How to write a research paper?

Writing a research paper is both a craft and a process. It blends critical thinking, methodological rigor, and clear communication—qualities that must be cultivated from project conception to publication. While the exact format can vary by discipline, most empirical papers follow the IMRaD structure (Introduction, Methods, Results, and Discussion) and adhere to broadly accepted reporting and publication standards [1,2]. This chapter walks you step-by-step through planning, drafting, revising, and submitting a research paper, with practical advice on structure, style, ethics, and journal strategy. It also highlights discipline-agnostic principles supported by reporting guidelines and publication ethics frameworks that will help your work be credible, reproducible, and impactful.

What this chapter will help you do

• Frame a compelling research question and design a feasible study.
• Use the IMRaD structure effectively and write each section with clarity and purpose.
• Report methods and results transparently and in line with recognized guidelines.
• Present tables and figures that enhance understanding rather than merely decorate.
• Engage ethically and strategically with peer review and publication processes.
• Avoid common pitfalls in style, statistics, and scholarly integrity.

Plan Before You Write

Start with the right question

The quality of a paper rests on the quality of the question. A useful heuristic is the FINER criteria—Feasible, Interesting, Novel, Ethical, and Relevant—often applied in clinical and health research but valuable across fields [22]. A question that is too broad will produce a vague paper; too narrow, and it may lack significance. Scope your question so your study can provide credible evidence and a clear contribution.

• Feasible: Do you have (or can you obtain) the data, skills, time, and resources?
• Interesting: Will it engage your target audience?
• Novel: Does it add, refine, or challenge existing knowledge?
• Ethical: Are risks and burdens justified (including for participants, if any)?
• Relevant: Does it matter to the field and to the journal’s readership?

Map the literature strategically

An effective literature review is not a catalogue; it is an argument for why your study is needed and how it extends prior work [3]. Begin by identifying clusters—what is known, what is contested, and what is missing. Organize your notes around themes rather than individual papers. As you draft, move from general background to the specific gap your study addresses; the “Create A Research Space” (CARS) model from genre analysis is particularly helpful here [4]. This rhetorical move-construction will later shape your Introduction.

Practical tips:

• Keep an annotated bibliography with brief critical evaluations of sources.
• Track seminal debates and methods, not just “latest papers.”
• Use forward and backward citation tracking to find relevant work.
• Start a living outline of your Introduction while you review; writing clarifies thinking.

Pre-register and plan for transparency

Before you collect or analyze data, preregister your study and analysis plan when appropriate. Many fields now expect pre-registration for clinical trials and encourage it for experiments and observational studies; it reduces selective reporting and strengthens credibility [2,18]. Clinical trials, for example, should be registered in recognized registries such as ClinicalTrials.gov before enrollment begins [2,26]. More broadly, adopt Open Science practices aligned with the Transparency and Openness Promotion (TOP) Guidelines: share data, code, and materials where feasible, with persistent identifiers and clear licenses [18].

Attend to ethics and authorship early

Define authorship and roles at project start and revisit them as contributions evolve. The ICMJE authorship criteria are widely used: substantial contributions; drafting or revising; final approval; and accountability for all aspects of the work [2]. Complement this with contributor role taxonomies like CRediT to specify who did what (e.g., conceptualization, data curation, formal analysis) [21]. Secure IRB/ethics approvals where needed, plan informed consent procedures appropriately, and anticipate conflicts of interest disclosures. Align data practices with FAIR principles—Findable, Accessible, Interoperable, Reusable—to ensure your research assets can be validated and reused [17].

Use the IMRaD Structure Effectively

Title and keywords: Accurate, specific, searchable

Your title should be specific enough to signal your contribution and method without being verbose. Avoid ambiguous terms and unnecessary jargon. Front-load key entities (population, intervention/exposure, comparator, outcome) where applicable. Select keywords that reflect established indexing terms to improve discoverability [1].

Abstract: A self-contained summary

Most journals require a structured abstract with headings (e.g., Background, Methods, Results, Conclusions), particularly in health and life sciences [2]. The abstract should:

• State the problem and objective.
• Summarize design, setting, participants, data sources, and primary measures.
• Report primary results with effect sizes and uncertainty (e.g., confidence intervals).
• Conclude with the main implication, not a restatement of results.

Do not include citations unless explicitly allowed. Ensure every claim in the abstract is supported in the paper.

Introduction: Motivate, locate, and state your aim

Using the CARS model, craft three moves [4]:
1) Establish the territory: What is the broader problem and why is it important?
2) Establish the niche: What is unknown, inadequate, or contested?
3) Occupy the niche: What is your research question/hypothesis and approach?

