Research integrity and misconduct

Textual Plagiarism

Textual plagiarism is unattributed copying of another person's written words — as distinct from plagiarism of ideas, where an author paraphrases someone else's argument or findings thoroughly enough that no wording matches but the underlying contribution still goes uncredited. The U.S. Office of Research Integrity (ORI) uses this exact two-way split — plagiarism of text vs. plagiarism of ideas — in its own research-integrity guidance. A passage counts as an instance of textual plagiarism when matching or near-matching wording appears without both a citation to the source AND quotation marks (or block-quote formatting) marking exactly which words are borrowed; a citation alone, without quotation marks around verbatim language, is not sufficient. Because it is defined by the words on the page rather than by intent, textual plagiarism is the category similarity-checking software (Turnitin, iThenticate, Crossref Similarity Check) is built to surface — these tools compare submitted text against a reference corpus and report string-level matches. They cannot detect plagiarism of ideas, since a thoroughly reworded passage produces no textual match at all.

IACUC Adverse Event and Unanticipated Outcome Plan

An IACUC adverse event and unanticipated outcome plan is the written contingency component of an approved animal-use protocol specifying: (1) trigger criteria for what counts as an adverse event or unanticipated outcome for that study (pain, distress, injury, illness, or a clinical sign exceeding the approved humane endpoints); (2) who is authorized to act immediately, in practice the Attending Veterinarian or delegated veterinary staff, who can order treatment, analgesia, or euthanasia without prior investigator consent when welfare requires it; (3) a defined internal reporting pathway and timeframe from the observer to veterinary staff to the IACUC office; and (4) a documented outcome record distinguishing an isolated clinical event from a pattern indicating the protocol itself needs IACUC-reviewed modification. Institutions holding a PHS Animal Welfare Assurance must additionally report specific categories of event -- serious or continuing noncompliance, serious deviation from the Guide for the Care and Use of Laboratory Animals, or an IACUC-ordered suspension -- to OLAW under PHS Policy IV.F.3.

TCPS 2 Chapter 7 (Conflicts of Interest)

TCPS 2 (the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans), Chapter 7, "Conflicts of Interest," is the chapter of Canada's tri-agency human-research ethics policy that governs how researchers, research ethics boards (REBs) and institutions identify, disclose and manage conflicts of interest arising in human-participant research. A conflict of interest under Chapter 7 exists whenever a person's or institution's duties toward research participants and the research record could be compromised, or could reasonably appear to be compromised, by a competing personal, financial, professional or institutional interest. The chapter treats real, potential and perceived conflicts as equally disclosable -- a conflict does not need to have actually influenced a decision to trigger Chapter 7's disclosure and management requirements, only to create a reasonable appearance that it might. Chapter 7 applies at three distinct levels: individual conflicts (a researcher's own competing interests), institutional conflicts (the institution's own financial or other stake in a study's outcome), and REB-member conflicts (a board member with a personal or professional stake in a protocol under that board's review).

IRB Approval and Expiration of Recruitment Materials

IRB approval and expiration of recruitment materials refers to the fact that flyers, advertisements, social-media posts, phone/email scripts, screening questionnaires, and other materials used to identify and approach prospective participants are themselves reviewed and approved by the IRB as part of the informed-consent process -- not treated as separate marketing collateral -- and that approval carries its own expiration date, tied to the study's current approval period, independent of whether the underlying protocol itself continues. A recruitment or screening document is an approved recruitment material only for the interval the IRB has authorized; once that interval lapses, the material cannot be used to recruit or screen new participants even if the protocol has not been suspended or terminated and even if continuing review is only pending, not yet complete.

AAALAC Accreditation Requirements

The specific set of standards and recurring obligations an institution must satisfy to achieve and then keep AAALAC International accreditation: evaluating its animal care and use program against AAALAC's Three Primary Standards (principally the Guide for the Care and Use of Laboratory Animals), submitting a detailed Program Description and hosting an on-site peer review before initial accreditation, then filing an Annual Report every year and undergoing a full site visit roughly every three years to keep that status.

Program of Veterinary Care (PVC)

A Program of Veterinary Care (PVC) is the documented set of policies, procedures, and staffing arrangements through which an institution's Attending Veterinarian provides adequate veterinary care to animals used in research, teaching, and testing. It is a required component of an animal-care-and-use program under Animal Welfare Act regulations (9 CFR 2.33) and, in parallel, under PHS Policy for PHS-supported vertebrate-animal research. A program qualifies as an adequate PVC only if it addresses, at minimum: preventive medicine (quarantine, disease surveillance, herd/colony health monitoring, vaccination and parasite control where applicable); appropriate methods for preventing, controlling, diagnosing, and treating disease and injury; guidance and oversight of surgery (pre-, intra-, and post-operative care) and of anesthesia and analgesia use, including a mechanism for the Attending Veterinarian to require or adjust pain relief; a means of communication so personnel promptly report animal health concerns; and 24-hour availability of emergency veterinary care. Where the Attending Veterinarian serves in a part-time or consultant capacity rather than full-time employment, 9 CFR 2.33 requires the arrangement to be formalized through a written program of veterinary care and regularly scheduled site visits, not an informal or on-call-only relationship.

TCPS 2 Article 3.1 (Free and Informed Consent)

TCPS 2 (the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans, 2022 edition) Article 3.1, in Chapter 3 ('The Consent Process'), sets out Canada's core requirement that participation in research involving human participants be based on free, informed, and ongoing consent. A consent process satisfies Article 3.1 when: (1) it is FREE -- given voluntarily, without manipulation, undue influence, or coercion; (2) it is INFORMED -- the prospective participant has enough understandable information about the research's purpose, procedures, reasonably foreseeable risks and potential benefits to make a genuine choice; and (3) it is ONGOING -- treated as a continuing process rather than a single signature, with researchers under a standing obligation to disclose material changes in risk or benefit as the study proceeds, and to reconfirm consent or offer withdrawal where warranted. Research Ethics Boards (REBs) are directed to apply particular scrutiny to trust or dependency relationships (for example, physician-patient or professor-student) where the relationship itself could create undue influence on a prospective participant's decision to take part.

Tri-Agency Framework: Responsible Conduct of Research (RCR)

The Tri-Agency Framework: Responsible Conduct of Research is the joint policy of Canada's three federal research funding agencies -- the Canadian Institutes of Health Research (CIHR), the Natural Sciences and Engineering Research Council of Canada (NSERC), and the Social Sciences and Humanities Research Council of Canada (SSHRC) -- that defines researcher, institution, and Agency responsibilities for research integrity, sets minimum requirements for institutional RCR policies, defines what constitutes a breach of Agency policies, and specifies the process the Agencies follow when addressing allegations of a breach. It applies to research funded in whole or in part by any of the three Agencies, and is administered jointly by the Secretariat on Responsible Conduct of Research (SRCR) and the Interagency Advisory Panel on Responsible Conduct of Research (PRCR). The current edition took effect in 2021, replacing the 2016 version.

