How to Write a Methodology Section: A Complete Guide to Clear, Replicable Research Methods

A methodology section is where a manuscript earns or loses a reviewer’s trust. Editors can forgive an awkward sentence in the introduction. They rarely forgive a methods description so thin that nobody could repeat the study from it. This guide walks through what belongs in a methodology section, how the major reporting guidelines (CONSORT, STROBE, PRISMA, and ARRIVE) shape what reviewers expect by study type, and where manuscripts most often lose points during peer review.

Key Takeaways

• A methodology section needs enough detail that another researcher could repeat your study from it alone.
• Reporting guidelines such as CONSORT, STROBE, PRISMA, and ARRIVE tell you exactly what a peer reviewer expects to see, organized by study design.
• CONSORT was substantially rewritten in 2025. Manuscripts still following the older 2010 checklist may be missing items reviewers now expect.
• Five elements form the backbone of almost every methodology section: design and rationale, participants, materials and procedures, data analysis, and ethical considerations.
• Vague statistical language, such as stating that “appropriate tests were used,” is one of the fastest ways to draw a revision request.

What Is a Methodology Section, and Why Does It Matter?

A methodology section is the part of a manuscript, thesis, or grant proposal that answers one question in full. How exactly did you do this study? That includes the design you chose, who or what you studied, the tools and procedures involved, how you analyzed the resulting data, and the ethical safeguards you followed. Reviewers use it to judge whether your conclusions are earned. Other researchers use it as a blueprint if they want to build on your work or attempt a replication.

Methods and methodology get used interchangeably in casual conversation, but they are not quite the same thing. Methods are the concrete steps you took, such as the survey instrument, the assay, the sampling frame, or the statistical test. Methodology is the broader logic behind those choices, the reasoning that connects your research question to the design you picked. A dissertation committee usually wants both. A journal editor mostly wants the methods, reported with enough precision that the logic behind them is self-evident.

Incomplete reporting has real consequences. Reporting guideline developers built CONSORT, STROBE, and their relatives specifically because editors and methodologists kept finding that published methods sections omitted details needed to judge a study’s validity or repeat it (Simera et al., 2010). That gap between what a study actually did and what its write-up describes is exactly what a strong methodology section closes.

The Five Building Blocks of a Strong Methodology Section

Almost every methodology section, regardless of field, rests on the same five elements. Journals differ in formatting and section headers, but reviewers are scanning for these five things whether the manuscript calls them “Methods,” “Materials and Methods,” or splits them across a “Methodology” chapter.

Study design shapes how much weight each block carries. A randomized trial leans heavily on procedures and allocation concealment. An observational study leans on how confounding was handled. A case report leans on a precise, chronological patient timeline. Before drafting, it helps to place your own study within the broader hierarchy of evidence, since that placement is often what tells you which reporting guideline applies.

Matching Your Methodology Section to the Right Reporting Guideline

Reporting guidelines are not design instructions. They will not tell you how to run a trial or structure a survey. What they do is specify, item by item, what a complete write-up of that study type needs to contain, based on years of evidence about what gets omitted when nobody sets a standard. Below are the guidelines Editverse editors reference most often, matched to the study designs they cover.

Randomized Controlled Trials: CONSORT 2025

CONSORT (Consolidated Standards of Reporting Trials) governs how parallel-group randomized trials should be written up. The guideline was substantially revised in 2025, replacing the widely used 2010 version, with the update co-published simultaneously in BMJ, The Lancet, JAMA, and Nature Medicine in April 2025 (Hopewell et al., 2025). If your manuscript or a template you are working from still cites “CONSORT 2010,” it is worth checking the current CONSORT statement checklist before submission, since some items and their wording have changed.

A CONSORT-compliant methods section needs the trial design and allocation ratio, full eligibility criteria, the interventions in enough detail to repeat them, pre-specified primary and secondary outcomes, the sample size calculation, the randomization method and allocation concealment mechanism, and who was blinded to what. Our CONSORT guidelines walkthrough breaks down the full checklist item by item. In the data analysis section, be explicit about whether your primary analysis was intention-to-treat or per-protocol, since reviewers treat this as a load-bearing methodological decision, not a footnote.

Observational Studies: STROBE

Cohort, case-control, and cross-sectional studies fall under the STROBE Statement, a 22-item checklist first published in 2007 and still the standard a decade and a half on (von Elm et al., 2008). STROBE asks for the setting and eligibility criteria, how exposures and outcomes were defined and measured, how confounding variables were identified and handled, and, importantly, a candid discussion of bias and its likely direction and size.

Because observational designs cannot randomize away confounding, reviewers scrutinize this section harder than they do the equivalent part of a trial report. Our STROBE tutorial walks through all 22 items in plain language. If your observational design uses genetic variants as instruments to strengthen causal claims, the relevant extension is STROBE-MR, covered in our guide to Mendelian randomization.

