CASRAI Dictionary

Tag: ethical misconduct in research

  • Academic Misconduct in Research: Two Separate Processes

    Academic misconduct is a student-facing breach of assessment rules — plagiarism, collusion, contract cheating or fabricated coursework data — handled by a university’s student disciplinary office. Research misconduct is a narrower, professional-conduct offence covering fabrication, falsification or plagiarism (FFP) in proposing, performing or reporting research, handled by a separate research-integrity committee with its own evidentiary standard.

    The two terms get used interchangeably in casual conversation, but treating academic misconduct in research settings as one problem is a real institutional risk: it routes cases to the wrong panel and, for postgraduate researchers straddling both worlds, can leave a gap in accountability.

    Research misconduct, in the definition adopted by the US Federal Policy on Research Misconduct and applied by the Office of Research Integrity under 42 CFR Part 93, is: “fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results.” Honest error and legitimate differences of scientific opinion are explicitly excluded.

    What is academic misconduct?

    Academic misconduct is any act that gains, or attempts to gain, an unfair advantage in formal assessment. The University of Cambridge’s Discipline Regulation 7 defines it as “gaining or attempting to gain, or helping others to gain or attempt to gain, an unfair academic advantage in formal University assessment, or any activity likely to undermine the integrity essential to scholarship and research” (Cambridge Statutes and Ordinances, Chapter II, 2024 edition).

    In practice, academic misconduct covers a defined, closed list of assessment offences:

    • Plagiarism — using another person’s words, ideas or data without acknowledgement.
    • Collusion — working with others on an assessment that must be completed individually.
    • Contract cheating — paying or arranging for a third party to produce assessed work.
    • Fabrication of coursework data — inventing results for a lab report, dissertation or dataset submitted for a mark.
    • Exam-related breaches — possession of unauthorised materials or impersonation during an examination.

    A course convenor, exam board or student conduct office decides these cases, with sanctions ranging from a capped mark to expulsion.

    What is research misconduct?

    Research misconduct is a narrower category applied to the professional conduct of research itself, regardless of job title. Imperial College London’s research governance office characterises it as actions that “fall short of the standards of ethics, research and scholarship required to ensure that the integrity of research is upheld,” noting it “can cause harm to people and the environment, wastes resources, undermines the research record and damages the credibility of research.”

    The internationally recognised core is fabrication, falsification and plagiarism (FFP):

    • Fabrication — inventing data or results and recording them as real.
    • Falsification — manipulating materials, equipment, processes or images so the research record misrepresents what was actually done.
    • Plagiarism — appropriating another researcher’s ideas, methods or results without credit, in a published or submitted work.

    Wider “questionable research practices” — gift or ghost authorship, salami-slicing, undisclosed conflicts of interest — sit outside the strict FFP definition but are increasingly folded into research-integrity policy. Robust use of the CRediT contributor role taxonomy, which CASRAI originated in 2014 and which NISO now stewards as ANSI/NISO Z39.104-2022, is one practical way institutions pre-empt gift-authorship disputes before they escalate.

    Why do institutions run two separate processes?

    Academic and research misconduct sit in different regulatory lineages, so institutions build separate committees and policies rather than one catch-all procedure.

    Dimension Academic misconduct Research misconduct
    Typical subject Registered student, any level Principal investigator, postdoc, established researcher
    Core acts Plagiarism, collusion, contract cheating, exam breaches Fabrication, falsification, plagiarism (FFP)
    Governing framework University disciplinary regulations / exam conventions Institutional research-integrity policy under the Concordat to Support Research Integrity (Universities UK, revised 2019); in the US, 42 CFR Part 93
    Decision body Exam board, student conduct office Dedicated research-integrity or research-misconduct committee, often with external assessors
    Typical sanctions Mark penalty, module fail, suspension, expulsion Correction or retraction, funding clawback, research ban, referral to a professional register
    External reporting Rarely reported outside the institution Often reportable to funders (e.g. UKRI), publishers and, for clinical research, regulators

    The UK has no statutory equivalent of the US Office of Research Integrity. Instead, UKRIO, founded in 2006, provides advisory guidance while institutions retain investigatory responsibility under their own research-misconduct policy, as required by the Concordat.

    Where does the line blur for postgraduate researchers?

    The clean split breaks down precisely where most CASRAI readers work: the PhD candidate who is simultaneously a registered student and an author generating primary research data.

    Cambridge’s own regulation illustrates the overlap directly. Discipline Regulation 7 folds research-conduct failures into the academic misconduct definition, explicitly including “fabrication, falsification or misrepresentation of data, results or other outputs or aspects of research” and “failure to meet legal, ethical and professional obligations in carrying out research.” That means conduct that would be classic FFP research misconduct if committed by a principal investigator is processed through the student disciplinary system when the person responsible is a registered student.

    This is not a Cambridge quirk. UKRIO’s April 2024 guidance, Research Integrity Matters Relating to Students, was published because institutions across the sector were struggling to decide whether a doctoral candidate’s data-fabrication allegation should go through student discipline or the research-misconduct committee — and because degrees already awarded can be affected by findings made years later.

