Glossary of Key Terms
| Term | Definition |
| Research question | The single, focused question that a study sets out to answer. It defines scope, guides methods, and shapes conclusions. |
| Research problem | The broader gap, contradiction, or unresolved issue in a field that justifies why the research question matters. |
| Hypothesis | A testable, proposed answer to a research question, often including a predicted relationship between variables. |
| FINER criteria | A checklist used to test a research question: Feasible, Interesting, Novel, Ethical, Relevant. |
| Sub-question | A smaller, more specific question that helps answer part of a larger main research question. |
| Variable | A factor, characteristic, or condition that can change or be measured within a study. |
| Literature review | A structured survey of existing scholarship on a topic, used to identify gaps and refine a research question. |
| Scope | The boundaries of a study: what is included and excluded in terms of population, time period, location, or variables. |
Key Takeaways
- A strong research question is specific, clear, feasible, researchable, and relevant to your field.
- Most research questions develop through five or six iterative steps: pick a topic, read broadly, narrow the focus, draft, and test against criteria such as FINER.
- A simple template, such as How does [variable A] affect [variable B] in [specific group or setting], makes it easier to draft a focused first version.
- Weak research questions are usually too broad, too vague, answerable with a simple yes or no, or already settled by existing literature.
- The right question type (exploratory, descriptive, comparative, experimental, qualitative, or quantitative) depends on what your study aims to do.
- Tools such as R Discovery can help you check whether a question has already been answered and spot unaddressed gaps before you commit to a direction.
- A thesis or dissertation may need more than one research question, but every question should tie back to a single, clearly stated research problem.
What Is a Research Question?
A research question is the single, focused question your study is designed to answer. It sets the direction of your investigation, defines what counts as relevant data, and shapes the conclusions you are able to draw.
Before researchers can begin a study or thesis, they need to pin down this question with care. A well-formed research question is built on background reading, not guesswork, which is why most researchers revise it several times as they read more deeply into their field.
Why Does a Strong Research Question Matter?
A strong research question matters because it keeps a study focused, feasible, and aligned with a real gap in knowledge, rather than a vague area of interest.
A clear question helps you do several things at once:
- Set the direction and purpose of the entire study
- Define the scope, including population, variables, and context
- Identify the methodology and data collection techniques you will need
- Help readers and reviewers immediately understand the purpose of your work
- Make it easier to evaluate whether your literature review is actually relevant
What Makes a Research Question Strong?
A strong research question is specific, clear, feasible, researchable, complex enough to merit real analysis, and relevant to your field or to society.
| Characteristic | What it means in practice |
| Specific | Clearly defines the variables, participants, and context. Leaves no room for ambiguity about what is being studied. |
| Clear | Stated in plain language. A reader should grasp the objective on first reading, without needing extra explanation. |
| Feasible | Answerable within your realistic time, budget, data access, and ethical constraints. |
| Researchable | Can be answered using primary or secondary sources, data collection, or existing scholarly literature. |
| Complex | Cannot be answered with a simple yes, no, or a single quick search. It requires synthesis and original analysis. |
| Relevant | Connects to a genuine gap, debate, or problem in your field, not just a topic you find interesting. |
How Do You Write a Research Question in Six Steps?
You write a research question by choosing a broad topic, narrowing it through reading, drafting a question, and testing it against criteria such as FINER before finalizing it.
Step 1: Identify a Broad Topic
Start with an area of interest within your field, something relevant enough to sustain months of work and specific enough to research properly. Choose a subject you are genuinely curious about, since you will spend a long time with it.
Step 2: Do a Thorough Literature Review
Read widely to understand what has already been studied and where the gaps sit. This stage is where most weak research questions get strengthened or abandoned. A tool such as R Discovery can speed this stage up by surfacing recent, relevant papers across major sources, so you can see quickly whether your idea has already been answered.
Step 3: Narrow the Focus
Narrow your topic by population, setting, variable, location, or time period. A topic broad enough to fill a book is still too broad for a single research question.
Step 4: Shortlist Possible Questions
Draft two or three candidate questions based on the gaps you found, or based on how you might build on or challenge previously published work.
