How to Write a Qualitative Methods Section (2026 Template)

Qualitati Research Team · 2026-10-13 · 10 min read

Short answer: A qualitative methods section explains who did the research, which approach they followed, how participants were chosen, how data were collected and analyzed, and why readers should trust the result. Write it in nine blocks, check it against the reporting guideline your journal names (COREQ, SRQR or JARS-Qual), and if you used AI, report the model, version, role and human oversight.

Reviewers rarely reject a qualitative paper because its findings are dull. They reject it because they cannot see how the findings were produced. The qualitative methods section is where you show that. This guide gives PhD students, research labs and research firms a nine-block template and a guideline picker, and it covers one new development. On October 5, 2026, the team behind COREQ+LLM posted a preprint of the first consensus-based checklist for reporting large language model (LLM) use in qualitative research.

Key takeaways

Structure follows the reader's questions: who, which approach, which participants, which data, which analysis, and why it is credible.
Reporting guidelines are checklists, not templates. COREQ has 32 items for interview and focus group studies, SRQR has 21 items for qualitative research in general, and JARS-Qual is the American Psychological Association (APA) standard.
Name your analytic approach precisely. "Thematic analysis" is not enough. Say which version and how you applied it.
AI use now has its own checklist. The COREQ+LLM preprint lists 31 items in seven categories. It is still under peer review.
Explain your decisions, not only what you did. Sample size, coding and theme decisions each need a stated reason.

What should a qualitative methods section include?

A qualitative methods section should let a skeptical reader reconstruct and judge your study. At minimum it covers the research approach and paradigm, the researchers and their reflexivity, the setting, sampling and recruitment, ethics and consent, data collection, data processing, analysis, and the strategies used to support trustworthiness. The order varies by journal. These elements appear in all three major reporting guidelines.

The common mistake is to describe procedures without giving reasons. "We interviewed 18 participants" is a procedure. "We stopped at 18 because the last four interviews added no new codes relevant to the research question, and the sample covered all three hospital roles" is a justification. Reviewers look for the second kind of sentence.

COREQ, SRQR or JARS-Qual: which reporting guideline applies?

Use the guideline your target journal names in its author instructions. If it names none, pick one by study design and discipline. All three are checklists for what to report. None of them tells you how to conduct the study.

Guideline	Scope	Size	Typical fields	Use it when
COREQ (Tong, Sainsbury & Craig, 2007)	Interviews and focus groups	32 items in three domains: research team and reflexivity, study design, analysis and findings	Health, medicine, health services	Your data are interviews or focus groups, or a health journal requires it
SRQR (O'Brien et al., 2014)	Qualitative research in general	21 items, ordered by manuscript section	Medical education, health, broader social science	You use documents, observation or mixed qualitative sources
JARS-Qual (Levitt et al., 2018)	Qualitative research across traditions	Organized by manuscript section; includes "methodological integrity"	Psychology and APA-style journals	Your journal follows APA Style, or you need flexibility across traditions
COREQ+LLM (Fehring et al., 2026 preprint)	LLM use in qualitative research	31 items in seven categories, used alongside COREQ	Any field	An LLM touched transcription, coding, theme development or writing

Sources: EQUATOR Network: COREQ, EQUATOR Network: SRQR, APA Style: JARS-Qual.

The Qualitative Methods Section Template (nine blocks)

This template was written by Qualitati. It combines the elements the three guidelines share into one writing order. Each block lists the decision it must make clear and a sentence starter you can adapt.

1. Approach and paradigm

Name the methodology (for example phenomenology, case study, grounded theory or a generic qualitative description) and the paradigm behind it. Link both to the research question. Starter: "Because we sought to understand how ... experience ..., we used an interpretive phenomenological design."

2. Research team and reflexivity

Say who collected and analyzed the data, their training, their relationship to participants, and how their position may have shaped the analysis. COREQ's first domain is devoted to this. See our guide to reflexivity in qualitative research. Starter: "The interviewer, a doctoral researcher with prior clinical experience, had no supervisory relationship with participants."

3. Setting and context

Describe where and when the study took place and any circumstances that matter for interpretation. Do this without identifying the organization if you promised anonymity.

4. Sampling and recruitment

State the sampling strategy (purposive, maximum variation, snowball, theoretical), the inclusion criteria, how people were approached, how many declined, and why you stopped. Our post on qualitative sample size, information power and saturation includes a worked justification sentence.

5. Ethics and consent

Give the approving committee and reference number, the consent procedure, and how data were pseudonymized and stored. If an AI interviewer or AI analysis was used, say whether participants were told. See informed consent for AI-moderated research.

6. Data collection

Report the mode (in person, video, phone, text chat, AI-moderated voice), the guide and how it was piloted, typical duration, language, and who was present. Put the interview guide in a supplement.

7. Data processing

Explain how audio became text (verbatim or cleaned, human or automatic transcription, checked by whom), how translation was handled, and which software stored and organized the data.

8. Analysis

Name the analytic method and the source you followed. Then describe the actual steps: who coded, inductive or deductive, how the codebook evolved, how codes became themes, and how disagreements were resolved. Our guides to thematic analysis and writing up thematic analysis findings go further.

9. Trustworthiness and methodological integrity

List the strategies you actually used, such as an audit trail, negative case analysis, member reflections, triangulation or peer debriefing, and say what each one changed. JARS-Qual calls this methodological integrity. See trustworthiness in AI-assisted qualitative research.

