Writing a Scientific Paper: IMRaD Structure for Undergraduates

You’ve spent weeks collecting data, running regressions, and staring at spreadsheets. Now you need to write a paper. The only thing you’re sure of is that you have no idea how to structure it.

That’s normal. Most undergraduates encounter their first formal research paper without any structural training. You might have read a dozen journal articles, but you never had someone explain what each section actually does or why the order matters.

Here’s the thing: academic writing has a structure that works. It’s called IMRaD — Introduction, Methods, Results, Discussion — and it’s used by journals across every discipline. Understanding this structure is the single biggest factor that separates a good paper from a poorly structured one.

In fact, a 2023 study found that 73% of undergraduate paper failures are structural errors — not content errors. Students had good ideas, but they didn’t know where to put them. When the research itself is solid but the writing isn’t, even strong work gets rejected.

This guide will walk you through every IMRaD section, what to write in it, how much to write, and common mistakes to avoid. By the end, you’ll know exactly how to organize your paper so reviewers don’t reject it on structural grounds.

• IMRaD stands for Introduction, Methods, Results, Discussion — it’s the default structure for empirical research papers across nearly all disciplines.
• The hourglass model is your mental map: start broad, narrow down to your specific data, then broaden back out to implications.
• The One-Sentence Test tells you whether a claim belongs in Results or Discussion: if it contains “suggests,” “implies,” or “means,” it goes in Discussion.
• Write in this order to avoid endless rewriting: Methods → Results → Discussion → Introduction → Abstract.
• Discipline conventions vary: natural sciences emphasize Methods reproducibility, social sciences expect detailed Literature Reviews, humanities often rename or restructure the sections entirely.

What Is IMRaD — And Why Does It Matter?

IMRaD stands for IMtroduction, Methods, Results, and Discussion. It’s not an acronym you made up — it’s the actual name of the structure used by thousands of peer-reviewed journals.

When you see a paper organized into four main sections (sometimes with an Abstract and/or Conclusion tacked on), that’s IMRaD. You’re not expected to deviate from it unless your discipline has a specific reason to do so.

The structure exists for a reason: it mirrors how scientific thinking works. You start by stating what’s known (Introduction), explain what you did (Methods), present what you found (Results), and explain what it means (Discussion).

The IMRaD structure isn’t just a formatting convention — it’s a logical argument. Your paper makes one continuous case: “Here’s a problem, here’s how I investigated it, here’s what I found, and here’s why it matters.”

Understanding IMRaD is why research papers get rejected. Most rejections aren’t about the quality of research; they’re about how the paper is written. When content is in the wrong section — interpretation in Results, raw data in Discussion — reviewers lose trust in the paper’s logic before evaluating the science. IMRaD eliminates this by giving every piece of content a defined home. The format that peer-reviewed journals expect is called IMRaD, and each section answers exactly one question. Getting that question right changes everything about how reviewers read your work.

The Hourglass Model — Your Mental Map

Before diving into each section, here’s the single most useful mental image you can carry into your writing:

The hourglass model.

Think of your paper as an hourglass:

• Top bulb (broad): Introduction — you start with the big picture, the field, the gap
• Narrow neck (specific): Methods and Results — you zoom all the way into your specific procedures and your specific data
• Bottom bulb (broad): Discussion — you zoom back out, connecting your findings to the broader literature and future directions

[Broad: Introduction]
↓
[Narrow: Methods + Results]
↓
[Broad: Discussion]

Why does this matter? Because most undergrads write the wrong way around. They start with broad context in the Discussion when they should be there at the end. They bury their most specific data too deep in a section that should be broad.

The hourglass isn’t just pretty. It’s why IMRaD works. The structure mirrors the logic of research itself: you state a question, describe how you answered it, report what you found, and explain what it means.

A practical way to think about it: Your Introduction moves from general to specific. Your Discussion moves from specific back to general. The single most common IMRaD mistake — interpretive language leaking into Results — is exactly what the hourglass prevents.

Introduction — The Top Bulb

The Introduction is where you convince readers why your paper matters. It serves four key jobs:

1. Establish the background and context
2. Summarize relevant prior work to build context
3. Identify the research gap
4. State your research question or hypothesis

The George Mason University Writing Center describes the Introduction as a structured argument with three core moves: establish the territory, expose the gap, and announce how this study fills it. Source

Word Count Guidance

For an undergraduate paper (8–15 pages total), aim for:

• Introduction: 15–20% of total word count (roughly 450–700 words for a 3,000-word paper)
• Methods: 20–25% (600–750 words)
• Results: 15–25% (450–750 words)
• Discussion: 20–30% (600–900 words)

How to Structure an Introduction (The “Funnel”)

Think of the Introduction as a funnel: start broad, narrow down, and end with your specific question.

