💬

How to Write a Research Paper: Step by Step

From selecting a research question to polishing your references — a practical walkthrough of every stage of writing a strong research paper.

📖 ~15 min read🔬 Research Writing✅ Updated 2025

What Makes a Good Research Paper?

A research paper is more than a longer essay. It requires you to formulate a focused question, locate and evaluate existing scholarship, construct an evidence-based argument, and contribute something — even modestly — to an ongoing academic conversation. Whether it is a 10-page undergraduate assignment or a 40-page graduate submission, the same core qualities define a strong research paper.

QualityWhat it looks like in practice
Clear research questionAnswerable, specific, and significant to the field
Credible sourcesPeer-reviewed, current, directly relevant
Logical argumentClaims supported by evidence; no logical leaps
Synthesis, not summarySources are compared, contrasted, and connected — not just reported
Correct citationEvery borrowed idea attributed; no plagiarism
Scholarly voiceFormal, precise, objective — hedging where appropriate

Unlike a personal essay, a research paper does not rely on your opinion alone. Every claim must be backed by evidence from the literature, your own data, or both. The strength of your argument depends on the quality and relevance of that evidence and on how clearly you connect it to your central thesis.

Choosing a Topic & Research Question

The single most common failure in research papers is a topic that is too broad, too vague, or too generic. "Climate change" is not a research topic. "The effect of carbon pricing policies on industrial emissions in the EU, 2015–2023" is a research topic. The difference is specificity: you can actually answer the second question; the first would require an entire library.

Moving from topic to question

Start with your general area of interest, then narrow it through a series of focusing questions:

  1. What aspect of this topic interests you most?
  2. What is already known about that aspect? (Initial literature scan.)
  3. What is debated, uncertain, or under-studied?
  4. Can you frame that gap as an answerable question?
  5. Is the question answerable within your word limit and time frame?
Topic → Research Question Examples
Too broad: The impact of technology on education
Focused: Does synchronous online instruction produce equivalent learning outcomes to in-person instruction in undergraduate introductory statistics courses?

Too broad: Mental health in the workplace
Focused: How do UK-based SME managers conceptualise and respond to employee disclosure of anxiety disorders?

Types of research questions

Evaluating Sources

Not all sources are equal. Wikipedia is a starting point, not a citation. A blog post from a practitioner has different authority to a peer-reviewed article in a top-tier journal. Knowing how to evaluate sources quickly is one of the most valuable research skills you can develop.

The CRAAP test

CriterionQuestions to ask
CurrencyWhen was it published? Is the information still current? (Science moves fast — aim for sources within 5–10 years unless citing foundational theory.)
RelevanceDoes it directly address your research question? Is the audience academic?
AuthorityWho are the authors? What are their credentials? Is this peer-reviewed?
AccuracyAre claims supported by evidence? Does it reference prior work? Can findings be replicated?
PurposeIs it informing, persuading, or selling? Who funded the research?

Best databases for academic research

Tip: Use the "cited by" feature

When you find one excellent paper, click "Cited by" in Google Scholar to find more recent work that built on it. Similarly, track the reference list backward to find foundational papers in the area. This forward-and-backward technique builds your literature map faster than keyword searching alone.

Creating an Outline

Skipping the outline is one of the most expensive time-saving shortcuts you can take. Writers who skip outlines typically write more, revise more, and produce weaker final papers. A solid outline turns a vague mess of notes into a clear argumentative map before you write a single sentence of prose.

Standard IMRaD structure

For empirical research papers in sciences and social sciences, the IMRaD format (Introduction, Methods, Results, and Discussion) is the standard. For humanities and theoretical papers, a more flexible argumentative structure is common.

1

Introduction

Hook → Background → Gap → Research question/thesis → Paper overview.

2

Literature Review / Theoretical Framework

What is known, what is contested, where your study fits.

3

Methodology (empirical papers)

Research design, data collection, analytical approach, limitations.

4

Results / Findings

What you found, presented objectively without interpretation.

5

Discussion

What findings mean, how they relate to existing literature, implications.

6

Conclusion

Summary of argument, direct answer to research question, future research.