Keep it concise. Funnel from wide relevance to your precise aim. Avoid a comprehensive literature review; prioritize studies that define the gap and contextualize your approach [3].

Methods: Make replication possible

The Methods should enable a competent researcher to reproduce your work. Organize logically (e.g., design, participants/data sources, variables, interventions/exposures, outcomes, sample size, procedures, statistical analysis). For health research, consult reporting guidelines via the EQUATOR Network to ensure completeness (e.g., CONSORT for randomized trials, PRISMA for systematic reviews, STROBE for observational studies) [5–8].

Essentials:

• Design and setting: Specify study design and context.
• Participants/data: Inclusion/exclusion criteria, recruitment, sampling frames, data sources, and dates.
• Variables/measures: Definitions, instruments (with validation references), and coding.
• Procedures and materials: Interventions, protocols, software versions, and equipment.
• Statistical analysis: Primary endpoints, model specifications, handling of missing data, multiplicity adjustments, and sensitivity analyses, guided by SAMPL [9].
• Ethics: Approvals, consent, trial registration identifiers, and data/protection procedures [2].
• Transparency: Where to find data, code, and materials (e.g., repositories), aligned with FAIR [17].

Results: Present facts clearly and logically

Report results in the same order as your Methods and prespecified analysis plan. Use text to highlight key findings and direct readers to tables and figures instead of duplicating them. Follow statistical reporting norms:

• Provide effect sizes with confidence intervals for primary outcomes; avoid sole reliance on p-values [9].
• Contextualize p-values—do not over-interpret threshold crossings [10,11].
• Report denominators, numbers at risk, and the flow of participants/data as relevant (e.g., CONSORT/PRISMA diagrams) [6,7].

Tables and figures should stand alone with descriptive titles and legends. Use visual encodings that minimize cognitive load and avoid chartjunk [15,16].

Discussion: Interpret, don’t inflate

Structure your Discussion to:

• Summarize key findings succinctly relative to your objective.
• Compare with prior literature: explain concordances and discrepancies.
• Assess strengths and limitations candidly (design choices, biases, generalizability, precision).
• Discuss implications for practice, policy, or theory—calibrated to evidence strength.
• Offer future directions without promising what your data cannot support.

Avoid common pitfalls such as causal language for observational designs, overgeneralization, and “spin.” Match claims to your design and analyses [1].

References: Accurate and consistent

Cite primary sources wherever possible and ensure factual accuracy. Use a reference manager to maintain consistency with the target journal’s requirements. For biomedical and many science journals, Vancouver style is standard; the National Library of Medicine’s Citing Medicine guide details formats for all source types [25]. Verify all DOIs/URLs and accession identifiers.

Reporting Guidelines and Checklists You Should Know

Find and follow the right checklist

Use the EQUATOR Network to identify the appropriate checklist for your study type and attach it during submission if requested [5].

• Randomized trials: CONSORT 2010 [6].
• Systematic reviews and meta-analyses: PRISMA 2020 [7].
• Observational studies (cohort, case-control, cross-sectional): STROBE [8].
• Case reports: CARE [27].
• Qualitative studies: COREQ or SRQR [28,29].

Checklists do not replace good writing; they ensure you do not omit critical information. Many journals require adherence to these guidelines and may send non-compliant manuscripts back without review.

Make Your Figures and Tables Work Hard

Principles of effective data display

Effective visuals clarify the message, reduce cognitive load, and reveal patterns. Aim for:

• Integrity: Show the data honestly—no truncated axes that mislead.
• Clarity: Use consistent scales, readable labels, and sufficient contrast for accessibility.
• Parsimony: Remove non-essential ink; prefer simple encodings (position, length) over color or area [15].
• Purpose: Each figure or table answers a specific question.

For pragmatic advice, see “Ten Simple Rules for Better Figures,” which covers planning, choosing appropriate chart types, and iteration with feedback [16]. Reserve tables for precise values and figures for patterns and relationships. Always define abbreviations and symbols in legends. If complex, provide a figure supplement with details such as model diagnostics or sensitivity analyses.

Flow diagrams and study profiles

Flow diagrams convey transparency in study selection or participant flow. Use PRISMA 2020 maps for screening and inclusion decisions in systematic reviews [7]. For randomized trials, CONSORT diagrams summarize allocation, follow-up, and analysis [6]. Include numbers at each stage and reasons for exclusions.