1. Opening hook — What’s the bigger problem? (1–2 paragraphs)
2. Literature review / background — What do we already know? What do we not know? (2–3 paragraphs)
3. Research gap — What’s missing in the literature? (1 paragraph)
4. Your study — What did you do to address that gap? What’s your hypothesis? (1 paragraph)

Common Mistake — Introduction: Writing the Introduction as a comprehensive review of every paper ever published on the topic, instead of a focused argument that leads logically to your specific research gap. If a paragraph doesn’t directly advance the case for your research question — cut it.

Tense Conventions

Example: Good Introduction Moves

Here’s a simplified version showing how the funnel works (adapted from CASRAI’s worked example):

Opening (broad): “Urbanization has transformed the way people access green spaces across North American cities in recent decades.”

Background (narrowing): “Numerous studies have linked urban green spaces to improved mental health outcomes, including reduced stress levels and lower rates of depression.”

Gap: “However, most of these studies focused on adult populations. Few have examined how park accessibility correlates with well-being among university students.”

Your study: “In this study, we surveyed 340 undergraduate students at [University Name] to examine the relationship between perceived park accessibility and self-reported well-being scores.”

Key takeaway: The last paragraph of your Introduction should state exactly what your paper does. Don’t make readers guess. The Introduction should end at the stated hypothesis, not the outcome.

Methods — The Narrow Neck (Part 1)

The Methods section answers one question: Could someone else replicate your study based on this description?

If a reader can’t replicate your study from your Methods alone, something is missing. Period. The George Mason University Writing Center emphasizes that a good Methods section lets a competent peer reproduce the study. Source

What to Include

1. Participants / sample — Who were they? How many? How did you recruit them?
2. Materials / instruments — What did you use? (surveys, equipment, software)
3. Procedure — What exactly did you do, step by step?
4. Data analysis — What statistical tests or qualitative methods did you use?

Tense Conventions

The Methods section should be written in past tense throughout. You’re describing what was done. Only exception: present tense when pointing a reader to a figure or table (“Table 1 shows the inclusion criteria”).

Common Mistake: Over-Reporting vs Under-Reporting

The Editage checklist for Methods includes: detailed enough for replication, reports sample size and sampling, names the exact statistical tests and software version, includes ethics approval and consent where relevant, written in past tense.

Pro tip: If you used an established instrument (like the PHQ-9, Big Five Personality Inventory, or a validated scale), cite the original paper and briefly describe how you administered it. You don’t need to reproduce the entire instrument.

Results — The Narrow Neck (Part 2)

This is where students make the most structural mistakes. The Results section has one job: report what you found, without interpretation.

What to Include

1. Descriptive statistics — Means, standard deviations, ranges
2. Inferential statistics — t-tests, ANOVAs, correlations, regression results
3. Tables and figures — Present them clearly with captions
4. Null results — If you found nothing significant, report it. Don’t hide them.
5. Effect sizes and confidence intervals — Report exact statistics, not just p-values

The One-Sentence Test (SAMVIK Research)

Here’s a practical rule that resolves the Results vs Discussion boundary:

Read every Results sentence. If it contains “suggests,” “implies,” “indicates,” “because,” or “therefore” — it belongs in Discussion.

This is the single most common IMRaD mistake. When the moment “suggests” or “indicates” appears in your Results section, you’ve crossed the line.

What to avoid: Do not explain why a result is significant in the Results section. That’s Discussion territory. Do not repeat every number from a table in the text. Instead, highlight the pattern the reader should notice, then point to the table or figure for the full detail.

Tense Conventions

Example: The One-Sentence Test in Action

The fastest way to understand IMRaD is to see what goes wrong and how the corrected version reads differently.

Wrong — Interpretation in Results: “The intervention group showed a 23% reduction in symptoms (p = 0.003), which suggests that the programme is effective.”

Correct — Data Only in Results: “The intervention group showed a statistically significant 23% reduction in symptoms compared to the control group (p = 0.003, 95% CI: 14–32%).”

The phrase “which suggests… should be considered” is interpretation — it belongs in Discussion, not Results.

Discussion — The Bottom Bulb

The Discussion is where you zoom back out of the hourglass. You’re answering: “What does all this data actually mean?”

What to Include

1. Summary of key findings (1 paragraph) — State your most important discovery directly and confidently.
2. Interpretation — How does your study relate to previous research? (2–3 paragraphs)
3. Limitations — What went wrong? What should be acknowledged? (1 paragraph)
4. Implications and future directions — What should happen next? (1–2 paragraphs)

The Discussion opens with the key finding, moves through comparison with prior work, then limitations, and closes with implications. This mirror-image of the Introduction, moving from specific back to general, gives readers a satisfying sense of closure.

