CASRAI Dictionary

Tag: annual research integrity statement

  • CRediT Taxonomy Author Contributions Example: Trial Consortia

    A credit taxonomy author contributions example for a 100+-author clinical trial consortium paper typically cannot assign all 14 CRediT roles to every named individual. Instead, most multi-site consortia assign roles to a small “writing committee,” then credit the remaining site investigators and staff as a collective group — a workable but imperfect compromise between transparency and practicality.

    The CRediT taxonomy author contributions example published by most journals — one paper, a handful of authors, each ticking a few of the 14 roles — is straightforward. It falls apart at scale. Multi-site clinical trial consortia routinely publish primary results papers with 50, 200, or even several hundred named contributors across dozens of hospitals, laboratories, and coordinating centres. Applying individual-level CRediT attribution to every one of them is rarely feasible, and the taxonomy itself offers no scaling guidance. This article examines how consortia actually resolve that gap, where the “writing committee” shortcut helps and where it hides real accountability problems, and what research administrators should check before signing off on a consortium submission.

    CASRAI originated the CRediT contributor role taxonomy in 2014. The standard is now stewarded by NISO as ANSI/NISO Z39.104-2022, an important distinction for any institution citing CRediT in policy documents.

    Contents

    What is the CRediT taxonomy and how is it meant to work?

    The CRediT (Contributor Roles Taxonomy) is a standardised list of 14 role categories — including Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Supervision, and the two Writing roles — used to describe what each named contributor to a research output actually did. Under ANSI/NISO Z39.104-2022, any of the 14 roles can be assigned to more than one contributor, and any contributor can hold more than one role. The taxonomy was designed around conventional author lists of perhaps two to twelve people, where a corresponding author can realistically survey everyone and compile an accurate statement.

    CRediT deliberately does not define who qualifies as an author — that remains the domain of criteria such as those published by the International Committee of Medical Journal Editors (ICMJE). CRediT only describes contribution once authorship, or collaborator status, has already been decided elsewhere.

    Why does individual-level CRediT attribution break down above 100 authors?

    Multi-site clinical trial consortia — platform trials, adaptive-design mega-trials, and large international collaborative groups — routinely list hundreds of contributors: principal investigators at each site, research nurses, statisticians, data monitors, and a central coordinating team. Surveying every one of them individually against 14 role definitions, reconciling disagreements, and keeping the record current through a multi-year trial is an administrative task few coordinating centres can sustain.

    Three practical failure points recur:

    • Collection burden. A corresponding author cannot manually chase 300 collaborators for role self-declarations before every manuscript revision.
    • Role granularity mismatch. Site-level staff often perform a genuinely narrow contribution (patient recruitment, sample handling) that maps to only one or two roles, making individual disclosure administratively disproportionate to its informational value.
    • Authorship-vs-collaborator ambiguity. Not every named contributor meets full authorship criteria, and CRediT provides no mechanism of its own for distinguishing the two — that decision is made upstream, under ICMJE or journal-specific rules.

    The ICMJE’s Recommendations on the role of authors and contributors state plainly: “When a large multi-author group has conducted the work, the group ideally should decide who will be an author before the work is started and confirm who is an author before submitting the manuscript for publication.” In practice, that decision — not the CRediT assignment — is what most consortia spend their governance effort on.

    How do multi-site consortia actually assign CRediT roles?

    Three models are in active use across large trial consortia, and each trades transparency against administrative load differently. The dominant compromise is a named writing committee that receives individual CRediT attribution, combined with a collective collaborative group byline (for example, “The [Trial Name] Collaborative Group”) that carries the remaining contributors without a role-by-role breakdown for each person.

    Model How it works Transparency Administrative load
    Full individual CRediT Every named author, however many, completes a role disclosure form Highest Unsustainable above roughly 30-50 authors
    Writing committee + collective group A small writing committee gets full CRediT roles; remaining contributors are credited as a named collective group, often with individual names and site affiliations in a supplementary appendix Moderate — accountable core, opaque periphery Manageable; used by most platform and mega-trials
    Hybrid tiered disclosure Writing committee gets full CRediT roles; site principal investigators get a single broad role (e.g. Investigation); frontline staff are acknowledged, not authored Higher than pure collective model Moderate, requires a pre-agreed authorship policy

    The ICMJE recommendations also clarify how this interacts with indexing: “the byline of the article identifies who is directly responsible for the manuscript,” and MEDLINE indexes as authors whichever names appear there, while non-author collaborators can still be individually listed and searchable if the journal provides an accompanying note. This means a consortium can preserve individual, searchable credit for site staff even when it does not extend full CRediT role disclosure to each of them — an option under-used by many trial groups.

    A pre-agreed authorship and contribution policy, set before a multi-site trial begins recruitment rather than at the manuscript stage, is the single factor that most reliably prevents disputes later. Waiting until submission to decide who was an “author” versus a “collaborator” — and who gets which CRediT role — is the most common cause of delay and disagreement in large consortium publications.

    Answer-first questions on CRediT and large author groups

    What are examples of author contributions?

    Typical author contributions include conceiving the study design, securing funding, recruiting patients, collecting or curating data, performing statistical analysis, writing the first draft, and critically reviewing the final manuscript. Under CRediT, each of these maps to one of 14 defined roles rather than a vague general description.

    What should substantial contributions include to be credited as an author?

    Per ICMJE criteria, a substantial contribution requires involvement in the work’s conception or design, or the acquisition, analysis, or interpretation of data, combined with drafting or critically revising the manuscript and final approval of the published version. Meeting only one element, such as data collection alone, typically warrants acknowledgement rather than authorship.

    How to write an author contribution in a case report?

    A case report contribution statement should name each author against the specific tasks they performed — for example, clinical assessment, literature review, drafting, and supervision — using plain, specific language rather than the fuller 14-role CRediT set, which is more suited to larger, multi-method studies with a genuinely divided workload.

    What this means for research administrators, funders, and publishers

    Research offices supporting multi-site consortium trials should treat CRediT and authorship decisions as a governance item from the protocol stage, not a manuscript-stage formality. A written policy — agreed by the steering committee before recruitment starts — should specify who sits on the writing committee, what threshold of involvement earns collective-group inclusion versus acknowledgement-only, and how the supplementary collaborator list will be maintained and version-controlled across a multi-year trial.

    Funders and institutions increasingly use CRediT statements as an input to research assessment, so an opaque “collective group” byline with no supplementary breakdown under-serves early-career site staff who did substantive work but receive no individually attributable, citable role. Publishers that support both a named writing committee and a searchable, named collaborator appendix — rather than a collective name alone — give institutions and funders a materially better evidence trail for exactly this reason.

    The underlying tension is not going away: CRediT was built for conventional author teams, and large trial consortia will keep testing its edges. Until a scaling mechanism is formally added to the taxonomy, the writing-committee-plus-named-collaborator-appendix model remains the most defensible practical compromise between individual accountability and administrative reality.

  • Research Misconduct vs Misbehaviour: FFP vs QRPs

    Research misconduct is a narrow, formally sanctionable transgression — fabrication, falsification, or plagiarism (FFP) committed intentionally, knowingly, or recklessly — while research misbehaviour is the much broader, informally policed category of questionable research practices (QRPs) that fall short of that threshold but still corrode trust in the research record. Confusing the two matters: one triggers a formal investigation and possible dismissal or retraction; the other typically triggers correction, training, or an editorial note.

    This distinction between research misbehavior and research misconduct is not academic hair-splitting. Regulators, universities, funders, and journal editors all apply a threshold test before they open a formal case, and where that line sits differs by jurisdiction. Research misconduct is fabrication, falsification, or plagiarism in proposing, performing, reviewing, or reporting research, committed with intent, knowledge, or recklessness — a definition codified in the United States under 42 CFR Part 93 and echoed, with local variation, across UK, European, and international frameworks.

    What is research misconduct? The formal FFP test

    Research misconduct has a narrow, legalistic definition. In the United States, the Office of Research Integrity (ORI) — part of the Department of Health and Human Services — defines it under 42 CFR Part 93 as fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results. Each term is precisely scoped: fabrication is making up data or results and recording or reporting them; falsification is manipulating research materials, equipment, or processes, or changing or omitting data so the research record is misrepresented; plagiarism is appropriating another person’s ideas, processes, results, or words without appropriate credit.

    Crucially, ORI’s definition explicitly excludes honest error or genuine differences of opinion. A finding of misconduct also requires that the act was committed intentionally, knowingly, or recklessly — not through a documented, defensible mistake. This intent threshold is what separates misconduct from misbehaviour, and it is the single fact most competitor explainers state without ever mapping it against a second jurisdiction.