Step 5: Test the Question Against FINER Criteria
Assess each candidate using FINER: is it Feasible, Interesting, Novel, Ethical, and Relevant. A question that fails on feasibility or novelty needs another round of narrowing.
Step 6: Formulate and Revise the Final Question
Write the final version so it is clear, well-defined, and includes the specificity, feasibility, and relevance discussed above. Expect to revise it again once data collection or further reading begins.
Research Question Format
A reliable research question format is: How, why, or what does [variable A] affect or relate to [variable B] in [specific group or setting].
Examples using this template:
- How does daily Instagram use affect self-esteem in teenagers aged 13 to 17?
- What is the relationship between sleep duration and academic performance among college students?
- To what extent does remote work policy influence employee productivity in small technology companies?
If a draft question still feels too broad once it is placed into this template, narrow one of the bracketed elements, the population, the setting, or the variable, rather than rewriting the whole question from scratch.
What Are the Main Types of Research Questions?
The six main types of research questions are exploratory, descriptive, comparative, experimental, qualitative, and quantitative, and each suits a different research purpose.
| Type | Purpose | Example |
| Exploratory | Investigate a topic that is broad, new, or under-studied | What are the potential benefits of mindfulness practices in the workplace? |
| Descriptive | Describe characteristics, trends, or patterns as they exist | What factors influence customer loyalty in the e-commerce industry? |
| Comparative | Compare two or more groups, conditions, or approaches | How does traditional teaching compare to online learning in mathematics? |
| Experimental | Test a cause-and-effect relationship under controlled conditions | Does a dietary intervention reduce weight among individuals with obesity? |
| Qualitative | Explore lived experiences, perceptions, or meaning in depth | What are the lived experiences of immigrants adapting to a new culture? |
| Quantitative | Measure and analyze numerical relationships between variables | Is there a relationship between sleep duration and academic performance? |
What Are Some Research Question Examples by Discipline?
Below are sample research questions across common academic disciplines, which you can adapt to your own population, setting, or variable.
| Discipline | Example research question |
| Psychology | How does daily social media use affect stress levels in university students? |
| Business | What factors drive customer loyalty in subscription-based companies? |
| Education | How does AI-supported feedback affect writing performance in high school students? |
| Healthcare and nursing | What barriers limit telehealth adoption among older adults? |
| Public policy | What is the impact of different healthcare policies on patient outcomes across countries? |
| Computer science and technology | What factors influence user trust in educational chatbots? |
| Environmental science | How has changing land use affected biodiversity in urban wetlands over the past decade? |
If you are exploring a discipline you do not know well, scanning recent literature through a platform such as R Discovery before you finalize a question can help you confirm that your chosen angle is still open, rather than already well covered.
What Do Weak and Strong Research Questions Look Like Side by Side?
Weak research questions are usually too broad, too vague, or answerable with a simple yes or no, while strong versions narrow the population, variable, and context.
| Weak research question | Stronger version |
| Does social media affect mental health? | How does daily Instagram use affect anxiety levels in teenagers aged 13 to 17? |
| What are the effects of climate change? | How has climate change affected agricultural yields in the American Midwest over the past decade? |
| Is AI useful in education? | How do AI-driven tutoring systems influence student engagement in online undergraduate courses? |
| What effect does social media have on people’s minds? | What effect does daily Twitter use have on the attention span of 16-year-old students? |
| Has homelessness increased in the United States? | How have economic and political factors affected patterns of homelessness in the United States over the past ten years? |
What Common Mistakes Weaken a Research Question?
The most common mistakes are choosing a question that is too broad, too vague, not researchable, answerable with yes or no, or built on data you cannot realistically access.
- Too broad: the question covers more ground than one paper, thesis, or dissertation can address
- Too vague: key terms are undefined, so two readers could interpret the question differently
- Not researchable: no realistic combination of data, sources, or methods could answer it
- Yes or no questions: closed questions do not leave room for analysis or discussion
- Value judgments: words such as good, bad, better, or worse do not give clear criteria for an answer
- Already answered: the question has a settled answer that a single search or article already provides
- Misaligned with available data: you do not have access to the population, archive, or instruments the question requires
Should You Use Sub-Questions?