Writing the Introduction

Your introduction must do four things: establish the context and significance of your topic, identify the problem or gap in existing knowledge, state your research question or thesis, and briefly preview how the paper proceeds.

The funnel structure

Think of the introduction as a funnel: start broad (the general field or issue), gradually narrow toward your specific question, then zoom in on your thesis. This creates natural momentum and helps readers understand where your specific question sits within the larger conversation.

Introduction Structure Example
Broad context (1–2 sentences): "Antibiotic resistance is one of the most pressing threats to global public health, responsible for an estimated 1.27 million deaths annually (Murray et al., 2022)."

Narrowing (2–3 sentences): "Despite extensive research into pharmaceutical interventions, less attention has been paid to the role of agricultural antibiotic use in accelerating resistance in human pathogens..."

Gap + thesis (1–2 sentences): "This paper argues that current EU agricultural antibiotic regulations remain insufficient to meaningfully reduce cross-resistance transmission, based on an analysis of ECDC surveillance data from 2018 to 2023."
Warning: Don't open with a dictionary definition

"According to Merriam-Webster, leadership is defined as..." is one of the weakest possible openings for an academic paper. It signals that you haven't read the scholarly literature on your topic. Open with context, a compelling statistic, a contested claim, or a knowledge gap instead.

Body Paragraphs & Evidence

Each body paragraph should develop one clear idea that advances your overall argument. A well-structured body paragraph follows the PEEL model: Point → Evidence → Explanation → Link back.

PEEL paragraph structure

P

Point

State the paragraph's main claim in one clear sentence. This is your topic sentence.

E

Evidence

Cite the source(s) that support this claim. Quote sparingly; paraphrase and summarise more.

E

Explanation

Explain what the evidence means and why it supports your point. Never let a quote speak for itself.

L

Link

Connect back to the paper's main argument or transition to the next paragraph.

Using evidence correctly

There are three ways to incorporate source material: direct quotation, paraphrase, and summary. Use quotation only when the exact wording matters — definitions, powerful statements, or contested claims where precise language is critical. Paraphrase and summary are generally preferred because they demonstrate your own understanding and keep the prose flowing in your voice.

Tip: Avoid "quote dumping"

Dropping a long block quote and moving on without explanation is quote dumping. Every piece of evidence — quoted or paraphrased — must be introduced, cited, and then explicitly explained in your own words. Your analysis is more important than the quotation itself.

Discussion & Analysis

The discussion section (or the analytical sections of a non-empirical paper) is where you demonstrate your scholarly judgment. Here you interpret evidence, compare findings across sources, identify patterns and contradictions, and explain what it all means in relation to your research question.

Discussion moves

Conclusion

The conclusion is not a summary dump. It should succinctly restate your thesis (in new words, not copy-pasted from the introduction), briefly synthesise the key findings or arguments that support it, note the most significant limitations, and point to directions for future research. A strong conclusion answers the "so what?" question: why does this paper matter?

Warning: Don't introduce new information in the conclusion

The conclusion is for synthesis, not new evidence or new arguments. If you find yourself wanting to add new content in the conclusion, that content belongs in the body of the paper.

Citations & References

Every factual claim, borrowed idea, statistic, or quotation requires a citation. The format depends on your discipline's preferred style — APA 7 is standard in social sciences and education; MLA in humanities; Chicago/Turabian in history and some humanities; Harvard in business and law; Vancouver in medicine.

Citation StyleCommon DisciplinesIn-text Format
APA 7thPsychology, Education, Social Sciences(Author, Year, p. XX)
MLA 9thHumanities, Literature, Languages(Author Page)
Chicago 17th (Notes)History, Art HistoryFootnote / Endnote
HarvardBusiness, Law, Sciences (UK/AUS)(Author, Year)
VancouverMedicine, Biomedical Sciences[Number]
Tip: Use reference management software from day one

Zotero (free), Mendeley (free), and EndNote are all excellent. They auto-format citations, store PDFs, and sync across devices. The 30 minutes you spend setting one up will save you hours of manual citation formatting and prevent the inconsistencies that markers penalise heavily.

Common Research Paper Mistakes to Avoid