HOW SINGAPORE EDUCATION WORKS · Narrowing a University Research Question
Find your next learning step
Choose the question closest to your situation, or read the full guide in order.
Full chapter index · Practice and checking · Sources and next reading · How Singapore Education Works
To narrow a research question, decide what you want to understand, which evidence you can legitimately use, and what conclusion that evidence could support. Specify the objects or people studied, the setting and the comparison where relevant. Then check the question against the time, word limit, methods and approvals of the actual assignment.
A small question can support substantial reasoning. A large question can remain vague even after many sources are collected. This guide helps university learners move from an interest to an answerable inquiry; it does not prescribe one research method or replace a supervisor's judgement. All assignment conditions and map examples below are fictional.
The choices at a glance
| What you notice | Possible issue | Useful next move |
|---|---|---|
| Every source seems relevant | The scope may lack boundaries | Specify the object, period or comparison |
| The claim requires inaccessible evidence | The proposed question outruns the data | Change the question or discuss access |
| Several different studies are combined | Too many jobs compete in one question | Choose the inquiry the assignment supports |
| The answer is an obvious lookup | The question may not fit the task's reasoning | Check the required comparison or explanation |
Full chapter index
Distinguish the question and available evidence
Work through a case and diagnose broad scope
Plan an answer and practise independently
Transfer across disciplines and answer questions
CHAPTER 1 OF 8 · Distinguish the question and available evidence
1. Distinguish a topic, a question and a claim
“Campus sustainability” names an area of interest. It does not yet identify what the student will investigate. “How do six supplied visitor maps identify water-refill points?” asks about a defined set of objects and a particular feature. “The maps use inconsistent symbols” is a possible finding that needs evidence.
Keep these roles separate. If the question already assumes the desired answer, the inquiry can become a search for confirmation. Ask what you would observe if the initial expectation were mistaken, and whether your method would allow that result to appear.
A useful question need not always take the form “Does X cause Y?” Describing a pattern, interpreting a set of texts and estimating a difference are different research jobs. Choose one that fits the discipline and brief. Narrowing is not merely removing words; it is making the relationship between the question and the evidence more precise.
For the broader project setting, see university capstones and final-year projects. This guide concentrates on the earlier decision about what the inquiry can answer.
CHAPTER 2 OF 8 · Distinguish the question and available evidence
2. Inspect the available evidence before promising a result
Read the assignment contract first. Is the student expected to analyse supplied sources, conduct fieldwork, build a prototype, compare published studies or develop an argument? Those tasks support different questions. A document-analysis assignment cannot automatically establish how people actually behave.
Make a short evidence inventory. Identify which sources exist, what period they cover, what units they describe and what access is permitted. Then run a preliminary search or inspect a sample, as appropriate. This can reveal that a seemingly precise question relies on information the student cannot obtain.
NUS Libraries' guide to focused research questions for systematic reviews discusses identifying key concepts and checking the available studies through preliminary searching. It presents PICO and alternatives within review planning. That context is useful, but PICO is not a compulsory template for every university subject.
If the work involves people, restricted records or institutional data, check the relevant arrangements with the course team before collecting information. A promising question does not itself confer access or permission. Sometimes the suitable repair is a different evidence source rather than a more elaborate recruitment plan.
CHAPTER 3 OF 8 · Work through a case and diagnose broad scope
3. A worked case: from sustainability to six maps
Assume a fictional course supplies six dated 2026 visitor maps and asks for a 1,800-word analysis using only that dataset. The student has two weeks. The maps are teaching materials that may be examined for the assignment; no visitor interviews or consumption data are available.
The initial question is, “Do water-refill stations make universities sustainable?” It asks about a broad outcome and implies a causal investigation. The supplied maps cannot show overall sustainability or the effect of installing stations.
A better-fitting question is, “How do the six supplied visitor maps use labels and legends to identify water-refill points?” The units are maps. The evidence is their visible labels, symbols and keys. The analysis can compare how identification is presented without claiming to measure actual visitor understanding or water use.
Define coding decisions before counting. A map with a clearly labelled refill symbol and a matching key can be distinguished from an unexplained icon. If an icon might mean something else, record it as unclear rather than treating it as confirmed. Keep the original map reference alongside each observation.
Suppose an invented coding exercise finds a clear legend explanation on three of the six maps. That is 3/6, or 50%, of this supplied dataset. It is not evidence that half of all university maps have clear legends, that the other campuses lack refill points, or that visitors cannot find water. Those are different questions requiring different evidence.
CHAPTER 4 OF 8 · Work through a case and diagnose broad scope
4. Diagnose the reason a question still feels too broad
If every source seems relevant, ask what evidence would be outside scope. A question that includes every aspect of campus life may need a defined object, period or comparison. Exclusion criteria should follow the question, rather than conceal inconvenient findings.
