You have been handed a topic — "climate change," "social media," "the Cold War" — and told to write a research paper. The topic is not the problem. The problem is that a topic is not yet a question you can actually answer.
Good research starts with a focused, arguable research question, and you can build one from any broad topic in a few deliberate steps. Get the question right and the rest of the paper gets dramatically easier.
What makes a research question good
Writing centers and research librarians tend to agree on five criteria. A strong research question is:
- Clear — it names the who, what, and where, with no vague terms.
- Focused — narrow enough to answer thoroughly in your page limit.
- Complex — it cannot be settled with a simple yes or no.
- Arguable — reasonable people could disagree about the answer.
- Researchable — you can actually find evidence for it in your timeframe.
Step 1: Map the landscape
Before you narrow, explore. Skim a few overview sources and ask: what are the major debates here? Who disagrees with whom, and why? You cannot find a good question inside a conversation you have not listened to yet.
Step 2: Find the tension
The best questions live where sources disagree or where something is unresolved. Look for a "but" — a place where the obvious answer runs into a complication. That friction is where an original argument becomes possible.
AI is a fantastic librarian and a terrible author. Use it to find the conversation, then join it in your own voice.
Step 3: Narrow with "how" and "why"
Yes/no questions are dead ends. Convert them into "how" or "why" questions that demand analysis. Watch a topic become a question:
Topic: social media and teenagers.
Too broad: "Is social media bad for teens?"
Focused question: "How do comparison-driven feed designs contribute to anxiety in middle-school students, and what does that imply for platform regulation?"
The focused version tells you what to research and what your thesis will argue.
Step 4: Pressure-test it
Run your question through the five criteria. Too broad? Add a population or a mechanism. Not arguable? Find the point of disagreement. Not researchable in time? Narrow the scope until the evidence is reachable.
Have a topic but don't know what to research?
If you want guided help turning a broad subject into a focused question, The Smarter Research Prompt Pack gives you ten copy-and-paste prompts to narrow your topic, generate better search vocabulary, map competing perspectives, and plan the evidence you actually need.
Verify everything AI suggests
AI is genuinely useful for mapping a topic and generating search terms — but treat every source, statistic, or name it produces as a lead to confirm, never a fact to cite. Language models frequently invent realistic-looking references that do not exist, so confirm each one in your library database before it enters your paper.
Common mistakes
- Confusing a topic with a question. "The Cold War" is a topic; "why did détente collapse in the late 1970s?" is a question.
- Starting to write before narrowing. A broad question produces a shallow paper.
- Citing AI-suggested sources unverified. Always check them yourself.
Once you have your question, turn it into a claim with our guide to writing a strong thesis, and keep your citations accurate as you gather sources.
Frequently asked questions
Can my research question change as I research?
Absolutely — expect it to. As you learn more, sharpen the question so it matches the evidence you are actually finding.
How narrow is too narrow?
If you cannot find enough credible sources to support a full paper, widen the scope slightly until the evidence is reachable.
Sources & further reading
- George Mason University Writing Center — How to Write a Research Question: writingcenter.gmu.edu
- Association of College & Research Libraries — Framework for Information Literacy: ala.org/acrl
