A research paper has a structure that has stabilized across fields for a reason. Each canonical section does a specific job, and readers — your instructor, your peers, a reviewer — expect those jobs to be done in the order they know. Departing from the structure is possible, but the bar is high: you need a good reason, and you need to signal what you are doing so the reader can follow. This guide is a reference for that canonical structure. It walks through the standard shape of a research paper, then gives you a section-by-section breakdown with word budgets, typical moves, and the questions each section must answer. Use it as a reference while you draft — consult each section when you hit it, rather than reading top to bottom. For a companion piece on outlining before you draft, see the research paper outline guide.
PaperDraft is a writing assistant, not a paper generator — the draft is your starting point, not your submission. You are responsible for editing, verifying sources, and following your school's academic integrity policy.
The canonical research paper structure
The canonical research paper across most fields follows a six-section structure: introduction, literature review, methods, results, discussion, conclusion. In empirical sciences and much of the social sciences, this is often formalized as IMRaD — introduction, methods, results, and discussion — with the literature review folded into the introduction and the conclusion sometimes folded into the discussion. In humanities and interpretive social sciences, the same functions appear but under different names: an introduction with embedded literature context, a methods or methodology section when relevant, an analysis or argument section in place of results plus discussion, and a conclusion.
Regardless of the labels, each canonical section does one of six jobs. The introduction orients the reader to the problem, the gap, and the thesis. The literature review maps the existing scholarly conversation the paper joins. The methods section explains how the work was done — the empirical procedures, the archival strategy, the interpretive framework. The results section presents findings without interpretation. The discussion interprets those findings, relates them to the literature, and addresses limitations. The conclusion summarizes the contribution and points toward implications or future work.
Understanding the job of each section is what lets you judge whether your draft's section is doing the work it should. A "methods" section that includes interpretation is out of place; a "results" section that jumps to implications is out of place. The structure is not bureaucratic — it is an information design that lets a reader efficiently evaluate whether your work supports its claims. Proportion matters too. In a typical 6,000-word research paper, the literature review and the discussion carry the most weight; the methods and conclusion are shorter. The section-by-section breakdown below gives target word budgets you can scale up or down.
Section-by-section breakdown
Introduction
Purpose. Orient the reader to the topic, establish why the problem matters, identify the gap in existing work, and state your thesis or research question.
Word budget. Roughly 8–12% of the paper. In a 6,000-word paper, aim for 500–750 words.
Typical moves. Open with a hook that establishes the topic's relevance — a real-world context, a significant debate, an unresolved puzzle. Narrow from the broad topic to the specific question your paper addresses. Name the gap: what existing work has not adequately addressed. State your thesis or research question precisely, in a single sentence if possible. Preview the paper's structure so the reader knows what to expect. See our research paper introduction guide for the detailed moves. Do not bury the thesis — a reader should know your claim by the end of the introduction.
Literature review
Purpose. Map the scholarly conversation your paper enters. Show what has been established, what is contested, and where the gaps lie.
Word budget. Roughly 20–30% of the paper. In a 6,000-word paper, aim for 1,200–1,800 words. Longer in dissertations and review articles; shorter in short empirical papers where the literature review collapses into the introduction.
Typical moves. Organize thematically or chronologically, not source-by-source. A literature review structured as "Source 1 said X. Source 2 said Y. Source 3 said Z." is an annotated bibliography, not a literature review. A proper review groups sources by theme, method, or position, and tells a story about how the field has developed. Identify tensions and unresolved questions. Situate your own contribution relative to these. Cite broadly but with verification — do not cite what you have not read. See how to write a literature review.
Methods (or Methodology)
Purpose. Describe how you did the work in enough detail that a qualified reader could evaluate its soundness or, in empirical work, replicate it.
Word budget. Roughly 15–20% of the paper for empirical work. In a 6,000-word paper, aim for 900–1,200 words. Shorter in non-empirical humanities papers, where methodology might be a few paragraphs describing the texts analyzed and the interpretive framework.
Typical moves. In empirical work: describe participants or sample, materials or instruments, procedures, and analytical approach — in that order. State any approvals (IRB, ethics committee). Report enough detail for replication. In archival or textual work: identify the sources, the selection criteria, and the interpretive lens. In mixed-method or case-study work: justify the method choice against alternatives. See how to write methodology. The methods section is the place for precision; write it matter-of-factly and do not slip into interpretation.
Results
Purpose. Present findings. No interpretation, no evaluation against prior literature — those belong in the discussion.
Word budget. Roughly 10–15% of the paper for empirical work. In a 6,000-word paper, aim for 600–900 words. Humanities papers typically do not have a separate results section; findings appear inside the analysis.
Typical moves. Report findings in the order the questions were posed in the introduction. Use tables and figures where they compress information more efficiently than prose. Report effect sizes and confidence intervals alongside statistical tests in quantitative work. In qualitative work, present themes with representative quotations. Do not editorialize — "Participants scored higher on condition A (M = 4.2, SD = 0.8) than on condition B (M = 3.1, SD = 0.9)" belongs in results; "this finding supports the hypothesis" belongs in discussion.