If the proposed answer depends on unavailable data, identify the missing link. The map case lacks measures of visitor behaviour and environmental outcomes. Adding more adjectives to the question does not provide those measures. Change the question, obtain appropriate evidence through approved arrangements, or discuss a feasible alternative with the supervisor.
If a question combines several distinct jobs, separate them. Identifying symbols, measuring visitor comprehension and estimating plastic-bottle use could each require a different method. A limited assignment may support one of them well rather than all three superficially.
If the question is so narrow that the answer is a single obvious lookup, ask what comparison or explanation the brief calls for. Do not add complexity merely to sound academic. A precise descriptive inquiry can be legitimate, while an argumentative essay may require a further interpretive issue. The assignment's purpose decides which is appropriate.
CHAPTER 5 OF 8 · Plan an answer and practise independently
5. Specify what would count as an answer
Write a short answer plan before collecting a large pile of notes. Name the evidence you would need, how you would compare it and what the final answer might look like. This is a plan for reasoning, not a prediction of the findings.
For the map question, an answer might include a table of the six maps, definitions of the coding categories, selected examples and an explanation of the patterns and exceptions. It should also state the limits of the dataset. That plan follows from the question's focus on representation.
Now test a counterexample. If all six maps used identical labels and keys, could you still answer the question honestly? Yes: consistency would be a finding. If the planned essay requires variation in order to have anything to say, it may be organised around a preferred conclusion rather than the inquiry.
Keep a record when the question changes. Note the previous wording, the reason for revision and the effect on the evidence plan. For work that already has a proposal or approval, check the course's process before changing scope. Informal refinement and a change to an approved study are not always the same action.
CHAPTER 6 OF 8 · Plan an answer and practise independently
6. Practise narrowing, then explain the evidence limit
For guided practice, assume an assignment provides four public museum websites and asks students to analyse how events are described. The proposed question is, “Do museums make everyone more interested in history?” Identify what the supplied evidence cannot establish.
Checking answer: the websites can show how those sites describe events; they do not by themselves measure everyone's interest or a causal effect on it. One suitable revision is, “How do the four supplied museum websites describe the intended audiences for their history events?” Other precise questions can also fit if they use the available evidence and meet the brief.
For an independent task, assume eight published course descriptions are supplied. A student proposes, “Which course gives graduates the best careers?” Write a question that can be answered from the descriptions alone, then name a claim the evidence would still not support.
Checking answer: “How do these eight course descriptions present opportunities to practise workplace communication?” is a feasible document-analysis question. It cannot establish which graduates obtain better jobs. A course's description of an opportunity is different from evidence of participation, learning or later outcomes.
Explain the repair aloud. If it only sounds better but still requires the same missing data, the question has not been made answerable.
CHAPTER 7 OF 8 · Transfer across disciplines and answer questions
7. Transfer the judgement across disciplines and countries
George Mason University's Writing Center guide to research questions emphasises clarity, focus and fit with the discipline. Its composition guidance is an international comparison, not a universal requirement that all scientific or descriptive questions must be argumentative essays.
The shared learning principle is alignment: the evidence and method must be capable of addressing the question. The form of that alignment differs. A laboratory study, literary analysis and historical source comparison have different objects, conventions and limits.
For a cross-country project, check whether the units and definitions actually match. Two documents called visitor maps may differ in purpose, date, scale and audience. These differences can be part of an analysis, but they should not disappear inside a league table of campuses or countries.
Later, return to a broad topic from another module and narrow it without consulting the old wording. Explain what evidence would answer it and what would remain unknown. Use academic reading and evidence to check whether the sources you find address that question rather than merely sharing its keywords.
CHAPTER 8 OF 8 · Transfer across disciplines and answer questions
8. Questions students ask about research scope
Does narrow mean easy or unimportant?
No. A well-defined question can require difficult comparison, interpretation or measurement. Its value depends on the purpose and the reasoning, not the number of broad topics mentioned in the title.
Can I change the question after reading?
Often refinement is part of learning, but follow your module's or project's process. Explain what the reading changed and how the new question affects the evidence and method. Check approval requirements where relevant.
Must I use PICO?
Use the framework appropriate to the discipline and task. PICO is useful in particular review and research contexts; it is not the only way to define concepts or formulate a question.
Can a peer check my question?
Within the permitted collaboration rules, ask them to identify the object of study, proposed evidence and expected kind of answer. If those are unclear, revise before requesting general reassurance. For a draft argument, continue to university writing peer review.
Sources and next reading
Checked 22 October 2026. NUS Libraries: Develop a Focused Research Question supplies the Singapore systematic-review context. George Mason University Writing Center: How to Write a Research Question supplies the discipline-sensitive international writing comparison. The assignments, map dataset and numerical coding result are fictional teaching cases.
Return to the How Singapore Education Works hub for related university and lifelong-learning guides.