    The ambiguity is not new. A widely cited 2008 study in Studies in Higher Education (Mitchell and colleagues, cited by 89 in Google Scholar) found that misconduct during doctoral study is difficult to resolve “because of lack of clarity in definitions, supervisor naïveté and failure to acknowledge students’ [research-conduct failures]” as distinct from ordinary academic misconduct. A 2024 conceptual review in the same journal (Harrad, cited by 22) found that framing a case as “academic integrity” foregrounds ethics and values, while framing it as “academic misconduct” foregrounds illegitimate advantage — a distinction that shapes which committee, and which burden of proof, a postgraduate case ends up under.

    Institutions that get this right build an explicit triage step into their research administration workflow: does the disputed conduct sit in assessed coursework (student discipline) or in the primary research record intended for publication or a thesis contribution (research-misconduct policy)? The same underlying act can trigger both processes in parallel, so the routing decision needs to be documented, not assumed.

    Answer-first Q&A

    What qualifies as academic misconduct?

    Academic misconduct is any act that gains, or helps another gain, an unfair advantage in formal assessment. It includes plagiarism, collusion, contract cheating, fabricated coursework data, and exam-related breaches such as possessing unauthorised materials, regardless of whether the student intended to cheat.

    What are examples of research misconduct?

    Classic examples include inventing data for a clinical trial that never took place, manipulating a Western blot image to fit a hypothesis, and plagiarising another researcher’s methods or results in a submitted manuscript. All three fall under the fabrication, falsification and plagiarism (FFP) definition used by research-integrity policy.

    What are the five unethical practices in research?

    A frequently cited framework lists falsification of data, failure to credit others, plagiarism, undisclosed conflicts of interest, and biased design or interpretation driven by outside influence. Only the first and third meet the strict FFP definition of research misconduct; the rest are treated as questionable research practices.

    How hard is it to prove academic dishonesty?

    Both academic and research misconduct cases in UK higher education are generally decided on the civil balance of probabilities, not the criminal standard. Proving intent is harder than proving the act itself, which is why panels focus on documentary evidence — submission logs, raw data files, similarity reports — rather than motive.

    What this means for institutions

    For research administrators and institutional leaders, a single “misconduct policy” covering both students and PIs under one procedure will consistently mis-route the hardest cases — postgraduate researchers — because it cannot specify which evidentiary standard and which committee applies to a thesis chapter versus an exam script.

    The forward-looking fix, already visible in UKRIO’s 2024 guidance and in policies revised against the Concordat to Support Research Integrity, is an explicit triage protocol naming both process owners and stating when a case is reportable to a funder such as UKRI, when it is reportable to a publisher under Committee on Publication Ethics (COPE) guidance, and when it stays internal. Clear authorship attribution at the outset of a project remains one of the cheapest ways to prevent disputes reaching either committee at all.

  • Research Integrity Annual Statement Requirements for UK Universities

    Research integrity annual statement requirements flow from Commitment Five of the UK Concordat to Support Research Integrity: every signatory employer of research — universities, government departments, and independent research organisations — must publish a short, public annual statement summarising its research-integrity activity and disclosing high-level data on formal misconduct investigations. A research integrity annual statement is a publicly available, governing-body-approved document reporting an institution’s progress against the Concordat’s five principles: honesty, rigour, transparency and open communication, care and respect, and accountability.

    What the Concordat actually requires

    The Concordat to Support Research Integrity was first published by Universities UK in 2019 and revised in 2025. Commitment Five requires signatory organisations to “publish an annual statement” describing how they support research integrity and to include data on formal research misconduct investigations conducted that year.

    There is no single legally mandated format. The UK Committee on Research Integrity (UKCoRI) and the Research Integrity Concordat Signatories (RICS) Group commissioned a common template — distributed via the UK Research Integrity Office (UKRIO) as a self-assessment tool — to standardise reporting, but its use remains voluntary. This is why statement quality and completeness still vary sharply across the sector.

    Required sections: what goes in the statement

    Whether an institution uses the RICS template or writes its own, a compliant statement needs to cover the same ground. The table below maps required content against what the recommended template adds on top of the bare Concordat obligation.

    Section Required by Commitment Five Added by the RICS/UKRIO template
    Governance and oversight Yes — named senior owner Named committee, reporting line to Council/Senate
    Culture and training activity Yes — summary of activity Structured prompts on induction, CPD, open research training
    Policies and systems in place Yes Cross-reference to ethics review, data management, authorship policies
    Misconduct investigation data Yes — numbers required Standardised categories (plagiarism, fabrication, falsification, other)
    Reflection and forward plan Implied (“summary of activities”) Explicit “next 12 months” section
    Sign-off and publication Yes — governing-body approval, public URL Standard approval statement wording

    Each statement should also name a senior point of contact for research integrity queries — a practice UKRIO’s guidance treats as effectively mandatory even though the Concordat text does not use that word. The statement must be approved by the institution’s governing body and hosted at a stable, publicly accessible URL, since UKCoRI and RICS collect these links centrally each year.

    How misconduct-case numbers are — and aren’t — disclosed

    Commitment Five requires “data on the number of research misconduct investigations conducted,” but it does not prescribe a taxonomy, a reporting period, or a denominator. In practice this produces wide variation: some institutions report raw investigation counts, others report allegations received versus investigations opened versus proven findings, and a minority report nothing more than a narrative sentence.