TCPS 2 Article 2.1 (REB Review Requirement)

Article 2.1 of Canada's Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (TCPS 2) is the threshold rule that triggers Research Ethics Board (REB) review: research involving living human participants, or research involving human biological materials (including embryos, fetuses, fetal tissue, reproductive materials, and stem cells related to human reproduction), must undergo REB review and receive REB approval before the research commences, and that approval must be maintained for the duration of the research. It is the operational test institutions apply to decide whether a proposed project needs to go to an REB at all -- if a project meets the Article 2.1 description and no Article 2.2-2.4 exception applies, REB review is mandatory; if it does not meet the description (for example, because it is not 'research' as TCPS 2 defines the term, such as certain quality-assurance, quality-improvement, or program-evaluation activities), REB review under this article is not required.

IEEE Plagiarism Policy and Prohibited Authors List (PAL)

IEEE (the Institute of Electrical and Electronics Engineers) investigates plagiarism complaints against its journals, transactions, and conference papers using a published five-level severity scale (Level One through Level Five) set out in IEEE's PSPB Operations Manual and its accompanying 'Plagiarism Levels and Corrective Actions' guidance. The scale grades a case by extent of uncredited or improperly credited verbatim copying, from Level One (an entire paper, or more than 50% of one, copied without credit, or verbatim reuse spanning more than one paper by the same author) down to Level Five (verbatim text copied from a credited source but not clearly marked as a direct quotation). Corrective actions scale with the level and with repeat-offense history, ranging from a required letter of apology through a suspension of an author's submission privileges to specific IEEE publications for a fixed period -- commonly cited ranges run from about one to five years -- up to a lifetime prohibition on publishing in any IEEE-copyrighted venue for the most serious or repeated Level One violations, administered at the discretion of the IEEE Publication Services and Products Board (PSPB) Chair. An author under an active submission suspension is added to IEEE's internal screening list -- long known as the Prohibited Authors List (PAL) and, per more recent PSPB Operations Manual language, formally relabeled the PPL -- which IEEE publication volunteers (editors, program chairs) check via an internal PAL Database tool before accepting a submission, to confirm a listed name is not attempting to submit to a venue it is currently barred from.

IEEE Conflict of Interest Disclosure

IEEE conflict of interest (COI) disclosure refers to the author- and reviewer-facing conflict-of-interest rules IEEE applies across its journals, transactions, magazines, and conferences, set out through IEEE Author Center guidance and the editorial policies of individual IEEE technical societies. It is distinct from a generic conflict-of-interest disclosure in two ways specific to IEEE's editorial process: IEEE defines a conflict along three tiers -- actual, perceived, and potential -- rather than a single financial-interest dollar threshold, and IEEE devolves the operational specifics (what a reviewer must disclose, how far back a prior collaboration counts, how long a mentorship cooling-off period lasts) to the sponsoring technical society or specific journal rather than fixing one uniform disclosure form across all of IEEE, the same way the specific mechanics of IEEE's peer-review model are set at the society level rather than uniformly.

Concordat to Support Research Integrity

The UK's sector-wide research integrity framework document, originally published by Universities UK in 2012, revised in 2019, and refreshed in 2025, that individual UK universities, funders, and sector bodies formally sign up to as named signatories. A signatory organisation commits to five thematic commitments and is expected to publish an annual public statement evidencing how it has acted on each one.

HSE Research Ethics Committee

An HSE Research Ethics Committee (REC) is one of the independent ethics review bodies operating within Ireland's Health Service Executive (HSE) and HSE-funded hospitals to provide ethical approval for health research conducted in the Irish public health system. HSE/Hospital RECs review studies that recruit participants through publicly funded health services, use personal healthcare data or biological samples from service users, involve health and social care staff as participants or researchers, or are hosted within publicly funded healthcare infrastructure. They are distinct from Ireland's National Office for Research Ethics Committees (NREC), a separate statutory body under the Department of Health that provides single national ethical approval only for regulated clinical trials of investigational medicinal products, regulated medical device investigations, and regulated performance studies of in-vitro diagnostic devices. Research falling outside NREC's narrower statutory scope -- including most observational, data-driven, and non-regulated clinical research -- is reviewed by an HSE or Hospital REC instead. The HSE is currently reforming this system, moving from a fragmented set of local committees toward six regional HSE Reference RECs aligned with Sláintecare Regional Health Areas, each operating to a Standard Code of Governance and Management and supported by the HSE Research Ethics Committees Support and Coordination Office, which reports to the National Committee for the Governance, Management and Support of Research.

NHMRC HREC Registration

NHMRC HREC registration is the process by which an Australian institution formally registers its Human Research Ethics Committee (HREC) with the National Health and Medical Research Council (NHMRC), submitting the committee's membership list, Terms of Reference, and Standard Operating Procedures for review against the composition and operating requirements set out in Section 5 of the National Statement on Ethical Conduct in Human Research. Registration is what makes an institution's HREC eligible to review NHMRC-funded research and is a prerequisite for -- though distinct from -- separate certification under NHMRC's National Certification Scheme for the Ethics Review of Multi-centre Research.

NIH Guidelines for Human Stem Cell Research

The NIH Guidelines for Human Stem Cell Research is the federal oversight framework, issued by the National Institutes of Health in 2009 (following Executive Order 13505) and amended in 2010, that governs whether federally funded research may use a given human embryonic stem cell (hESC) line. A research activity falls under the Guidelines whenever a US institution seeks NIH funding for work involving hESCs derived from embryos donated after in vitro fertilization (IVF) treatment. Under the Guidelines, two conditions must both be satisfied before federal funds may be used: (1) the specific hESC line must be listed on the NIH Human Pluripotent Stem Cell Registry (hPSCR), which NIH adds a line to only after confirming the embryo was donated voluntarily, with specific informed consent, by IVF patients whose decision to donate was kept separate from their clinical fertility care and was not induced by payment; and (2) many institutions additionally require the protocol to clear review by an institutionally convened Embryonic Stem Cell Research Oversight (ESCRO) committee before the work proceeds, a review that is separate from and in addition to standard Institutional Review Board (IRB) review of the original donor-consent process. The Guidelines govern use of already-derived hESC lines; they do not authorize federal funding of embryo derivation itself, which remains separately restricted by the Dickey-Wicker Amendment, an annual appropriations rider unaffected by the 2009 Guidelines.