Write with Clarity and Cohesion

Sentence and paragraph craft

Clarity arises from aligning grammatical structure with the reader’s expectations. A classic insight is to put the “topic” (old information) early and the “stress” (new, important information) at the end of the sentence; align subjects and verbs with the story’s main characters and actions [12,13]. At the paragraph level:

• Lead with a topic sentence that frames the point.
• Develop logically through evidence and reasoning.
• End by linking back to your point or forward to the next paragraph.

Prefer concrete verbs over nominalizations, cut unnecessary qualifiers, and favor active voice when it clarifies agency. Avoid long strings of nouns and undefined acronyms. For the broader narrative, Schimel emphasizes telling “the story of your science” with a coherent arc, even within IMRaD constraints [14].

Hedging responsibly

Scientific prose benefits from calibrated language—enough hedging to avoid overclaiming, but not so much that you obscure the message. Qualify claims with evidence strength (e.g., “is associated with,” “may suggest”). Match certainty to design and analysis (e.g., causal claims typically require randomized or strong causal inference designs).

Style guides and economy

Internalize a small set of rules that deliver large gains: omit needless words, place emphatic words at the end of sentences, and keep related words together [23]. Consistency, not perfection, underpins readability and professionalism.

Statistics and Reproducibility

Plan analyses and sample sizes

Design your study with statistical power and precision in mind. Conduct sample size and power analyses for primary outcomes before data collection; document assumptions about effect sizes, variability, and error rates [24]. For observational work, plan strategies to address confounding, missing data, and multiplicity. Predefine primary and secondary analyses to limit selective reporting [2,18].

Report estimates with uncertainty

Follow SAMPL guidance when reporting statistics [9]:

• Present effect estimates with confidence intervals.
• Clearly define the analysis population and units.
• Report exact p-values to two or three decimal places; avoid “p = 0.000.”
• Describe methods for handling missing data and model assumptions.
  The ASA’s statements caution against using p < 0.05 as a bright line for truth; emphasize effect sizes, uncertainty, design quality, and plausibility in interpretation [10,11].

Enable verification and reuse

Make your data and code findable and accessible when possible. Deposit them in trusted repositories, assign DOIs, and include READMEs and metadata that conform to FAIR principles [17]. State any constraints (e.g., sensitive data) and provide access procedures or de-identified datasets. Transparency is increasingly mandated and always beneficial for credibility [18].

Choose the Right Journal and Avoid Pitfalls

Target journals strategically

Select a journal before writing your final draft. Consider:

• Scope and audience: Does your study fit the journal’s aims?
• Article types and length limits: Are there word/figure constraints?
• Indexing and visibility: Is it indexed in databases your audience uses?
• Open access and fees: Are APCs acceptable and funding available?
• Data and reporting policies: Can you comply with checklists and sharing requirements?
  Use the journal’s recent articles as stylistic models. Many authors use frameworks like Think. Check. Submit. to verify journal legitimacy and fit [20]. Check the editorial board, peer-review process, and policy transparency.

Beware predatory journals

Predatory outlets solicit manuscripts aggressively, charge fees, and provide little or sham peer review. Red flags include vague scope, misleading metrics, and opaque policies. Use COPE and Think. Check. Submit. resources to assess journals; when in doubt, consult mentors or librarians [19,20].

Format, cover letter, and submission

• Follow the Instructions for Authors precisely: structure, word limits, reference style, figure formats.
• Write a brief cover letter: state the article type, novelty, and why it fits the journal; disclose related manuscripts and conflicts; confirm that all authors approve the submission [2].
• Suggest qualified reviewers and list unsuitable ones if permitted, providing institutional emails.

Preprints and prior publication

Preprints can accelerate dissemination and invite feedback. Check your target journal’s policy (many allow, some encourage) and ensure you disclose preprint deposition on submission. For clinical research, be cautious about media coverage ahead of peer review and follow community norms and ethical guidance [2,18].

Peer Review, Revision, and Beyond

Understand peer review models

Journals may use single-blind, double-blind, or open review. Each has advantages and trade-offs. COPE provides practical guidance on ethical peer review and editorial processes [19]. Read decision letters carefully; triage comments into major and minor issues.

Respond effectively

• Be courteous and specific. Respond point-by-point, quoting or summarizing each reviewer comment.
• Indicate exactly what changed (with page/line numbers) or justify why a change was not made, with evidence.
• Where you disagree, do so respectfully and provide data or citations.
• Revise the manuscript thoroughly, not just the parts explicitly mentioned; often, a comment signals a deeper clarity issue.