    What is research misbehaviour? Where QRPs fit

    Research misbehaviour is an umbrella term for actions that deviate from responsible research practice without meeting the fabrication-falsification-plagiarism bar. It is most often used interchangeably with “questionable research practices” (QRPs) — a term formalised in the UK’s Concordat to Support Research Integrity, first published by Universities UK in 2012 and revised in 2019, which defines QRPs as “minor infractions or research practices, including avoidable errors, which fall short of the definition of intentional research misconduct.”

    Common QRPs include selective reporting of favourable results, “p-hacking” data until a significant finding emerges, HARKing (presenting a post-hoc hypothesis as though it were pre-registered), salami-slicing a single dataset into multiple papers, and inappropriate authorship allocation. A 2009 meta-analysis by Daniele Fanelli, published in PLoS ONE, found that on average only 1.97% of surveyed scientists admitted to fabricating or falsifying data at least once — but up to 33.7% admitted to other questionable research practices, and the figure rose sharply when respondents were asked about colleagues’ behaviour rather than their own. That roughly seventeen-fold gap is the empirical case for treating misbehaviour as a distinct, much larger risk surface than misconduct.

    Authorship disputes sit squarely in this grey zone. CASRAI originated the CRediT contributor role taxonomy in 2014. The standard is now stewarded by NISO as ANSI/NISO Z39.104-2022, and consistent use of its CRediT contributor roles is one of the more effective structural fixes institutions have for the ghost- and guest-authorship misbehaviours that recur in authorship disputes.

    How do regulators draw the line?

    No single global definition governs the misconduct/misbehaviour boundary. Each major framework sets its own threshold language, and the differences are consequential for cross-border collaborations and multi-national author teams.

    Framework / body Core threshold language Legal or policy basis Standard of intent
    US Office of Research Integrity (ORI) Fabrication, falsification, or plagiarism (FFP) 42 CFR Part 93 Intentional, knowing, or reckless
    UK Concordat to Support Research Integrity “Behaviours that deliberately or recklessly fall short of the standards expected” Universities UK Concordat (2012, rev. 2019) Deliberate or reckless
    World Health Organization “Wrongdoing” — FFP plus misrepresentation and other fraudulent behaviour WHO Code of Conduct for Research Intentional, knowing, or reckless
    Committee on Publication Ethics (COPE) Overlapping guidance distinguishing retraction-triggering misconduct from correction-triggering QRPs COPE flowcharts and guidelines for editors Proportionate to severity and evidence

    Notice the pattern: every framework anchors misconduct to intent or recklessness, and every framework treats everything short of that — errors, sloppy practice, grey-zone shortcuts — as a separate, lower-tier category, whether it is called a QRP, an “avoidable error,” or simply “wrongdoing” of a lesser kind.

    What happens after a finding? Sanctions compared

    The practical consequence of the misconduct/misbehaviour distinction is procedural. A misconduct finding typically follows a formal, panel-based investigation and can result in retraction, funding debarment, employment termination, or referral to a professional regulator. A misbehaviour or QRP finding more commonly results in a correction to the published record, mandatory training, enhanced supervision, or an editorial expression of concern — remedial rather than punitive action.

    • Misconduct outcomes: retraction, debarment from funding, dismissal, professional sanction, referral to law enforcement in extreme cases.
    • Misbehaviour/QRP outcomes: correction or erratum, mandatory research-integrity training, revised authorship credit, closer supervisory oversight.
    • Shared consequence: both can damage institutional reputation and require correction of the scholarly record, which is why UKRIO and equivalent bodies investigate both under a shared procedural umbrella even though the findings differ.

    Research institutions and research administration offices increasingly triage complaints against this two-tier structure before deciding whether a matter warrants a full misconduct panel or a lighter-touch integrity review — a proportionality principle that COPE explicitly recommends to journal editors handling post-publication concerns.

    Common questions about misconduct and misbehaviour

    What are the three types of research misconduct?

    Under the US federal definition, the three types are fabrication (inventing data or results), falsification (manipulating processes or altering data so the record misrepresents what occurred), and plagiarism (appropriating others’ ideas, words, or results without credit). Together these form the FFP standard used by ORI and mirrored internationally.

    What are the five unethical practices most often cited in research ethics literature?

    A widely cited framework lists falsification of data, failure to credit others, plagiarism, undisclosed conflicts of interest, and biased design or interpretation driven by outside influence. Some of these meet the formal misconduct threshold; others, like undisclosed conflicts, more often sit in the misbehaviour category depending on intent and severity.

    What is research misbehaviour, precisely?

    Research misbehaviour is any deviation from responsible research conduct — including questionable research practices and avoidable errors — that falls short of intentional fabrication, falsification, or plagiarism. It is a broader, less formally policed category than misconduct, typically addressed through correction and training rather than disciplinary panels.

    What actions count as research misconduct in UK institutional policy?

    UK institutional policies, following the Concordat to Support Research Integrity, typically list plagiarism, fabrication, falsification, breaching ethical or legal research requirements, proceeding without required approvals, and failing to manage conflicts of interest as research misconduct, provided the conduct was deliberate or reckless rather than an honest error.

    Implications for institutions and research administrators

    For research administrators, publishers, and funders, the practical task is triage: distinguishing an honest error or a QRP from conduct that meets the intent threshold for a formal misconduct investigation, before committing to a resource-intensive panel process. Institutions that document this distinction clearly in their code of practice — and train staff and early-career researchers on where the line sits — reduce both the number of misdirected formal investigations and the risk of under-reacting to genuine misconduct.

    As research-integrity offices, funders, and publishers converge on shared vocabulary for this spectrum, consistent taxonomy work — from CRediT-style contributor attribution to standardised institutional definitions — will keep doing more to prevent misbehaviour from escalating into misconduct than any single enforcement action can.

  • Predatory Journal Checker vs Plan S Compliance

    A Plan S compliant journal is not automatically screened by a predatory journal checker: Plan S tests open-access licensing and Directory of Open Access Journals (DOAJ) registration, not editorial integrity or peer-review conduct. The two checks answer different questions, and treating DOAJ/Plan S clearance as proof a journal is legitimate leaves a real compliance gap that research administrators need to close separately.

    A predatory journal checker is a tool, checklist, or reference list — such as the Think. Check. Submit. checklist, Cabells’ Predatory Reports, or the archived Beall’s List — used to test whether a journal’s peer review, editorial board, and fee practices are genuine rather than a vehicle for harvesting article-processing charges.

    What is a predatory journal checker?

    A predatory journal checker evaluates the operational and editorial conduct of a journal rather than its licensing terms. It looks at whether peer review actually happens, whether the editorial board is real and contactable, whether article-processing charges are disclosed upfront, and whether the publisher’s indexing claims can be verified.

    Common red flags that these tools and checklists are built to catch include:

    • Unsolicited, aggressive email invitations promising rapid publication
    • No transparent article-processing-charge (APC) schedule until after acceptance
    • An editorial board listing academics without their knowledge or consent
    • A journal scope so broad it covers unrelated disciplines
    • Fabricated or unverifiable impact-factor and indexing claims

    These are the criteria a checker tests. None of them is what Plan S compliance actually checks — which is the source of the confusion this article addresses.

    Does Plan S compliance screen for predatory journals?

    Not directly. Plan S is a funder mandate — led by cOAlition S — requiring that publicly funded research be published open access under specific licensing terms. Its technical requirements state that a fully open-access journal must be listed in the Directory of Open Access Journals (DOAJ), or apply for DOAJ listing within one year of publishing its first article, to count as a compliant venue.

    DOAJ listing is a proxy signal, not a predatory-publishing audit. cOAlition S guidance separately points authors toward the Think. Check. Submit. checklist for journal-selection due diligence — a clear indication that cOAlition S itself does not treat DOAJ/Plan S clearance as a substitute for a dedicated predatory check. Responsibility for the final journal-selection decision sits with the researcher and their institution, not with the funder’s compliance rule.

    DOAJ listing vs a dedicated predatory journal checker

    DOAJ vetting and a predatory journal checker overlap in intent — both aim to exclude disreputable venues — but they differ in scope, update frequency, and what they miss. DOAJ’s 2014–2016 re-application process is a useful illustration: it removed roughly 3,300 previously listed journals that failed revised inclusion criteria, which shows DOAJ listing is a snapshot assessment, not a continuously monitored guarantee.

    Mechanism What it verifies What it misses Best used for
    Plan S / DOAJ listing Open-access licence terms; baseline transparency criteria at time of listing Ongoing editorial conduct; peer-review quality after listing Confirming funder-mandate eligibility
    Predatory journal checker (Think. Check. Submit., Cabells, Beall’s archive) Editorial board authenticity, peer-review conduct, fee transparency Funder licensing compliance Author-level due diligence before submission
    Scopus / Web of Science journal check Active indexing status, citation metrics, discontinued-title flags Newer or non-English-language legitimate journals not yet indexed Cross-checking indexing claims a journal makes about itself
    Publisher/journal finder tools Journal-manuscript fit by scope and audience Legitimacy screening entirely — these tools assume the candidate pool is already vetted Narrowing a shortlist of already-verified journals

    What does a Scopus journal check add?

