How to Write a Research Paper: IMRaD Structure Guide

To write a research paper, organise it into IMRaD: an Introduction that states the problem, Methods that let others repeat the work, Results that report what you found, and a Discussion that explains what it means. A precise title, an informative abstract, specific keywords and accurate references complete the manuscript.

Most journals in microbiology, food science, clinical research and pharmaceutical sciences expect this structure. Learn it properly once and it serves your MSc dissertation, your PhD chapters and every manuscript after, because editors and reviewers know exactly where to look for each piece of information.

What IMRaD is and why journals use it

IMRaD became the dominant format in biomedical journals during the second half of the twentieth century. It mirrors the logic of an experiment: a question, a method, the evidence and an interpretation. The discipline that matters most is keeping these apart. Methods do not contain results, Results do not contain interpretation, and the Discussion does not introduce new data.

What goes in each section

Title, abstract and keywords

The title should state what was studied, in whom and how. The abstract is often the only part a reader sees, so many journals ask for a structured abstract with a strict word limit. Give key numbers, and choose keywords a researcher would actually type into PubMed or Scopus.

Introduction

Think of a funnel: what is known, what is not, why that gap matters, and finally your aim or hypothesis in the last paragraph. Three to five paragraphs is usually enough. Cite recent primary studies rather than textbooks.

Methods

Write Methods so a competent scientist in another laboratory could repeat the work. State the design, setting and dates, how samples were selected, media and instruments, the standard followed with its version, quality control strains, and the statistical tests and software. Add ethics approval wherever human participants or animals are involved.

Results and Discussion

Present results in the same order as the Methods, highlight the main patterns in the text and always report denominators with percentages. Open the Discussion with the main finding, compare it with earlier studies, including those that disagree, state limitations honestly and claim only what your data support.

Write it in this order, not front to back

1. Figures and tables: decide which few displays tell the story
2. Methods: the easiest section, often adapted from your protocol
3. Results: describe what the displays show
4. Discussion: interpret what you found
5. Introduction: set up the gap your results address
6. Abstract: summarise the finished paper
7. Title and keywords: last, once the message is settled

Use a table when readers need exact values and a figure when the pattern matters more. Each display should stand alone, with a clear title, units, n values and defined abbreviations.

Reporting guidelines and the EQUATOR Network

Reporting guidelines are checklists that define what a complete report of a given study design must contain. The EQUATOR Network (Enhancing the QUAlity and Transparency Of health Research) hosts a searchable library of them, and many journals ask for a completed checklist at submission.

• CONSORT: randomised controlled trials
• STROBE: observational studies such as cohort, case-control and cross-sectional designs
• PRISMA: systematic reviews and meta-analyses (PRISMA 2020)
• ARRIVE: research using animals (ARRIVE 2.0)

Referencing, plagiarism and AI disclosure

Zotero is free and open source, Mendeley is free and owned by Elsevier, and EndNote is a paid Clarivate product that many universities license. All three insert citations into Word and reformat them when you change journal.

Journals routinely run submissions through similarity-checking software. Close paraphrasing, or reusing large parts of your thesis or earlier papers without acknowledgement, can lead to rejection or retraction. Cite every source of ideas, data and figures.

Policies on generative AI tools differ between journals and are still evolving. Many state that such tools cannot be authors and that any use must be disclosed, so check the target journal's policy before you start. For authorship, the ICMJE criteria are widely used: substantial contribution, drafting or critical revision, final approval, and accountability for the work.

Worked example: a Kerala MSc thesis becomes a paper

An MSc microbiology graduate in Thrissur wrote a thesis on antibiotic resistance in Escherichia coli from retail chicken bought in three Kerala districts. It ran to 90 pages, including a 25-page literature review. Her supervisor suggested turning it into a journal paper.

• Focus: one question, the prevalence and resistance profile of E. coli, including ESBL producers
• Methods: sampling months, sites and numbers, identification methods, disc diffusion following CLSI with the edition used, and E. coli ATCC 25922 as the quality control strain
• Results: fourteen thesis tables became two tables and one figure, with n and percentages
• Discussion: comparison with other Indian poultry studies, One Health implications, and limitations such as convenience sampling and no molecular confirmation of resistance genes
• Reporting: STROBE as a guide for the cross-sectional design
• Authorship: her supervisor and the co-guide who ran confirmatory tests met the ICMJE criteria

Choosing a journal and writing the cover letter

Choose the journal before you finalise formatting. Match its scope and article types to your study, and confirm that it is genuinely indexed and peer reviewed; our guide on how to avoid predatory journals shows how to check.

The cover letter is a short, formal note to the editor. Give the title and article type, the main finding and why it suits the journal's readers, and confirm that the work is original, not under consideration elsewhere and approved by all authors. Declare any conflicts of interest.

Common mistakes

• Starting with the Introduction and losing weeks to the literature review
• Mixing results and interpretation
• Methods too thin to replicate, such as missing standard versions
• An abstract whose numbers do not match the paper
• Ignoring the journal's instructions for authors
• Adding gift authors or leaving out people who earned authorship
• Submitting to a journal found through an unsolicited email

Key takeaway: Structure the paper as IMRaD, but build it in the order Methods, Results, Discussion, Introduction, Abstract and Title. Use the matching reporting guideline, tie every claim to your data, and follow the target journal's instructions exactly.

Frequently Asked Questions

What does IMRaD stand for?

IMRaD stands for Introduction, Methods, Results and Discussion. It is the standard structure for original research articles in biomedical and life science journals. The core sections are preceded by a title, abstract and keywords, and followed by declarations and references. Some journals rename or add sections, so follow the instructions for authors.

Which section of a research paper should I write first?

Most authors start with the Methods, because it describes what you actually did and can be adapted from your protocol or thesis. Then prepare your figures and tables, write the Results around them, and follow with the Discussion and Introduction. Write the abstract and title last, once the main message of the paper is settled.

Can I publish my MSc or PhD thesis as a paper?

Yes, and many good papers start as theses. You will need to shorten the work heavily, focus on one clear question, turn the literature review into a brief Introduction and present only key results. Check the journal's policy on prior publication and your university's rules, and agree authorship with your supervisor early.

Can I use AI tools when writing a research paper?

Journal policies vary and are still changing. Many publishers do not allow generative AI tools to be listed as authors and require you to disclose how they were used, for example for language editing. You remain fully responsible for accuracy, originality and every citation, so read the target journal's current policy and declare any use honestly.

Next steps

To build these skills in a structured way, explore the Scientific Writing & Research Methodology course, which Innoryx offers online to learners in India and the GCC.

Before you submit, read How to spot and avoid predatory journals, and see ALCOA data integrity explained for keeping raw research data trustworthy. Unsure which course suits your stage? Speak to a counsellor.