How to Write Research Questions: A Canadian Student’s Guide

The research question is the spine of your dissertation. Get it right and every chapter — literature review, methodology, results, discussion — flows from one clear purpose. Get it wrong and you will spend months writing material your examiners do not believe answers a coherent question. This guide walks you through the question-writing process Canadian graduate supervisors expect: starting from a broad topic, narrowing through three established frameworks, testing the result against the FINER criteria, and finishing with worked examples across nursing, education, business, and STEM.

What a Research Question Actually Is

A research question is the single sentence your dissertation answers. It is not a topic (“AI in education”), it is not a hypothesis (“students using ChatGPT score lower”), and it is not a thesis statement (“AI degrades academic integrity”). It is a question framed so precisely that a methodology can be designed to answer it and a reader can judge whether the answer is correct.

Three properties separate a research question from everything else:

• Specific — narrow enough that a finite study can answer it. “What are the effects of AI on universities?” is not specific. “How does ChatGPT use during essay drafting affect first-year University of Toronto undergraduates’ argument quality?” is specific.
• Answerable — the data needed to answer it must be collectable. A question that requires interviewing every PhD student in Canada is not answerable. One that requires interviewing twelve at three universities is.
• Gap-linked — the question must address something the existing literature has not already settled. Your literature review proves the gap exists; your research question fills it.

From Topic to Question: The Five-Step Funnel

Most students arrive at their supervisor with a topic, not a question. The work between those two is a deliberate narrowing process. Use this funnel:

1. Start with a broad interest area — “climate communication”, “AI in healthcare”, “rural-urban education gaps”.
2. Locate the existing scholarship — read 15–25 recent peer-reviewed articles. Note where authors say “further research is needed”.
3. Identify one specific gap — the studies you read have not addressed a particular population, context, intervention, or outcome.
4. Frame the gap as a question — use one of the formal frameworks (PICO, PEO, SPIDER) covered below to structure the wording.
5. Test against the FINER criteria — Feasible, Interesting, Novel, Ethical, Relevant. If any criterion fails, refine.

A topic like “AI in education” might become “What are the effects of ChatGPT-assisted brainstorming on first-year argumentative essay structure among Canadian-university undergraduates?” That is a question your committee will accept.

The Three Frameworks: PICO, PEO, and SPIDER

Each of these frameworks turns a topic into a structured question. PICO dominates clinical and quantitative research; PEO is preferred for qualitative and exploratory work; SPIDER bridges both with a specifically qualitative emphasis. Pick whichever fits the type of evidence you intend to collect.

PICO — for quantitative and intervention studies

• Population — who are you studying?
• Intervention — what action, treatment, or exposure are you testing?
• Comparator — what does the intervention compare against (no treatment, alternative treatment, placebo, baseline)?
• Outcome — what measurable result will you observe?

Example: In Canadian first-year university students (P), does ChatGPT-assisted essay outlining (I) compared with traditional outlining (C) affect argumentative essay rubric scores (O)?

PEO — for qualitative and exposure studies

• Population
• Exposure (the experience, context, or condition of interest)
• Outcome (often an experience, perception, or process rather than a measurement)

Example: How do Indigenous nursing students (P) experience cultural-safety training during clinical placements (E) in terms of their professional-identity development (O)?

SPIDER — the qualitative-leaning bridge

• Sample — equivalent to Population.
• Phenomenon of Interest — the experience or behaviour you study.
• Design — the type of qualitative study (interview, focus group, ethnography).
• Evaluation — the angle from which you analyse the data (themes, lived experience, processes).
• Research type — qualitative, mixed-methods, or quantitative.

Framework Comparison Table

The FINER Test: Five Quality Criteria

Once you have a candidate question, run it through the FINER test before showing your supervisor:

• Feasible — you can collect the data with the time, budget, and access you actually have. A question that requires data from twenty federal departments is not feasible for a Master’s thesis.
• Interesting — to you, your supervisor, and the broader research community. Boring questions do not get funded or published.
• Novel — it adds something the literature does not already settle. Replication is fine; rehashing is not.
• Ethical — it can pass Research Ethics Board (REB) review under TCPS 2. Questions about vulnerable populations need explicit justification.
• Relevant — the answer matters to policy, practice, or theory.