    UKCoRI’s 2025 review of statements published between 2022 and 2024 — the largest analysis of its kind, following an earlier 2023 review of 280 statements from 2019/20–2021/22 — found that annual statements were available for 78% of institutions for the 2022/23 cycle and 75% for 2023/24, meaning roughly a quarter of signatories had no traceable statement at all in the most recent year reviewed. Where investigation data was reported, the review found the top three cited reasons for misconduct allegations were consistently:

    • Plagiarism
    • Failure to meet legal, ethical, or professional obligations
    • Misrepresentation

    Government departments and independent research organisations that sign the Concordat — the UK Health Security Agency and the Ministry of Defence among them — tend to publish shorter, more standardised statements against a customised template, according to UKCoRI’s companion analysis of non-HEI signatories. HEI statements, by contrast, show more variation in depth: the review found monitoring and evaluation of stated activities remains limited even where the narrative content is strong.

    Common questions, answered directly

    What is the Concordat to Support Research Integrity?

    The Concordat to Support Research Integrity is a UK sector agreement, first published by Universities UK in 2019 and updated in 2025, under which signatory universities, funders, and research organisations commit to five principles — honesty, rigour, transparency, care and respect, and accountability — and to publishing an annual statement on their progress.

    Is there an official template for the annual statement?

    There is a recommended template, commissioned by the RICS Group and distributed by UKRIO as part of a self-assessment tool, but no institution is legally required to use it. Most Russell Group and post-92 universities now use it or a close variant to aid year-on-year comparison.

    Who checks whether universities comply with the Concordat?

    No single regulator audits compliance. UKCoRI periodically commissions independent reviews — most recently covering statements published 2022–2024 — that assess coverage and quality across the sector, but publication itself is on an honour system enforced by peer visibility rather than sanction.

    What counts as research misconduct under the Concordat?

    The Concordat frames research misconduct as fabrication, falsification, plagiarism, and other serious breaches of the honesty and rigour principles, alongside failures to meet legal, ethical, or professional obligations — the three categories UKCoRI’s analysis found most frequently cited across published statements.

    Best practice: what the Open University does differently

    Several structural choices separate stronger statements from weaker ones. The Open University is a useful reference point: it names its Pro-Vice-Chancellor, Research and Innovation as the explicit institutional point of contact for research integrity matters and as the office responsible for investigating misconduct allegations — addressing the named-contact expectation that many statements leave implicit. It also maintains a public archive of its statements dating back to 2020 at a stable URL, satisfying the transparency and accountability principles beyond the single-year minimum.

    Common gaps against this standard include: publishing the statement only internally rather than at a public URL; omitting a named senior contact; reporting a single aggregate misconduct figure with no breakdown by category or outcome; and failing to link the statement back to the institution’s research ethics or authorship policies, which weakens the “systems in place” section that the RICS template expects.

    Implications for research administrators

    For research administrators, the annual statement is not a compliance afterthought — it is one of the few Concordat-derived documents external funders, publishers, and prospective staff can actually inspect. Institutions that adopt the RICS/UKRIO template structure, publish consistently year-on-year, and disaggregate misconduct data by category give funders and collaborators a materially clearer signal than a narrative paragraph can. As UKCoRI continues its periodic sector reviews, the gap between the roughly three-quarters of institutions with a traceable statement and full sector coverage is likely to remain the most visible measure of Concordat compliance in the years ahead.

    Research administration teams building or auditing a statement should also review related governance obligations set out in institutional research administration policy, and consult standard definitions of research-integrity terminology in the CASRAI research dictionary when aligning statement language with sector norms.

  • Scottish Funding Council Research Integrity Policy: What Changes From September 2026

    The Scottish Funding Council Research Integrity Policy takes effect on 1 September 2026 and applies to every Scottish higher education institution that receives SFC research and innovation funding. Its central change is a new, mandatory duty for institutions to report the outcomes of formal research misconduct investigations to SFC within one month of conclusion — a reporting relationship that did not exist under the Concordat to Support Research Integrity alone.

    Research integrity, as SFC adopts the definition from the Concordat, exists “when research is carried out according to the principles of the Concordat, and in a way that is trustworthy, ethical, and responsible.” The policy, published on 26 May 2026 by SFC’s Investment and Research Directorate, sets out what Scottish universities and SFC-funded individuals must do differently, and what SFC itself commits to doing when misconduct is proven.

    What does the policy require of Scottish universities?

    The policy applies to Scottish higher education institutions receiving research and innovation funding from SFC, and to research activities directly funded by SFC. It sets out general responsibilities drawn from the Concordat to Support Research Integrity, of which SFC is a signatory, and adds SFC-specific misconduct-reporting requirements that institutions must meet from 1 September 2026.

    Institutions must maintain transparent, robust and fair research-integrity policies, publish a governing-body-approved annual statement, and make named senior research-integrity contacts publicly available. Universities receiving SFC’s core research grants already link to that annual statement via their Research Assurance and Accountability (RAA) return; the new policy layers a live, case-by-case reporting duty on top.

    SFC is explicit about the limits of its role: it is not an appeals body for individual cases, cannot support parties during investigations, and does not grant ethical clearance or advise on research ethics. Assessment of institutional research-integrity infrastructure under the Research Excellence Framework sits outside the policy’s scope.

    What counts as research misconduct versus a questionable research practice?

    SFC uses the Concordat’s five key principles — honesty, rigour, transparency and open communication, care and respect, and accountability — as its benchmark, then draws a deliberate line between two categories of shortfall that many institutional policies still conflate.