Vertebrate Animals Section (VAS)

The Vertebrate Animals Section (VAS) is the required narrative component of an NIH grant application or cooperative-agreement proposal covering any proposed use of live vertebrate animals. Per NIH guidance (originally NOT-OD-10-027, simplified by NOT-OD-16-006), it must describe the proposed procedures and species/strains/ages/sex/numbers; justify the choice of species and why a non-animal alternative cannot achieve the research goals; describe interventions minimizing pain and distress, including analgesia, anesthesia, and humane endpoints; and state whether the planned euthanasia method is consistent with the AVMA Guidelines for the Euthanasia of Animals. It is evaluated by NIH Scientific Review Groups as part of merit review and is distinct from, and does not substitute for, the institution’s own IACUC protocol approval and PHS Assurance.

IEEE Double-Blind Peer Review

IEEE's double-blind (also called double-anonymized) peer review is a submission-review model, adopted by specific IEEE journals, transactions, and conferences rather than applied IEEE-wide, in which the manuscript sent to reviewers has been stripped of author names, affiliations, acknowledgments, funding sources, and other directly identifying material, and reviewer identities are withheld from authors throughout review. A submission counts as following IEEE double-blind policy only if the venue itself has formally adopted the model (checkable in that publication's author guidelines) and the author has met its specific anonymization requirements at submission -- most IEEE publications instead use IEEE's default single-anonymized model, where the author's identity is visible to reviewers but not vice versa, so double-blind status is never assumed from the IEEE name alone.

Guest Editor Vetting

The process a journal or publisher uses to confirm that an individual invited to serve as a guest editor for a themed special issue actually holds the institutional affiliation, independent publication record, and editorial independence they claim, before that person is given authority to invite submissions, select peer reviewers, and recommend accept/reject decisions. Vetting is present only when a publisher independently confirms these facts (for example against ORCID, institutional directories, or a prior publication history in the field) rather than accepting a guest editor's self-reported credentials or an unsolicited nomination at face value.

NISO CREC (Communication of Retractions, Removals, and Expressions of Concern)

NISO CREC refers to NISO RP-45-2024, the Communication of Retractions, Removals, and Expressions of Concern Recommended Practice published by the National Information Standards Organization (NISO) on June 27, 2024 (DOI 10.3789/niso-rp-45-2024). CREC is not a new editorial rule about *when* something should be retracted — that judgment remains the domain of editorial-ethics guidance such as COPE's Retraction Guidelines. It is a communication standard about *how* a retraction, removal, or expression of concern (EoC), once decided, gets represented consistently in metadata and on-page display so that both human readers and machine systems (indexers, citation databases, reference managers, library discovery layers, aggregators) can detect and propagate the status automatically, rather than relying on someone noticing a notice.A publisher, platform, or content system is applying CREC when it implements RP-45-2024's paired recommendations: (1) metadata — structured, machine-readable status elements attached to the item's record (classified in the Recommended Practice as Essential, Essential if Available, or Recommended), covering the original publication, the retraction/removal/EoC notice itself, dates, and reasons; and (2) display — visible, human-facing signals, including prepending “RETRACTED:” to the title, watermarking and labelling the content in both HTML and PDF renderings, and linking a distinct, easily located notice to the affected item rather than burying it in an unrelated editorial page.

Hijacked journal

A journal is operating as a hijacked journal when a website solicits and accepts manuscripts, typically for a fee, under the name, ISSN, and/or editorial identity of a specific real, previously or currently legitimate journal, without authorization from that journal's actual publisher or editorial board. The operational test is genuine continuity with, or authorization from, the real publisher of record, checkable against the ISSN Portal, the journal's own site, or its listing in the index it claims.

Peer Review Fraud

The deliberate corruption of the manuscript peer-review process to obtain a favorable or falsely legitimate-looking editorial decision, via fabricated reviewer identities, reciprocal peer-review rings, or coercive demands unrelated to manuscript merit. Distinguished from good-faith reviewer error, ordinary editorial disagreement, or unconscious reviewer bias by the presence of deliberate intent to subvert the review process itself.

Acceptable Similarity Percentage

There is no single, official percentage that marks a similarity/originality report as acceptable or unacceptable. Turnitin, Crossref Similarity Check, and iThenticate all publish a numeric similarity index, but none of the vendors, COPE, or ORI define a universal cutoff -- individual institutions, journals, and graduate schools sometimes set an informal internal benchmark (commonly cited figures cluster around 15% or 20%) purely as a triage trigger for closer human review, not as a formal misconduct threshold. What actually determines whether a score is a problem is what the flagged text consists of: properly quoted and cited material, a shared reference list or bibliography, standard field terminology, and (for theses) a candidate's own previously deposited coursework all inflate the percentage without indicating misconduct, so a report has to be opened and read, not just scored, before any percentage is meaningful.

Self-Citation

Citing one's own previously published work (an article, dataset, preprint, or other output) within a new manuscript. Self-citation is legitimate, and often necessary, when it substantively supports the current work — building on prior results, pointing to a previously validated method instead of repeating it, or situating a study within a sustained research programme. It becomes problematic — citation manipulation rather than ordinary scholarly practice — when self-citations are added primarily to inflate personal or journal-level citation metrics rather than because the cited work is genuinely relevant, including reciprocal citation-stacking arrangements between authors or journals.

MDPI (Multidisciplinary Digital Publishing Institute)

MDPI (Multidisciplinary Digital Publishing Institute) is a Basel, Switzerland-headquartered academic publisher, founded in 1996, that publishes exclusively open access: every article it accepts is made freely readable on publication, funded by an author-side Article Processing Charge (APC) rather than reader subscriptions. MDPI operates several hundred peer-reviewed journals across the sciences, social sciences, and humanities -- examples include International Journal of Molecular Sciences (IJMS), Sustainability, and Nutrients -- structured heavily around themed 'Special Issues' with guest editors, and is known for comparatively fast median times from submission to first editorial decision and to publication relative to many subscription journals. MDPI is a member of the Committee on Publication Ethics (COPE) and the Open Access Scholarly Publishing Association (OASPA). It is a real, currently operating publisher, not a defunct or fictitious one -- but the label 'MDPI journal' does not, by itself, tell you the editorial rigor of any specific title, which varies journal by journal and has been the subject of genuine, documented scrutiny from indexing bodies and parts of the research community (see below).

ICH E9 (Statistical Principles for Clinical Trials)

ICH E9, "Statistical Principles for Clinical Trials," is the International Council for Harmonisation guideline (finalized 1998) that sets out the core statistical methodology expected of a confirmatory clinical trial submitted to a regulatory authority. A trial's design and statistical analysis plan (SAP) are ICH E9-consistent when they are pre-specified before unblinding and address, at minimum: the trial's overall design and hypothesis structure (superiority, non-inferiority, or equivalence, each with its own margin and testing logic); the randomization and blinding scheme; the sample-size/power justification; the choice of analysis populations, most centrally the Full Analysis Set (FAS, an operationalization of the intention-to-treat principle) and the Per-Protocol Set (PPS); the handling of missing data; and the control of Type I error across multiple endpoints, interim looks, or comparisons (multiplicity). ICH E9 was substantially extended, not replaced, by its 2019 addendum ICH E9(R1), which introduced the estimand framework to address a gap the original guideline left largely implicit: how a trial's statistical objective should be translated into a precisely defined treatment-effect question before an analysis method is chosen to answer it.

