COREQ Checklist Explained: All 32 Items, With Examples

Reporting gaps are among the easiest things for a reviewer to flag in a qualitative manuscript: the interviewer’s training, the number of people who declined, and whether anyone double-coded the transcripts are often simply not stated. The COREQ checklist exists to stop that: 32 items, published by Tong, Sainsbury and Craig in 2007, that many journals ask you to complete and upload with interview and focus-group studies. This is the item-by-item guide, with what reviewers actually want for each item and a worked example answer from a fictional study, so you can fill the table in once and avoid avoidable revision requests.

What is the COREQ checklist?

The COREQ checklist is a 32-item reporting guideline for qualitative studies that use in-depth interviews or focus groups, published by Allison Tong, Peter Sainsbury and Jonathan Craig in the International Journal for Quality in Health Care in 2007. Its stated purpose was “to develop a checklist for explicit and comprehensive reporting of qualitative studies,” because at the time no consolidated reporting framework existed for any qualitative design (Tong et al., 2007).

The method behind it matters for how you read the items. The authors searched Cochrane and Campbell protocols, Medline, CINAHL, systematic reviews of qualitative studies, and journal author and reviewer guidelines, found 22 existing checklists, and compiled 76 items from them. They then removed duplicates and items that were ambiguous, too broadly defined, or impractical to assess, which left 32 (Tong et al., 2007). So COREQ is a consolidation of what other people had already asked for, filtered for what a reviewer can actually check on the page.

COREQ’s 32 items are grouped in three domains:

1. Research team and reflexivity (items 1–8): who interviewed, and what they brought to the room.
2. Study design (items 9–23): framework, participant selection, setting, and data collection.
3. Analysis and findings (items 24–32): how the data were coded and how the results are shown.

The split is lopsided on purpose. Fifteen of the 32 items are about design and data collection, because that is where interview studies most often leave a reader unable to judge what was done.

Which journals require a COREQ checklist?

Many health, nursing and health-services journals ask for a completed COREQ or SRQR checklist with every qualitative submission, uploaded as a separate file, and the EQUATOR Network lists COREQ as the reporting guideline for “qualitative research interviews and focus groups.” PLOS ONE’s submission guidelines state that qualitative studies should be reported in accordance with the COREQ checklist or, alternatively, SRQR. No single body makes COREQ mandatory across the field; each journal decides.

Two practical consequences follow. First, read the author instructions of the target journal before you write the methods section, because it is far cheaper to write “the interview guide was piloted with two nurses” into a first draft than to retrofit it at revision. Second, the checklist is a reporting tool, not an approval tool: a completed COREQ table does not make a study good, and a missing one usually just triggers an editorial request. Reviewers, though, do read it, and a column full of “see Methods” is a signal that you did not.

Domain 1: who interviewed, and what did they bring? (items 1–8)

Domain 1 asks you to put the interviewer on the page: their name, credentials, occupation, gender, training, and any prior relationship with the participants. Eight items that authors often leave thin, because they feel personal rather than methodological. They are methodological. An interview is a social event, and what a participant says to a clinician they know is not what they say to a stranger with a PhD.

The worked answers below come from a fictional study we will use throughout: 18 semi-structured interviews with ward nurses about night-shift handover, in two hospitals, conducted by a doctoral candidate we will call MR.

Item	Topic	What reviewers want to see	Example answer (fictional study)
1	Interviewer/facilitator	Which named author ran the interviews or focus groups, not 'the research team'.	"All interviews were conducted by the first author (MR)."
2	Credentials	The interviewer's qualifications at the time of the study (PhD, MD, RN, MSc).	"MR holds an MSc in nursing and is a PhD candidate."
3	Occupation	Their job or role when the data were collected.	"MR was a full-time doctoral researcher, not employed by either hospital."
4	Gender	The interviewer's gender, because it can shape disclosure on some topics.	"MR is a woman."
5	Experience and training	Prior qualitative interviewing experience and any training received.	"MR had conducted 40 interviews in a previous project and completed a 2-day qualitative interviewing course."
6	Relationship established	Whether any relationship existed with participants before the study.	"MR had no prior relationship with the participants."
7	Participant knowledge of the interviewer	What participants were told about the researcher's goals and reasons for the research.	"Participants were told the study aimed to improve handover and that MR was a former ward nurse."
8	Interviewer characteristics	Assumptions, interests or biases that could shape the data, and how you handled them.	"MR's own handover experience was recorded in a reflexive journal and discussed in monthly supervision."