    A Scopus journal check confirms whether a title is actively indexed, flags titles that have been discontinued from Scopus for quality reasons, and surfaces citation-based metrics. This is a useful cross-check against a journal’s own indexing claims — predatory titles frequently claim indexing status they do not have — but Scopus coverage is not designed as a predatory-publishing screen and does not evaluate peer-review conduct directly.

    It is also asymmetric: a legitimate new journal may not yet be Scopus-indexed, so absence from Scopus is not itself proof of a predatory operation. Administrators should treat a Scopus check as one data point in a layered process, not a standalone verdict.

    How should administrators layer both checks?

    Institutions handling funder-mandate compliance and research-integrity screening as two separate workstreams should merge them into one journal-selection workflow. A practical sequence:

    1. Confirm funder eligibility first. Check DOAJ listing (or ROAD registration) to establish Plan S / open-access mandate compliance.
    2. Run a dedicated predatory check second. Apply the Think. Check. Submit. checklist, or consult Cabells’ Predatory Reports where the institution has a subscription, against the same candidate journal.
    3. Cross-check indexing claims. Verify any Scopus, Web of Science, or PubMed indexing the journal advertises against the indexing service’s own database.
    4. Escalate ambiguous cases to the institution’s research-integrity office or library scholarly-communications team rather than relying on a single automated pass/fail signal.
    5. Record the outcome in the researcher’s submission file, since funders and REF-style assessment exercises increasingly expect an audit trail of due-diligence steps, not just a final compliance flag.

    This sequencing matters because each mechanism fails differently: DOAJ/Plan S can clear a journal on licensing grounds while missing recent editorial decline; a predatory checker can flag conduct issues DOAJ has not yet caught up with; Scopus can catch a false indexing claim that neither of the other two checks is built to test.

    Common questions about predatory journal screening

    How do you check if a journal is predatory?

    Run the Think. Check. Submit. checklist against the journal, verify the editorial board members individually, confirm the article-processing charge is disclosed before submission, and cross-check any indexing claims (Scopus, DOAJ) directly against the indexing service rather than trusting the journal’s own website.

    What is a red flag for a predatory journal?

    Aggressive, unsolicited invitation emails promising unusually fast peer review are the most cited red flag. Other consistent signals include an editorial board that cannot be independently verified, a scope spanning unrelated disciplines, and article-processing fees disclosed only after acceptance.

    How do you check if a journal is reputable?

    Confirm active listing in DOAJ or an equivalent recognised index, verify the publisher belongs to COPE or a comparable ethics body, check that peer-review policy is published and specific, and confirm the editorial board’s affiliations independently rather than trusting journal-supplied contact details.

    What is considered a predatory journal?

    A predatory journal is one that charges publication fees while failing to provide the genuine editorial and peer-review services legitimate scholarly journals promise, prioritising revenue from article-processing charges over publication quality and research integrity, per definitions developed by COPE and reflected in Frandsen et al.’s peer-reviewed literature review.

    Implications and the path forward

    For institutional research offices, the practical implication is procedural: a single “is this journal Plan S compliant?” check cannot double as a research-integrity sign-off, and treating it that way creates audit risk when a funder or REF-style exercise later asks how a submission venue was verified. Layering a DOAJ/Plan S check with a dedicated predatory journal checker and an indexing cross-check is not duplicative effort — each step tests a distinct failure mode that the others do not cover.

    As open-access mandates expand and predatory operations grow more sophisticated at mimicking legitimate indexing and DOAJ-style transparency signals, the gap between funder-mandate compliance and editorial-integrity verification is likely to widen rather than close. Institutions that formalise the layered workflow now — rather than relying on DOAJ/Plan S status as an implicit predatory-publishing seal of approval — will be better positioned as funders tighten reporting expectations around journal-selection due diligence.

  • What Is Research Governance? Beyond NHS Ethics

    Research governance is the institutional system of standards, delegated responsibilities and accountability mechanisms that ensures research is sponsored, conducted, resourced and reported to a consistent standard of quality and integrity — a system that spans sponsorship, data protection, financial probity and research integrity, not just the ethics approval most people associate with an NHS Research Ethics Committee.

    The phrase is frequently reduced, in searches and in institutional shorthand, to “getting NHS ethics sign-off.” That collapses a much wider accountability structure into a single procedural step. Research governance is the umbrella; ethics review is one component operating underneath it.

    What Is Research Governance?

    Research governance is the set of rules, standards and lines of accountability an institution puts in place to control how research is initiated, resourced, conducted and reported. NHS Research Scotland, whose remit covers governance across Scottish health boards, describes it as concerned with “setting standards to improve research quality and safeguard the public.” That is the safeguarding function. But governance is also an administrative control system: it determines who is legally and financially answerable when something goes wrong, long before any ethical question is raised.

    ARMA (the UK’s Association of Research Managers and Administrators) frames it more structurally, describing effective research governance as “the implementation of a fit-for-purpose decision-making framework under which an institution” operates. That decision-making framing matters: governance is not a checklist a researcher completes once. It is an ongoing institutional control system — the same category of function as financial governance or clinical governance, applied to the research enterprise.

    What Does Research Governance Actually Cover?

    A governance system that only covered ethics would be incomplete. In practice, institutional research governance operates across four interlocking strands, each with its own named accountable party and its own failure mode if neglected.

    • Sponsorship and legal accountability — the sponsor (usually the employing institution or funder) takes on the legal responsibility for a study’s initiation, management and financial arrangements, distinct from the researcher’s day-to-day conduct of it.
    • Data governance — how participant data, tissue samples and research datasets are collected, stored, shared and protected, governed alongside UK GDPR and institutional data protection policy.
    • Financial governance — probity in the use of grant and contract funds, adherence to funder terms and conditions, and audit trails for how public or charitable money was spent.
    • Research integrity — the honest conduct, reporting and attribution of research, including handling allegations of misconduct such as fabrication, falsification or plagiarism.

    Health and safety oversight and intellectual property management sit alongside these four strands in most institutional frameworks, particularly for laboratory-based or commercially exploitable research.

    Research Governance vs Research Ethics Review: What’s the Difference?

    Ethics review answers one question: is this specific study, as designed, ethically acceptable to run? Governance answers a broader one: does the institution have the systems in place to sponsor, resource, monitor and be accountable for research generally? A study can pass ethics review and still fail governance requirements — for example, if the sponsor has not confirmed indemnity and insurance arrangements, or if data-sharing agreements are not in place.

    Aspect Research Ethics Review Research Governance
    Core question Is this study design ethically acceptable? Can the institution be accountable for this research?
    Typical body Research Ethics Committee (REC) Sponsor, R&D office, research administration function
    Scope Participant welfare, consent, risk-benefit balance Sponsorship, data, finance, integrity, health & safety, IP
    Timing Pre-approval, one-off per protocol Continuous, across the study lifecycle
    Applies beyond NHS? Only where human participants/data/tissue are involved Yes — to all disciplines and funding types

    An institution’s own research administration function typically holds the governance oversight role, coordinating sponsor sign-off, data agreements and financial compliance across a study’s life, while the ethics committee’s involvement is generally concentrated at the design and approval stage.

    Who Is Responsible for Research Governance?

    Responsibility is distributed, not centralised in one office. The sponsor carries overall legal and financial accountability. The Chief Investigator is responsible for day-to-day conduct in line with the approved protocol. The employing institution provides the administrative infrastructure — contracts, insurance, data protection compliance — that makes sponsorship possible. Funders, including UK Research and Innovation (UKRI), attach their own governance conditions through grant terms and conditions, requiring institutions to demonstrate integrity and financial-probity safeguards as a condition of funding.

    Under international clinical trial standards such as ICH Good Clinical Practice (ICH-GCP), sponsor obligations are made explicit and legally binding — a level of formality that has increasingly influenced how non-clinical research governance is structured, even where GCP itself does not strictly apply.

    What Frameworks Define Research Governance in the UK?

    The UK’s foundational document was the 2005 Research Governance Framework for Health and Social Care, issued separately by the four UK nations. The Health Research Authority (HRA) subsequently consolidated these into a single UK-wide document — the UK Policy Framework for Health and Social Care Research — which, per the HRA’s own record, replaced “the separate Research Governance Frameworks in each UK country with a single, modern set of principles for the whole UK,” co-developed with the health departments of Northern Ireland, Scotland and Wales. The HRA’s published record shows this framework was most recently updated on 10 January 2025, reflecting a living document rather than a static one.