Generative AI use and disclosure· Contribution→

NSF AI Policy

NSF AI policy is not a single rule but a set of related National Science Foundation positions on artificial intelligence that a proposer or awardee needs to track separately: (1) guidance, currently framed as encouraged rather than mandatory, on disclosing generative-AI use in preparing a proposal; (2) an explicit statement that fabrication, falsification, or plagiarism committed with the assistance of AI-based tools counts as NSF research misconduct under the Proposal & Award Policies & Procedures Guide (PAPPG); and (3) NSF's own AI research-infrastructure investments, chiefly the National AI Research Resource (NAIRR) and the NSF AI Institutes program, which are funding/access policy rather than proposal-conduct policy. Treat these as three distinct tracks that happen to share the same agency and the same underlying 2023-2025 policy push, not one unified 'AI rule.'

Appearance of a Conflict of Interest

An appearance (or perceived) conflict of interest exists when a reasonable, informed outside observer could conclude that a researcher's, reviewer's, or official's outside relationships or interests might improperly influence their professional judgment -- even though, on closer examination, no actual conflict is present (there is no real financial or professional stake in the specific decision, or the relationship in fact has no bearing on it). It is judged from the observer's vantage point, not the individual's own confidence in their objectivity, which is precisely why most institutional, funder, and journal conflict-of-interest policies require an appearance-level situation to be disclosed and managed on essentially the same terms as an actual conflict: unmanaged appearances erode trust in research even when no bias actually occurred.

Attending Veterinarian (AV)

The Attending Veterinarian (AV) is the credentialed veterinarian (DVM/VMD) with direct or delegated authority and responsibility for the health and well-being of all research, teaching, and testing animals at an institution. Required under the Animal Welfare Act (9 CFR 2.33) and PHS Policy, the AV is a voting IACUC member and holds documented authority to provide or order veterinary treatment, require appropriate anesthesia/analgesia, and suspend a procedure or study causing unrelieved animal pain or distress.

Retraction Statement

The formal notice a journal or publisher issues to retract a published article — a standalone document, not the article itself, that identifies the article, states who initiated the retraction and why, and is bidirectionally linked to the retracted paper. A publication counts as a retraction statement only when the journal publishes it under its own editorial authority and explicitly labels it as a retraction, distinct from a correction, erratum, or expression of concern.

Exception from Informed Consent (EFIC) for Emergency Research

EFIC (Exception from Informed Consent) is the regulatory pathway under 21 CFR 50.24 that permits FDA-regulated clinical investigations of a drug, biologic, or device to enroll subjects in a life-threatening emergency without prospective informed consent from the subject or a legally authorized representative (LAR). A study qualifies as EFIC only when all of the following are true: (1) subjects are in a life-threatening situation for which available treatments are unproven or unsatisfactory; (2) obtaining informed consent is not feasible because the condition requires intervention before consent from the subject or LAR can reasonably be obtained; (3) participation holds the prospect of direct benefit to the subject; (4) the research could not practicably be carried out without the waiver; and (5) additional protections are in place, most notably community consultation and public disclosure conducted before the study begins, and an independent data monitoring committee. IRB approval of the full protocol, including the community consultation and disclosure plan, is required before enrollment can start under the exception.

OLAW (Office of Laboratory Animal Welfare)

OLAW is the office within NIH's Office of the Director responsible for implementing and interpreting the PHS Policy on Humane Care and Use of Laboratory Animals: it negotiates and approves each institution's Animal Welfare Assurance, evaluates reports of noncompliance from Assured institutions, and runs a nationwide PHS Policy education program. It governs any live vertebrate animal used in PHS-supported (including NIH-funded) research, testing, or training. OLAW does not replace an institution's own IACUC, and it does not conduct the on-site facility inspections that USDA/APHIS carries out under the Animal Welfare Act -- OLAW's authority runs through the PHS funding relationship, not independent statutory inspection power.

PRIM&R (Public Responsibility in Medicine and Research)

PRIM&R (Public Responsibility in Medicine and Research) is a US-based 501(c)(3) nonprofit membership organization, founded in 1974 and headquartered in Boston, Massachusetts, that supports the ethical conduct of research involving human participants and animals. An organization or individual falls within PRIM&R's operational scope when their work touches research oversight -- Institutional Review Boards (IRBs) reviewing human-subjects research, or Institutional Animal Care and Use Committees (IACUCs) overseeing animal research -- rather than the conduct of a specific study itself. PRIM&R does not review protocols or approve studies; it trains and credentials the professionals and administrators who staff the committees that do.

Ethical Approval

Ethical approval is the prospective, documented authorization a research ethics body issues before a study involving human participants (or, in a parallel but separate system, live animals) may begin — confirming the protocol's risk-benefit balance, consent process, and participant protections meet the applicable regulatory and ethical standard. It is granted by whichever committee has jurisdiction in a given country, which is why the same underlying requirement is described using different national names.

Questionable research practices

A questionable research practice (QRP) is a design, analysis, or reporting choice -- such as p-hacking, HARKing, selective outcome reporting, or salami-slicing -- made after seeing data or results and then presented as if it had been planned in advance, or otherwise used to manufacture a favourable result, without inventing or altering the underlying data itself. QRPs are distinct from research misconduct (fabrication, falsification, plagiarism -- FFP), which does involve inventing or altering data or appropriating others' work; QRPs sit below that legal/regulatory threshold but are still recognised as damaging to research integrity.

Ag Guide (Guide for the Care and Use of Agricultural Animals in Research and Teaching)

An institution follows the Ag Guide (Guide for the Care and Use of Agricultural Animals in Research and Teaching, FASS, 4th edition 2020) when agricultural animals -- livestock, poultry, and other food/fiber species -- are used in research or teaching whose purpose is agricultural (improving nutrition, breeding, management, production efficiency, or food/fiber quality). The determining factor is the purpose and context of the work, not the species alone: the same species used instead for biomedical purposes (e.g., a pig as a cardiovascular-device model) is typically reviewed under the Guide for the Care and Use of Laboratory Animals rather than the Ag Guide.