Common Discussion Mistake: Overstating Findings

Pro tip: Start the Discussion by answering the research question, not by restating the objective or summarizing the Methods. Write your Discussion’s opening sentence last — after the section is complete, come back and write one confident sentence about your key finding.

Tense Conventions

The “So What?” Paragraph

Every Discussion needs a strong final paragraph answering: Why should anyone care?

For undergraduates, this might look like:

“Our findings add to a growing body of evidence linking urban green space access to student well-being. For university administrators, this suggests that investing in campus green spaces may not only improve student satisfaction but also contribute to mental health outcomes. Future research should examine whether the quality of green spaces — not just their accessibility — predicts well-being.”

Common Undergraduate Mistakes (And How to Fix Them)

Mistake 1: Putting Interpretation in Results

Wrong: “The correlation of r = .34 suggests that park access improves well-being.”

Correct: “Park accessibility and well-being were positively correlated, r = .34, p < .01.”

Then move the “suggests” sentence to Discussion. The phrase “which suggests… should be considered” is interpretation — it belongs in Discussion, not Results.

Mistake 2: Reproducing Your Results in the Discussion

Don’t just restate the results in the Discussion. The GMU Writing Center lists “Discussing the findings” as a distinct move from “Presenting the results.” The Discussion should feel like a conversation with the literature, not a repetition of what you already wrote.

Mistake 3: Ignoring Limitations

Every study has limitations. Hiding them doesn’t make them go away. Acknowledge your sample size, your measurement methods, your generalizability constraints.

What to avoid: Don’t sound defensive. Frame limitations as opportunities for future research. The Editage checklist flags “burying limitations” as a common problem.

“While our convenience sample limits generalizability, the effect size (r = .34) is comparable to previous studies using representative samples (Li et al., 2020), suggesting the association may be robust across sampling strategies.”

Mistake 4: Weak Literature Review in the Introduction

If your Introduction only references 2–3 sources and doesn’t address the broader literature, reviewers will question whether you understand the field. The CASRAI worked example explicitly warns against previewing results in the Introduction and instead requires a focused narrowing to the gap.

Fix: Aim for at least 8–12 peer-reviewed sources in your Introduction. You can build on our literature review guide to strengthen this section.

Mistake 5: Abstract That Doesn’t Match the Paper

A common rookie mistake: writing the paper first, then the abstract from memory, and then realizing the abstract doesn’t accurately reflect what you actually wrote.

Fix: Write the abstract last. It should be a precise, standalone summary of every section. See our abstract writing guide for specifics. The GMU Writing Center recommends 150–300 words covering background, methods, results, and conclusion.

Discipline Variants — IMRaD Isn’t One-Size-Fits-All

While IMRaD is nearly universal, the way it’s expressed changes by discipline. Understanding these shifts will save you from formatting mistakes before you even start writing. CASRAI maps how the skeleton itself bends, gets renamed, gets split, or gets dropped entirely across different fields.

Natural Sciences (Biology, Chemistry, Physics)

• Structure: Standard IMRaD
• Heavy emphasis: Methods reproducibility (detailed, precise)
• Typical additions: Supplementary materials for large datasets
• Note: Results are usually heavily tabulated; Discussion is concise
• Naming variant: Many use “Materials and Methods” rather than “Methods” — same content, different label

Social Sciences (Psychology, Sociology, Education)

• Structure: IMRaD with an expanded Literature Review section (sometimes merged into Introduction, sometimes standalone)
• Heavy emphasis: Theoretical framing in the Introduction
• Typical additions: Ethics statement, limitations section
• Note: Effect sizes and confidence intervals are expected in Results, not just p-values
• Qualitative research: Frequently replaces “Results” with “Findings,” organized thematically rather than by hypothesis

Engineering / Computer Science

• Structure: Often IMRaD-style but sometimes labeled differently
• Heavy emphasis: System architecture and technical design
• Typical additions: A dedicated “Related Work” section after the Introduction (not folded into Introduction as biomedical IMRaD does)
• Naming variants: “Methods” becomes “Approach,” “System Design,” or “Implementation”; “Results” becomes “Evaluation” or “Experiments”
• Note: Discussion is often compressed or folded into the Evaluation/Results section, followed directly by a Conclusion

Humanities

• Structure: Often not IMRaD. Humanists frequently write essays with fewer labeled sections.
• Typical approach: Thematic sections with arguments developed through evidence
• Note: IMRaD rarely applies. Most humanities scholarship does not use IMRaD at all, and this isn’t an omission to correct. IMRaD is built for reporting original empirical research with a discrete methods-generates-results-generates-interpretation process, and a thesis-driven argumentative essay doesn’t have that process.

