“COPE data” most plausibly means one of two things: what the Committee on Publication Ethics (COPE) actually requires or recommends about research data specifically, or a garbled search for its Core Practices generally. This guide covers the first, real sense — COPE’s Data and Reproducibility Core Practice, the separate (and still incomplete) data-sharing guidance COPE has been developing, and how COPE’s flowcharts and case database handle suspected data fabrication in a submitted or published article. For COPE’s full guidance library across all ten Core Practice areas, see CASRAI’s COPE Guidelines Explained; for the plagiarism, duplicate-submission, authorship-dispute, and image-manipulation flowcharts specifically, see COPE Flowcharts Explained. This guide does not restate either — it’s the data-specific layer neither one covers in depth.
What COPE actually requires on research data: the Data and Reproducibility Core Practice
Among COPE’s ten Core Practice categories (the framework introduced in 2017 that member journals and publishers are expected to have a documented policy or process for — see the COPE Guidelines Explained guide for the full list and its current transitional status), one is specifically about data: Data and reproducibility. COPE’s own framing of this category is that journals should include policies on data availability and should encourage — or, depending on the journal, require — the use of reporting guidelines and registration of clinical trials and other study designs, according to standard practice in the relevant discipline.
In practice this means a COPE-aligned journal is expected to have a stated position on three related things: whether and how authors must make underlying data available (a data availability statement, a repository deposit requirement, or an explicit exemption process), whether a discipline-appropriate reporting guideline applies (for example CONSORT for randomized trials, PRISMA for systematic reviews — see CASRAI’s PRISMA flow diagram entry), and whether prospective study registration is expected before data collection begins. None of this is COPE inventing new data-management standards from scratch — it is COPE requiring that a journal have some documented position, appropriate to its field, rather than leaving data availability to informal convention. A CASRAI Data Management Plan (DMP), which a researcher typically produces earlier, at the funding-proposal stage, is a distinct document written to a funder’s or institution’s requirements; COPE’s Core Practice operates later, as a journal-level publication policy, and the two are not interchangeable even though they cover overlapping ground.
COPE’s dedicated data-sharing guidance: further along than a Core Practice line item, still not a finished standalone policy
Beyond the Core Practices summary above, COPE has separately been developing more detailed guidance specifically on data sharing, published as an open topic-discussion document rather than a finished policy. That discussion explicitly frames data sharing as a continuum — how far a journal should go in encouraging, requiring, or actively enforcing data sharing — rather than a single fixed rule, and it flags several open problems any journal-level policy has to contend with: data-sharing infrastructure is unevenly developed across countries and disciplines, a stated policy is only as effective as a journal’s willingness and ability to enforce it, and researchers whose data are reused by others don’t routinely get adequate credit for that reuse. It also acknowledges directly that different research communities have very different norms and comfort levels around sharing data, which is part of why COPE’s own Core Practice language above defers to “standard practice in the discipline” rather than a single cross-field rule.
The practical takeaway for a research administrator or author: don’t assume COPE membership implies a specific, mandatory data-sharing standard identical across every journal. It implies the journal has committed to having a documented data-availability policy under the Data and Reproducibility Core Practice — the specifics of what that policy actually requires (open deposit in a named repository, data available on request, a stated exemption for sensitive or proprietary data) vary by journal and discipline, and should be checked against that specific journal’s own author guidelines, not inferred from COPE membership alone. Check publicationethics.org directly for the current state of COPE’s dedicated data-sharing guidance before treating any specific enforcement mechanism as settled COPE policy.
How COPE handles suspected data fabrication or falsification
Separately from the availability/sharing question above, COPE also publishes scenario-specific guidance for editors who suspect data was fabricated or falsified — a research-integrity concern, not a data-sharing-policy one. As with COPE’s other flowcharts (see COPE Flowcharts Explained), the guidance for a published article follows a specific pattern: a reader or reviewer raises a concern about the integrity of reported data, the editor investigates and typically seeks an independent second opinion, and the author is contacted and given a genuine opportunity to explain or clarify before any formal action is taken. Depending on what the investigation finds, the recommended outcomes range from no further action, to a published expression of concern while the matter is unresolved, to retraction, to referral to the authors’ institution or, where relevant, a regulatory body. COPE’s case database also documents real (anonymized) Forum discussions of suspected data fabrication, including cases involving suspected systematic fabrication across multiple papers or a research group — these function as applied precedent alongside the flowchart’s more abstract decision tree.