Post-acceptance responsibilities

Proofs must be checked promptly and carefully; this is your last chance to catch errors. Share accepted versions in line with journal policies. Deposit data/code if not already done and ensure your ORCID and acknowledgments are correct. Promote your work responsibly (press releases, social media threads, plain-language summaries), sticking to the evidence and avoiding overstatements.

Ethics and Integrity

Plagiarism, redundancy, and honest authorship

Plagiarism includes uncredited use of text, ideas, or data and is a serious breach. Self-plagiarism—reusing substantial parts of your own published text without citation—is also problematic. Familiarize yourself with COPE’s Core Practices and flowcharts for handling ethical issues [19]. Use plagiarism-detection tools as aids, not crutches, and keep careful notes on sources.

Ensure authorship follows ICMJE criteria and that contributorship is transparent (e.g., CRediT) [2,21]. Declare funding and conflicts of interest comprehensively.

Data integrity and corrections

Maintain raw data securely with version control and clear metadata. If errors are discovered post-publication, work with the editor to issue corrections or retractions as appropriate. Reporting honest limitations and uncertainties increases trust in your work.

Practical Workflow for Productivity

Develop a writing habit

Regular, scheduled writing sessions produce more and better text than binge-writing. Protect short, daily blocks for drafting and revision. Start with low-friction tasks (e.g., methods drafting, figure legends) to build momentum. Outline before you draft; revise after you rest. Use checklists (EQUATOR, SAMPL) as end-of-draft audits [5,9]. Write to communicate, then edit to conform to style and format.

Use tools wisely

• Reference managers (e.g., Zotero, EndNote) to handle citations and format conversions.
• Version control (e.g., Git) or shared documents with tracked changes for collaboration.
• Reproducible analysis ecosystems (e.g., R Markdown, Jupyter) to integrate code and narrative.
• Templates from target journals to streamline formatting.

Common Pitfalls and How to Avoid Them

Vague questions and scope creep

Refine your research question and analysis plan before you begin. Resist adding exploratory analyses to the main narrative; label them clearly as exploratory and consider preregistered secondary analyses in follow-up work [2,18,22].

Underpowered studies and overconfident claims

Underpowered designs inflate false-positive and false-negative risks. Plan sample size and interpret uncertainty honestly. Emphasize effect sizes and confidence intervals over dichotomous p-value thresholds [9–11,24].

Opaque methods and irreproducible results

Write Methods for a competent colleague outside your team. Share code and data where feasible and provide enough detail to replicate (software versions, parameter settings, seeds). Align with FAIR and TOP practices [17,18].

Poor visuals

Avoid clutter, misleading axes, and ambiguous legends. Design figures to answer specific questions, and iterate with colleagues until the message is unmistakable [15,16].

Style issues that obscure meaning

Watch for:

• Long, abstract sentences laden with nominalizations.
• Buried verbs and missing agents.
• Shifting terms for the same concept.
  Revise with the reader in mind; clarity is a service to your audience [12–14,23].

A Mini Checklist Before Submission

Content completeness

• Title and abstract accurately reflect the work.
• Introduction motivates a clear, gap-focused objective [3,4].
• Methods enable replication; ethics and registration included [2,5–7,17,26].
• Results align with prespecified analyses; estimates and uncertainty reported [9–11].
• Discussion interprets without overclaiming; limitations explicit [1].
• References complete and consistent with Vancouver style [25].

Transparency and compliance

• Relevant EQUATOR checklist completed (CONSORT, PRISMA, STROBE, etc.) [5–8,27–29].
• Data/code availability statements included; repositories linked [17,18].
• Authorship and contributorship clear; conflicts disclosed [2,21].
• Journal scope and formatting requirements met; figures/tables compliant [2,20].

Conclusion

Writing a research paper is a disciplined journey from question to credible claim. Begin with a focused, consequential question; design and preregister your study to answer it; write with rhetorical clarity and methodological transparency; report statistics responsibly; and choose a journal that reaches your intended audience. Use reporting checklists to ensure completeness, visuals to underscore your message, and ethical frameworks to guide decisions. Above all, remember that clear writing reflects clear thinking. Cultivate both, and your work will travel further and age better.