    Beyond health research, the Concordat to Support Research Integrity — developed under the auspices of Universities UK — sets out institutional commitments to rigour, transparency, accountability and support for researchers across all disciplines, not solely clinical fields. Attribution and authorship disputes, a recurring integrity concern under governance, connect to contributorship standards such as the CRediT taxonomy, which CASRAI originated in 2014 and which is now stewarded as ANSI/NISO Z39.104-2022 — a reminder that even a narrow-looking standard can sit inside a much larger governance accountability chain.

    Common Questions About Research Governance

    Why is research governance important?

    Research governance is important because it protects participants, safeguards public and funder trust, and creates a clear accountability chain when something goes wrong — financially, ethically or scientifically. Without it, institutions have no defined mechanism for assigning responsibility across sponsors, investigators and funders, increasing legal and reputational exposure.

    Is research governance the same as clinical governance?

    No. Clinical governance covers the quality and safety of patient care delivery within a healthcare organisation, while research governance covers the conduct, sponsorship and accountability of research activity itself. They overlap in NHS settings but apply to different organisational functions and different named accountable roles.

    What is a sponsor in research governance?

    A sponsor is the organisation — typically the employing institution, a university, or a funder — that takes on legal responsibility for confirming a study is properly designed, resourced, insured and managed before it begins. The sponsor role is distinct from the researcher’s role and cannot be left undefined.

    Does research governance apply outside the NHS?

    Yes. Research governance applies across all disciplines — social sciences, engineering, humanities and commercially funded research — wherever an institution sponsors, funds or hosts a research activity, not only where NHS patients, tissue or data are involved.

    Implications and Outlook

    For institutional leaders, the practical implication is structural: governance cannot be delegated entirely to an ethics committee, nor treated as a one-time approval gate. It requires standing infrastructure — a research administration function capable of tracking sponsorship status, data agreements, financial compliance and integrity casework concurrently, across every live project, not just those with NHS involvement.

    As funders including UKRI tie funding conditions more tightly to demonstrable integrity and financial-probity safeguards, and as the HRA continues to revise the UK Policy Framework, institutions that treat governance as an accountability system — rather than an ethics-approval formality — will be better positioned to withstand funder audits, data protection scrutiny and misconduct investigations alike.

  • Research Misconduct Allegation Phases (3 Stages)

    A research misconduct allegation moves through three formal stages — inquiry, investigation and adjudication — each with a distinct purpose, evidentiary threshold and, under the US federal model, a defined deadline. The inquiry screens whether a full investigation is warranted; the investigation builds a factual record; adjudication delivers the finding and any sanctions. This article walks through each phase in order, then contrasts the US federal timetable with the more variable UK institutional model.

    Research misconduct is formally defined, under the US Public Health Service policy (42 C.F.R. Part 93), as fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results — a definition that explicitly excludes honest error and legitimate differences of scientific opinion.

    What Is a Research Misconduct Allegation?

    A research misconduct allegation is a formal statement, made by a complainant to a designated institutional official — typically a Research Integrity Officer (RIO) — that a named respondent has fabricated, falsified or plagiarised research. The allegation triggers a duty on the institution to make a preliminary assessment before any formal phase begins. This assessment is not itself one of the three formal phases; it simply determines whether the allegation, if true, would fall within the scope of research misconduct.

    Phase One: The Inquiry

    The inquiry is a preliminary, confidential fact-check, not a determination of guilt. Its sole question is whether the allegation has enough substance to warrant a full investigation.

    • A small committee or designated official gathers documents and research records.
    • The respondent and complainant may be interviewed informally.
    • A written inquiry report records the findings and a recommendation on whether to proceed.

    Under 42 C.F.R. §93.307(g), US institutions receiving Public Health Service funding must complete the inquiry within 60 calendar days of initiation unless the delay is documented and justified. If the inquiry finds no substance to the allegation, the matter closes at this point; the majority of allegations do not proceed past inquiry.

    Phase Two: The Investigation

    If the inquiry concludes an investigation is warranted, the process becomes formal. The respondent is notified in writing, and under federal guidance the institution must notify the Office of Research Integrity (ORI) within 30 days of that determination.

    The investigation committee, typically composed of subject-matter experts without conflicts of interest, must establish three things:

    1. Whether there was a significant departure from accepted practices in the relevant field.
    2. Whether the conduct was committed intentionally, knowingly, or recklessly.
    3. Whether the allegation is proven by a preponderance of the evidence — the standard set out in 42 C.F.R. §93.106, a lower bar than criminal “beyond reasonable doubt.”

    US institutions are expected to complete the investigation within 120 days of initiation, covering evidence review, interviews, drafting the report, and giving the respondent an opportunity to comment, per 42 C.F.R. §93.311. Extensions are routinely requested and granted where cases are complex.

    Phase Three: Adjudication

    Adjudication is the decision phase. An institutional Deciding Official — someone who played no role in the inquiry or investigation — reviews the investigation report, the respondent’s comments, and any procedural objections, then issues a final institutional finding.

    Where federal funding is involved, ORI conducts its own oversight review of the institution’s finding before any Public Health Service administrative action is imposed. According to ORI’s own published process, a respondent facing formal action retains the right to request a hearing before an Administrative Law Judge before sanctions take effect. Possible outcomes range from correction of the research record to debarment from federal funding and referral for further action.

    How Do UK and US Procedures Compare?

    The US model is unusual in having a single statutory timetable enforced through 42 C.F.R. Part 93. The UK has no equivalent statutory regulator; instead, individual institutions run their own codes of practice, informed by non-binding sector guidance.

    Feature United States (PHS-funded research) United Kingdom
    Governing instrument 42 C.F.R. Part 93 Institutional codes of practice, informed by UKRIO and UKRI guidance
    Federal/national oversight body Office of Research Integrity (ORI) None statutory; UK Research Integrity Office (UKRIO) is advisory
    Inquiry deadline 60 calendar days Not centrally mandated; varies by institution
    Investigation deadline 120 calendar days Not centrally mandated; varies by institution
    Final appeal route Hearing before an Administrative Law Judge Institutional appeal stage (naming varies, e.g. “Stage 3”)

    UKRI’s April 2025 guidance for research organisations describes an assessment stage that “refers to the process of reviewing the nature of an allegation of research misconduct” and establishing scope — functionally equivalent to the US inquiry, but without a fixed clock. UK institutional codes, such as those published by the University of Surrey and University of Staffordshire, typically layer a similar inquiry-investigation-decision logic across numbered stages, though the labelling and timeframes differ institution to institution. This is separate from journal-level handling: COPE’s flowcharts guide editors on suspected misconduct in submitted or published papers, and a publisher may pause or flag a work while the institutional process above runs in parallel.

    Answer-First Questions About Misconduct Allegations

    What are the phases of a research misconduct allegation?

    The formal process runs through three phases: an inquiry that screens whether an allegation has substance, an investigation that establishes the facts against a defined evidentiary standard, and adjudication, where a Deciding Official issues the final finding and any sanctions.

    What is the first phase in response to a research misconduct allegation?

    The first formal phase is the inquiry — a confidential, preliminary review of documents and testimony conducted to determine whether the allegation warrants a full investigation. It is not a determination of wrongdoing and, under US federal policy, must normally conclude within 60 days.

    Who investigates allegations of research misconduct?

    The employing or funded institution conducts the inquiry and investigation, typically through a Research Integrity Officer and an appointed committee of subject experts. For US federally funded research, the Office of Research Integrity then reviews the institution’s finding before any federal sanction is applied.

    What is a research misconduct allegation?

    A research misconduct allegation is a formal claim that a named individual has fabricated, falsified, or plagiarised research. It excludes honest error or genuine scientific disagreement, and it obliges the receiving institution to make a documented preliminary assessment before any formal phase begins.

    Implications and Next Steps

    For research administrators, the practical lesson is procedural discipline: documenting the preliminary assessment, meeting inquiry and investigation deadlines where they apply, and keeping the Deciding Official separate from earlier stages protects the institution against later appeal. For researchers named as respondents, understanding which phase they are in — and what standard of proof applies — clarifies what rights and evidence obligations attach at each step.

    As UK institutions continue to align local codes with UKRI’s 2025 guidance, expect greater convergence in stage naming, though a single UK statutory deadline regime, of the kind ORI enforces in the US, remains absent. Institutional research administration teams should treat the inquiry-investigation-adjudication sequence as the baseline procedural map, then check their own code of practice for the specific labels and timeframes that apply locally.