Pro tip: The reliable answer to “what structure should I use?” is never “pick the version you prefer.” It’s: (1) the target journal’s or conference’s current author guidelines, (2) any reporting guideline that governs your study type, (3) 2–3 recently published papers from that venue, and only then (4) standard IMRaD as a fallback.

Ethical AI Use — What Your Professor Really Wants

A 2024 analysis of undergraduate writing identified Mistake #24: unauthorized AI use as one of the most common academic integrity violations. Students are increasingly using AI for writing assistance, but the line between ethical support and academic dishonesty is blurry.

Many major publishers and journals allow AI as a writing aid but prohibit listing it as an author, since it can’t take responsibility for the work. You remain accountable for everything: verify every citation, and never let AI generate data, results, or analysis.

The Boundary

Essays-Panda’s Mistake #24 highlights the consequences: when students use AI beyond tutoring or grammar checking, they risk academic integrity violations. Many universities already have AI-use policies in place.

The Safe Path

1. Check your department’s policy. Some departments allow AI as a drafting tool with attribution; others ban it entirely.
2. If permitted, cite AI use. Most universities require you to list AI tools used in an acknowledgment section.
3. Never submit AI-generated content without attribution. If a sentence doesn’t sound like you, it probably isn’t.

Recommended Writing Order

Here’s the counterintuitive advice every graduate student and professor wishes they’d known as an undergraduate:

Write in this order: Methods → Results → Discussion → Introduction → Abstract

Why? Because you can’t write a coherent Introduction until you know what you actually did (Methods) and what you found (Results). Writing the Introduction first means you’ll rewrite it at least three times as you discover what your paper actually says.

This approach can save hours of rewriting. Trust it. The Editage and SAMVIK guides both recommend this order for experienced researchers.

Structural coherence test: Place your Introduction’s final research objective sentence beside your Conclusion’s final sentence. They should feel like a matched pair — question and answer. If they don’t align, your paper has a gap reviewers will find.

FAQ — Questions Students Ask About IMRaD

Can I merge Methods and Results?

Some disciplines (especially social sciences) combine them. Check with your advisor. In natural sciences, they’re usually separate. Some journals and paper types allow or require a combined “Results and Discussion” section, particularly for multi-experiment papers reporting one study at a time.

Does the Discussion always come at the end?

Yes. In IMRaD, the Discussion is the final major section before the Conclusion or References. Some journals add a separate Conclusion heading after the Discussion.

What if my paper isn’t empirical?

If your paper is theoretical, philosophical, or literary analysis, IMRaD likely doesn’t apply. Most humanities scholarship does not use IMRaD at all.

Should I include negative results?

Yes. Null results and unexpected findings are valid and worth reporting. They can be as informative as positive findings.

How long should each section be?

For a 3,000-word paper: Introduction ~500 words, Methods ~700 words, Results ~500 words, Discussion ~600 words. Total should roughly follow the hourglass ratio. A common split is roughly Introduction 15%, Methods 25%, Results 30%, and Discussion 30% of the word count.

What is the difference between IMRAD and IMRaD?

There is no real difference. IMRaD simply lowercases the “a” from “and,” which is not a section. Both spellings describe the same Introduction, Methods, Results, and Discussion structure.

In what order should I write IMRaD sections?

Most experienced researchers write Methods first (the study is already done), then Results, then Discussion, then Introduction, and finally the Abstract. Writing Introduction last produces a sharper argument because you know your findings before building the case for them.

Next Steps — Wrapping Up

IMRaD isn’t a rigid template. It’s a logical argument about what you investigated, how you investigated it, what you found, and what it means.

Your paper follows the hourglass: broad → narrow → broad. Your Introduction starts with the field, your Methods and Results zoom into your specific study, and your Discussion zooms back out.

The single biggest structural mistake students make is not knowing where to put information. With the One-Sentence Test — “suggests, implies, belongs in Discussion” — and the recommended writing order — Methods → Results → Discussion → Introduction → Abstract — you have the tools to write a paper that won’t trip on structure.

Bottom line: If you follow IMRaD, your paper will pass the structural check. Everything else — content, analysis, writing quality — is secondary to getting the organization right.

Need help turning your research into a polished, submission-ready paper? Our team of qualified writers can help you craft every IMRaD section with discipline-appropriate conventions. Visit Place-4-Papers.com for 24/7 academic writing support.