The dictionary distinction matters here: fabrication is making up data or results that were never actually collected or generated; falsification is manipulating real research materials, equipment, processes, or data so the record no longer accurately represents what happened. COPE’s guidance addresses both, and an editor’s flowchart path can differ depending on which is suspected — falsification more often surfaces as image or figure manipulation, covered by COPE’s separate image-manipulation flowchart, while fabrication more often surfaces as data or results that can’t be reproduced or independently verified at all.
Journal role vs. institutional role: who actually investigates a data-fabrication allegation
COPE is explicit that a journal’s editorial office is not the body that determines whether misconduct occurred. An editor’s job under COPE’s guidance is to assess whether a concern is credible enough to warrant action on the published record (correction, expression of concern, or retraction) — the actual investigation into whether an individual researcher committed fabrication or falsification is the responsibility of that researcher’s institution, which has investigative powers (access to raw data, lab notebooks, interviews with the individuals involved) no journal has. This is the same institution-vs-journal division CASRAI’s research misconduct investigation guide covers in full from the institutional side; COPE’s data-fabrication flowchart is the trigger that, in many real cases, leads a journal to refer a concern to the institution in the first place. See also CASRAI’s research misconduct case studies guide for what real investigations of this kind have actually looked like in practice.
What this means for researchers and research administrators
For a researcher preparing a submission to a COPE-member (or COPE-guidance-following) journal, the practical implications are: check that journal’s specific data-availability and reporting-guideline requirements early, since they vary by discipline and journal even though the underlying Core Practice expectation is common across COPE members; keep the underlying data, code, and lab records well-organized and retrievable, since that’s exactly what an editor’s fabrication inquiry or an institutional investigation will ask for if a concern is ever raised; and understand that a data availability statement or repository deposit is a publication-ethics expectation as much as a funder requirement — the two obligations (funder DMP, journal data-availability policy) usually point the same direction but come from different sources and should each be checked on their own terms. For a research administrator supporting compliance more broadly, COPE’s Data and Reproducibility Core Practice is a useful reference point precisely because it is deliberately discipline-agnostic at the framework level — it tells you a journal should have a policy, not what that policy must say — so it’s not a substitute for reading the specific journal’s own instructions to authors.
Related CASRAI guidance
For COPE’s full guidance library and Core Practices framework, see COPE Guidelines Explained. For the plagiarism, duplicate-submission, authorship-dispute, and image-manipulation decision trees, see COPE Flowcharts Explained. For the institutional process a journal-level concern can trigger, see How a Research Misconduct Investigation Actually Works and Research Misconduct Case Studies. For the editorial process once retraction is the likely outcome, see How a Retraction Actually Happens. For the cluster this guide sits in, see the Research Integrity & Compliance pillar page.
Frequently asked questions
Does COPE require journals to make authors share their underlying research data?
Not as a single, uniform rule. COPE’s Data and Reproducibility Core Practice requires member journals to have some documented data-availability policy and to encourage discipline-appropriate reporting guidelines and study registration — but the specifics (mandatory open deposit, data-on-request, or a stated exemption) are set by the individual journal, not dictated uniformly by COPE. COPE has separately been developing more detailed, dedicated data-sharing guidance, published as an open topic discussion rather than a finished standalone policy as of this writing — check publicationethics.org directly for its current status.
What happens if a journal editor suspects data fabrication in a published article?
COPE’s guidance for editors follows a consistent pattern: investigate the concern, seek an independent second opinion, and contact the author for an explanation before taking formal action. Depending on what’s found, the outcome can range from no action, to a published expression of concern, to retraction, to referral to the authors’ institution or a relevant regulatory body. The editor assesses the published record; determining whether an individual researcher actually committed misconduct is the institution’s role, not the journal’s.
Is COPE’s data guidance the same thing as a data management plan?
No. A Data Management Plan is typically produced earlier, at the funding-proposal stage, to a funder’s or institution’s specification. COPE’s Data and Reproducibility Core Practice is a journal-level publication-ethics expectation that applies later, at submission and publication. The two often point in a compatible direction but come from different sources with different requirements, and satisfying one doesn’t automatically satisfy the other.
What’s the difference between data fabrication and data falsification under COPE’s guidance?
Fabrication is inventing data or results that were never actually collected or generated. Falsification is manipulating real materials, equipment, processes, or data so the record misrepresents what actually happened. COPE addresses both, though falsification more often surfaces as image or figure manipulation (covered by a separate COPE flowchart) while fabrication more often surfaces as data that can’t be reproduced or independently verified.
Do I need to be a COPE member to see its data-related guidance?
No. COPE’s Core Practices, flowcharts, and topic-discussion documents (including its data-sharing discussion) are published openly on publicationethics.org at no cost. COPE membership provides additional case-consultation support through COPE’s Forum and eligibility for member listing, not exclusive access to the published guidance itself.