  • Stem Cell Research Scientific Misconduct Legacy

    Stem cell research scientific misconduct is best defined by two landmark cases: Hwang Woo-suk’s fabricated human cloning papers (South Korea, 2004-2006) and the STAP cell falsification scandal (Japan, 2014). Both involved fabricated data published in top journals, both were exposed through failed replication and image forensics, and both reshaped how institutions oversee stem cell research integrity today.

    Research misconduct is formally defined as fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting results. In stem cell science specifically, stem cell research scientific misconduct has produced two of the most consequential fraud cases in modern science, each triggering journal retractions, criminal or disciplinary proceedings, and lasting changes to how laboratories, journals, and funders verify extraordinary claims.

    What counts as stem cell research scientific misconduct?

    Stem cell research scientific misconduct covers fabricated data, falsified images, and unethical procurement of biological materials in studies involving embryonic, induced pluripotent, or somatic-cell-derived stem cell lines. The field is unusually exposed to this risk because of intense media attention, national prestige stakes, and the technical difficulty of independently verifying claims about pluripotency or successful cloning.

    Both cases examined here meet the formal definition applied by journals and integrity offices: invented results presented as genuine experimental findings, subsequently confirmed by institutional investigation and retracted from the scientific record.

    How did the Hwang Woo-suk cloning fraud unravel?

    Hwang Woo-suk, a veterinary scientist at Seoul National University, published two papers in Science in 2004 and 2005 claiming to have created the first cloned human embryonic stem cell lines through somatic cell nuclear transfer, including patient-matched lines for individuals with injuries or disease. The claims made Hwang a national figure in South Korea and were treated as a milestone toward personalised regenerative medicine.

    The case collapsed on two fronts simultaneously. First, journalists and whistleblowers within Hwang’s own laboratory raised concerns that junior female researchers had donated their own eggs for the experiments, a practice that breached informed-consent norms because of the coercive power dynamics involved. Second, a Seoul National University investigative panel examined the underlying data in December 2005 and January 2006 and found that none of the eleven claimed patient-specific stem cell lines existed; the single verified cell line was later shown to have arisen through parthenogenesis rather than cloning.

    • Science formally retracted both papers in January 2006.
    • Hwang was dismissed from his university post and indicted for fraud, embezzlement of research funds, and bioethical violations.
    • A South Korean court convicted him in 2009 and imposed a two-year suspended prison sentence, a verdict the South Korean Supreme Court upheld on appeal in 2014.

    What happened in the STAP cell scandal?

    In January 2014, RIKEN researcher Haruko Obokata and co-authors published two papers in Nature describing “stimulus-triggered acquisition of pluripotency” (STAP) — a claim that ordinary adult cells could be reprogrammed into a stem-cell-like state simply by exposing them to stress, such as a weak acid bath. The method promised a dramatically simpler alternative to existing induced pluripotent stem cell techniques.

    Independent laboratories worldwide were unable to replicate the results, and close scrutiny of the published figures revealed duplicated and manipulated images alongside plagiarised text from earlier work. RIKEN convened a formal investigation committee, which in April 2014 found Obokata guilty of falsification and fabrication. Nature retracted both papers in July 2014, and a subsequent verification experiment — conducted by Obokata herself under RIKEN supervision — failed to reproduce STAP cells by the end of that year, at which point she resigned.

    The human cost was severe. Obokata’s supervisor and senior co-author, RIKEN Center for Developmental Biology (CDB) deputy director Yoshiki Sasai, died by suicide in August 2014 amid the fallout. Waseda University revoked Obokata’s doctorate in 2015 after she failed to correct the thesis within a set deadline, and RIKEN CDB itself was dissolved and reorganised into the RIKEN Center for Biosystems Dynamics Research in 2018, partly in response to the reputational damage.

    How do the two cases compare?

    Despite occurring a decade apart and in different countries, the Hwang and STAP cases share a common failure pattern: extraordinary claims, inadequate internal verification before publication, and exposure driven by replication failure rather than routine peer review.

    Factor Hwang Woo-suk case STAP cell case
    Journal Science (2004, 2005) Nature (2014)
    Claimed method Cloned human embryonic stem cells via somatic cell nuclear transfer Pluripotency induced by external stress (acid bath)
    Detection trigger Whistleblower reports plus egg-donation ethics concerns Failed replication plus image duplication analysis
    Institutional finding Seoul National University panel, Dec 2005-Jan 2006 RIKEN investigation committee, April 2014
    Retraction date January 2006 July 2014
    Consequence for researcher 2009 conviction, two-year suspended sentence Resignation; doctorate revoked 2015

    What reforms followed these scandals?

    Both scandals functioned as forcing events for the wider research integrity infrastructure, well beyond the two institutions directly involved.

    • Image-forensics screening became standard practice at major journals after 2014, with publishers adopting software-assisted duplication and manipulation detection for every submitted figure, not just those flagged by reviewers.
    • The Committee on Publication Ethics (COPE) expanded its guidance on image manipulation and data fabrication, giving journal editors a shared, referenceable framework for handling suspected figure manipulation as research misconduct.
    • Institutional oversight of extraordinary claims tightened, with more stem cell laboratories requiring independent, blinded replication of headline results before submission.
    • Retraction Watch, founded in 2010, has since built a public database that made both cases — and thousands of subsequent retractions — searchable and citable as case-study evidence for research misconduct case studies used in training and policy work.
    • Egg-donation and biospecimen ethics protocols were tightened across stem cell research consent frameworks following direct scrutiny of the coercive donation practices in the Hwang case.

    Frequently asked questions

    What are the problems with stem cell research?

    Beyond the underlying ethical debate over embryo use, stem cell research carries elevated misconduct risk because pluripotency and cloning claims are technically hard to verify quickly, media and funding pressure reward speed over replication, and image-based evidence is easy to manipulate before independent scrutiny occurs.

    What is the controversy with stem cell research?

    The scientific controversy extends beyond embryo ethics to research integrity: the Hwang and STAP cases showed that landmark claims in prestigious journals could be entirely fabricated, undermining public trust and forcing funders and institutions to demand stronger pre-publication verification.

    Is stem cell research banned in the UK?

    No. The United Kingdom permits regulated human embryonic stem cell research under the Human Fertilisation and Embryology Authority framework, one of the more permissive regimes in Europe, though all work remains subject to licensing and ethical review distinct from the misconduct issues in the Hwang and STAP cases.

    What does this legacy mean for research integrity oversight?

    For research administrators, publishers, and funders, the enduring lesson of these two cases is structural, not personal: misconduct was caught by replication failure and whistleblowing, not by peer review at the point of publication. Institutional research integrity offices, journal editorial teams, and funder due-diligence processes now build in image screening, raw-data deposition requirements, and independent replication checkpoints specifically because peer review alone did not catch either fraud before publication.

    Two decades after Hwang and more than a decade after STAP, both cases remain the reference points cited whenever a stem cell claim looks too clean, too fast, or too convenient — a durable legacy for a field whose credibility depends on distinguishing genuine breakthroughs from fabricated ones.

  • ORI Research Misconduct Case Summaries Database: What It Publishes and Removes

    The ORI Case Summaries database is the U.S. Office of Research Integrity’s public register of confirmed research misconduct findings involving Public Health Service-funded research. Each entry names the respondent, describes the fabrication, falsification, or plagiarism found, and lists the administrative sanctions imposed — but, unlike a permanent archive, a case is removed from the live list once its sanction period expires.

    The ORI research misconduct case summaries register is the closest thing the U.S. federal research-integrity system has to a public “misconduct docket.” For research administrators, publishers, and compliance officers, understanding exactly what it discloses — and, just as importantly, what it stops disclosing over time — is essential to using it correctly as a due-diligence tool.

    The Office of Research Integrity (ORI) is the HHS component that oversees research integrity for Public Health Service-supported research, a remit that in practice covers the great majority of federally funded biomedical science, including nearly all NIH-funded work.

    What is the ORI Case Summaries database?

    The ORI Case Summaries database is a page maintained at ori.hhs.gov listing individuals against whom ORI has made a formal finding of research misconduct and imposed an administrative action. It is not a docket of allegations or open investigations — a case only appears once an inquiry and full investigation have concluded and ORI has accepted the finding.

    Research misconduct itself is narrowly defined under 42 CFR Part 93 as fabrication, falsification, or plagiarism (FFP) in proposing, performing, or reviewing research, or in reporting research results. Disputes about interpretation, authorship credit, or research design are explicitly excluded from this definition, which is why the register is narrower than the broader universe of retracted papers.

    ORI itself, established in 1992 within HHS, does not conduct most investigations directly. Institutions receiving PHS funds are required to investigate allegations under their own procedures and report outcomes to ORI, which then reviews the institutional finding before a case summary is published.

    What information is disclosed in each case summary?