Systematic Reviews and Meta-Analyses: PRISMA 2020

The PRISMA statement governs how a review of the literature gets written up once it moves beyond a narrative summary into a reproducible synthesis. The 2020 update expanded the checklist to 27 items and introduced a redesigned flow diagram that separates database records from other sources and documents exclusions at each stage with reasons attached (Page et al., 2021).

A PRISMA-compliant methods section needs a full search strategy for at least one database, explicit eligibility criteria, the number of reviewers who screened records and whether they worked independently, the risk-of-bias tool used for included studies, and the synthesis method, including how heterogeneity was assessed. Our PRISMA 2020 checklist guide and PRISMA flow diagram walkthrough cover both pieces in detail. Undetected publication bias and unexplained statistical heterogeneity between pooled studies are two problems that come up constantly in review manuscripts. Address both explicitly rather than letting a reviewer discover them first.

Animal Research: ARRIVE 2.0

In vivo animal studies follow ARRIVE, the Animal Research Reporting of In Vivo Experiments guidelines, now in its second version. ARRIVE 2.0 split its checklist into the “ARRIVE Essential 10,” a non-negotiable minimum covering the study design, sample size, randomization, blinding, outcome measures, statistical methods, and animal details, plus a longer “Recommended Set” for full transparency (Percie du Sert et al., 2020). The ARRIVE guidelines site hosts both checklists along with a free compliance-checking tool many journals now require before submission.

What If Your Study Design Is Not Listed Here?

CONSORT, STROBE, PRISMA, and ARRIVE cover the bulk of what Editverse editors see, but they are far from the only reporting guidelines in active use. Diagnostic accuracy studies follow STARD, case reports follow CARE, and dozens of other designs, from qualitative research to clinical practice guidelines, have their own checklists. Rather than memorize all of them, most researchers start at the EQUATOR Network, a searchable directory that indexes reporting guidelines by study type and links directly to the current version of each one.

Writing the Participants and Sample Description

Reviewers read this part looking for one thing above all, whether the sample can actually answer the research question being asked. Start with a clear definition of your target population, then explain the sampling method, whether that is random sampling, consecutive enrollment, purposive sampling for a qualitative study, or something else, and why that approach fits your design.

State your final sample size and, for hypothesis-testing studies, show how you arrived at it. A power calculation does not need to be lengthy, but it needs the assumed effect size, the significance level, the desired power, and the software or formula used to compute it. Skipping this is one of the fastest ways to draw a “please justify your sample size” comment from a reviewer. Report demographic characteristics relevant to your outcome, such as age, sex, and clinically relevant comorbidities, rather than an exhaustive list that adds length without adding information.

Describing Materials, Instruments, and Procedures

Every instrument, questionnaire, assay kit, or piece of equipment needs enough specification that someone else could source or build the same thing. For a validated psychometric scale, cite the validation study and report the reliability statistic (such as Cronbach’s alpha) if you calculated one for your own sample. For laboratory equipment, include the manufacturer, model, and any settings that affect the result. For software, list the exact version, since defaults and algorithms change between releases in ways that can shift results.

Procedures should read as a timeline, not a list of loosely related activities. Walk through what happened first, second, and later, including any pilot testing, training given to those collecting data, and the interval between measurements. If two or more people collected or coded data, state how consistency between them was checked. If randomization or blinding was part of your design, describe the actual mechanism, whether that was a computer-generated sequence or sealed envelopes, rather than simply asserting that the study was randomized or blinded.

Reporting Data Analysis and Statistical Methods

This is the subsection reviewers with a statistics background read most critically, and it is also where thin manuscripts fail most visibly. Name your software and its version. List your descriptive statistics and the inferential tests you ran, matched to the type of data each was applied to. State your significance threshold before you report a single result, and report effect sizes and confidence intervals alongside p-values rather than p-values alone, since a p-value alone tells a reader almost nothing about the size or practical relevance of an effect (Lang and Altman, 2015).

Explain how you handled missing data. “Complete case analysis,” “multiple imputation,” and “last observation carried forward” are not interchangeable, and reviewers increasingly expect a stated reason for whichever one you chose. If you ran subgroup or sensitivity analyses, say whether they were planned in advance or added after seeing the results, since that distinction affects how much weight a reader should give them.

Handling Ethical Considerations and Approvals

Human-subjects research needs a named ethics committee or Institutional Review Board, along with the approval or protocol reference number, not just a sentence stating that “ethical approval was obtained.” Describe how informed consent was collected and from whom, particularly for studies involving minors, cognitively impaired participants, or emergency care where consent procedures are less straightforward. Research involving identifiable clinical data or human tissue should also reference the ethical principles guiding the work, most commonly the Declaration of Helsinki.

