Most first drafts fail on focus: they are really three questions wearing one coat. Narrowing feels like losing content, but in practice it is what makes deep, well-evidenced work possible — and it makes the next decision, choosing a research methodology, dramatically easier, because each question shape points toward suitable methods.
Matching Question Shape to Method
“How many” and “to what extent” questions call for measurement: surveys, tests, existing statistics. “How” and “why” questions about experience call for depth: interviews, focus groups, case studies. “What works” questions call for comparison or evaluation designs. You do not need to finalise your methodology while drafting the question, but a question that no method could answer is a warning sign. If you cannot picture the evidence that would settle your question, the question needs more work.
Refining the Final Wording
Write the question on its own line and polish it like a title. Name the population or context (“first-year nursing students,” not “students”). Prefer precise verbs — examine, compare, explore, evaluate — over vague ones like “look at.” Check that the question is genuinely a question, not a statement with a question mark. Then park it for two days and reread it cold; distance reveals ambiguity that familiarity hides. Share it with your supervisor early, before it hardens into a proposal.
Done well, this single sentence becomes the spine of your whole project: it shapes your literature search, your methodology, your analysis and eventually the way you defend the work under peer review or present it, as our guide to presenting research at a student conference explains. Spend the hour the question deserves now, and it will repay you in every chapter that follows.
Examples: Weak Questions Made Strong
The fastest way to learn the craft is to watch weak questions get repaired. “What is the impact of technology on education?” fails nearly every test: unbounded population, undefined impact, no context. Narrow it: “How do teachers in rural primary schools use tablet devices to support reading instruction?” — now there is a population, a setting, a practice and an observable focus. Or take “Why do students drop out?” — too vast to answer honestly. Repaired: “What reasons do first-year university students give for considering withdrawal during their first term?” The repaired versions are not smaller in value; they are larger in usefulness, because evidence can actually be gathered against them. Try the same exercise on your own draft: circle every vague noun, name the population, add a context, and watch the question sharpen.
Sub-Questions: The Supporting Cast
Many projects carry one main question plus two or three sub-questions that break it into investigable parts. A main question about how novice teachers build classroom confidence might carry sub-questions about which experiences they find most formative, what support they use, and how confidence changes across the first year. Sub-questions are not loopholes for smuggling in extra topics; each must feed the main question directly. They earn their keep in your structure: they often become your chapter divisions, your interview themes, or the sections of your survey. If a sub-question could stand alone as a separate project, it does not belong — park it in a “future research” note and keep the present project focused.
Keep Exploring
If this guide was useful, these related articles go deeper into the skills and ideas it touches on:
- Understanding Peer Review: How Academic Publishing Works — why a precise question matters to reviewers
- How to Choose a Research Methodology for Your Dissertation — turning a sharp question into a workable design
- How to Apply Critical Thinking to Analyze Academic Material — stress-testing your question before fieldwork
- How to Develop Problem-Solving Skills Through Structured Practice — working through a vague question systematically
- How Reflection Helps Students Track Progress — reviewing and refining your question over time
- How Time Management Apps Can Improve Student Focus — planning the drafting stages that follow