HREC (Human Research Ethics Committee)

The Australian term for the committee that reviews and approves research involving human participants before it may lawfully or ethically proceed. An HREC is constituted and registered under the National Health and Medical Research Council (NHMRC)'s National Statement on Ethical Conduct in Human Research, and its review is the Australian institutional counterpart to an IRB (Institutional Review Board) in the United States or an REC (Research Ethics Committee) in the UK/EU. A study is subject to HREC review whenever it involves human participants, their identifiable data, or their tissue and is designed to produce generalisable knowledge; the depth of that review is graded by risk rather than uniform for every protocol.

iThenticate

iThenticate is a text-matching (similarity-checking) service, owned and operated by Turnitin, that is marketed to publishers, journal editorial offices, funding agencies, and research institutions rather than to classroom instructors. A document qualifies as having been "run through iThenticate" when it has been submitted to the service's own web interface or via an API/editorial-management integration, compared against iThenticate's reference corpus (current and archived web content, previously submitted documents, and licensed academic publications, journals, and periodicals), and returned as a similarity report with a percentage-based similarity index and a source-by-source match breakdown. iThenticate is the same underlying detection engine Turnitin uses for its education product, packaged and licensed separately for non-classroom, professional research contexts: manuscript submission, grant proposal screening, and pre-publication or pre-submission self-checks. Crossref Similarity Check is a related but distinct service that Crossref operates on top of the iThenticate engine for its member publishers; a document checked through Crossref Similarity Check is, technically, still being matched by iThenticate's underlying engine, just accessed through Crossref's integration rather than a direct iThenticate license.

3Rs (Replacement, Reduction, Refinement)

The 3Rs -- Replacement, Reduction, and Refinement -- are the founding ethical framework for the humane use of animals in research. A protocol satisfies the 3Rs when it documents, for the specific study: (1) Replacement -- whether a non-animal or less-sentient alternative (in vitro, in silico, or lower-order organism) could answer the scientific question instead; (2) Reduction -- that the number of animals proposed is the minimum needed for a statistically and scientifically valid result, no more; and (3) Refinement -- that procedures, housing, and endpoints are designed to minimize pain, distress, and suffering for the animals that are still used. The 3Rs are not a single compliance checkbox but a standing analytic lens applied at every protocol-review stage by the Institutional Animal Care and Use Committee (IACUC).

APA Authorship Guidelines

APA authorship guidelines are the authorship-determination rules set out in Standard 8.12 ("Publication Credit") of the American Psychological Association's Ethical Principles of Psychologists and Code of Conduct. They govern two separate questions: who qualifies as an author at all (Standard 8.12a: only someone who has actually performed the work or substantially contributed to it — formulating the problem or hypothesis, structuring the experimental design, conducting the statistical analysis, or interpreting the results), and how author order and credit should be assigned once the author list is settled (Standard 8.12b: order must accurately reflect each person's relative scientific or professional contribution, never institutional rank or seniority alone). A separate provision, Standard 8.12c, addresses student work specifically: except in exceptional circumstances, a student is listed as principal (first) author on any multi-authored article substantially based on that student's doctoral dissertation. The guidelines apply within the ethical code psychologists are bound by professionally — they are not a citation-formatting rule, and are distinct from the APA Publication Manual's separate reference-list and in-text citation conventions.

Generative AI use and disclosure· Contribution→

ACM Policy on Authorship

The ACM Policy on Authorship is the Association for Computing Machinery's publisher-wide rule set, applied across its journals, transactions, magazines, and 70+ conference proceedings, defining who may be listed as an author on an ACM-published work, prohibiting generative AI tools from being listed as authors, and requiring authors to disclose any use of generative AI tools and technologies in producing the work — with a narrow exemption for basic word-processing aids such as spelling and grammar checkers.

Simultaneous Submission

Simultaneous submission (also called multiple or concurrent submission) occurs when an author submits the same manuscript, or manuscripts substantively reporting the same data or findings, to more than one journal at the same time, so that it is under active review at two or more venues simultaneously. It is prohibited by virtually all journal policies and by ICMJE's Recommendations, which state that authors 'should not submit the same manuscript, in the same or different languages, simultaneously to more than one journal.' A manuscript counts as under simultaneous submission from the point it is submitted to a second journal until it is formally withdrawn from, or rejected by, all but one of the venues. It is distinct from duplicate publication (also called redundant or self-plagiarized publication), which concerns a manuscript that substantially overlaps with work that has already been published, rather than work that is merely under review elsewhere at the same time.

Deferential Vulnerability (Research Ethics)

A type of research-subject vulnerability, identified in Kenneth Kipnis's taxonomy commissioned for the U.S. National Bioethics Advisory Commission (NBAC, 2000/2001), arising from a hierarchical or authority relationship between a prospective subject and the person recruiting or conducting the research -- formal or, more often, informal (e.g., professor/student, employer/employee, treating physician/patient, military superior/subordinate) -- in which the subject may feel unable to freely decline or withdraw from participation, whether from fear of offending the authority figure and incurring retribution or from a genuine desire to please a respected other. It is distinguished from juridic vulnerability (formal legal authority, as with incarceration) by the informality of the compelling relationship, and from population-based vulnerability categories (age, pregnancy, incarceration) by being relational and situational rather than a fixed characteristic of the subject.

Research Integrity Officer (RIO)

The institutional official a research institution formally designates, per 42 CFR § 93.233, to administer its written research-misconduct policy: receiving allegations, conducting the initial assessment of whether they meet the fabrication/falsification/plagiarism definition and fall within institutional jurisdiction, deciding whether to open a formal inquiry, sequestering relevant research records, and appointing inquiry/investigation committees. Institutions that apply for or receive PHS support (NIH, CDC, FDA, and other Public Health Service funding) must designate an RIO as a condition of that support.

Retraction vs. Redaction

Retraction and redaction are frequently confused because both involve a publisher acting on already-published content, but they answer different questions and have different consequences for the reader. A RETRACTION withdraws an entire published work from the reliable literature: the journal or publisher determines the findings are unreliable (due to error, misconduct, duplication, or other invalidating factors), issues a public retraction notice explaining why, and the article is prominently marked as retracted while typically remaining accessible for the scholarly record (per COPE's Retraction Guidelines). A REDACTION removes or obscures specific sensitive content within a document -- personal data, classified or export-controlled information, or other legally protected material -- while the document as a whole remains published and in circulation, usually with a note indicating that specific content has been withheld and why. Retraction is a judgment about the validity or integrity of the whole work; redaction is a targeted removal of particular information from an otherwise valid, still-published work. The two are not mutually exclusive: a paper can, in principle, be both redacted (to remove sensitive data) and later retracted (if a separate integrity issue emerges), but neither implies the other.

Cabells Predatory Reports

Cabells Predatory Reports (formerly Cabells' Blacklist, renamed in June 2020) is a commercial subscription database, published by the scholarly-analytics company Cabells, that lists journals its review team has determined exhibit deceptive or predatory editorial and publishing practices. A journal's inclusion is determined by evaluating it against a published, versioned set of behavioral indicators — v1.0 (2017) contained 64 criteria and v1.1 (2019) expanded to 74, organized by category (integrity, peer review, business practices, indexing claims, and similar) and by severity — rather than by a single reviewer's informal judgment. Unlike Beall's List, which is a static snapshot frozen in January 2017, Predatory Reports is actively maintained: journals are added, and the underlying criteria themselves have been revised, on an ongoing basis. Access is by paid subscription, typically arranged at the institutional level through a university library, either as a standalone product or bundled with Cabells' companion Journalytics subscription.

