How to Write a Research Paper Structure: Complete Guide (IMRaD Framework)

A research paper follows the IMRaD structure (Introduction, Methods, Results, Discussion), which is the standard framework used by 80% of scientific journals worldwide. Understanding this structure isn’t just about following rules — it’s about learning how to tell a compelling scientific story that your readers can follow from start to finish.

Key Takeaways

• The IMRaD structure (Introduction, Methods, Results, Discussion) follows a wineglass/hourglass shape: broad context → specific data → broad implications
• Most students draft in the wrong order. The recommended sequence is Methods → Results → Discussion → Introduction → Abstract
• The biggest structural mistake students make? Putting interpretation and analysis in the Results section instead of the Discussion
• Literature review placement varies by discipline — embedded in the Introduction for journal papers, standalone for theses and dissertations

The IMRaD Structure Explained

IMRaD stands for Introduction, Methods, Results, and Discussion. This structure didn’t always dominate scientific writing. In fact, it took half a century to become the universal standard.

The peer-reviewed historical survey by Sollaci and Pereira (2004) tracked 1,297 articles across four major medical journals — BMJ, JAMA, The Lancet, and NEJM — from 1935 to 1985. They found that IMRaD reached 80% adoption by the 1970s and became the only pattern adopted by the 1980s. The New England Journal of Medicine (NEJM) mandated the format in 1975, followed by the BMJ in 1980 and JAMA and The Lancet in 1985. This wasn’t a gradual trend — it was a revolution^[1].

Today, IMRaD isn’t just a format preference. Many journals require it explicitly. Understanding why this structure became standard helps you write a paper that feels natural rather than forced.

Wineglass Model: Why Structure Matters

The IMRaD structure follows what researchers call the wineglass model (or hourglass structure). It’s called this because of its distinctive shape: broad at the top, narrow in the middle, broad again at the bottom.

The top section (Introduction) starts broad with general context and narrows down to your specific research question. The middle section (Methods and Results) is narrow and highly specific to your study. The bottom section (Discussion) starts narrow, focused on your specific findings, then widens to discuss broader implications.

This “top-bottom symmetric shape” represents the symmetry of your story^[2]. Your Introduction asks, “Why should anyone care about this question?” Your Methods ask, “How did I investigate it?” Your Results ask, “What did I find?” And your Discussion asks, “What does it all mean?”

The wineglass model isn’t decorative. It explains why papers structured this way are easy to follow. Your reader starts with general context they understand, zooms into the specific work you did, then zooms back out to see how your work fits into the bigger picture.

The 7 Parts of a Research Paper

A complete research paper has nine distinct parts, even though the IMRaD framework highlights four main sections. Here’s the full sequence:

1. Title — Clear, descriptive, and specific enough that readers can determine relevance
2. Abstract — A standalone summary of the entire paper, including purpose, methods, key results, and conclusions. Despite appearing first, it’s written last (as discussed below)
3. Introduction — Background context, research gap, and your specific research question or hypothesis
4. Literature Review — A synthesis of existing research relevant to your study. Placement varies by discipline (more on this below)
5. Methods — Detailed description of how you conducted your study, including materials, procedures, and analysis
6. Results — Objective presentation of your findings without interpretation
7. Discussion — Interpretation of results, comparison with existing literature, limitations, and implications
8. Conclusion — Summary of key findings, implications, and suggestions for future research
9. References — Complete list of all sources cited in the paper

Each part serves one specific purpose. Trying to do multiple things in one section creates confusion. Keep the boundaries clean.

Introduction

The Introduction is where you establish why your research matters. According to the Swales 3-move framework — widely taught in university writing centers — a strong Introduction has three moves^[3]:

1. Establish Territory — Provide background context and show why the topic is important
2. Establish Niche — Identify the gap in existing research or the problem your study addresses
3. Occupy the Niche — State your research question, hypothesis, or the purpose of your paper

For the Introduction section, we have a dedicated guide that walks through each move with examples and sample language. If you want to learn more about crafting a compelling Introduction, check out our guide on how to write a research paper introduction.

The Introduction typically takes 10–15% of your total word count. Don’t rush through it, but don’t let it swallow the rest of the paper either. A common mistake is making the literature review too long when it’s folded into the Introduction — this crowds out your actual research question.

Methods

Your Methods section is the blueprint of your study. Anyone reading it should be able to replicate your work exactly. The GMU Writing Center emphasizes that this section is about transparency, not creativity^[4].

Include:

• Study design and rationale
• Participants or materials (with sampling details)
• Procedures and instruments used
• Data collection methods
• Statistical or analytical approach

The Methods section typically takes 10–15% of your word count. It should be detailed enough for replication but concise enough to maintain reader engagement.

For an in-depth look at crafting this section, see our guide on writing your research paper methodology section.

Results

Here’s where most students stumble. The Results section has one purpose and one purpose only: present your findings without interpretation.

The GMU Writing Center puts it clearly: “In the Results section, you simply report your findings. In the Discussion section, you comment on them.” This is the single most important structural boundary in a research paper^[5].

What goes in Results:

• Statistical outputs (t-tests, ANOVA, regression coefficients)
• Descriptive statistics (means, standard deviations, ranges)
• Tables and figures that summarize your data
• Raw numerical findings

What does NOT go in Results:

• Interpretation of what the findings mean
• Comparison with previous studies
• Speculation about why you got these results
• Discussion of limitations

The Results section takes 15–20% of your word count. Let your data speak. Save the commentary for the Discussion.

Discussion

If the Results section is “what happened,” the Discussion section is “what it means.” This is where you connect your specific findings back to the broader context — widening the wineglass from narrow to broad.

