How to Write a Research Paper

A research paper is an argument built on evidence – the work is in narrowing the question, finding good sources, and making them serve a thesis of your own.

A research paper presents an argument or answers a question using evidence gathered from credible sources. It is more than a report that collects facts: it makes a claim of your own and defends it with research. The process rewards planning – a focused question, careful source-gathering, and a clear structure make the writing far easier than diving in and hoping a paper takes shape on its own.

Choose and narrow a topic

Start with a subject that interests you and that fits the assignment, then narrow it until it is specific enough to cover in the space you have. Broad topics produce shallow papers; a narrow topic lets you go into real depth. Turn your topic into a research question – the question your paper will set out to answer – which gives your reading a purpose and keeps you from drowning in material.

Find credible sources

Gather sources from reliable places: peer-reviewed journals, scholarly books, reputable databases, and authoritative reports. Be cautious with general web pages, and always check who wrote something, when, and why. Take notes as you read, recording full citation details for every source immediately – hunting them down later wastes hours. Aim for enough sources to support your argument from more than one angle.

Develop a thesis

Once you have read enough to form a view, write a thesis statement: a clear, arguable answer to your research question that the rest of the paper will support. A strong thesis is specific and takes a position. Expect to refine it as you write – your argument often sharpens once you see how the evidence actually lines up.

• Too broad: “Social media affects mental health.”
• Arguable thesis: “Among adolescents, the harm of social media stems less from time spent than from social comparison, which suggests interventions should target design, not screen limits.”

Outline before you draft

Organize your evidence into an outline before writing full paragraphs. Group related points, decide the order that builds your argument most convincingly, and check that every section supports the thesis. An outline turns a pile of notes into a structure, and it exposes gaps – places where you need more evidence – while they are still easy to fix.

Integrate evidence and cite it

In the draft, use sources to support your points, not to replace your own reasoning. Introduce each piece of evidence, explain how it supports your claim, and cite it properly. Paraphrase more often than you quote, and reserve direct quotation for wording that matters. Every borrowed idea – quoted or paraphrased – needs a citation in the style your course requires. Careful citation is also your protection against accidental plagiarism, which is treated seriously.

Structure the paper clearly

Most research papers open with an introduction that presents the question and thesis, continue with body sections that each develop one part of the argument with evidence, and close with a conclusion that draws the findings together and reflects on what they mean. Use clear transitions so the reader can follow how each section builds on the last, and let the thesis remain visible throughout.

Guard against plagiarism as you write

Plagiarism is often accidental – a paraphrase that stays too close to the original, a note copied without a source attached, a citation forgotten in the rush to finish. Guard against it from the start by recording where every idea and quotation came from as you take notes, and by paraphrasing properly: set the source aside, restate the idea in your own words and structure, then check that you have genuinely rewritten it rather than swapped a few synonyms.

Cite not only direct quotations but any idea, finding, or argument that is not your own or common knowledge. When in doubt, cite. A paper that is careful with attribution reads as more trustworthy, and it protects you from an academic-integrity problem that can undo an otherwise strong piece of work.

Revise, edit, and check

Revise for the big things first – is the argument clear, is every claim supported, does the structure flow? Then edit for clarity and concision, and proofread for grammar and citation errors. Read the paper against the assignment brief and the marking criteria to be sure you have done what was asked. A careful revision pass is often the difference between a paper that is competent and one that earns top marks.

Move from topic to answerable question

A topic is an area; a research paper needs a question. The transition usually means specifying a relationship, a population and a context. “Remote work” becomes “does the shift to remote work explain the change in reported job satisfaction among knowledge workers between 2019 and 2023?” — which tells you what evidence you need.

Test the question against the word count. A question requiring a book cannot be answered in five thousand words, and narrowing early saves the painful mid-draft realisation that the argument cannot be completed.

Search systematically, not opportunistically

Random searching produces a bibliography of whatever surfaced first. Work from a small set of terms, use the databases your discipline relies on, and then mine the reference lists of the two or three most central recent papers. That backward search finds the foundational work far faster than keywords.

• Record the full citation at the moment you read, not later.
• Note the finding, the method and the limitation in your own words.
• Mark verbatim material clearly to avoid accidental plagiarism.
• Keep a running note of how each source relates to your question.

Outline from the argument, not the sources

Build the outline around the claims you intend to make, then slot sources beneath them. Outlining by source produces a paper that reads as an annotated bibliography. If a source does not fit under any claim, it either needs a new claim or does not belong.

Write the introduction last, or twice

Introductions promise what the paper delivers, and you cannot know that until it is written. Either draft it last, or draft it early and rewrite it fully at the end. An introduction that promises an argument the paper did not make is among the most common structural faults, and it is entirely avoidable.

Integrate sources rather than stacking them

Every source should be doing identifiable work: establishing context, supplying evidence, providing a method, or representing a position you contest. A paragraph containing three citations and no interpretation has collected rather than argued.

Watch the balance of paraphrase to quotation. In most disciplines, paraphrase should dominate heavily — quotation is for wording that carries weight the paraphrase would lose.

Leave time for structural revision

Most improvement in a research paper comes from moving and cutting, not from polishing sentences. On a revision pass, read only your topic sentences in order: they should form a coherent argument. Where they do not, the problem is structural and no amount of line editing will fix it.

Reserve the final pass for citation accuracy and formatting, which are mechanical, checkable, and a routine source of avoidable lost marks.