A research paper differs from an essay in where its authority comes from. An essay argues a position; a research paper answers a question by systematically engaging with existing scholarship, and sometimes with data you analyse yourself. The question drives everything: your search terms, your structure, your analysis and your conclusion.
This guide follows the whole process in working order, from reading the assignment to submitting the final draft. If you are on a deadline, the single most useful thing you can do is settle the research question before you start reading widely.
Understand the research assignment
Before anything else, establish what kind of paper is being asked for. A literature-based paper synthesises published work. An empirical paper reports data you collected or analysed. A secondary-analysis paper works with an existing dataset. These have different structures, and writing the wrong one is not recoverable in revision.
Then pin down the boundaries: word count, the number and type of sources required, whether a specific method or theoretical framework is expected, the referencing style, and whether any ethics process applies if you are collecting data. If your brief is ambiguous on any of these, ask your instructor rather than guessing.
Choose a focused research question
A good research question is specific, answerable with available evidence, and genuinely open. Move from topic to question by naming the population or context, the variables or concepts you care about, and the relationship you are examining.
Then stress-test it. If a database search returns nothing relevant, it may be unanswerable at your level. If it returns a settled answer everywhere, there is nothing to examine. If answering it honestly needs primary data you cannot collect, reframe it as a question about the literature instead.
- Too broad: "What are the effects of childhood trauma?"
- Too narrow: "How many participants were in one 2019 study?"
- Workable: "How does childhood trauma shape adult social relationships according to sociological research since 2010?"
Search and evaluate the literature
Turn your question into search terms: the key concepts, plus their synonyms and alternative spellings. Search your library databases rather than relying on a general web search, and keep a log of the databases, terms and date limits you used. That log takes minutes and saves an afternoon of re-searching later.
Once you have a small set of highly relevant papers, work outward: follow their reference lists backward for foundational work and check what has cited them for recent developments. Stop when new searches keep returning sources you already have.
Screen efficiently. Read the abstract to decide relevance, then the method to judge quality, then the discussion for how the authors frame the debate. Read the full paper properly only for sources you will actually cite. A summariser can help you triage long readings into notes, but read anything you cite in full before citing it.
- Is it peer reviewed, and is the venue credible in this field?
- Is the method appropriate to the claim, and is the sample adequate?
- How recent is it relative to the debate you are entering?
- Does it support, qualify or challenge your emerging answer?
Take notes you can actually write from
Organise notes by theme rather than by source. A note file per source produces a paper that walks through sources one at a time, which reads as a list. Grouping by theme produces synthesis, which is what research papers are marked on.
For every note record the full reference, the page, the claim in your own words, and a line on how it relates to your question and to other sources. Marking your own commentary distinctly from the source's claim is what protects you against accidental plagiarism at the drafting stage.
Build a working outline
Outline before drafting, with one line per section stating what that section establishes. Most literature-based papers follow a predictable shape, and knowing the job of each section stops content drifting into the wrong place.
| Section | Its job |
|---|---|
| Introduction | The problem, why it matters, the gap in existing work, your research question, and the shape of the paper. |
| Literature review | What is already known, organised by theme, and where the disagreement or gap lies. |
| Method or approach | How you answered the question: the sources, data or analytical framework, and why they were appropriate. |
| Findings or analysis | What you found, presented in a logical order, without interpretation. |
| Discussion | What the findings mean, how they relate to prior work, and what limits them. |
| Conclusion | A direct answer to the research question, its implications, and what should be studied next. |
Write the literature review as synthesis
A literature review is an argument about the state of knowledge, not a catalogue. Organise by theme, debate or methodology, and let each paragraph make a claim about the literature that several sources support, qualify or contest.
A practical test: if most paragraphs begin with an author's name, you are summarising. If they begin with a claim about the field and cite several sources in support, you are synthesising. Close the review by naming the gap your question addresses, so the reader sees why the paper exists.
Draft the paper in a sensible order
Do not draft in reading order. Write the method and findings first, because they are the most factual and least dependent on the rest. Then write the discussion, then the literature review now that you know which threads matter, then the introduction, and the abstract last, once there is something to summarise.
Draft to get the content down; do not edit sentences while drafting. Leave placeholders where a citation or figure is needed and keep moving. Getting a complete rough draft is worth far more than a polished half.
Present evidence and analysis clearly
Keep findings and interpretation apart. In the findings, report what the evidence shows. In the discussion, say what it means, how it fits the literature and where it is limited. Mixing the two makes it impossible for a marker to see what you found versus what you inferred.
Use tables and figures only when they communicate better than prose, number them, caption them, refer to them in the text and never present the same information twice.
Write limitations honestly. Naming the boundaries of your evidence demonstrates judgement, and overstating certainty is one of the fastest ways to lose credibility with a marker.
Cite consistently as you write
Add every citation at the moment you use the source. Reconstructing references from memory the night before submission is how mismatches, missing page numbers and orphaned list entries happen.
Use the referencing style your department specifies, and use it consistently for in-text citations, the reference list, and any tables or figures. If you are working in APA, you can format individual APA 7 references with our free citation generator; for other styles, follow the official manual or your library guide.
Revise, edit and proofread
Revise in layers. First check that the paper answers the question it asked and that each section does its own job. Then check the argument: does the analysis follow from the evidence, and is every claim supported? Then work at paragraph level on topic sentences and transitions. Only then proofread.
Two final passes catch most remaining problems: read the introduction and conclusion back to back to check they agree, and read the reference list against the text in both directions.
Common research paper mistakes
The recurring problems in student research papers are structural, which means they are fixable before submission if you look for them deliberately.
- A topic instead of a question, so the paper has nothing to answer.
- A literature review that summarises source by source rather than by theme.
- A method section that lists what was done without justifying it.
- Interpretation mixed into the findings.
- A conclusion that never answers the research question directly.
- Claims stronger than the evidence supports, with no limitations stated.
- References added at the end, producing mismatches and gaps.
Research paper checklist
- The research question is specific, answerable and stated in the introduction.
- The literature review is organised by theme and ends by naming the gap.
- The method or approach is described and justified, not just listed.
- Findings are reported separately from interpretation.
- The discussion connects findings to prior work and states limitations.
- The conclusion answers the research question directly.
- Every source cited has been read in full.
- In-text citations and the reference list match in both directions.
- Formatting, word count and referencing style follow the brief.
- The introduction and conclusion have been read together and agree.
If your research paper has stalled
Research papers usually stall at the question or at the literature review, and both are easier to unblock with someone who works in your field. A subject specialist can help you narrow the question, organise the literature by theme, or review a draft and explain what the analysis is missing.