    Questionable research practices are minor infractions or avoidable errors falling short of the five principles without evident intent to deceive, often arising from negligence or attention-to-detail failures. Research misconduct covers deliberate actions falling short of the principles at any point in the research lifecycle: fabrication, falsification, plagiarism, breach of legal or ethical obligations, misrepresentation, and improper handling of allegations — including inappropriate use of non-disclosure agreements to silence concerns. Honest errors and legitimate methodological disagreement are explicitly excluded. Institutions comparing these categories against their own glossaries can cross-check definitions in the CASRAI dictionary of research-integrity terms.

    How must institutions report misconduct investigations to SFC?

    This is the operative change from September 2026. Institutions in scope must inform SFC of the outcome of any formal research misconduct investigation, with reporting due immediately after conclusion and no later than one month afterward. Preliminary or initial assessments that do not proceed to a formal investigation do not need to be reported.

    Each confidential report to SFC must include:

    • A factual statement of the allegation’s nature, disciplinary area and time period, and which misconduct category it falls under (fabrication, falsification, plagiarism, legal/ethical/professional breach, misrepresentation, improper handling, or another category)
    • The investigation’s findings — upheld (fully or partially) or not upheld
    • A timeline of the investigation’s key stages, benchmarked against the institution’s own policy timescales
    • Actions taken where allegations are upheld, such as disciplinary referral, publisher retraction requests, or mitigation for affected research
    • Any resulting changes to institutional policy or practice, and how their effectiveness will be assured
    • What, if anything, will be communicated publicly about the case

    Institutions must also disclose if an investigation touches one of five named SFC strategic investments — Alliances for Research Challenges, the Hydro Nation Scholars Programme, the Scottish Graduate Schools for Arts and Humanities or Social Science, or Royal Society of Edinburgh–supported research — or a member of SFC’s Research & Knowledge Exchange Committee or Board.

    Personal-data protections run in parallel: respondent names, special-category data, criminal-offence information and third-party (e.g. complainant) details must never be routinely sent to SFC, though redacted disclosure can be requested in high-risk or strategic-investment cases. Looking ahead, from academic year 2027-28, SFC will cross-check upheld allegations in institutions’ annual Concordat statements against what has actually been reported to it — a compliance backstop that did not previously exist.

    How does the SFC policy compare with the UK Concordat and UKRI’s good-practice policy?

    SFC’s policy does not replace the UK-wide Concordat to Support Research Integrity, first published in 2012 and refreshed in 2025; it operationalises SFC’s own responsibilities as a signatory and funder. It draws explicitly on UKRI’s Policy on the Governance of Good Research Practice, reused with permission, and was developed with input from the Scottish Research Integrity Network.

    Framework Scope Misconduct reporting to the body Effective / published
    Concordat to Support Research Integrity All UK research employers (signatory-based) Annual statement only, no case-level reporting 2012; refreshed 2025
    UKRI Policy on the Governance of Good Research Practice UKRI-funded research and grant holders UKRI-specific notification duties for funded projects Ongoing, periodically updated
    SFC Research Integrity Policy Scottish HEIs receiving SFC funding; SFC-funded research activity Mandatory outcome reporting within one month of a formal investigation’s conclusion Published 26 May 2026; effective 1 September 2026

    The practical distinction for Scottish institutions is that the Concordat’s annual-statement cycle is now supplemented by live, case-triggered reporting — closing the “lag” SFC identifies between when misconduct is resolved and when it would otherwise surface in annual publications.

    Answer-first Q&A: research integrity principles and misconduct

    What are the 5 principles of research integrity?

    Under the Concordat definition SFC adopts, the five principles are honesty, rigour, transparency and open communication, care and respect, and accountability. Research has integrity when it is carried out according to these principles in a way that is trustworthy, ethical and responsible, at every stage of the research lifecycle.

    What are the violations of research integrity?

    Violations range from serious research misconduct — fabrication, falsification, plagiarism, breach of legal or ethical obligations, misrepresentation, and improper handling of allegations — to lesser questionable research practices such as avoidable errors or negligent attention to detail. Honest errors and genuine methodological disagreement are not violations.

    What should Scottish universities change before 1 September 2026?

    SFC acknowledges that existing institutional policies may not currently permit disclosure of investigation outcomes to an external funder, which would limit reporting on cases already underway. Research administrators and research integrity leads should treat this as a compliance gap to close now, not after September.

    • Review misconduct policies for clauses that would block or delay disclosure of outcomes to SFC, including confidentiality wording that predates this requirement
    • Confirm named contacts and senior oversight roles for research integrity are published and current, as the policy expects
    • Align internal misconduct categorisation with the Concordat’s fabrication/falsification/plagiarism/legal-ethical-breach/misrepresentation/improper-handling taxonomy so outcome reports map cleanly to SFC’s required fields
    • Check that non-disclosure agreements are not used to close out misconduct findings, which the policy treats as improper handling in itself
    • Flag any live or upcoming investigation touching SFC strategic investments (Alliances for Research Challenges, Hydro Nation Scholars, the two Scottish Graduate Schools, or RSE-supported work) for the additional disclosure requirement

    These changes matter beyond Scotland: institutions with cross-UK portfolios should map SFC’s requirements against equivalent obligations to UKRI and other funders, so research administration teams work from one internal reference of reporting triggers and timescales rather than several conflicting ones.

    What happens next?