    Each published case summary follows a broadly consistent disclosure pattern, even though length and detail vary case by case. This consistency is what makes the register usable as a structured compliance-checking resource rather than a set of one-off press notices.

    • Respondent identity — the named individual found to have committed misconduct, plus their institution at the time of the conduct.
    • Nature of the misconduct — which category of fabrication, falsification, or plagiarism was found, and in what context (grant application, published paper, data submitted to PHS).
    • Affected research — the specific publications, images, datasets, or grant materials implicated, often naming the journal or funding mechanism.
    • Administrative actions — the sanctions ORI imposed, which can include debarment from receiving federal funds, exclusion from PHS advisory or peer-review service, and supervision or certification requirements for any future PHS-supported research.
    • Corrective steps — where applicable, a note that the respondent must request retraction or correction of the affected literature.

    What the summaries generally do not disclose is the full institutional investigation report, interview transcripts, or the raw evidentiary record — those remain with the institution and are only obtainable, if at all, through a separate public-records request.

    How long do case summaries stay public?

    This is the detail most guidance on the ORI register omits, and it materially changes how the database should be used. ORI’s own case summary page states that the list only includes respondents who currently have an administrative action in effect — once a debarment, supervision period, or certification requirement expires, the case summary is removed from the live online list.

    In practice, this means the ori.hhs.gov register functions as an active-sanctions list, not a permanent misconduct archive. A researcher sanctioned for three years in 2020 will typically no longer appear on the current page by 2026, even though the underlying finding was never reversed. Older findings persist instead in ORI’s historical newsletters and in contemporaneous Federal Register and NIH Guide for Grants and Contracts notices, which are not removed once published.

    For anyone conducting due diligence — a journal editor vetting a submission, a hiring institution, or a funder — checking the live case summaries page alone is not sufficient to establish that a researcher has a clean misconduct history; the historical notice archive must also be checked.

    How does it compare with other misconduct registers?

    Research administrators often conflate the ORI register with other misconduct- and retraction-tracking resources. They serve different purposes and, critically, have very different persistence rules.

    Register Scope Named respondent? Persistence
    ORI Case Summaries (ori.hhs.gov) PHS/NIH-funded research with a confirmed misconduct finding and an active sanction Yes Removed once the sanction period expires
    ORI historical notices (NIH Guide / Federal Register) Same findings, at time of original publication Yes Permanent archival record
    Retraction Watch Database Retracted or corrected papers from any funder, any cause (misconduct, error, or dispute) Sometimes, via linked retraction notice Permanent
    Institutional investigation reports Single institution, single case Varies by institution’s public-records policy Retained per institutional records policy, rarely public by default

    The practical takeaway: the ORI register and the Retraction Watch Database answer different questions. ORI tells you whether someone is currently under federal sanction; Retraction Watch tells you whether a specific paper has been retracted or corrected, regardless of whether any individual was ever sanctioned by ORI.

    Common questions about ORI case summaries

    Are researchers named in ORI case summaries?

    Yes. Once ORI has finalised a finding of research misconduct and imposed an administrative action, the respondent’s name and affiliated institution are published in the case summary. This differs from the confidentiality that generally applies while an inquiry or investigation is still open and unresolved.

    What happens to a case summary once the sanction ends?

    The entry is removed from the live online list at ori.hhs.gov once the administrative action — debarment, supervision, or certification requirement — expires. The finding itself is not reversed; it persists instead in ORI’s historical newsletters and Federal Register notices, which remain permanently accessible.

    Where else are ORI misconduct findings published besides the website?

    Beyond the case summaries page, ORI findings have historically been announced through the NIH Guide for Grants and Contracts, the ORI Newsletter, and Federal Register notices. These channels provide an archival record that outlasts the rolling, sanctions-only website list.

    What this means for research administrators and publishers

    For institutional research integrity officers, the practical implication is clear: the ORI Case Summaries page is a screening tool for active sanctions, not a comprehensive misconduct history check. A clean search result today does not confirm a researcher has never been found to have committed misconduct — only that no sanction is currently in force.

    A 2024 modernisation of the underlying regulations at 42 CFR Part 93, with a 1 January 2026 effective date, updated investigation and reporting procedures between institutions and ORI, but did not change this fundamental “currently sanctioned” design of the public case summary list. Institutions building author-vetting, hiring, or peer-reviewer-screening workflows should pair a live ORI search with a check of ORI’s historical notice archive and, where the concern involves a specific publication, the Retraction Watch Database.

    As research-integrity oversight continues to modernise, the case for a permanent, jointly maintained misconduct record — rather than a rolling active-sanctions list — is likely to grow, particularly as funders and publishers increasingly expect due diligence to extend beyond an individual’s current sanction status.

    For related definitions and standards context, see the CASRAI Dictionary and the broader research administration resource hub.

  • Research Misconduct Lawyer: Do You Need One?

    A research misconduct lawyer is a solicitor, barrister, or (in the US) an attorney who advises a researcher through an allegation of fabrication, falsification, or plagiarism — from the initial screening through formal investigation, any disciplinary hearing, and appeal. Legal counsel becomes advisable once an allegation is put in writing, and becomes necessary where funding, employment, professional registration, or a potential fraud referral is at stake.

    Research misconduct is fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results — a definition set out in US federal regulation and mirrored across UK institutional policies aligned to the Concordat to Support Research Integrity. Honest error and legitimate differences of scientific opinion are explicitly excluded.

    Do you need a research misconduct lawyer?

    Not every allegation requires a lawyer. A colleague’s informal query about a figure in a preprint, resolved by sending the underlying data, rarely needs legal input. A written allegation that triggers a formal screening or inquiry under an institution’s misconduct procedure is a different matter entirely.

    Engage counsel as soon as any of the following applies:

    • The allegation has been formally logged and a named investigation or inquiry panel has been convened.
    • Federal or funder money is involved, raising the possibility of debarment, clawback, or a fraud referral.
    • The outcome could affect employment status, tenure, or professional registration.
    • The institution’s own HR or disciplinary process is running in parallel with the research-integrity process.
    • Findings, if upheld, would likely require retraction of published work or notification to a journal or funder.

    A research misconduct lawyer does not replace the institution’s own procedure. Their role is to make sure that procedure is followed correctly, that evidence is preserved and properly interpreted, and that the respondent’s side of the record is documented at every stage — screening, inquiry, investigation, and any appeal.

    What happens during a research misconduct investigation?

    Most frameworks — UK and US alike — separate a misconduct case into a preliminary screening stage and a fuller inquiry, followed by a formal investigation and an appeal if findings are contested. The exact labels and thresholds differ by jurisdiction and by institution, which is itself a source of confusion for respondents.

    In the UK, there is no single statutory regulator equivalent to the US Office of Research Integrity (ORI). Each research organisation runs its own procedure, historically modelled on the UK Research Integrity Office (UKRIO) template. UKRIO’s 2023 publication, Procedure for the Investigation of Misconduct in Research, was formally sunset and replaced in December 2025 by a revised document titled Detailed procedure for investigating breaches of research integrity — a change institutional policy owners should check their references reflect. UK Research and Innovation (UKRI) also published dedicated guidance for research organisations on investigating research misconduct in April 2025, which states that institutional procedures “should be developed and reviewed in light of, and be consistent with, the Concordat to Support Research Integrity.”

    In the US, misconduct involving Public Health Service-funded research is governed by federal regulation at 42 CFR Part 93, which sets a three-stage process of inquiry, investigation, and adjudication, with ORI oversight and a defined right of appeal.

    Aspect US (PHS/ORI-funded research) UK (institution-led)
    Governing framework 42 CFR Part 93; ORI oversight No single regulator; institutional procedures aligned to the Concordat to Support Research Integrity and UKRI guidance
    Typical phases Inquiry → investigation → adjudication → appeal Screening → inquiry/initial assessment → formal investigation → appeal
    Standard of proof Preponderance of the evidence Balance of probabilities (the standard civil test used across most institutional procedures)
    Right to be accompanied Institution-dependent; counsel’s role at interviews is often advisory only Statutory right to be accompanied at a formal disciplinary hearing (Employment Relations Act 1999)

    The practical consequence for a respondent is definitive: know which framework applies to your case before assuming what your lawyer can and cannot do in the room.

    What rights does a respondent have?

    Due process protections for an accused researcher exist precisely because a misconduct finding can end a career. A fair procedure gives the respondent the opportunity to see and answer the case against them before any finding is made.

    Core respondent rights typically include:

    • Written notification of the specific allegations and the evidence supporting them.
    • A named point of contact and a reasonable timeframe to prepare a response.
    • The opportunity to submit evidence, nominate witnesses, and comment on a draft report before it is finalised.
    • Confidentiality of the process, so far as is consistent with a fair and thorough investigation.
    • A right of appeal against adverse findings or disproportionate sanctions.