You should use sub-questions when your main research question is too complex to answer in one step, but they are not mandatory for a simple, well-scoped study.
Good sub-questions share three qualities:
- They are less complex than the main question
- Each one focuses on a single type of research or data source
- They are presented in a logical order, usually building toward the main question
As a rule of thumb, keep sub-questions to four or five at most. If you need more, your main question is probably not specific enough yet, and it is worth revisiting your research problem before adding further sub-questions.
How Do You Apply the FINER Criteria?
You apply FINER by checking your draft question against five tests: is it Feasible, Interesting, Novel, Ethical, and Relevant, before you commit time to it.
| Criterion | Question to ask yourself |
| Feasible | Do I have the time, budget, sample access, and skills to answer this? |
| Interesting | Will this question hold my attention and that of my supervisor or reviewers for the full project? |
| Novel | Does this confirm, extend, or challenge existing findings, rather than repeat them? |
| Ethical | Can this be approved by an ethics board, and does it avoid harm to participants? |
| Relevant | Does this matter to my field, to policy, or to practice, beyond personal curiosity? |
Running a quick literature check, for example through R Discovery, before finalizing your question is one of the fastest ways to confirm the Novel and Relevant criteria, since it shows you what has already been published in your specific niche.
Frequently Asked Questions
How many research questions should a thesis or dissertation have?
Most theses and dissertations work best with one central research question and, where needed, three to five sub-questions. Having too many separate main questions usually signals that the research problem itself is not focused enough.
Can a research question change after you start collecting data?
Yes, and this is normal. Many researchers refine their question once early reading, pilot data, or fieldwork reveals that the original scope was too broad, too narrow, or already answered elsewhere. Document why the change happened, since supervisors and committees often ask about it.
What is the difference between a research question and a hypothesis?
A research question asks something you intend to investigate. A hypothesis is a testable, proposed answer to that question, often stating a predicted relationship between variables. Not every study needs a hypothesis, but every study needs a research question.
How specific is too specific for a research question?
A question becomes too specific when it can only be answered by a single, narrow data set with no room for broader discussion, for example a single classroom on a single day. If the answer would not generalize or interest anyone beyond that exact case, broaden the population, time period, or setting slightly.
Is it acceptable to use AI tools when drafting a research question?
Many researchers use AI tools to brainstorm angles, check clarity, or test wording, and this is generally acceptable as long as the final question reflects your own judgment and original thinking. Always confirm your institution’s policy on AI assistance before relying on it for graded or examined work.
Why did my advisor say my research question was not researchable?
This usually means no realistic combination of data, sources, or methods can answer the question within your constraints, often because the population is inaccessible, the concept is too abstract to measure, or the question is fundamentally a matter of opinion rather than evidence.
How long should it take to finalize a research question?
Most researchers spend several weeks to a few months refining a research question, since it typically goes through multiple drafts as background reading deepens. Rushing this stage is one of the most common causes of restarting a literature review later.
What should you do if your research question has already been answered?
Narrow the question by location, population, time period, or method to find an unanswered angle, or pivot to comparing your context against the existing finding. A focused literature search, for instance through R Discovery, can quickly confirm whether a near-identical study already exists before you invest further time.
This article was originally published on July 7, 2023, and updated on June 20, 2026.
R Discovery is a literature search and research reading platform that accelerates your research discovery journey by keeping you updated on the latest, most relevant scholarly content. With 250M+ research articles sourced from trusted aggregators like CrossRef, Unpaywall, PubMed, PubMed Central, Open Alex and top publishing houses like Springer Nature, JAMA, IOP, Taylor & Francis, NEJM, BMJ, Karger, SAGE, Emerald Publishing and more, R Discovery puts a world of research at your fingertips.
Try R Discovery Prime FREE for 1 week or upgrade at just US$72 a year to access premium features that let you listen to research on the go, read in your language, collaborate with peers, auto sync with reference managers, and much more. Choose a simpler, smarter way to find and read research – Download the app and start your free 7-day trial today!
