Will my research question survive the committee?
Score your question against FINER and PICO across eight criteria. You get a score, and fixes that name the specific element to sharpen — not general advice about being more focused.
Rate your question on each of the eight criteria below. Be honest — the fixes are only as useful as your answers. Then press Score my question.
Targeted fixes
Looking strong
Every criterion scored well. Pressure-test it once more against your data sources and your timeline, then start drafting your proposal.
Free: the Comparative Research Question Matrix
A worksheet for drafting several candidate questions side by side and comparing them on feasibility, novelty and measurability before you commit. We’ll email you the PDF.
The two lenses examiners apply to a research question
Everything downstream rests on the question. Get it right and the design, the methods and the analysis follow almost mechanically. Get it wrong and no amount of sophisticated statistics will rescue the thesis — which is why proposal defences spend so much time here, and why a weak question is one of the most expensive problems to discover late.
Two frameworks do complementary work. FINER asks whether the question is worth doing and can be done: Feasible, Interesting, Novel, Ethical, Relevant. PICO — or PECO for observational work — asks whether it is specific enough to answer, by forcing you to name each moving part:
- P — Population. Who exactly? Name the setting, the age range, the condition, the sampling frame.
- I / E — Intervention or Exposure. What are you doing, giving, or observing?
- C — Comparison. Compared with what? Usual care, an unexposed group, a benchmark, or an explicit absence of comparison.
- O — Outcome. What will you measure, and how? A defined, observable endpoint — not a concept you have not yet operationalised.
The single most useful test: if you cannot point to how each concept will be measured, you still have a topic rather than a question. “How does social media affect student wellbeing?” is a topic. “Among first-year undergraduates at one UK university, is daily self-reported Instagram use associated with higher scores on the WHO-5 Wellbeing Index over one semester?” is a question.
Frequently asked questions
What actually makes a research question good?
It is specific, answerable and measurable. It names the population, the variable or exposure, and the outcome; it is framed as a focused question rather than a broad topic; and every key concept can be observed or measured with an instrument you can name. On top of that it has to be feasible within your real timeline and resources, novel enough to contribute, ethical, and significant to your field.
What is FINER?
A five-part appraisal checklist. Feasible: answerable with the time, sample and resources you actually have. Interesting: worth the years it will take. Novel: adds something beyond what is already established. Ethical: can be approved and conducted responsibly. Relevant: matters to your field, to practice, or to future research. A strong question satisfies all five, and a question that fails Feasible fails no matter how well it scores elsewhere.
What is PICO, and what is PECO?
PICO structures a comparative question into Population, Intervention, Comparison and Outcome. PECO swaps Intervention for Exposure and suits observational and epidemiological work, where you observe rather than assign. Naming each element is what forces the question from vague to specific, which is exactly what the validator checks.
My study is qualitative. Does PICO still apply?
Not in that form, and forcing it would distort your question. Use a qualitative equivalent instead: PICo (Population, phenomenon of Interest, Context) or SPIDER (Sample, Phenomenon of Interest, Design, Evaluation, Research type). The FINER half of this check applies perfectly well to qualitative work, and so does the underlying principle — name your population, your phenomenon and your context precisely.
How narrow should the question be?
Narrow enough that you can state exactly what data would answer it, and broad enough that the answer would interest someone other than you. Most doctoral questions fail on the first count rather than the second. If your question contains 'and' joining two different outcomes, it is probably two questions — pick the primary one and demote the other.
Is my question stored anywhere?
No. The validator asks you to self-assess against eight criteria and computes the score in your browser. Neither your question nor your answers are transmitted or saved.