    From 1 September 2026, SFC expects to act on most reports without further intervention, reserving formal letters and improvement plans — and, for systemic governance failures, restrictions on strategic funding applications or suspension of all SFC research funding — for the most serious or repeated cases. From AY 2027-28, its cross-check between reported outcomes and annual Concordat statements becomes the enforcement backstop, giving universities roughly one full reporting cycle to embed compliant processes first.

    Institutions treating September 2026 as a wording exercise rather than a change to reporting workflows, confidentiality clauses and NDA practice risk falling foul of that 2027-28 check.

  • How to Avoid Research Misconduct: PhD Checklist

    How to avoid research misconduct as an early-career researcher comes down to five habits: keep verifiable records from day one, agree authorship and CRediT roles before writing starts, retain data for the period your funder requires, disclose every conflict of interest, and know your institution’s reporting route before you ever need it. Research misconduct is the fabrication, falsification, or plagiarism of data, results, or other people’s work in proposing, performing, reviewing, or reporting research — a definition used by the US Office of Research Integrity and echoed by the UK Research Integrity Office (UKRIO).

    PhD students and postdocs face specific pressure points that senior investigators rarely encounter in the same way: unclear authorship expectations, ambiguous data ownership when moving labs or institutions, and thin knowledge of funder-specific retention rules. This guide sets out an individual-level checklist — not an institutional policy document — for the two career stages where most avoidable misconduct risk actually sits.

    What is research misconduct?

    The US Office of Research Integrity defines research misconduct as fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results — and explicitly excludes honest error or genuine differences of scientific opinion. UKRIO uses a broader formulation: behaviours that deliberately or recklessly fall short of the standards expected in the conduct of research.

    • Fabrication — inventing data or results that were never generated.
    • Falsification — manipulating materials, equipment, processes, or data so the research record misrepresents what actually happened.
    • Plagiarism — using another person’s ideas, words, or results without appropriate credit.

    Below these three formal categories sit questionable research practices (QRPs) that rarely trigger a formal misconduct finding but do real damage: salami-slicing a single dataset into multiple papers, gift or ghost authorship, undisclosed conflicts of interest, and selective reporting of results. Early-career researchers are disproportionately exposed to QRPs because they typically have the least authority to push back when a supervisor or collaborator proposes one.

    The early-career checklist: how to avoid research misconduct

    Prevention at the individual level works best as a small number of fixed habits applied consistently, rather than a one-off training module. The checklist below groups those habits into the three areas where early-career researchers are most exposed: record-keeping, authorship, and disclosure.

    Data retention and record-keeping

    Keep a dated, tamper-evident record of every experiment, dataset version, and analysis script — electronic lab notebooks and version-controlled repositories are far easier to defend under scrutiny than a personal laptop folder. Retention periods are not optional extras; they are usually written into your funding agreement. UKRI’s grant Terms and Conditions require standard research data to be retained for a minimum of ten years from the date the last researcher on the project drew on it, with longer retention commonly required for clinical or patient-identifiable data. Before you leave a lab or finish a contract, confirm who takes custodial responsibility for the data and archive it through your institutional repository rather than a personal drive.

    Authorship sign-off using CRediT

    Disputed or inflated authorship is one of the most common sources of misconduct allegations against early-career researchers, usually because it was never discussed until submission. Raise authorship expectations at the start of a project, not the end, and record who did what using a recognised contributor taxonomy. CASRAI originated the CRediT contributor role taxonomy in 2014; the standard is now stewarded by NISO as ANSI/NISO Z39.104-2022. Alongside CRediT, the International Committee of Medical Journal Editors (ICMJE) sets four authorship criteria — substantial contribution, drafting or critical revision, final approval, and accountability for accuracy — all of which should be satisfied and confirmed in writing before a manuscript is submitted. See CASRAI’s overview of the CRediT taxonomy and individual contributor roles, and the broader guidance on authorship practice.

    Disclosure habits: conflicts of interest and funding

    Declare funding sources, consultancy income, and any relationship that could be perceived to bias your work — on every manuscript, grant application, and peer-review invitation, not just the ones where a conflict feels obvious. Undisclosed conflicts of interest are one of the categories of unethical practice identified in the research-ethics literature alongside fabrication, falsification, plagiarism, and biased design or interpretation. Build a habit of updating your disclosure statement annually rather than reconstructing it from memory when an editor asks.

    Practice area PhD student focus Postdoc focus
    Data management Agree a data management plan and storage location with your supervisor at project start Confirm retention complies with the funder’s terms (e.g. UKRI’s minimum ten-year rule) and archive before your contract ends
    Authorship Discuss authorship order and CRediT roles with your supervisor before drafting begins Lead authorship conversations for your own team and confirm every contributor’s CRediT role before submission
    Disclosure Declare your funding source and any external roles on every submission Disclose conflicts of interest on grant applications and review invitations; update annually
    Raising concerns Know your university’s research integrity or whistleblowing contact Signpost concerns to UKRIO’s confidential advisory service or the relevant COPE flowchart

    Common questions on avoiding research misconduct

    How can research misconduct be prevented?

    Prevention combines individual habits with institutional support: complete responsible conduct of research training, keep a verifiable data record from the outset, agree authorship and disclosure expectations early, and use a named mentor to check difficult decisions before they become irreversible.

    What are the 5 unethical practices in research?

    The research-ethics literature commonly lists five categories: fabrication of data, falsification of results, plagiarism of others’ work, failure to credit contributors, and undisclosed conflicts of interest that bias design or interpretation.

    How can we prevent unethical research?