Good Publication Practice (GPP)

Good Publication Practice (GPP) is a named, industry-authored set of ethics guidelines for planning, writing, and publishing company-sponsored biomedical research -- primarily pharmaceutical, biotechnology, and medical-device studies. It is maintained by a GPP Steering Committee working with ISMPP (the International Society for Medical Publication Professionals) and published as a peer-reviewed article in Annals of Internal Medicine. The current version is GPP 2022, which superseded GPP3 (2015); GPP3 itself superseded the original GPP2 (2009) and GPP (2003). A study, manuscript, or publication process is GPP-compliant when: authorship is assigned strictly against ICMJE's four authorship criteria (not honorary, guest, or ghost authorship); the funding source and the sponsor's role in study design, analysis, and the decision to publish are disclosed; any professional medical writer's involvement is named and their support disclosed, typically in the acknowledgements; named authors have full access to the data (or a summary sufficient to vouch for it) and control over interpretation and the final decision to publish; and there is a documented intent to publish results regardless of outcome, including negative or inconclusive findings.

Editorial Independence

Editorial independence is the principle, formalized by COPE (Committee on Publication Ethics), that a journal's editorial decisions -- to accept, reject, or request revision of a manuscript -- are made solely on scholarly merit and are insulated from influence by the journal's owner, publisher, advertisers, sponsors, or any other commercial or non-scientific interest. It requires that editors hold final, unoverridable authority over content decisions, that advertising/sponsorship functions operate organizationally separate from editorial functions, and that any relationship between the journal and a commercial or sponsoring party is governed by a declared, public policy rather than an informal arrangement.

European Code of Conduct for Research Integrity

The European Code of Conduct for Research Integrity is a specific, named document — not a general principle or a professional network — published by ALLEA (All European Academies), the federation of European academies of sciences and humanities. First issued in 2011 (jointly with the European Science Foundation), it was revised in 2017 and again in 2023. It sets out four principles — reliability, honesty, respect, and accountability — and translates them into good research practices covering research design, data management, methodology, authorship, publication, and peer review, alongside a taxonomy of research misconduct and other research integrity violations. A document or practice is 'covered' by the Code when it is cited as the reference standard in a funding agreement, institutional policy, or national code — most consequentially, the European Commission recognises the Code as the primary research-integrity standard for Horizon Europe, and it is incorporated by reference in the Horizon Europe Model Grant Agreement, making adherence a binding condition of EU research funding rather than a voluntary aspiration.

Knowledge equity, diversity, global-south inclusion· Assessment→

Beall’s List

Beall's List refers to the self-published blacklist of publishers and standalone journals that University of Colorado Denver librarian Jeffrey Beall judged to be "potential, possible, or probable" predatory open-access operations, posted on his blog Scholarly Open Access between 2008 and January 15, 2017. It was a single-author evaluation against Beall's own evolving criteria (fake or absent peer review, fabricated editorial boards, false indexing claims, deceptive fee disclosure, and similar red flags), not a criteria-based process run by a standards body, publisher coalition, or institution. Beall removed the entire site in January 2017 and it has not been updated since, so any reference to "Beall's List" today points to a frozen, decade-plus-old snapshot (or to one of several unofficial community mirrors of it) rather than to a live, current resource.

Reproducibility and computational research· Data & methods→

Open Science Badges (Open Data, Open Materials, Preregistered)

The Open Science Badges are a set of journal-awarded visual icons, created and administered by the Center for Open Science (COS), that certify a published article met a disclosed openness standard for one of three practices: publicly posting the study's underlying data (Open Data), publicly posting the materials needed to reproduce the reported procedure and analysis (Open Materials), or registering the study design -- and, for the Preregistered+Analysis Plan variant, the analysis plan -- in a public registry before data collection began (Preregistered). A journal awards a badge either on a simple author self-disclosure statement or after independent editorial verification, and displays the badge icon alongside the article together with a link to the disclosed data, materials, or registration.

Research Animal

An animal that is used, or intended for use, in research, research training, testing, or experimentation, and that therefore falls under one or both of two overlapping US regulatory frameworks: the Animal Welfare Act (which reaches warm-blooded species other than purpose-bred rats, mice, and birds) and PHS Policy on Humane Care and Use of Laboratory Animals (which reaches any live vertebrate used in PHS/NIH-funded work, including species the AWA excludes). Whether a specific animal is a "research animal" for compliance purposes depends on which framework applies to it and to the funding/registration status of the institution using it.

Nuremberg Code

The Nuremberg Code is a ten-point statement of ethical principles for research involving human subjects, set out in the August 19, 1947 verdict of the U.S. military tribunal in United States v. Karl Brandt et al. (the 'Doctors' Trial') at Nuremberg. It is the operative reference point for a claim that a given human-subjects study meets the minimum ethical floor for legitimate research: the study is judged against the Code when its consent process, scientific justification, and risk controls are being assessed retrospectively as a historical or foundational benchmark, distinct from applying a current regulation such as the Common Rule or an IRB's own protocol-specific determination. Its first and most influential point establishes that the voluntary, informed consent of the human subject is 'absolutely essential' -- the earliest internationally recognized statement of that requirement, and the direct conceptual ancestor of informed-consent requirements in every later research-ethics code and regulation.

Risk-Benefit Analysis

Risk-benefit analysis is the determination an IRB or REC must make, under the Belmont Report's Beneficence principle and its codification at 45 CFR 46.111(a)(1)-(2), that (1) risks to research participants have been minimized through sound research design, and (2) the residual risks are reasonable in relation to the study's anticipated benefits to participants, if any, and the importance of the resulting knowledge. It is a mandatory precondition for IRB approval, made by the board itself before a study proceeds -- distinct from informed consent, which discloses those already-evaluated risks and benefits to participants so each can decide individually whether to accept them.

Reproducibility and computational research· Data & methods→

Research Transparency

Research transparency is the practice of making the specific decisions, materials, data, and code behind a study visible and checkable at the point they are used, not the broader movement to open up research outputs generally (that is <a href='/dictionary/term/open-science'>open science</a>). A study demonstrates research transparency to the degree it discloses, and ideally shares and cites, what it did: whether hypotheses and analysis plans were specified before or after seeing the data, where materials and data can be obtained, whether analytic code is available, and whether the design and reporting follow a recognized standard. The most widely adopted operational framework for this is the Transparency and Openness Promotion (TOP) Guidelines, published by the Center for Open Science (COS), which turns 'be transparent' into eight checkable standards — citation, data transparency, analytic methods (code) transparency, research materials transparency, design and analysis transparency, study preregistration, analysis plan preregistration, and replication — each ratable at one of three levels of stringency, from disclosure through requirement to independent verification.

