Explain how to frame a research question for the Independent Study, distinguishing good from poor questions and refining scope, contestability and answerability

What this dot point is asking

SEAB wants you to understand how to turn a broad interest into a research question fit for the Independent Study: the individual inquiry that culminates in a dissertation. The research question is the single most important decision in the project, because everything else, the methodology, the sources, the argument, flows from it. Your task is to explain what makes a question good, how to refine a topic into one, and the faults that sink poor questions.

The answer

A topic is not a question

The first move is to distinguish a topic from a research question. "Artificial intelligence," "free will," "social media" are topics: areas of interest, not things that can be answered. A research question asks something specific that an inquiry could settle, at least provisionally. Beginning with a topic is fine; the work is to narrow it into a question that has an answer worth arguing for.

The marks of a good question

A good research question has five features. It is focused: narrow enough to be answered within the scope and time available, rather than sprawling. It is answerable: there is evidence or argument that could settle it, so it is not a question that no inquiry could resolve. It is contestable: it admits more than one defensible answer, so there is a genuine thesis to argue rather than a fact to look up. It is clear: its key terms are defined, so the inquiry knows what it is asking. And it is significant: answering it matters, or illuminates something beyond the narrow case. A question that meets all five is the engine of a strong study.

Refining a topic into a question

Refinement proceeds by a series of narrowing moves. Begin with the topic, then fix the specific aspect or variable of interest, then bound the scope (a particular context, period or domain), then phrase the result so that a defensible thesis is possible. For example, "social media" narrows to "the credibility of claims on social media," then to a contestable, bounded question such as "To what extent does the framing of a claim on social media affect whether users judge it credible?" Each step trades breadth for the feasibility and sharpness that make a study answerable.

Contestability and the right kind of question

Knowledge and Inquiry rewards questions with a conceptual or evaluative edge, not bare head-counts. A purely descriptive question ("how many students use a given tool?") may be answerable but thin, lacking the contestability and significance the subject prizes. The strongest questions either weigh competing positions ("does principle A or principle B better account for our judgements about X?") or assess a contestable claim ("to what extent does Y?"), so that the inquiry must reason, not merely report.

Common faults

Poor questions fail in characteristic ways. Too broad: "what is justice?" is unbounded and unanswerable in the scope of a study. Vague or loaded: "is climate change bad?" leaves the key term undefined and presupposes its answer. Not contestable: a question with only one defensible answer leaves nothing to argue. Unanswerable: a question no available evidence or argument could settle. Diagnosing which fault a draft question commits is the quickest route to improving it, and refining the question early is repaid through every later stage of the study.

Examples in context

Example 1. From "happiness" to a researchable question. A student fascinated by happiness cannot study "happiness" as such. Narrowing to the relationship between wealth and reported wellbeing, bounding to a defined context, and phrasing for contestability yields "To what extent does additional income raise reported wellbeing once basic needs are met?" The result is focused, answerable from evidence, contestable, and significant, where the original topic was none of these. The example shows the narrowing moves in action.

Example 2. Fixing a loaded question. A draft asks "Should we ban harmful content online?" The word harmful is undefined and the question half-presupposes its answer. Refined, it becomes "On what criteria, if any, can restricting online content be justified without unacceptable costs to free expression?" This defines the contested terms, makes the question genuinely open, and turns a slogan into an inquiry. The case illustrates diagnosing and repairing the loaded-question fault.

Try this

Q1. State the five marks of a good research question. [6 marks]

• Cue. Focused (answerable within scope), answerable (evidence or argument could settle it), contestable (more than one defensible answer), clear (key terms defined), and significant (answering it matters).

Q2. Explain why a purely descriptive question may be a weak choice for a Knowledge and Inquiry study. [6 marks]

• Cue. It may be answerable but thin, a head-count lacking contestability and significance; the subject rewards questions with a conceptual or evaluative edge that require reasoning, not mere reporting.

Q3. Refine the topic "censorship" into a focused, contestable research question and explain your moves. [8 marks]

• Cue. Fix an aspect (e.g. censorship of misinformation), bound the scope (a defined platform or context), and phrase for a thesis: "On what criteria can removing misinformation be justified without unduly restricting legitimate debate?" Note narrowing for feasibility and defining terms for clarity.