    Reduce bias systematically: pre-register study design where possible, separate data collection from data interpretation roles, declare every conflict of interest, and keep records detailed enough that a colleague could reconstruct your analysis without asking you a single question.

    What to do if you suspect misconduct

    Early-career researchers are frequently the first to notice irregularities — and often the least sure how to act on them without risking their own position. UKRIO operates a free, confidential advisory service specifically to help individuals think through a concern before deciding whether to raise it formally, independent of any single university’s internal process. The Committee on Publication Ethics (COPE) publishes decision flowcharts that editors and institutions use to work through suspected data fabrication, image manipulation, and authorship disputes; reading these in advance demystifies what a formal process actually involves.

    • Raise concerns informally with a trusted mentor first, where safe to do so.
    • Use your institution’s named research integrity contact for a formal report.
    • Contact UKRIO’s advisory service for independent, confidential guidance if you are unsure whether a concern meets the threshold for misconduct.
    • Document your concern in writing, with dates, before raising it — memory fades and paper trails protect everyone involved.

    Building a habit of research integrity

    None of the five habits above require institutional permission to start. A PhD student can open a data management plan, propose a CRediT-based authorship discussion, and file a disclosure statement without waiting for a policy update. What changes between PhD and postdoc stage is not the principle but the stakes: postdocs increasingly lead the projects, supervise the juniors, and sign the disclosures that PhD students are only learning to draft. Starting the habit early is what keeps it intact when the stakes rise.

  • UKRIO Code of Practice for Research: The Institutional Checklist

    The UK Research Integrity Office Code of Practice for Research is a voluntary, non-statutory framework — first published in 2009 and updated to Version 3.5 in July 2025 — that sets out principles and a practical checklist for how researchers and institutions should plan, conduct and report research responsibly. It tells institutions what a robust research-integrity policy should contain, but it does not carry the force of law: adoption, in full or in part, is left to each organisation’s discretion.

    UKRIO’s Code of Practice for Research is a nationally recognised benchmark that defines standards and responsibilities institutions and researchers are expected to meet across the research lifecycle, from planning through to publication.

    This walkthrough sets out what the Code asks institutions to do, what changed in the July 2025 update, and how its voluntary status differs from a statutory obligation.

    What is the UKRIO Code of Practice for Research?

    The UK Research Integrity Office (UKRIO) is an independent, UK-based advisory body supporting researchers and institutions on integrity, ethics and misconduct handling. Its Code of Practice for Research is UKRIO’s flagship guidance document, first issued in 2009 and periodically revised.

    The Code is deliberately not a rulebook but a benchmark for good practice. Organisations can adopt the Code in full, adopt it in part, or use it simply as a reference point when writing or reviewing their own research governance policies. The current Code (Version 3.5) is organised into three parts:

    • Section 1 — Recommended Checklist for Researchers: a one-page, non-technical, step-by-step summary covering actions before, during and after a research project.
    • Section 2 — Commitments: responsibilities for researchers and organisations mapped against the five key commitments of the Concordat to Support Research Integrity.
    • Section 3 — Standards for Organisations and Researchers: detailed, discipline-agnostic expectations covering the full research lifecycle.

    What does the Code ask institutions to do?

    Read from the institutional side rather than the individual-researcher side, the Code functions as an implied policy checklist. An organisation working through it should be able to demonstrate each of the following:

    • A stated culture of research integrity that treats identifying problems as routine, not reputationally damaging.
    • Clear, accessible governance policies for how research is authorised, supervised and reported.
    • A documented misconduct procedure with due process and proportionate sanctions — UKRIO’s companion Procedure for the Investigation of Misconduct in Research covers this.
    • Training and supervision structures, including guidance for those supervising early-career researchers.
    • Ethical and legal compliance mechanisms, such as functioning research ethics committees.
    • Data management aligned with the FAIR principles (Findable, Accessible, Interoperable, Reusable).
    • Defined lines of accountability, including support for research-enabling staff.
    • Periodic self-review using UKRIO’s Self-Assessment Tool.

    None of this is enforced by an external inspector. The checklist works because institutions choose to hold themselves to it — usually because funders and governing bodies expect them to.

    How did the July 2025 update (Version 3.5) change institutional obligations?

    UKRIO republished the Code and its Recommended Checklist for Researchers on 8 July 2025 as Version 3.5. This is a materially different document from the 2023 edition still cited by many secondary sources, and it adds obligations institutions had not previously needed to address explicitly. The three headline changes are:

    • Alignment with the revised Concordat (April 2025), recognising research-enabling staff — technicians, data managers, research development professionals — as parties with integrity responsibilities, not just principal investigators.
    • A new section on AI and emerging technologies, UKRIO’s first endorsed reference point for responsible AI use in research design, analysis and writing.
    • Destigmatising misconduct language, informed by UKRIO’s 2024 report on barriers to investigating research misconduct, with terminology covering the fuller spectrum of Questionable Research Practices (QRPs), not just fabrication, falsification and plagiarism.

    For a policy team, this means a Code adopted or benchmarked before mid-2025 is now out of date on at least two fronts: AI governance and the treatment of research-enabling staff. A review against Version 3.5 is a reasonable near-term action for any institution referencing the Code in its governance documents.

    No. The UKRIO Code of Practice for Research is guidance, not legislation, and UKRIO is not a statutory regulator with powers to inspect, license or sanction institutions. Compliance is voluntary, and there is no legal penalty for an institution that never adopts the Code.