Tuskegee Syphilis Study

The Tuskegee Syphilis Study refers specifically to the U.S. Public Health Service's Study of Untreated Syphilis in the Negro Male, conducted in Macon County, Alabama from 1932 to 1972, in which roughly 600 Black men (399 with latent syphilis, 201 as an uninformed control group) were enrolled without being told their actual diagnosis or the study's purpose, and were denied treatment -- including penicillin once it became the standard cure in the 1940s -- so researchers could observe the untreated disease's natural progression. The term is invoked today as the specific, documented U.S. case of research conducted without informed consent, through active deception, targeting a vulnerable population defined by race, and involving the deliberate denial of available treatment; its 1972 public exposure is the direct legislative trigger for the National Research Act of 1974 and, through the National Commission it created, the 1979 Belmont Report.

Mutual Support Authorship

An arrangement in which two or more researchers agree, explicitly or informally, to reciprocally list each other as authors on each other's papers regardless of whether each individual meets the substantive ICMJE authorship criteria, in order to inflate publication counts on both sides. Unlike gift authorship, which typically flows in one direction, mutual support authorship is a two-way trade of unearned credit.

AAHRPP (Association for the Accreditation of Human Research Protection Programs)

AAHRPP is an independent, non-profit accrediting body that reviews and certifies an institution's entire Human Research Protection Program (HRPP) -- not an individual researcher, and not a single IRB committee in isolation, but the full organizational system of policies, procedures, and oversight structures that protects human research participants. Accreditation is voluntary and organizational-level: an institution (a university, hospital, independent IRB, government agency, or research organization) applies, undergoes a self-assessment against AAHRPP's published standards, and is then peer-reviewed by a site visit team of experienced research-protection professionals before AAHRPP's Council on Accreditation makes a determination. A program either holds current AAHRPP accreditation or it does not -- there is no equivalent to an individual passing an exam. This is the key distinction from an individual credential such as CIP (Certified IRB Professional) or from completing required human-subjects training through a platform like CITI: those attach to a person; AAHRPP accreditation attaches to the institution's program as a whole.

Placebo-Controlled Study Design

A placebo-controlled study is an interventional design in which one or more comparator arms receive an inert or sham intervention (a placebo) instead of, or in addition to, the intervention under investigation, so that any outcome difference between arms can be attributed to the study intervention itself rather than to the act of receiving treatment, participant or investigator expectation, natural disease fluctuation, or regression to the mean. It is one of the standard concurrent-control-group categories set out in ICH E10, alongside no-treatment control, dose-response control, active (positive) control, and external/historical control, and it is most often paired with randomization and blinding so that neither participants nor assessors know which arm a given participant is in.

Mentorship, training, and career stages· Contribution→

Coercive Authorship

Coercive authorship occurs when a person with evaluative or supervisory authority over a researcher — typically a principal investigator, department chair, or thesis supervisor — uses that power imbalance to compel the inclusion of an undeserving author (often themselves) or the exclusion of a deserving one, rather than authorship being negotiated on the basis of ICMJE/CRediT contribution criteria. The defining feature is not the resulting byline, which can look identical to a gift-authorship byline, but the mechanism: the junior researcher's assent is not freely given, because dissent carries real, foreseeable career risk the senior party controls — a weaker reference, delayed degree progress, or reduced access to resources or future collaboration.

Journal Editor

A journal editor is a person holding operational editorial authority over a journal or over specific submissions to it: authority to select and manage peer reviewers, and to accept, reject, or request revision of a manuscript. This is distinct from serving on the journal's editorial board in a non-operational capacity, and from acting as an invited peer reviewer on a single manuscript. The role commonly splits into editor-in-chief (overall journal direction and final authority on contested decisions) and associate/handling/section editors (day-to-day management of individual submissions through peer review).

New Approach Methodologies (NAMs)

A technology or method (in vitro assay, organ-on-chip/microphysiological system, in silico/computational model, or human-relevant cell-based model) developed or applied specifically to generate safety/efficacy data in place of, or alongside, an animal study a regulatory pathway would otherwise require or expect. The "new approach" framing is relative to the existing animal-testing default in a given regulatory pathway (principally FDA drug/biologic review), not a general claim of scientific novelty for the underlying technology itself.

Self-Citation Rate

The self-citation rate is the proportion of an entity's total citations that point back to its own prior output. At the author level, it is the share of citations to a researcher's work that were made by that same researcher (in a later paper, citing an earlier one). At the journal level, it is the share of citations counted toward a journal's metrics (most visibly its Journal Impact Factor in Journal Citation Reports, or its CiteScore in Scopus) that come from articles published in that same journal. In both cases the rate is a simple ratio: self-citations divided by total citations received, usually expressed as a percentage over a defined citation window (JCR uses the same two- and five-year windows it uses for the Journal Impact Factor itself). A nonzero self-citation rate is normal and expected — researchers build on their own prior work, and journals publish follow-on articles in an ongoing research area. The metric becomes a research-integrity concern only when the rate is unusually high relative to field and venue norms, which is treated as a signal of possible citation manipulation rather than of an intrinsic quality problem.

IRB Chair

The IRB Chair is the IRB member designated to preside over the board's convened meetings and to personally conduct, or assign to an experienced board member, expedited-review determinations under 45 CFR 46.110 / 21 CFR 56.110 -- distinct from a regular voting IRB member (no presiding or independent expedited-review authority) and from the institution's Designated Institutional Official / FWA-signing Institutional Official (an institution-wide compliance-commitment role, not a board-review role).

Inducements in Research (Undue Inducement)

An inducement is any payment, benefit, or opportunity a researcher offers to encourage a prospective participant to enroll in or remain in a study. Offering compensation is not itself a problem: the Belmont Report's respect-for-persons and beneficence principles, and their regulatory implementation at 45 CFR 46.116(a)(1) (Common Rule) and 21 CFR 50.20 (FDA), treat reasonable payment for time and burden as compatible with voluntary consent. An inducement becomes an undue inducement when its size or structure is large enough, or contingent enough, to impair a prospective participant's ability to weigh the study's risks and benefits rationally -- functioning as pressure to enroll or stay enrolled rather than as compensation for burden. Under 45 CFR 46.111(b), an IRB must determine that consent will be sought under circumstances that minimize the possibility of coercion or undue influence; review of the payment plan (amount, schedule, and any completion contingency) is the concrete mechanism through which that determination gets made for offers of payment specifically.

Reproducibility and computational research· Data & methods→

SCORE (Systematizing Confidence in Open Research and Evidence)

A DARPA-funded research program, run by DARPA's Defense Sciences Office from 2019 to roughly 2022, that tested whether quantitative 'confidence scores' -- generated from human forecasting/prediction markets, expert surveys, and automated algorithms -- could reliably estimate how likely a published social and behavioral science (SBS) claim was to replicate. The program sampled around 3,000 research claims from roughly 60 journals published between 2009 and 2018, had forecasters and algorithms score them before any new data collection, then commissioned independent replications on a subsample to check which forecasting method came closest to the actual outcome. It is now complete; DARPA lists it as inactive and retained for reference.