Item 8 is the one that separates a competent methods section from a thoughtful one. “The interviewer kept a reflexive journal” is a sentence; “the interviewer’s eight years as a ward nurse made her sympathetic to accounts of blame, which the second coder challenged during analysis” is reflexivity. Reviewers can tell the difference.

Domain 2: how was the study designed and the data collected? (items 9–23)

Domain 2 is the largest, with 15 items across four sub-sections: theoretical framework (item 9), participant selection (items 10–13), setting (items 14–16), and data collection (items 17–23). It is where most “please clarify” comments originate, because the information is usually in the authors’ heads and in the ethics application, but not in the manuscript.

Item	Topic	What reviewers want to see	Example answer (fictional study)
9	Methodological orientation and theory	The named approach that underpins the analysis (grounded theory, phenomenology, thematic or content analysis, ethnography), stated once and used consistently.	"We used reflexive thematic analysis within a constructivist stance."
10	Sampling	How participants were selected: purposive, convenience, consecutive, snowball, maximum variation.	"Purposive sampling for shift pattern, seniority and ward type."
11	Method of approach	How people were contacted: face to face, phone, email, letter, via a gatekeeper.	"Nurses were invited by email through the ward manager, then contacted directly by MR."
12	Sample size	The number of participants who actually took part.	"18 nurses took part."
13	Non-participation	How many declined or dropped out, and why, if known.	"26 were invited; 6 declined citing workload and 2 did not reply."
14	Setting of data collection	Where the interviews happened: home, clinic, workplace, online.	"Interviews took place in a private room on the ward or by video call (n = 5)."
15	Presence of non-participants	Whether anyone other than participants and researchers was present.	"No one else was present."
16	Description of sample	The sample's important characteristics and the data-collection dates, usually a table.	"Table 1: age, years qualified, ward; data collected March–June 2025."
17	Interview guide	Whether a guide, questions or prompts were provided, and whether it was pilot tested.	"The semi-structured guide (Supplement 1) was piloted with two nurses and revised."
18	Repeat interviews	Whether any participant was interviewed more than once, and how many times.	"No repeat interviews were conducted."
19	Audio/visual recording	Whether audio or video recording was used.	"All interviews were audio-recorded."
20	Field notes	Whether field notes were taken during or after the interviews or focus groups.	"MR wrote field notes within an hour of each interview."
21	Duration	How long the interviews or focus groups lasted, as a range or mean.	"Interviews lasted 38–71 minutes (median 52)."
22	Data saturation	Whether saturation was discussed, and how it was judged.	"No new codes appeared after interview 14; we conducted 4 more to confirm."
23	Transcripts returned	Whether transcripts went back to participants for comment or correction.	"Transcripts were not returned to participants."

Three COREQ domain 2 items deserve a second look: sampling non-participation (13), saturation (22) and transcript return (23). Item 13, non-participation, is the item most often answered “not applicable,” and it almost never is: if you invited 26 people and 18 took part, the other 8 are the answer. Item 22, data saturation, is the one reviewers are increasingly suspicious of, because “data were collected until saturation was reached” has become a ritual sentence with nothing behind it.

State the rule you used and the interview at which it was met; our thematic saturation calculator produces that sentence from your coding log, or justifies a sample size before you collect anything. Item 23, transcript return, is where honesty beats aspiration: “not returned” is a complete answer, and reviewers rarely penalize it.

Domain 3: how were the data analyzed and the findings reported? (items 24–32)

Domain 3 covers nine items in two sub-sections: data analysis (items 24–28) and reporting (items 29–32). This is where the checklist overlaps with the quality of the analysis itself, and where the gap between what teams did and what they wrote is widest. Almost everyone double-codes something; far fewer say how many coders, on how much of the data, and what happened when they disagreed.

Item	Topic	What reviewers want to see	Example answer (fictional study)
24	Number of data coders	How many people coded the data, and on what share of it.	"Two authors (MR, JL) coded all 18 transcripts independently."
25	Description of the coding tree	Whether a coding tree or codebook is described or supplied.	"The final codebook (34 codes, 5 themes) is in Supplement 2."
26	Derivation of themes	Whether themes were set in advance, derived from the data, or both.	"Themes were derived inductively; two sensitizing concepts from the handover literature were declared a priori."
27	Software	Any software used to manage or analyze the data, with the version.	"Coding was managed in NVivo 15."
28	Participant checking	Whether participants gave feedback on the findings.	"A one-page summary was sent to all 18 participants; 7 responded and none disputed the themes."
29	Quotations presented	Whether quotations illustrate the themes, and whether each is identified (participant number, role).	"Each quotation carries a participant ID and shift pattern, e.g. (P07, nights)."
30	Data and findings consistent	That the quoted data actually support the findings claimed.	"Every theme is illustrated by at least three quotations from different participants."
31	Clarity of major themes	The major themes are presented clearly and are distinguishable from each other.	"Five themes are presented with a definition, boundary and example each."
32	Clarity of minor themes	Diverse cases and minor themes are described, not just the dominant story.	"Two divergent accounts that contradicted theme 3 are reported in section 3.4."

