In brief
- A research question is the specific, answerable question your study is designed to address.
- Quantitative questions name variables and relationships; qualitative questions explore experiences, meanings and processes.
- The FINER criteria test whether a question is feasible, interesting, novel, ethical and relevant.
- PICO and its variants turn a broad health or policy topic into a precise, searchable question.
- Most strong questions are narrowed from a broad topic through several rounds of revision.
What is a research question? ¶
A research question is a clear, focused question that a study sets out to answer. It names what you will investigate, in whom or what, and often in what context. It grows out of a research problem, determines which methods and data are appropriate, and defines what counts as an answer. Most theses have one main question and two to four sub-questions.
Research question example
How do first-generation university students in the UK describe their sense of belonging during their first year of study?
This question names the population (first-generation students), the setting (UK universities, first year) and the focus (sense of belonging), and its wording ("how do... describe") signals a qualitative design. Research questions in research papers work the same way: readers should be able to predict your method from the question.
Types of research questions ¶
Questions differ by what they ask you to establish. The type you choose shapes your design and analysis.
| Type | Asks | Example |
|---|---|---|
| Descriptive | What is happening, how much or how often? | What proportion of nurses in public hospitals in Kenya report symptoms of burnout? |
| Comparative | How do groups differ? | Do online and in-person statistics courses differ in final exam performance among first-year psychology students? |
| Relational (correlational) | How are variables associated? | What is the relationship between daily screen time and sleep duration in adolescents aged 13 to 17? |
| Causal (explanatory) | Does X cause or affect Y? | Does a 6-week mindfulness program reduce test anxiety in secondary school students compared with a waitlist control? |
| Exploratory | What is going on in an under-studied area? | What strategies do small-scale farmers in Morocco use to adapt to repeated drought? |
| Interpretive | How do people experience or make sense of something? | How do early-career teachers experience their first parent–teacher conferences? |
| Evaluative | How well does something work, for whom and why? | How effective is a city's bike-share scheme in reaching low-income neighborhoods, and what limits its uptake? |
Descriptive, comparative, relational and causal questions are usually quantitative. Exploratory and interpretive questions are usually qualitative. Evaluative questions often call for mixed methods.
What makes a good research question? The FINER criteria ¶
A widely used test from clinical research (Hulley and colleagues, discussed by Farrugia et al., 2010) is FINER. It works in any field.
| Criterion | Question to ask yourself |
|---|---|
| Feasible | Do I have the time, participants, data access, skills and funding to answer it? |
| Interesting | Will the answer matter to me, my supervisor and readers in the field? |
| Novel | Does it confirm, refute or extend previous findings, or fill a gap? |
| Ethical | Can it be answered without undue risk to participants, and would an ethics committee approve it? |
| Relevant | Will the answer inform theory, practice or policy? |
Beyond FINER, a good research question is clear (one interpretation), focused (narrow enough to answer fully), researchable (answerable with data, not values alone) and complex enough to need analysis rather than a yes-or-no lookup.
Is this a good research question? If you could answer it with a single search or a dictionary, it is too simple. If you would need a whole career, it is too broad.
Research Proposal Help
Unsure your research question is focused and feasible? Get field-matched feedback on your proposal, with comments explaining every suggestion.
Research question examples: weak vs strong ¶
| Weak | Problem | Strong |
|---|---|---|
| Is social media bad? | Vague and value-laden | Is time spent on image-based social media associated with body dissatisfaction among girls aged 12 to 15 in Ireland? |
| What is remote work? | Answerable by definition | How does the number of in-office days per week relate to professional isolation among early-career hybrid employees? |
| How can we fix climate change? | Far too broad | Which household factors predict adoption of rooftop solar panels in suburban areas of Spain? |
| Does homework help? | Missing population, outcome and context | Is the amount of homework assigned in Grade 8 associated with mathematics achievement, after controlling for prior attainment? |
| Why do students drop out? | Too broad for one study | What reasons do part-time mature students at one community college give for withdrawing in their first year? |
Descriptive: What is the [prevalence/level/pattern] of [variable] among [population] in [context]? Comparative: Do [group A] and [group B] differ in [outcome] in [context]? Relational: What is the relationship between [variable X] and [variable Y] among [population]? Causal: Does [intervention/exposure] affect [outcome] in [population], compared with [comparison], over [time]? Qualitative: How do [participants] experience/describe/make sense of [phenomenon] in [setting]? Mixed methods integration: How do [qualitative findings] help explain [quantitative results]?