    What creates real-world pressure to comply is funding conditionality, not law. UK Research and Innovation (UKRI) requires funded organisations to be signatories to the Concordat to Support Research Integrity — the sector agreement the UKRIO Code is designed to help institutions operationalise. An institution ignoring the Code’s implied checklist risks falling short of Concordat commitments it has separately, formally signed. The Code itself has no legal force, but funder conditions and funding-linked assessment exercises give voluntary adoption practical teeth.

    How does the Code compare with other integrity frameworks?

    The Code sits inside a wider landscape of UK and international integrity instruments with very different enforcement characters. The table compares the mechanisms most often confused with one another.

    Framework Status Published by Enforcement mechanism
    UKRIO Code of Practice for Research Voluntary, non-statutory guidance UK Research Integrity Office Adopted at institutional discretion; no external penalty for non-adoption
    Concordat to Support Research Integrity Voluntary sector agreement, funder-linked UKRI / Universities UK (signatory bodies) Signatory institutions must publish annual integrity statements; tied to UKRI funding eligibility
    REF (Research Excellence Framework) Funding-linked assessment exercise Research England and the UK funding bodies People, Culture and Environment element is scored and affects QR funding allocation
    US Office of Research Integrity (ORI) Statutory, federal regulation US Department of Health and Human Services Legal findings of research misconduct under 42 CFR Part 93 for PHS-funded research

    The contrast with the US model is instructive: ORI can make a formal, legally consequential misconduct finding against a researcher; UKRIO cannot. UKRIO’s role is advisory and supportive rather than adjudicative. Research administration teams on UK-US collaborations should hold both models in mind, since a US collaborator’s misconduct exposure is legally different from a UK one’s.

    Common questions about the UKRIO Code

    What is UKRIO?

    UKRIO, the UK Research Integrity Office, is an independent advisory body supporting UK research organisations and researchers on integrity, ethics and misconduct. It does not fund research and holds no statutory regulatory powers; its principal output is guidance, including the Code of Practice for Research.

    What is the difference between the UKRIO Code and the Concordat to Support Research Integrity?

    The Concordat is the sector-wide policy agreement UK funders and universities formally sign; the UKRIO Code is a practical, operational document helping institutions meet the Concordat’s five commitments. Signing the Concordat is a funding condition; adopting the Code is not.

    Do universities have to follow the UKRIO Code of Practice for Research?

    No university is legally compelled to adopt the Code. Most UK research organisations use it — in full or as a benchmark — because it maps onto Concordat commitments that are a condition of UKRI funding, and gives institutions a defensible standard to cite if a misconduct case is scrutinised.

    What changed in the 2025 version of the UKRIO Code?

    Version 3.5, published 8 July 2025, added a section on responsible AI use, aligned the Code with the revised 2025 Concordat (recognising research-enabling staff), and rewrote misconduct language to reduce stigma around reporting concerns.

    What this means for institutional policy teams

    Three actions follow from the July 2025 update. First, benchmark governance policy against Version 3.5 specifically — the AI section and research-enabling-staff language are new obligations in substance, even if not in law. Second, use UKRIO’s Self-Assessment Tool to find gaps before a funder audit or misconduct case forces the issue. Third, treat the Code’s implied checklist as a floor, not a ceiling: institutions with mature research administration functions go further, especially on AI governance, where the Code is a starting reference rather than a complete framework.

    The Code’s non-statutory status is unlikely to change. UKRIO operates as an advisory body by design, and practical enforcement of research integrity in the UK runs through funder conditionality — via the Concordat and, increasingly, funding-linked exercises such as the REF — rather than direct regulation. Institutions treating Code adoption as funding-risk management, not box-ticking, are best placed for whatever the next revision requires.

  • How UKRIO Defines Research Misconduct: UK Guide

    UKRIO defines research misconduct as behaviours that deliberately or recklessly fall short of the standards expected in the conduct of research, from the initial idea through to reporting outcomes and peer review. This is a working, principles-based definition rather than a closed list of banned acts, and it is deliberately broader than the US Office of Research Integrity’s fabrication-falsification-plagiarism (FFP) standard.

    In UKRIO’s own words: research misconduct is “behaviours that deliberately or recklessly fall short of the standards expected in the conduct of research.” That single sentence, not a statute, is the operative UK working definition research offices build their procedures around.

    What is UKRIO’s working definition of research misconduct?

    UKRIO does not create a statutory definition of research misconduct. It adopts the definition set out in the Concordat to Support Research Integrity, published by Universities UK in 2019, under Commitment 4 and Annex A. UKRIO’s own guidance and its model Procedure for the Investigation of Misconduct in Research (Version 2, March 2023) state explicitly that the definition used throughout is “taken from the Concordat to Support Research Integrity.”

    The Concordat-derived definition requires two elements: a departure from expected standards, and either intent or recklessness behind it. UKRIO describes the boundary as “a spectrum rather than a hard and fast line” — a deliberate design choice that keeps the definition workable across every discipline UK research offices oversee, from clinical trials to computational modelling.

    Which behaviours does UKRIO’s definition capture?