    In the UK, employees have a statutory right under the Employment Relations Act 1999 to be accompanied at a formal disciplinary hearing by a colleague or trade union representative; a legal representative is not automatic and depends on the institution’s own procedure and the seriousness of the case. This is a materially different arrangement from many US institutional processes, where legal counsel may attend interviews but is frequently restricted to an advisory role rather than direct advocacy before the panel.

    Whose responsibility is it to report misconduct?

    Under the Concordat to Support Research Integrity, responsibility for identifying and reporting suspected research misconduct is shared, not delegated to a single office. Researchers, supervisors, co-authors, and institutions each carry a duty to raise a genuine concern through the correct channel rather than resolve it informally or ignore it.

    This shared-responsibility model has a direct consequence for anyone accused: the person who raised the concern may be a co-author, a journal editor acting on a reader’s tip, a funder’s compliance team, or an internal whistleblowing route — each of which triggers a different entry point into the institution’s procedure, and each of which a research misconduct lawyer will want identified early, since it shapes what evidence already exists before the respondent is even notified.

    Common questions about research misconduct allegations

    Who investigates allegations of research misconduct?

    In the UK, the employing research organisation investigates, typically through an inquiry panel or investigation committee constituted under its own procedure. In the US, institutions conduct the inquiry and investigation, with the federal Office of Research Integrity overseeing PHS-funded research and adjudicating contested findings.

    What counts as research misconduct?

    Research misconduct is fabrication, falsification, or plagiarism in proposing, conducting, reviewing, or reporting research. Some institutional frameworks add serious deviation from accepted practice or failure to secure required ethical approvals, but honest error and legitimate differences of scientific opinion are explicitly excluded from every recognised definition.

    What is considered the most serious form of research misconduct?

    Fabrication — inventing data outright — and falsification — manipulating materials, equipment, or data so the record is inaccurate — are generally treated as the most serious categories, because they corrupt the research record itself rather than misattributing credit, which is the harm caused by plagiarism.

    Whose responsibility is it to identify and report research misconduct?

    Under the UK’s Concordat to Support Research Integrity, every researcher, supervisor, and institution shares responsibility for identifying and reporting suspected misconduct. There is no single national regulator, so each employer’s own procedure — typically aligned to UKRI guidance — governs how, and to whom, a concern must be formally raised.

    For an accused researcher, the practical implication of all this is straightforward: engage legal counsel as soon as a formal allegation exists, confirm which procedural framework governs the case, and insist on the written notice, response time, and appeal rights that framework guarantees. As UKRIO’s December 2025 procedure revision and UKRI’s 2025 guidance both illustrate, the rules themselves continue to evolve — a respondent’s best protection is a lawyer who tracks which version currently applies, not the version that applied when the institution’s policy was last printed.

    For related definitions and process context, see the CASRAI Dictionary and the broader research administration resource hub.

  • Francesca Gino Research Misconduct Case Study

    The Francesca Gino research misconduct case is the 2021–2025 dispute in which Harvard Business School investigated, and ultimately dismissed, a tenured professor after concluding she had fabricated data in four published studies — a process that also produced a $25 million lawsuit, a Harvard countersuit, and one of the rare tenure revocations in the university’s modern history.

    Research misconduct, in the definition used by US federal policy and echoed by bodies such as the Committee on Publication Ethics (COPE), is the fabrication, falsification, or plagiarism of data in proposing, performing, or reviewing research. The Gino case is unusual not for the underlying allegation but for how visible the institutional machinery became: an 18-month internal inquiry, a 1,200-page report, a public unsealing order, and two overlapping lawsuits that together offer a rare, document-level view of how one major research university actually runs a misconduct investigation.

    What is the Francesca Gino research misconduct case?

    Francesca Gino was the Tandon Family Professor of Business Administration at Harvard Business School, where her research on honesty and ethical behaviour made her one of the most cited figures in behavioural science. Concerns about her data first surfaced around 2020, when a doctoral student’s replication attempt failed to reproduce a widely publicised Gino networking study. That failure led to a wider audit that eventually implicated four separate papers.

    Harvard’s internal investigation committee — three senior Harvard Business School faculty, assisted by an outside forensic firm — concluded that Gino had committed research misconduct intentionally, knowingly, or recklessly. The university placed her on unpaid administrative leave in June 2023 and, in May 2025, revoked her tenure and ended her employment before her two-year suspension had even run its course.

    How did Harvard’s misconduct investigation unfold?

    The Gino case shows a misconduct investigation moving through distinct, document-traceable stages rather than a single disciplinary event. Each stage generated its own record, several of which later became public through litigation.

    • 2020–2021: Doctoral candidate Zoé Ziani fails to replicate a Gino personal-networking study and raises concerns internally.
    • Autumn 2021: The Data Colada team — Uri Simonsohn, Leif Nelson, and Joseph Simmons — contacts Harvard Business School about anomalies in four Gino papers.
    • 2021–2023: Harvard conducts an internal investigation described by the HBS dean as an “18-month” process, producing a 1,200-page report under Case RI21-001.
    • June 2023: HBS places Gino on unpaid administrative leave; Data Colada simultaneously publishes its “Data Falsificada” blog series detailing the alleged anomalies.
    • August 2023: Gino files a $25 million lawsuit against Harvard, HBS Dean Srikant Datar, and the three Data Colada researchers, alleging defamation and gender discrimination.
    • March 2024: Judge Myong J. Joun orders the unsealing, with redactions, of Harvard’s 1,200-page investigation report.
    • September 2024: The court dismisses Gino’s defamation and privacy claims against both Harvard and the Data Colada defendants in full; breach-of-contract and gender-discrimination claims are allowed to proceed.
    • May 2025: The Harvard Corporation revokes Gino’s tenure and terminates her employment.
    • September 2025: Harvard sues Gino for defamation, alleging she submitted a falsified dataset to the university during the dispute.

    Notably, under US federal research-integrity rules, the Office of Research Integrity (ORI) only has jurisdiction over misconduct in Public Health Service-funded research. Much of Gino’s behavioural-science work fell outside that remit, meaning the entire investigation design — committee composition, evidentiary standard, and appeal rights — was governed solely by Harvard’s own internal policy rather than a codified federal or funder-mandated process.

    What evidence did investigators find in the retracted papers?

    Four papers sit at the centre of the case. All four have since been retracted, though the first was flagged for an unrelated data issue before the wider investigation began.

    Paper Journal Year published Retraction status
    “Signing at the beginning makes ethics salient…” (Shu, Mazar, Gino, Ariely, Bazerman) Proceedings of the National Academy of Sciences 2012 Retracted September 2021
    “Evil Genius? How Dishonesty Can Lead to Greater Creativity” (Gino, Wiltermuth) Psychological Science 2014 Retracted 2023
    “The Moral Virtue of Authenticity…” (Gino, Kouchaki, Galinsky) Psychological Science 2015 Retracted 2023
    “Why Connect? Moral Consequences of Networking…” (Gino, Kouchaki, Casciaro) Journal of Personality and Social Psychology 2020 Retracted 2023

    According to Harvard’s unsealed report, Gino offered investigators two explanations for the data irregularities: honest error by her or her research assistants, or tampering by a malicious third party with access to her files. The committee found neither explanation plausible, writing that her “repeated and strenuous argument” for a bad-actor scenario across four separate studies undermined the credibility of her broader testimony.

    A separate, self-organised accountability effort — the Many Co-Authors Project — later reviewed 56 papers naming Gino as involved in data collection. Its contributors reported that for roughly 60% of those papers, responding co-authors said they had never had access to the underlying raw data, a data-provenance gap that goes beyond the four papers formally retracted.

    What happened in the Gino v. Harvard litigation?

    Litigation is what made this a document-level case study rather than a private disciplinary matter. Gino’s August 2023 suit sought $25 million and alleged defamation, gender discrimination under Title IX, and breach of contract. Harvard’s decision to submit its full 1,200-page report as court evidence — and a subsequent judicial order to unseal it — meant the investigative record itself became a matter of public record, rather than remaining confidential under standard university misconduct procedure.

    Did Harvard sue Francesca Gino for defamation?

    Yes. In September 2025, Harvard filed its own defamation suit against Gino, alleging she submitted a falsified dataset to the university in an attempt to prove she had not committed data fraud. This followed her original 2023 defamation claim against Harvard, which a federal judge dismissed in September 2024.

    Does Francesca Gino still work at Harvard?

    No. The Harvard Corporation, the university’s top governing board, revoked Gino’s tenure and terminated her employment in May 2025, ending her unpaid suspension before it was due to expire. Harvard described tenure revocation as an extremely rare step not used at the institution for decades.