Research question examples by field ¶
Education
- Does peer tutoring improve reading fluency among Grade 3 students compared with teacher-led small-group instruction?
- How do secondary teachers decide when to allow generative AI tools in coursework?
Psychology
- Does self-compassion moderate the relationship between perfectionism and depressive symptoms in university students?
- How do adults describe the experience of returning to work after a period of burnout?
Health and nursing
- Is night-shift frequency associated with medication administration errors among hospital nurses?
- What barriers do rural women report in accessing cervical cancer screening?
Business and management
- Does CEO turnover predict changes in corporate environmental disclosure among listed firms?
- How do small retail businesses decide whether to adopt mobile payment systems?
Environmental science
- How has urban tree canopy cover changed in a single metropolitan area since 2000, and is the change associated with neighborhood income?
Computer science
- Does adding retrieval of source documents reduce factual errors in summaries produced by large language models for clinical notes?
Sociology and political science
- How do young voters in Brazil use messaging apps to discuss elections, and how does this shape their trust in news?
Examples by method: quantitative, qualitative and mixed ¶
Quantitative questions name measurable variables and the relationship you will test. They usually begin with "Is there", "Does", "What is the relationship" or "To what extent".
Quantitative
To what extent do perceived workload and supervisor support predict intention to leave among junior doctors?
Qualitative questions are open and exploratory. They usually begin with "How" or "What" and focus on meanings, experiences or processes.
Qualitative
How do junior doctors make sense of the decision to leave hospital medicine?
Mixed methods studies often have a quantitative question, a qualitative question and an integration question that links them.
Mixed methods
- Quantitative: What is the prevalence of intention to leave among junior doctors in three regional hospitals?
- Qualitative: How do junior doctors explain their intentions to stay or leave?
- Integration: How do the explanations from interviews help account for differences in intention to leave across hospitals?
Your question drives your design, so see our research methodology examples next.
Research problem, research questions and hypotheses ¶
The research problem (or statement of the problem) describes the gap or issue; the research question asks what you will find out; the hypothesis predicts the answer in quantitative studies.
Problem, question and hypotheses
Problem: Although hybrid work is now common, little is known about how it affects professional isolation among employees early in their careers.
Research question: Is the number of in-office days per week associated with professional isolation among employees in their first 3 years of work?
H0: There is no association between in-office days and professional isolation.
H1: More in-office days are associated with lower professional isolation.
Qualitative studies normally do not state hypotheses; they state aims or guiding questions instead. If you are building a proposal around these elements, our research proposal help gives field-matched feedback on your problem statement, questions and design. A clear question also makes your thesis statement much easier to write.
How to write a research question in 6 steps ¶
- Choose a broad topic you care about and can access data on.
- Read the recent literature to see what is known and where findings conflict or stop. Our guide to writing a literature review explains how to spot gaps.
- Identify a specific gap or problem: an untested population, an unresolved contradiction, a new context or a method not yet applied.
- Draft the question naming the population, the variables or phenomenon and the context.
- Test it against FINER and, where relevant, PICO or PEO.
- Get feedback and revise. Ask your supervisor whether the question is answerable within your timeline and resources.
Sources and further reading
- Farrugia, P., et al. (2010). Practical tips for surgical research: Research questions, hypotheses and objectives. Canadian Journal of Surgery, 53(4), 278-281
- How to Write a Research Question, The Writing Center, George Mason University
- The Research Problem/Question, Organizing Your Social Sciences Research Paper, USC Libraries