Item 24 is where qualitative reporting meets reliability reporting. COREQ only asks how many coders there were, but a reviewer who sees “two coders” will want to know what they did with their disagreements, and a growing number will ask for an agreement statistic. If you report one, do it properly: coefficient, unit of analysis, confidence interval, software, as in our template for reporting inter-rater reliability. If your coding lives in NVivo, MAXQDA or ATLAS.ti, the intercoder reliability calculator reads the export and writes that paragraph.

A short, explicit codebook for item 25 is the single document that makes items 24, 26 and 30 easy to answer; a published coding scheme such as the Gottschalk-Gleser content analysis scales or the VR-CoDES cues and concerns codes shows what a complete one looks like: every code with a definition, an inclusion rule, and an anchor example.

COREQ vs SRQR vs ENTREQ: which checklist do you need?

Use COREQ for primary studies built on interviews or focus groups, SRQR for any other qualitative design, and ENTREQ when the paper is a synthesis of other people’s qualitative studies. Journals sometimes name one and sometimes let you choose; when they let you choose, pick by design, not by length.

	COREQ	SRQR	ENTREQ
Full name	Consolidated Criteria for Reporting Qualitative Research	Standards for Reporting Qualitative Research	Enhancing Transparency in Reporting the Synthesis of Qualitative Research
Source	Tong, Sainsbury & Craig, 2007	O'Brien, Harris, Beckman, Reed & Cook, 2014	Tong, Flemming, McInnes, Oliver & Craig, 2012
Items	32 in 3 domains	21 items	21 in 5 domains
Built for	Interview and focus-group studies	The whole spectrum of qualitative research, any paradigm or method	Syntheses of qualitative studies (meta-ethnography, thematic synthesis, etc.)
Strongest on	The interviewer, the relationship with participants, data collection logistics	Paradigm and approach, ethics, trustworthiness techniques, integration with prior work	Search, selection, appraisal and synthesis methods
Silent on	Ethics approval, funding, conflicts of interest; no explicit paradigm item	Interviewer gender and credentials, repeat interviews, saturation as a separate item	Primary data collection (not its job)
How it was built	22 checklists consolidated into 76 items, cut to 32	Sources through July 2013, iterative item coding, five external reviewers	Guidance and 40 published syntheses, piloted and refined
Use it when	Your data are interviews or focus groups and the journal names COREQ	Observational, document, arts-based or mixed qualitative designs, or when the journal lets you choose	Your paper is a qualitative evidence synthesis

The two checklists for primary research are not nested: SRQR is not “COREQ minus eleven items.” SRQR asks explicitly for ethical issues, a research paradigm, techniques to enhance trustworthiness, conflicts of interest and funding, which COREQ does not ask for as items (COREQ’s transcript return and participant checking, items 23 and 28, touch on trustworthiness only indirectly), and it collapses COREQ’s eight reflexivity items into one (O’Brien et al., 2014). COREQ, in turn, asks about repeat interviews, duration, non-participants in the room and transcript return, which SRQR leaves to the author’s judgment. A manuscript written to satisfy SRQR will usually still have gaps against COREQ, and vice versa.

How do you fill in the “reported on page” column?

To fill in the “reported on page” column, write the page number plus the section heading and paragraph where the manuscript reports each of the 32 items, and mark any item you did not report as “Not done” or “Not applicable” with a short reason. Each row of the official form has three cells: the item, a guide question, and that location column. Make it precise enough that a reviewer can check it in ten seconds: a page number plus the section heading and paragraph is ideal, since page numbers shift between the submitted file and the typeset proof.