    Did Harvard revoke Francesca Gino’s tenure over falsifying data allegations?

    Yes. Harvard’s investigation committee, made up of three senior HBS faculty, concluded after an 18-month inquiry that Gino had committed research misconduct “intentionally, knowingly, or recklessly,” a finding that directly preceded the termination proceedings completed in 2025.

    What does the case reveal about institutional misconduct processes?

    For research administrators, the Gino case is less a story about one professor than a stress test of how an elite institution structures a misconduct inquiry when there is no external regulator compelling a specific procedure. Several implications stand out.

    • Investigation length is a real institutional risk. An 18-month internal inquiry, followed by two further years of litigation before a final personnel decision, shows how misconduct cases without a codified stage-gate process (unlike, for example, COPE’s published flowcharts) can extend for years and generate parallel legal exposure.
    • Confidentiality and public interest can collide. Harvard initially treated its 1,200-page report as confidential; a court order — not university policy — forced its unsealing. Institutions relying purely on internal confidentiality norms should anticipate that litigation can override them.
    • Peer self-auditing is emerging as a parallel accountability layer. The Many Co-Authors Project shows co-authors organising their own data-provenance review independently of the university process, filling a gap that formal institutional investigation did not cover at scale.
    • Whistleblower exposure has financial consequences. Data Colada’s team faced personal litigation risk for reporting anomalies, prompting outside researchers to crowdfund legal costs — a chilling-effect dynamic that institutional research-integrity policy rarely addresses directly.

    Frameworks such as the UK’s Concordat to Support Research Integrity and COPE’s core practices exist precisely to standardise the stages this case worked through ad hoc: initial concern, preliminary assessment, formal investigation, evidentiary report, and sanction. Institutions without an equivalent codified process should expect that, absent clear stage-gates, a misconduct case can default to years of litigation to resolve what a documented procedure might settle in months.

    The case remains open in part: Gino’s breach-of-contract and gender-discrimination claims against Harvard, and Harvard’s own defamation suit against Gino, were both still active as of late 2025. The eventual rulings will further shape how far US courts are willing to scrutinise a university’s internal misconduct-investigation process.

    For broader context on how institutions structure research-integrity roles and terminology, see CASRAI’s research administration resources and the CASRAI Dictionary.

  • Sanctions for Research Misconduct: Debarment, Funding Bans and Criminal Prosecution Compared

    Sanctions for research misconduct fall into three distinct enforcement tracks: institutional and federal debarment (led in the United States by the Office of Research Integrity), funder-imposed bans such as those applied by UKRI and Wellcome, and, in rare cases, criminal prosecution for fraud involving public funds. Each track has its own decision-maker, standard of proof, and range of outcomes, and the three are frequently confused in general explainers.

    Research misconduct is formally defined under US federal regulation as fabrication, falsification, or plagiarism (“FFP”) committed in proposing, performing, or reviewing research, or in reporting research results (45 CFR § 93.103). No equivalent single statutory definition exists in the United Kingdom, where funders and employers apply their own policies against the same broad FFP framework.

    What sanctions exist for research misconduct?

    Three separate mechanisms can be triggered by a single proven case of research misconduct, and they are not mutually exclusive. An institution can dismiss a researcher, a federal or national funder can bar them from future awards, and — only in the most serious cases involving fraud against public money — a prosecutor can bring criminal charges.

    The table below compares the three tracks as they currently operate.

    Enforcement track Who imposes it Typical sanction range Standard of proof
    Federal debarment (US) HHS Office of Research Integrity / federal agency Supervision or funding restriction (commonly around three years) up to lifetime exclusion from federal funds Preponderance of the evidence
    Funder sanctions (UK) UKRI, Wellcome, NIHR and other funders individually Grant termination, repayment demands, time-limited bar on future applications Funder’s own investigation findings
    Criminal prosecution National courts (e.g. US Department of Justice) Fines, restitution, imprisonment Beyond reasonable doubt

    Institutional actions — reprimand, retraction requests, termination of employment — typically happen first and independently of whether a funder or prosecutor later gets involved.

    How does ORI debarment work in the United States?

    ORI oversees misconduct findings in Public Health Service–funded research under 42 CFR Part 93, and can recommend administrative actions ranging from a certification requirement to full debarment. Debarment itself is executed through the government-wide exclusion mechanism at 2 CFR Part 180, which bars a researcher from receiving any federal award, not only from the agency that funded the original research.

    Debarment periods are not fixed by statute. ORI case records and published analyses of federal misconduct findings indicate that a supervised-research period or funding restriction of around three years is the most common outcome, with longer bars — up to a lifetime exclusion — reserved for the most serious or repeated cases. The National Institutes of Health can additionally exclude a debarred individual from serving on grant review panels.

    • Certification of research integrity for a fixed period
    • Supervision requirements on future federally funded work
    • Time-limited exclusion from federal funding and review committees
    • Lifetime debarment in the most severe or repeat cases

    How do UKRI and UK funders sanction research misconduct?

    The UK has no single federal debarment register equivalent to the US system. Instead, individual funders investigate and sanction cases against their own grant conditions, and institutions handle the underlying employment consequences separately.

    Wellcome’s grant conditions state that where an organisation upholds a research misconduct allegation, sanctions “may vary in length, depending on the severity of the misconduct” — which can include grant termination, repayment demands, and a time-limited bar on future applications to that funder. UKRI applies a comparable case-by-case approach across its research councils rather than a single published tariff, and the UK Research Integrity Office supports institutions in running consistent, fair investigation processes rather than issuing sanctions itself. Plagiarism findings that concern disputed authorship credit are frequently resolved at institutional level even before a funder becomes involved.

    Because UK sanctions are funder-specific rather than government-wide, a researcher barred by one funder is not automatically barred by another — a structural difference from the US debarment model that institutional research administration offices need to track carefully when advising on eligibility.

    When does research misconduct lead to criminal prosecution?

    Criminal prosecution is rare and reserved for cases where misconduct overlaps with fraud against public funds, false statements to a federal agency, or theft — not for FFP findings alone. The evidentiary bar is far higher than for an administrative debarment finding, which is one reason so few cases reach a courtroom.

    Two US cases illustrate the outer edge of this track. In 2005, University of Vermont researcher Eric Poehlman became the first American scientist sentenced to prison over research misconduct-related fraud, receiving one year and a day and an order to repay roughly $180,000 after fabricating data in federal grant applications. In 2015, former Iowa State University researcher Dong-Pyou Han was sentenced to 57 months in federal prison for making false statements in NIH-funded HIV vaccine research — one of the longest custodial sentences on record for research fraud in the US.

    No comparable UK criminal prosecutions for research misconduct itself exist on the public record; related conduct there is more commonly pursued, if at all, under general fraud law rather than a research-specific statute.

    Frequently asked questions about sanctions for research misconduct

    What are the penalties for research misconduct?

    Penalties include institutional actions (reprimand, retraction, termination), funder sanctions (grant termination, repayment, time-limited application bans), and federal debarment from all future government funding. Criminal penalties — fines and imprisonment — apply only in rare cases involving fraud against public money.

    What is the most severe sanction for research misconduct?

    The most severe administrative sanction is lifetime federal debarment, which permanently excludes a researcher from receiving any US government funding or serving on review panels. Where fraud against public funds is proven, imprisonment is the most severe outcome overall, though it is far less common than debarment.

    Can you go to jail for research misconduct?

    Yes, but only in cases that cross into criminal fraud, such as fabricating data to obtain federal grant money. Documented examples include Eric Poehlman (one year and a day, 2005) and Dong-Pyou Han (57 months, 2015), both convicted over federally funded research fraud rather than FFP findings alone.

    What are the three types of research misconduct?

    Under the US federal definition, the three types are fabrication (inventing data), falsification (manipulating data, materials, or processes), and plagiarism (appropriating others’ ideas or words without credit) — collectively known as FFP under 45 CFR § 93.103.

    What these sanctions mean for institutions and researchers

    The practical implication for research administrators is that a single misconduct finding can trigger overlapping consequences on different timelines: an institution may act within weeks, a funder’s sanction may follow months later, and a criminal referral — where warranted — can take years to resolve. Advising researchers or institutional leadership requires tracking all three tracks separately rather than assuming one outcome implies another.

    For publishers and standards bodies, the growing use of AI-assisted fabrication and image manipulation is pushing debarment and funder-sanction frameworks to adapt faster than criminal law can follow, since prosecution requires proof of intent to defraud rather than mere data manipulation. Institutions should expect funders on both sides of the Atlantic to keep tightening administrative sanctions — debarment periods and funder bans — well ahead of any corresponding shift in criminal enforcement.