Scientific misconduct in research is not defined the same way twice: the UK Research Integrity Office (UKRIO) uses a broad, non-statutory standard covering fabrication, falsification, plagiarism and procedural breaches; the US Office of Research Integrity (ORI) applies a narrower federal test limited to fabrication, falsification and plagiarism (FFP); and COPE, the Committee on Publication Ethics, issues no single definition at all, instead guiding journal editors through case-by-case flowcharts. For multinational collaborations, this divergence means the same conduct can be misconduct under one regime and merely a “questionable research practice” under another.
Research misconduct, at its narrowest and most widely cited, is fabrication, falsification, or plagiarism committed intentionally, knowingly, or recklessly in proposing, performing, reviewing, or reporting research — the definition set out in US federal regulation at 42 CFR Part 93. Every wider or narrower national and institutional definition builds outward from, or diverges from, that FFP core.
- How does UKRIO define research misconduct?
- How does the US Office of Research Integrity define misconduct?
- How does COPE approach scientific misconduct?
- Where do the three definitions overlap and diverge?
- What does this mean for multinational collaborations?
- Frequently asked questions
- Outlook
How does UKRIO define research misconduct?
UKRIO’s Code of Practice for Research defines misconduct as behaviour that deliberately or recklessly falls short of the standards expected in the conduct of research. This is a materially broader standard than the US federal test, because it extends beyond fabrication, falsification and plagiarism to include:
- Misrepresentation of data, interests, or authorship contributions
- Breach of duty of care to research participants, animals, or the environment
- Improper handling of confidential information or peer-review material
- Failure to follow agreed protocols or obtain proper approvals
Crucially, UKRIO is an advisory body, not an investigator. Under the Concordat to Support Research Integrity (overseen by Universities UK since 2012 and last revised in 2019), responsibility for investigating an allegation sits with the employing institution — UKRIO’s Research Integrity Advisory Service, including its confidential helpline, provides independent guidance to institutions and individuals but has no power to adjudicate cases itself.
How does the US Office of Research Integrity define misconduct?
The ORI’s definition is narrower and legally codified. Under 42 CFR Part 93, research misconduct means fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results. Three conditions must all be met before a finding is made:
- The conduct represents a significant departure from accepted practices of the relevant research community
- It was committed intentionally, knowingly, or recklessly
- The allegation is proven by a preponderance of the evidence
ORI’s own guidance is explicit that honest error or differences of opinion are excluded from the definition entirely. Unlike UKRIO, ORI has direct oversight authority — but only over research funded by the US Public Health Service, principally the National Institutes of Health. Misconduct in research funded by other US federal agencies, such as the National Science Foundation, falls to that agency’s own Office of Inspector General, not ORI.
How does COPE approach scientific misconduct?
COPE, the Committee on Publication Ethics, is a membership body for journal editors and publishers, and it deliberately does not issue a single formal definition of research misconduct. Instead, its Core Practices and topic-specific flowcharts guide editors through individual cases of suspected fabrication, falsification, plagiarism, image manipulation, authorship disputes, and undeclared conflicts of interest.
This has a practical consequence: COPE’s guidance explicitly recognises a wider category of “questionable research practices” that may fall short of intentional misconduct but still compromise the reliability of the published record — for example, salami-slicing a dataset across multiple papers, or selective outcome reporting. COPE is not an investigative or regulatory body; it directs editors back to the authors’ institution when an allegation requires formal investigation.
Where do the three definitions overlap and diverge?
All three frameworks agree that fabrication, falsification, and plagiarism sit at the core of research misconduct. Where they diverge is scope, legal weight, and who has authority to act.
| Dimension | UKRIO | US ORI | COPE |
|---|---|---|---|
| Legal status | Non-statutory advisory body | Federal regulation (42 CFR Part 93) | Non-governmental membership body for journals |
| Definition scope | Broad — FFP plus misrepresentation, breach of duty of care, procedural failures | Narrow — FFP only | No single definition — FFP plus “questionable research practices” |
| Intent threshold | Deliberate or reckless departure from expected standards | Intentional, knowing, or reckless; significant departure; preponderance of evidence | Not codified — case-by-case editorial judgement via flowcharts |
| Who investigates | The employing institution; UKRIO advises only | ORI itself, for Public Health Service-funded research | The journal or publisher, with referral to the institution where needed |
| Jurisdiction | UK research organisations under the Concordat to Support Research Integrity | US Public Health Service-funded research (principally NIH) | Journals and editors globally holding COPE membership |
What does this mean for multinational collaborations?
Because none of the three bodies has exclusive jurisdiction, a single alleged incident on a collaborative, internationally co-funded project can trigger three separate, non-mutually-exclusive accountability pathways at once. A UK-US project with NIH co-funding and a journal output governed by a COPE-member publisher could face: an institutional investigation under UKRIO guidance for the UK partner, a formal ORI review for the US-funded component, and independent editorial action by the journal under COPE’s Core Practices — each applying a different definition, threshold, and evidentiary standard to the same facts.
For institutions managing multinational grants, the practical implication is that data management plans and collaboration agreements should specify in advance which misconduct framework governs which output, rather than assuming a single national standard will apply throughout the life of the project. Research administrators, in particular, should map funder-specific misconduct clauses against journal policies before submission, not after an allegation surfaces.
Frequently asked questions
What are the three types of scientific misconduct?
The three internationally recognised categories are fabrication (inventing data or results), falsification (manipulating materials, equipment, processes, or data so the research record is inaccurate), and plagiarism (using another person’s ideas, processes, results, or words without credit). All three national and publisher frameworks compared here treat FFP as the common core.
What is considered research misconduct?
Under the US federal definition, research misconduct is fabrication, falsification, or plagiarism in proposing, performing, reviewing, or reporting research, provided it is a significant, intentional or reckless departure from accepted practice. UKRIO’s broader UK definition additionally covers misrepresentation and breaches of duty of care.
What are the 5 unethical practices in research?
Beyond FFP, commonly cited unethical practices include failure to credit others’ contributions, undisclosed conflicts of interest, and biased design or interpretation driven by outside influence. COPE’s guidance groups many of these under “questionable research practices” rather than formal misconduct.
What are examples of scientific misconduct?
Documented examples include inventing experimental results never actually collected, altering images or statistical outputs to support a desired conclusion, and copying text or data from another author’s unpublished grant application or manuscript. Retraction Watch’s public database catalogues thousands of such cases resolved through the three pathways described above.
Outlook: toward interoperable standards
None of the three frameworks is converging toward a single global standard, and none is likely to. UKRIO’s advisory model, ORI’s statutory federal authority, and COPE’s editorial guidance each serve distinct institutional, governmental, and publishing constituencies with different legal powers. What is changing is the expectation that funders and publishers align their own misconduct clauses explicitly to one or more of these frameworks at the contracting stage, rather than leaving jurisdiction ambiguous until an allegation forces the question. For research administrators overseeing cross-border projects, treating “research misconduct” as a single settled term — rather than three overlapping regimes — remains the single most avoidable compliance error.