A few conventions that keep the table from coming back:

• Point at a sentence, not a section. “Methods” is not a location. “p. 6, Data collection, para. 2” is.
• Write “Not done” when you did not do it. Transcript return (item 23) and participant checking (item 28) are the usual cases. The checklist records reporting, so the honest sentence in the manuscript plus a pointer to it satisfies the item.
• Reserve “Not applicable” for design-level impossibilities, and add a five-word reason. Repeat interviews in a single-contact design; non-participants in an online one-to-one interview.
• Match the manuscript to the table, not the table to the manuscript. If an item has no location, the fix is a sentence in the methods, not a creative pointer.
• Upload it as a separate file in the format the journal specifies, usually a Word table or the journal’s own PDF form, and name it so the editor finds it (“COREQ_checklist.docx”).

Our COREQ builder does the mechanical part: it scans the manuscript for sentences that plausibly address each item, you confirm or reject each match and add the location, and it exports the table. Treat its suggestions the way you would a junior co-author’s first pass.

Why does a COREQ checklist get bounced?

Editors return COREQ checklists for five recurring reasons, and all of them are cheaper to fix before submission than during revision. Five problems recur in editorial feedback on COREQ tables:

1. Blank or “see Methods” cells. The table is supposed to save the reviewer a search; a vague pointer tells them you did not do the search yourself.
2. Domain 1 skipped entirely. Authors treat items 1–8 as optional biography. Reviewers treat a blank domain as a reflexivity gap, and in interview research that is a methodological finding about your study.
3. Non-participation (item 13) answered “N/A.” If anyone was invited and did not take part, the item applies.
4. Saturation (item 22) asserted, not shown. “Until saturation” without the rule or the interview number is now a routine query.
5. The checklist says yes and the manuscript says nothing. Items 25 (coding tree), 27 (software) and 29 (identified quotations) are the usual offenders: the codebook is on the author’s laptop, the software is in the lab, and the quotation IDs were stripped during copy-editing.

One more that is not a bounce but a quiet downgrade: pointing item 24 at “two authors coded the data” with no account of disagreement. It passes the checklist and loses the reviewer’s confidence at the same time. Our comparison of qualitative analysis software covers which tools make the coding audit trail easy to export, and if you are choosing a tool now, the NVivo alternatives guide weighs them on exactly that.

Is COREQ a measure of good qualitative research?

COREQ is a reporting floor, not a definition of rigor, and its own construction has been challenged. Buus and Perron (2020) attempted to replicate the four-phase process by which the 76 candidate items became 32 and reported that they “were rarely able to replicate them completely because of uncertainty about the actual procedures,” concluding that the coding of the 22 source checklists was not trustworthy and that items were decontextualized and their meaning distorted along the way. They call it “alarming” that a checklist so widely endorsed by journals rests on a process that cannot be reproduced.

That critique does not change what you should do at submission, because the journal will still ask for the table. It does change how you should read the items. Treat COREQ as a list of things a reader needs to know to interpret an interview study, which it is, and not as a scorecard where 32 of 32 means good science.

A phenomenological study with a single deeply reflexive interviewer and no codebook can be excellent and score badly on domain 3; a mechanical thematic analysis can tick every box and say nothing. Narrative methods such as the Labov and Waletzky structural analysis sit uneasily with items 25 and 26 and are none the worse for it.

What should you do before submitting a COREQ checklist?

Write the methods section with the COREQ checklist open beside it, in the first draft, and the table takes far less time at submission than at revision. Name the interviewer and what they brought, count the people who said no, state the saturation rule and the interview it was met at, say how many coders and what they did with disagreement, and give every quotation an ID. If you change one thing, change this: stop writing “N/A” for things you simply did not do. “Not done” is honest, complete, and far better received than a blank cell.

References

• Buus N, Perron A. The quality of quality criteria: Replicating the development of the Consolidated Criteria for Reporting Qualitative Research (COREQ). International Journal of Nursing Studies. 2020;102:103452. doi:10.1016/j.ijnurstu.2019.103452
• O’Brien BC, Harris IB, Beckman TJ, Reed DA, Cook DA. Standards for reporting qualitative research: a synthesis of recommendations. Academic Medicine. 2014;89(9):1245–1251. doi:10.1097/ACM.0000000000000388
• Tong A, Flemming K, McInnes E, Oliver S, Craig J. Enhancing transparency in reporting the synthesis of qualitative research: ENTREQ. BMC Medical Research Methodology. 2012;12:181. doi:10.1186/1471-2288-12-181
• Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care. 2007;19(6):349–357. doi:10.1093/intqhc/mzm042
• EQUATOR Network. Consolidated criteria for reporting qualitative research (COREQ) and Standards for reporting qualitative research (SRQR). Accessed October 2026.
• PLOS ONE. Submission guidelines: qualitative research. Accessed October 2026.