Worked Examples Across Disciplines

Six examples showing topic → question transformation. Each uses one of the three frameworks plus survives the FINER test.

Hire A Professional Academic Writer

• PhD-qualified Canadian writer
• AI-free, plagiarism-free guaranteed
• On-time delivery, every time
• Unlimited free revisions 14 days
• 24/7 chat support

Order Now Learn More

Use code FIRST10 for 10% off your first order.

Central Question Plus Sub-Questions

Canadian dissertation committees almost universally expect one central research question (CRQ) plus two to four sub-questions (SRQs). The central question states the overall purpose; sub-questions break it into investigable pieces that map to your data sources or analysis steps.

Structure them so each sub-question corresponds to one chapter or one major data source. A common pattern:

• CRQ — the overall puzzle.
• SRQ 1 — what does prior literature say? Answered by the literature review chapter.
• SRQ 2 — what does new empirical data show? Answered by your results chapter.
• SRQ 3 — what mechanism explains the result? Answered by your discussion.

Common Mistakes Examiners Flag

• Yes/no questions — “Does ChatGPT affect grades?” is too binary. Reframe with a magnitude or mechanism: “By how much do ChatGPT-assisted drafts shift argument-quality scores compared with unassisted drafts?”
• Two questions in one — “What are the effects of AI on student learning and on academic integrity?” Split into two studies or two papers.
• Unmeasurable concepts — “How does culture influence learning?” needs an operational definition of both “culture” and “learning” before it becomes researchable.
• Pre-empting the answer — “Why is online learning worse than in-person?” assumes the conclusion. Reframe neutrally: “How do learning outcomes differ between online and in-person delivery of first-year statistics?”
• Topic without question — submitting “AI in education” as your research question. Your committee will return it unread.
• Mismatched methodology — asking “how do they experience…” then designing a survey. Phenomenological questions need interviews; “how many” questions need surveys.

Iterative Refinement: From Rough Draft to Examiner-Ready

Few research questions arrive fully formed. Treat your first draft as raw material. Most strong questions go through four or five rounds of refinement before they survive both your supervisor and the FINER test. Below is a worked iteration showing how one student turned a vague topic into a defensible doctoral question over four passes.

Each iteration tightens one element. v1 to v2 added a verb. v2 to v3 added an intervention. v3 to v4 added population specificity, dose, comparator, and outcome measurement. The final version is what your committee approves.

Quantitative vs Qualitative Question Phrasing

The verbs you use signal what kind of evidence you intend to collect. Examiners read the question and form an expectation about your methodology before they reach chapter three. Mismatching phrasing and methods is one of the most common reasons for major revisions.

• Quantitative verbs — “what is the effect of…”, “how does X affect Y”, “what is the relationship between…”, “what proportion of…”, “to what extent does…”. These signal hypothesis testing, statistical comparison, surveys with closed-ended items, or experimental designs.
• Qualitative verbs — “how do participants experience…”, “what meanings do…”, “in what ways do…”, “how do processes unfold…”, “what are the perspectives of…”. These signal interviews, focus groups, ethnography, narrative analysis, or grounded theory.
• Mixed-methods verbs — “what is the effect of X, and how do participants experience it?”, typically split into linked CRQ + SRQ pairs.

If you find yourself drafting a quantitative-sounding question but planning interviews, the question is wrong — not the method. Reframe.

Stating Your Question in the Dissertation

The research question belongs in three places:

1. Abstract — one sentence stating the central question; one sentence answering it.
2. Introduction chapter — in the last section of chapter one, after you have framed the problem and motivated the gap. Set the CRQ and SRQs in bold or set off as a numbered list.
3. Conclusion chapter — restate the questions and answer them point by point. Examiners will check that every question you posed in chapter one has been answered in the conclusion.

Quick Checklist Before You Submit

• One central research question (CRQ) stated in a single sentence.
• Two to four sub-questions (SRQs), each addressable by one chapter.
• Question structured via PICO, PEO, or SPIDER.
• Survives FINER (Feasible, Interesting, Novel, Ethical, Relevant).
• No yes/no phrasing.
• Population, intervention/phenomenon, and outcome are all explicitly named.
• Methodology in chapter three matches the question type (qualitative phrasing → qualitative methods; quantitative phrasing → quantitative methods).

Frequently Asked Questions