The Discussion should include:

• A brief summary of key findings (without repeating every statistic)
• Interpretation: what do these results actually tell us?
• Comparison with existing literature (does this confirm or contradict prior work?)
• Limitations of your study
• Implications for practice or future research

The Discussion typically takes 20–25% of your word count — often the longest section. Resist the urge to overclaim. Your Discussion should be measured, cautious, and grounded in what your data actually show.

Results vs Discussion: The Critical Boundary

Confusing these two sections is the single most common structural error. If you’re writing “This finding suggests that…” — that belongs in the Discussion, not the Results.

Writing Order: The Counterintuitive Sequence Students Should Use

Here’s a piece of advice most students never hear until after they’ve already written three terrible drafts: write in reverse order.

The intuitive (but wrong) sequence is:

• ❌ Abstract → Introduction → Methods → Results → Discussion

The recommended drafting sequence is:

• ✅ Methods → Results → Discussion → Introduction → Abstract

Why? The American Medical Writers Association (AMWA) explains this clearly: writing Methods first forces you to know exactly what you did. Writing Results next forces you to present what you found. Then the Discussion lets you interpret what you’ve already established^[6].

The trap most students fall into is writing the Introduction first. They craft a grand, sweeping opening that promises more than their data actually deliver. By the time they reach the Discussion, they realize the Introduction overpromised. This creates awkward padding, hand-waving, and a paper that reads as inconsistent.

Here’s what we recommend: start with the Methods. It’s the section that describes what you actually did — there’s nothing to fabricate or embellish. Then write the Results. You’re literally just listing what you found. Then the Discussion, where you connect those findings to the broader field. Finally, write the Introduction and Abstract — now that you know exactly what the paper contains.

This isn’t just theory. Multiple university writing centers and academic writing services (including Ref-n-Write and the AMWA) explicitly recommend this sequence^[7]. It’s the single most practical structural insight you’ll learn in this article.

When to Deviate from IMRaD

IMRaD is dominant — but not universal. Some disciplines use entirely different structures:

Humanities papers often follow a thematic or argumentative structure. Instead of Methods and Results, they organize around themes, claims, or chronological periods. The structure serves the argument, not the research design.

Law papers typically follow the IRAC framework (Issue, Rule, Analysis, Conclusion) or a similar argument-driven structure. Legal analysis doesn’t follow empirical methodology.

Social science papers sometimes use a hybrid. Qualitative studies may blend Methods and Discussion. Some fields use a “discursive” structure that’s more essay-like.

If you’re unsure which structure your professor expects, check the assignment guidelines. Many humanities and social science assignments explicitly state “no IMRaD required” or “traditional essay format.” Don’t force an IMRaD structure where a thematic approach is more appropriate.

Common Student Mistakes

Here are the structural errors we see most frequently in undergraduate research papers:

1. Putting analysis in Results — Already covered above, but it’s so common it bears repeating. If you’re interpreting, you’re in the Discussion.

2. Making the Introduction too long — The Introduction should be 10–15% of the paper, not 30–40%. Most students pad it with excessive literature review when that material belongs either in the Literature Review section or the Discussion.

3. Repeating Results in Discussion — Don’t restate every statistic. Summarize the key finding in one sentence, then move to interpretation.

4. Writing the Abstract before finishing the paper — You can’t summarize a story you haven’t written. Save the Abstract for last.

5. Neglecting to link back — The Discussion should echo the Introduction. If your Introduction raised a specific question, the Discussion should answer it explicitly.

6. Including raw data — Raw datasets belong in appendices or supplementary materials, not in the Results section. Present summarized data.

7. Overstating conclusions — If your sample size was 50 students from one university, don’t claim “this proves that…” Say “this suggests” or “these findings indicate.”

Word Count Distribution

If you’re writing a 5,000-word paper, here’s a practical breakdown of how the word count should distribute across sections:

These ranges come from structured academic writing guides (ThesisAI, Editage, and UCSD Psychology) and represent the most common proportional distribution for empirical papers^[8].

What to Avoid

Here’s a quick checklist of structural pitfalls to avoid:

• Don’t put your hypothesis in the Methods — it belongs in the Introduction
• Don’t discuss limitations in the Results — that’s Discussion territory
• Don’t add new information in the Conclusion — the Conclusion summarizes, not introduces
• Don’t cite sources in the Methods section unnecessarily — cite sources when justifying your methodology choices, but not when describing standard procedures
• Don’t use first-person pronouns indiscriminately — many disciplines prefer passive voice in Methods, but active voice is increasingly acceptable
• Don’t mix Results and Discussion — keep them separate, even when results and interpretation feel intertwined
• Don’t forget to mention ethical approvals — include IRB approval, consent, and anonymity statements in the Methods

Final Thoughts

The IMRaD structure isn’t an arbitrary set of rules. It’s a story framework — one that mirrors how humans naturally process information: context first, specifics next, implications last. Understanding the wineglass model and drafting in the recommended reverse order will save you hours of revision.

If you’d like dedicated guidance on any specific section, check out our guide on how to write a research paper introduction or our research paper methodology section guide.

And if you’re feeling overwhelmed — which is completely normal — remember that Essays-Panda has expert writers across all academic disciplines. We’ve written thousands of research papers for students just like you. Visit our services page to learn more or place an order at https://essays-panda.com/order for a paper that meets your professor’s exact expectations.

Your paper structure matters. It’s not just about getting the sections right — it’s about telling your research story clearly, honestly, and compellingly. Stick to the IMRaD framework, follow the wineglass model, and draft in reverse order. You’ve got this.