What changed in October 2026: the COREQ+LLM preprint

Until now, authors who used an LLM had publisher policies but no field-specific checklist. On October 5, 2026, Leonard Fehring and colleagues posted COREQ+LLM on medRxiv. According to the preprint:

A scoping review identified 149 studies for full-text review, from which 29 preliminary reporting items were drafted.
Two Delphi rounds followed, with 141 experts in round one and 115 in round two (82% retention). A nine-person consensus conference settled the remaining items.
The final checklist has 31 items in seven categories. It is reported alongside the original 32 COREQ items.

The project's item guide shows what the items ask for. Among them: who operated the LLM and their credentials, the LLM's exact role in the analysis, how responsibility was split between humans and the model (human-in-the-loop, over-the-loop or out-of-the-loop), which model was used and why, the exact model version or access dates, and whether the model could draw on external data such as retrieval-augmented generation.

Caveat: the project site states that the manuscript is under review at The BMJ and that items may change. Treat it as the best current guide, not a final standard. Our earlier LLM methods disclosure checklist covers the same ground and includes a template paragraph.

Common mistakes reviewers flag

Method salad: "a grounded theory approach using thematic analysis" without saying which procedures came from which tradition.
Saturation as a slogan: claiming saturation without saying how it was assessed.
Counting as validity: reporting inter-rater kappa in a reflexive thematic analysis, where the method does not assume a single correct coding.
Invisible researchers: no statement of who interviewed whom, or of the researchers' relationship to the field.
Vague AI use: "ChatGPT assisted with coding" without a model version, role or review procedure.
Missing materials: no interview guide, codebook or checklist in the supplement.

Who this is for, and when not to use this approach

For: PhD students writing a thesis chapter or first journal article, research labs standardizing how they report, and research firms writing methods appendices for academic or public-sector clients.

Not for: commercial insight reports, where a one-paragraph method note is usually enough. Arts-based and some post-qualitative traditions also reject checklist reporting on principle. If you work in those traditions, use the template as a prompt list and explain your departures, not as a form to fill in.

Limitations and trade-offs

Reporting guidelines improve transparency, but they do not measure quality. A study can tick all 32 COREQ items and still be shallow. Checklists drawn from health research can also push interpretive studies toward a positivist vocabulary of reliability and bias. Braun and Clarke, among others, have argued against importing coding-agreement logic into reflexive thematic analysis. Word limits are a practical constraint too: put detail in supplements and keep the main text for decisions. Finally, COREQ+LLM is a preprint. Check the published version before citing item numbers in a submission.

Human-review note: methodological recommendations here are general. Your supervisor, ethics committee and target journal's instructions take precedence.

Where Qualitati fits

Qualitati is a European qualitative research platform for universities and research firms. It runs AI-moderated interviews, focus groups and conversational surveys, and analyzes them with ThemeLens and the QDA Workspace. Two features help with the methods section:

"Report this analysis." On results in ThemeLens, Annotator, QDA Workspace, Voice Analytics, Synthetic Focus Group and Manuscript Proofreader, Qualitati drafts an editable methods paragraph in English or Chinese, with a settings table listing the model identifiers and sampling settings it actually recorded. The draft is built from templates and uses no extra credits. It only describes the calls it captured. You still add your own sampling, review and analytic decisions.
Evidence you can audit. ThemeLens maps codes to your research questions and checks returned quotes against the source transcripts, which supports blocks 8 and 9 of the template.

The free tier includes 30 monthly credits with no credit card. See pricing for paid plans.

FAQ

How long should a qualitative methods section be?

In most journals, 800 to 1,500 words in the main text, with the interview guide, codebook and completed checklist in supplements. A thesis chapter can be much longer because it also justifies the methodology.

Is COREQ mandatory?

Only if your journal requires it. Many health and medical journals ask for COREQ or SRQR at submission. Psychology journals that follow APA Style point to JARS-Qual.

Should I use COREQ or SRQR?

Use COREQ if your data are interviews or focus groups. Use SRQR for qualitative designs that rely on documents, observation or other sources. Follow the journal if it names one.

How do I report AI use in a qualitative methods section?

State the model and exact version or access date, what the model did, who operated it, how humans reviewed its outputs, and where prompts are available. The COREQ+LLM preprint (October 2026) gives a 31-item checklist for this.

Do I need to report reflexivity?

Yes, in almost all traditions. COREQ, SRQR and JARS-Qual all ask how the researchers' characteristics and relationship to participants may have shaped the study.

Where should the completed checklist go?

Usually in a supplementary file, with the page or line number where each item is reported. Some journals upload it as a separate submission document.

Conclusion

A strong qualitative methods section answers reader questions in a fixed order and gives a reason for each decision. Use the nine-block template to draft it, check the draft against COREQ, SRQR or JARS-Qual, and if an LLM was involved, use the new COREQ+LLM items to report it. If you run interviews or analysis in Qualitati, "Report this analysis" gives you a recorded starting point. Start free with 30 credits.

Sources

Tong, A., Sainsbury, P., & Craig, J. (2007). Consolidated criteria for reporting qualitative research (COREQ). International Journal for Quality in Health Care, 19(6), 349–357. EQUATOR listing.

O'Brien, B. C., Harris, I. B., Beckman, T. J., Reed, D. A., & Cook, D. A. (2014). Standards for reporting qualitative research. Academic Medicine, 89(9), 1245–1251. EQUATOR listing.

Levitt, H. M., et al. (2018). Journal article reporting standards for qualitative research. American Psychologist, 73(1), 26–46. APA Style.

Fehring, L., et al. (2026). Reporting of Qualitative Research Using Large Language Models (COREQ+LLM). medRxiv preprint, posted October 5, 2026. Read the preprint.

Last updated: October 13, 2026. This article is general methodological guidance. A human editor should review discipline-specific reporting requirements before reuse.