    UKRIO’s published guidance lists specific behaviours that fall within its working definition. These are illustrative, not exhaustive, but they anchor most institutional investigations:

    • Fabrication — making up data, results, or other research information such as patient consent.
    • Falsification — manipulating or selecting processes, materials, equipment, or data to present a false impression or outcome.
    • Plagiarism — using others’ ideas, intellectual property, or work without permission or acknowledgement.
    • Misrepresenting data or other information in proposing, performing, or reporting research.
    • Breaching legal, ethical, or professional requirements — for example proceeding without the necessary ethics approval or informed consent for human participants, animals, or human tissue.
    • Proceeding without necessary permissions or approvals being in place.
    • Failing to declare or manage conflicts of interest appropriately.

    Because the definition covers “all stages in the research cycle,” it extends beyond data collection into peer review conduct and reporting — a scope UK research offices sometimes overlook when they draft misconduct policies focused narrowly on data integrity.

    How does UKRIO’s definition compare with ORI and UKRI Research England?

    Three UK/US frameworks use visibly different formulations, and the differences matter for cross-border collaborations and grant compliance:

    Body / framework Core formulation Scope Governing source
    UKRIO (adopting the Concordat) “Behaviours that deliberately or recklessly fall short of the standards expected in the conduct of research” Open, behaviour-based; covers the whole research cycle, including peer review Concordat to Support Research Integrity, Universities UK, 2019
    US Office of Research Integrity (ORI) Fabrication, falsification, or plagiarism (FFP) Closed, three-act list; scoped to Public Health Service (PHS)-funded research 42 CFR Part 93 (US federal regulation)
    UKRI / Research England (funder condition) Fabrication, falsification, plagiarism, “or other serious deviations from commonly accepted practices” Closed FFP core plus an open “serious deviations” clause UKRI grant terms and conditions

    ORI’s PHS-scoped FFP standard is the narrowest of the three because it is a statutory list tied to federal funding. UKRI’s funder-condition definition sits in the middle: it names FFP explicitly but adds an open “serious deviations” clause. UKRIO’s Concordat-based definition is the broadest of the three, because it is anchored to a standard of expected behaviour rather than to a closed list of acts — which is precisely why UK research offices need their own institutional Code of Practice to give it operational teeth.

    Where does UKRIO’s definition deliberately stay vague?

    UKRIO’s definition leaves at least three things unsettled by design, and understanding why matters as much as knowing the wording.

    First, UKRIO does not itself specify what “the standards expected” are. It states that these standards “are those that your organisation will normally have put in place either in a Code of Practice for research or research handbook” — pushing the substantive content of the definition down to each institution, funder, or learned society.

    Second, the Concordat draws a case-by-case line between research misconduct and questionable research practice (QRP), defined in Annex A (page 13) as “minor infractions or research practices, including avoidable errors, which fall short of the definition of intentional research misconduct.” Genuine errors and reasonable methodological disagreement, however strongly disputed, sit outside the misconduct definition entirely — but distinguishing an “avoidable error” from “reckless” conduct is a judgement made on the facts of each case, not a bright-line test.

    Third, UKRIO frames misconduct as “a spectrum rather than a hard and fast line,” which is workable across disciplines precisely because it resists codification into a fixed act list — the trade-off UK research offices absorb in exchange for that flexibility.

    What does this mean for UK research offices?

    Because UKRIO’s definition is principles-based rather than a closed statute, UK research offices carry three practical obligations. They must maintain their own Code of Practice for Research (or equivalent) that states, in terms specific to their disciplines, what “standards expected” means locally — the Concordat requires this under Commitment 4. They must document any departure between their institutional definition and the Concordat wording, since an open definition raises dispute risk if a respondent argues the standard applied was undefined or inconsistently applied. And investigation panels must still make a fact-specific finding on intent or recklessness for every allegation, rather than concluding misconduct from a poor outcome alone.

    UKRIO’s model Procedure for the Investigation of Misconduct in Research operationalises this: it imports the Concordat definition verbatim and sets out common features most UK institutional procedures share — an initial assessment, a panel investigation with external academic input, a written report, and an appeal stage. The definition and the procedure are designed to work together, not as separate standards.

    As UK research integrity oversight matures — with more institutions publishing named-officer processes and more funders cross-referencing the Concordat in grant conditions — the vagueness UKRIO deliberately preserves in the definition is likely to remain the deliberate trade-off for a system built on institutional self-regulation rather than statutory command.

    Common questions about research misconduct

    What are the three types of research misconduct?

    The three internationally recognised core acts are fabrication, falsification, and plagiarism (FFP) — the categories used by the US ORI. UKRIO recognises these same three acts as core examples but extends its working definition beyond them to cover ethics breaches, undeclared conflicts of interest, and unauthorised research activity.

    What is an example of research misconduct?

    A researcher inventing patient consent records, or manipulating raw data to produce a more favourable result, are both clear examples. Under UKRIO’s definition, proceeding with a human-participant study without the required ethical approval also counts as misconduct, even without fabricated data.

    What is research misconduct in the UK?

    In the UK, most institutions and funders anchor their definition to the Concordat to Support Research Integrity (Universities UK, 2019), which UKRIO also adopts. UKRI’s grant conditions separately define misconduct as FFP “or other serious deviations from commonly accepted practices.”

    What counts as research misconduct?

    Under UKRIO’s definition, conduct counts as misconduct only where it involves intent or recklessness and falls short of the standards an institution’s Code of Practice sets out. Genuine errors and honest methodological disagreements are explicitly excluded, however serious their consequences.

    For related standards work on contributor accountability and research governance, see CASRAI’s research administration resources and research integrity dictionary terms.