For clinical trials, prospective registration in a public registry is an expectation from the International Committee of Medical Journal Editors, and the registration number belongs in your methods section, not buried in supplementary material. Our guide to the IRB approval process covers timelines and common submission mistakes if you are still preparing your protocol.

A Worked Example: Rewriting a Thin Methodology Section

Consider a hypothetical cohort study examining sleep duration and blood pressure. A thin first-draft version might read something like this. “Participants were recruited and their blood pressure was measured. Data was analyzed using appropriate statistical tests.” That sentence tells a reviewer almost nothing they could act on.

Here is a STROBE-aligned rewrite of the same study. “We recruited 214 adults aged 40 to 65 years from three primary care clinics between March and September 2025, using consecutive sampling. Eligible participants had no prior hypertension diagnosis and were not taking antihypertensive medication. Blood pressure was measured in triplicate with an automated sphygmomanometer after a five-minute rest period, and the mean of the second and third readings was used in analysis. Self-reported sleep duration was captured with the Pittsburgh Sleep Quality Index. We used multivariable linear regression in Stata version 18, adjusted for age, sex, body mass index, and smoking status, with a two-sided significance threshold of p<0.05.”

Notice what changed. The sample, setting, and timeframe are specific. The measurement protocol is repeatable by someone who has never seen the study before. The statistical approach names the test, the software, the version, and the covariates. Nothing in the second version is more impressive research. It is the same study, described so that a reviewer, and a future researcher trying to build on it, can actually evaluate what was done.

Common Methodology Section Mistakes We See in Manuscript Editing

A few problems recur often enough across disciplines that they are worth calling out directly.

• Vague statistical language. “Data were analyzed using appropriate statistical methods” is a sentence that answers no question a reviewer actually has. Name the test.
• Unversioned software. “Analyzed in SPSS” without a version number leaves a reader unable to check whether a default setting affected the result.
• Missing sample size justification. A sample size with no calculation or rationale behind it invites a reviewer to assume the study was simply underpowered.
• Uncited instruments. Naming a psychometric scale or assay kit without a validation reference makes it impossible for a reviewer to judge whether it measures what you claim.
• Procedures out of order. A methods section that jumps between recruitment, intervention, and follow-up without a clear sequence forces the reader to reconstruct the timeline themselves.
• Design labels that do not match the analysis. Calling a study “randomized” when allocation was not actually random, or “prospective” when key data were collected retrospectively, is one of the fastest ways to trigger a methodological red flag.
• Ethics approval without a reference number. A blanket statement that the study was ethically approved, with no committee name or protocol number, reads as unverifiable.

If you are still shaping your first draft rather than revising one, our companion piece on writing a clear methods section works through the drafting process from a different angle and pairs well with the checklist-driven approach here.

Frequently Asked Questions

Conclusion: Write a Methodology Section Reviewers Trust

A strong methodology section is not the most exciting part of a manuscript to write, but it is often the part that decides whether reviewers trust everything that follows it. Anchor your draft in the five building blocks covered here, check it against the reporting guideline that matches your design, and replace any sentence a stranger could not act on with one they could.

If you want a second pair of eyes before submission, Editverse’s editors review methodology sections against CONSORT, STROBE, PRISMA, and ARRIVE requirements as part of our broader publication support services, alongside statistical review and journal-specific formatting checks.

References

1. Hopewell S, Chan AW, Collins GS, et al. CONSORT 2025 statement: updated guideline for reporting randomised trials. BMJ. 2025;389:e081123. doi:10.1136/bmj-2024-081123
2. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-349. doi:10.1016/j.jclinepi.2007.11.008
3. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi:10.1136/bmj.n71
4. Percie du Sert N, Hurst V, Ahluwalia A, et al. The ARRIVE guidelines 2.0: updated guidelines for reporting animal research. PLoS Biol. 2020;18(7):e3000410. doi:10.1371/journal.pbio.3000410
5. Gagnier JJ, Kienle G, Altman DG, et al. The CARE guidelines: consensus-based clinical case reporting guideline development. BMJ Case Rep. 2013. doi:10.1136/bcr-2013-201554
6. Bossuyt PM, Reitsma JB, Bruns DE, et al. STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. BMJ. 2015;351:h5527. doi:10.1136/bmj.h5527
7. Lang TA, Altman DG. Basic statistical reporting for articles published in biomedical journals: the SAMPL guidelines. Int J Nurs Stud. 2015;52(1):5-9. doi:10.1016/j.ijnurstu.2014.09.006
8. Simera I, Moher D, Hoey J, Schulz KF, Altman DG. Transparent and accurate reporting increases reliability, utility, and impact of your research: reporting guidelines and the EQUATOR Network. BMC Med. 2010;8:24. doi:10.1186/1741-7015-8-24