Discussion
Purpose. Interpret the findings, relate them to the literature reviewed earlier, and acknowledge limitations.
Word budget. Roughly 20–25% of the paper. In a 6,000-word paper, aim for 1,200–1,500 words. The discussion is often the longest single section.
Typical moves. Restate the main finding in plain language. Relate it to each body of literature reviewed — does it support, extend, contradict, or complicate prior work? Explain unexpected findings. Address limitations honestly — sample size, generalizability, method constraints, alternative interpretations. Suggest implications for theory, practice, or policy as the scope warrants. Avoid overclaiming — a discussion that says more than the results can support weakens the paper.
Conclusion
Purpose. Summarize the contribution, restate the thesis in light of what the paper established, and point forward.
Word budget. Roughly 5–8% of the paper. In a 6,000-word paper, aim for 300–500 words.
Typical moves. Restate the research question and the answer the paper developed — but in new words, not by copying the introduction. Summarize the main contribution in one or two sentences. Point to future directions: unanswered questions, follow-up work, extensions to new contexts. Close with a sentence that frames the stakes or implications. Do not introduce new evidence or arguments in the conclusion. See how to write a research paper conclusion.
PaperDraft scaffolds a structured starting draft — thesis, outline, the canonical sections — for you to revise. Start this paper — free.
Variations by field
Humanities. The canonical six-section structure is less formal. A literature paper typically has an introduction that embeds scholarly context, a body organized thematically or around a developing argument, and a conclusion — methods and results rarely appear as named sections. When a methods section does appear, it is usually a brief description of the archive, texts, or interpretive framework. The thesis-driven structure is what matters; keep it explicit.
Social sciences. Most social-science papers follow the full six-section IMRaD-plus-literature-review structure. Quantitative papers separate methods, results, and discussion cleanly. Qualitative papers often merge results and discussion into a single "findings" or "analysis" section because interpretation is less separable from presentation.
STEM (natural sciences, biomedicine). IMRaD is tightly enforced, often with journal-specific word limits. The literature review is typically folded into the introduction's first several paragraphs. Methods must support replication. Results are graphical and tabular as much as textual. The discussion addresses each hypothesis in turn.
Engineering. Similar to STEM but often with an additional "problem formulation" or "system design" section between the literature review and the methods. Results often include engineering metrics specific to the system evaluated. Discussions frequently address trade-offs and design implications.
For the full paper-type context, see research papers.
Common mistakes
- Putting interpretation in the results section. Results present findings; discussion interprets them. Mixing the two reduces a reader's confidence in both.
- Literature reviews organized source-by-source. A proper literature review organizes thematically and tells a story about the field's development. See how to write a literature review.
- Underweighting the discussion. The discussion is where the paper's contribution is made explicit. A short discussion signals that the author has not thought through what the findings mean.
- Introducing new evidence or arguments in the conclusion. The conclusion summarizes and points forward; it does not add material. New material in the conclusion suggests poor structural planning.
- Ignoring section word budgets. A paper where the introduction is 40% of the word count and the methods are 5% is structurally unbalanced. Reviewers and graders read with rough expectations about proportion.
How PaperDraft helps you start
PaperDraft scaffolds a research paper in the canonical six-section structure with the proportions the field expects. The introduction previews a thesis; the literature review is organized thematically; the methods, results, discussion, and conclusion each do their expected job at roughly the expected length. The scaffold is a starting point — you revise thesis, add the specific findings from your data, verify and expand the literature discussion, and refine the argument until it reflects your work. PaperDraft is a writing assistant, not a paper generator — the draft is your starting point, not your submission. You are responsible for editing, verifying sources, and following your school's academic integrity policy.
Frequently asked questions
Does every research paper need all six sections?
Not always. Short empirical papers often fold the literature review into the introduction and the conclusion into the discussion, yielding an IMRaD structure. Humanities papers typically omit named methods and results sections. Check the conventions of your field and your assignment prompt.
How long should my paper be in total?
Follow the assignment. Typical undergraduate research papers run 2,500 to 6,000 words; master's-level work 6,000 to 15,000; dissertation chapters 8,000 and up. The section proportions above scale linearly.
Should I write the sections in order?
No — most writers do not. A common order is methods first (while the procedures are fresh), then results, then introduction and literature review (once you know exactly what you need to motivate), then discussion, then conclusion. The abstract is always written last. See the outline guide for sequencing.
Can I use headings and subheadings?
Yes — in most fields they are expected. Use your style guide's conventions. APA has a structured five-level heading system; MLA uses headings more sparingly; Chicago and Harvard vary by field. Consistency within the paper matters more than the specific style.
How do I know if my structure is working?
Read only your headings and the first sentence of each paragraph. If that skeleton tells a coherent story — problem, gap, method, findings, interpretation, implications — your structure is working. If it jumps or repeats, revise the structure before polishing the prose.