References

1. Day RA, Gastel B. How to Write and Publish a Scientific Paper. 8th ed. Santa Barbara (CA): Greenwood; 2016.
2. International Committee of Medical Journal Editors. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals [Internet]. Updated 2024 [cited 2025 Aug 31]. Available from: http://www.icmje.org/recommendations/
3. Booth WC, Colomb GG, Williams JM. The Craft of Research. 4th ed. Chicago (IL): University of Chicago Press; 2016.
4. Swales JM. Genre Analysis: English in Academic and Research Settings. Cambridge (UK): Cambridge University Press; 1990.
5. EQUATOR Network. Enhancing the QUAlity and Transparency Of health Research (EQUATOR) Network [Internet]. Oxford (UK): EQUATOR Network; c2006–2025 [cited 2025 Aug 31]. Available from: https://www.equator-network.org/
6. Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 Statement: Updated Guidelines for Reporting Parallel Group Randomised Trials. BMJ. 2010;340:c332.
7. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews. BMJ. 2021;372:n71.
8. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies. PLoS Medicine. 2007;4(10):e296.
9. Lang TA, Altman DG. Basic Statistical Reporting for Articles Published in Biomedical Journals: The SAMPL Guidelines. In: Smart P, Maisonneuve H, Polderman A, editors. Science Editors’ Handbook. Oxford (UK): European Association of Science Editors; 2013.
10. Wasserstein RL, Lazar NA. The ASA’s Statement on p-Values: Context, Process, and Purpose. The American Statistician. 2016;70(2):129-133.
11. Wasserstein RL, Schirm AL, Lazar NA. Moving to a World Beyond “p < 0.05”. The American Statistician. 2019;73(sup1):1-19.
12. Gopen GD, Swan JA. The Science of Scientific Writing. American Scientist. 1990;78(6):550-558.
13. Williams JM. Style: Lessons in Clarity and Grace. 12th ed. New York (NY): Pearson; 2017.
14. Schimel J. Writing Science: How to Write Papers That Get Cited and Proposals That Get Funded. New York (NY): Oxford University Press; 2012.
15. Tufte ER. The Visual Display of Quantitative Information. 2nd ed. Cheshire (CT): Graphics Press; 2001.
16. Rougier NP, Droettboom M, Bourne PE. Ten Simple Rules for Better Figures. PLoS Computational Biology. 2014;10(9):e1003833.
17. Wilkinson MD, Dumontier M, Aalbersberg IJ, et al. The FAIR Guiding Principles for Scientific Data Management and Stewardship. Scientific Data. 2016;3:160018.
18. Nosek BA, Alter G, Banks GC, et al. Promoting an Open Research Culture. Science. 2015;348(6242):1422-1425.
19. Committee on Publication Ethics (COPE). COPE Core Practices [Internet]. Hampshire (UK): COPE; 2017 [cited 2025 Aug 31]. Available from: https://publicationethics.org/core-practices
20. Think. Check. Submit. Choose the Right Journal for Your Research [Internet]. London (UK): Think Check Submit; 2015 [cited 2025 Aug 31]. Available from: https://thinkchecksubmit.org/
21. Brand A, Allen L, Altman M, Hlava M, Scott J. Beyond Authorship: Attribution, Contribution, Collaboration, and Credit. Learned Publishing. 2015;28(2):151-155.
22. Hulley SB, Cummings SR, Browner WS, Grady D, Newman TB. Designing Clinical Research. 4th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2013.
23. Strunk W Jr, White EB. The Elements of Style. 4th ed. New York (NY): Pearson; 1999.
24. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988.
25. Patrias K, Wendling DL. Citing Medicine: The NLM Style Guide for Authors, Editors, and Publishers [Internet]. 2nd ed. Bethesda (MD): National Library of Medicine (US); 2007 [updated 2015; cited 2025 Aug 31]. Available from: https://www.ncbi.nlm.nih.gov/books/NBK7256/
26. ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US); 2000– [cited 2025 Aug 31]. Available from: https://clinicaltrials.gov/
27. Gagnier JJ, Kienle G, Altman DG, Moher D, Sox H, Riley D; CARE Group. The CARE Guidelines: Consensus-based Clinical Case Reporting Guideline Development. Journal of Clinical Epidemiology. 2014;67(1):46-51.
28. Tong A, Sainsbury P, Craig J. Consolidated Criteria for Reporting Qualitative Research (COREQ): A 32-item Checklist for Interviews and Focus Groups. International Journal for Quality in Health Care. 2007;19(6):349-357.
29. O’Brien BC, Harris IB, Beckman TJ, Reed DA, Cook DA. Standards for Reporting Qualitative Research: A Synthesis of Recommendations. Academic Medicine. 2014;89(9):1245-1251.

I hope this additional information is helpful in understanding the format of a research paper. Is there anything else you would like to know?