A quantitative methodology research paper section that satisfies reviewers is built on reporting completeness, not writing flair. Every choice should leave an audit trail.
Across thousands of peer-review reports, the dominant rejection reason in quantitative work is not "you chose the wrong test." It's "you didn't tell us enough about what you did." Underreporting reads as either carelessness or concealment — neither is acceptable in a Q1 or Q2 journal.
This supporting guide breaks down exactly what to report in a quantitative methodology, in the order most journals expect, with reviewer-tested phrasing for each element. For the wider treatment of methodology writing, return to our pillar on how to write a methodology section that reviewers respect.
The seven reporting categories
Every quantitative methodology, whether for a 6-page short report or a 9,000-word full study, can be mapped to seven reporting categories. Reviewers scan for each. Missing any one of them produces a comment; missing two or three produces a rejection.
| Category | What to report |
|---|---|
| Design | Cross-sectional, longitudinal, experimental, quasi-experimental — with one-line justification |
| Population & sample | Target population, sampling frame, sampling technique, final sample size with response rate |
| Measures | Construct, source, item count, response scale, reliability (Cronbach's α or McDonald's ω), validity evidence |
| Procedure | Timeframe, mode of data collection, incentives, pilot test, follow-up protocol |
| Data preparation | Missing data treatment, outlier screening, normality, common method bias diagnostics |
| Statistical analysis | Tests named, mapped to hypotheses, software and version stated, significance thresholds set |
| Ethics & rigour | Approval body, approval number, consent, data handling, conflicts of interest |
The order above is the order most journals expect. Some social-science journals merge measures and procedure; some clinical journals split sample into eligibility and recruitment. The categories themselves don't change.
How reporting depth changes by design
The level of detail expected varies by study type. Cross-sectional surveys can be reported in 800–1,000 words; longitudinal panel studies need more on attrition and timing; experiments need the full apparatus of manipulation checks, randomisation procedure, and blinding.
Standard reporting
Sampling frame, technique, response rate, instrument psychometrics, common method bias diagnostics, one-shot analysis plan.
Adds timing & attrition
Wave structure, time lags justified, attrition analysis comparing dropouts to completers, missing-by-design handling.
Adds manipulation detail
Randomisation procedure, manipulation checks, blinding, condition assignment ratios, pre-registration if applicable.
For every method you used, write one sentence that names it, one sentence that justifies it, and one sentence that operationalises it. If you can produce all three, the reporting depth is right. If you can't justify it, the choice is suspect; if you can't operationalise it, the description is too thin.
Statistical analysis — what reviewers want to see
Three things separate a credible analysis subsection from a vulnerable one. First, each statistical test must be matched to a specific hypothesis or research question — never listed in the abstract. Second, software and version must be named (SPSS 28, AMOS 26, R 4.3, Mplus 8.9). Third, the threshold for significance, the effect-size measure to be reported, and any correction for multiple comparisons must be stated before results are presented.
For models with latent variables, fit indices need to be specified in advance — typically CFI > 0.90, RMSEA < 0.08, SRMR < 0.08 for acceptable fit, with stricter cutoffs for confirmatory work. Our deeper guide on SEM and CFA reporting walks through what indices to include and how to defend the cutoffs you use.
Reporting assumption checks matters more than authors realise. Stating that normality, multicollinearity, and homoscedasticity were tested — and naming the tests used (Shapiro-Wilk, VIF, Breusch-Pagan) — converts your section from descriptive to defensible. Most reviewers won't comment positively on the presence of these checks; they'll comment negatively on their absence.
The four omissions that cost authors most often
Four reporting gaps appear in nearly every weak quantitative methodology. Patching these alone moves most drafts up a tier.
1. Response rate without denominator. "The survey received 327 responses" is a number, not a response rate. The denominator — total invitations sent, total eligible respondents — must be reported. Without it, reviewers cannot judge sampling quality or non-response bias.
2. Reliability without validity. Reporting Cronbach's alpha alone tells reviewers your scale is internally consistent. It tells them nothing about whether it measures what you claim. Convergent and discriminant validity evidence is expected for any multi-item construct used in inferential testing.
3. Sample size without justification. "A sample of 412 was collected" is a description. "Power analysis using G*Power 3.1 indicated a minimum sample of 395 to detect a medium effect (f² = 0.15) at α = .05 with power = 0.80; the final sample of 412 exceeded this threshold" is a justification. See our supporting guide on how to justify sample size in research for the full method.
4. Analysis without mapping. If you state three hypotheses and run four tests, or four hypotheses and run three, reviewers notice. Every test should map to a specific question. Add a sentence at the start of the data analysis subsection that lists the test-to-hypothesis correspondence.
Under-reporting reads as either careless or evasive. Reviewers assume the second. Over-report by 20% and you remove their best ammunition.
Style: confident, specific, replicable
Quantitative methodology should be written in past tense, active voice where possible, and short declarative sentences. Avoid hedging language ("we attempted to ensure," "it was largely felt that") — these signal weakness without removing the weakness itself. State what you did, why, and how, then move on.
The methodology section is also the wrong place for theoretical justification of constructs (that belongs in the literature review) or for interpretation of results (that belongs in the discussion). Reviewers flag methodology sections that drift into either territory.
If your work uses structural equation modelling, confirmatory factor analysis, mediation, or moderation analyses, the reporting standards become more specific. See our companion guide on SEM and CFA reporting requirements for those details.
One growing expectation in Q1 journals is methodological transparency that extends beyond the manuscript itself. Pre-registration of hypotheses, sharing the analysis script or syntax, and depositing anonymised data in a public repository are increasingly treated as default rather than optional. Even when your target journal doesn't require these, mentioning them where they exist signals confidence and converts your methodology from a closed account into an audit-ready record. Reviewers reward authors who behave as if their work will be checked.
Check Your Quantitative Methodology
Upload your draft and get a structured scorecard across the seven reporting categories — design, sample, measures, procedure, data preparation, analysis, and ethics. Takes 60 seconds.
Run Free Health Check → Or read the case study: How structured CFA/SEM guidance secured acceptance in Scientific Reports