Conflict of Commitment (COC)

A conflict of commitment (COC) exists when the time and effort a researcher or faculty member devotes to outside professional activities -- consulting, a second job, board service, an external appointment, or a startup role -- interferes with, or exceeds institutional limits on, the effort owed to their primary institutional obligations. It is a time-allocation problem, not a financial-bias problem: a COC can exist with no money involved (an unpaid but time-consuming board seat), and a significant financial conflict of interest can exist with no conflict of commitment at all (passive equity with no time commitment). Institutions typically manage COC through annual outside-activities disclosure, defined time caps, and cross-checking against sponsored-award effort reporting, distinct from the financial-disclosure process used for conflict of interest.

Human Research Protection Program (HRPP)

A Human Research Protection Program (HRPP) is the complete organizational structure an institution operates to protect the rights and welfare of human research participants -- not a single committee, but the full system of one or more Institutional Review Boards (IRBs), written policies and procedures, an institutional compliance/administrative office, researcher training requirements, conflict-of-interest management, and ongoing monitoring of active research, all operating under the institution's Federalwide Assurance (FWA) with HHS's Office for Human Research Protections (OHRP). The IRB is the HRPP's core review committee, not the whole program; AAHRPP is the external body that voluntarily accredits an institution's HRPP as a unit. An institution operates an HRPP whether or not it ever seeks AAHRPP accreditation -- running a compliant HRPP under the Common Rule (45 CFR 46) and applicable FDA regulations is a baseline legal requirement for institutions conducting covered human-subjects research; AAHRPP accreditation is an optional, additional layer of external, third-party verification on top of it.

Reproducibility and computational research· Data & methods→

Rigor and Transparency Index (RTI)

The Rigor and Transparency Index (RTI) is the average SciScore — an automated score of methods-rigor and resource-identification reporting — across a journal's or institution's articles in a given year, published by SciCrunch as a benchmarking metric for reporting quality. It is calculated independently of citation-based metrics such as Journal Impact Factor and is meant to be compared only within the same field or subfield.

PHS Policy (Public Health Service Policy on Humane Care and Use of Laboratory Animals)

The binding NIH policy, administered by the Office of Laboratory Animal Welfare (OLAW) under the Health Research Extension Act of 1985 (42 U.S.C. section 289d), that requires any institution receiving Public Health Service (PHS) funding to hold a written Animal Welfare Assurance accepted by OLAW before conducting or supporting activities involving any live vertebrate animal. It requires the institution's IACUC to review protocols and inspect facilities as OLAW's agent, and requires the animal care program to be based on the Guide for the Care and Use of Laboratory Animals and to comply with USDA's Animal Welfare Regulations as an absolute minimum. It is distinct from the Animal Welfare Act (a separate, funding-independent federal statute with narrower species coverage) and from AAALAC accreditation (a voluntary, non-mandatory private accreditation that does not substitute for the Assurance).

Expedited Review (IRB)

Expedited review is an IRB review pathway, available only when a proposed research activity (or a minor change to already-approved research) presents no more than minimal risk to participants and falls into one of the categories on OHRP's expedited-review category list. Instead of going to a convened meeting of the full board, it is reviewed by the IRB Chairperson personally, or by one or more experienced reviewers the Chairperson designates from the board's membership, under 45 CFR 46.110 (and the parallel FDA provision, 21 CFR 56.110). An expedited reviewer may exercise all the authorities of the full IRB except one: they may not disapprove the research -- a study can only be disapproved through full-board review at a convened meeting. ’Expedited’ describes who conducts the review (chair or designee, not the full board), not a guaranteed fast turnaround.

NIH Financial Conflict of Interest (FCOI) Policy

The federal regulatory framework (42 CFR Part 50, Subpart F for PHS grants/cooperative agreements; 45 CFR Part 94 for PHS contracts) requiring any institution receiving Public Health Service or NIH funding to maintain a written FCOI policy, collect financial disclosures from every Investigator on a covered project, have a designated official review and manage any identified conflict, and report it to the funding agency before award funds are expended.

AAALAC International

A private, nonprofit accrediting body (founded 1965, headquartered in Frederick, Maryland) that grants voluntary, triennial peer-review accreditation to animal research, teaching, and testing programs worldwide against three primary standards (the Guide, the Ag Guide, and ETS 123), confirming a program meets or exceeds legally required animal-welfare standards -- distinct from the mandatory Animal Welfare Act (USDA-enforced statute) and PHS Policy (NIH funding condition), neither of which AAALAC accreditation replaces or is required by.

Desk Rejection

A desk rejection is an editorial decision made by the handling editor (or an in-house editorial team) to decline a submitted manuscript before it is sent out for external peer review. It is issued at the initial screening stage, prior to any reviewer being invited, and is typically driven by one or more of: a scope mismatch with the journal's aims, insufficient apparent novelty or significance for the target venue, failure to meet basic formatting, language-quality, or file-completeness requirements, or failure to meet the journal's baseline reporting standards (missing ethics statements, required disclosures, or a data-availability statement, for example). Because no reviewer has evaluated the manuscript's methods or findings, a desk rejection reflects a judgment about fit and readiness, not a verdict on scientific merit. Some journals and publishers report a desk rejection rate -- the share of submissions declined at this screening stage -- though the figure varies enormously by journal prestige and field and has no universal benchmark.

Erratum

An erratum is a formal, published correction notice that a journal or publisher issues to fix a factual error introduced during the publication process itself (e.g. a typesetting mistake, a mislabeled figure, an author name spelled wrong, a column transposed in a table) rather than an error in the authors' own research, analysis, or reporting. It is linked bidirectionally to the original article's metadata (Crossref lists 'erratum' as one of its enumerated update types) so that the correction is discoverable wherever the article is indexed or cited. An erratum does not withdraw or invalidate the article's findings -- by definition, the underlying science is not in question, only how it was rendered in print or online.

Engagement, impact, and SDG alignment· Assessment→

Race Equality Charter (REC)

The Race Equality Charter (REC) is an award-mark scheme run by Advance HE, the UK sector body for higher education and research quality, that recognises universities and research institutes for identifying and addressing structural barriers to race equality among staff and students. Modelled on the same self-assessment/award-mark structure as Advance HE's Athena SWAN gender-equality charter, REC requires a member institution to convene a self-assessment team, analyse ethnicity-disaggregated staff and student data across recruitment, promotion, pay and attrition, gather qualitative evidence of lived experience, and submit an application with a costed, time-bound action plan for independent peer review by an Advance HE panel. Institutions that meet the bar receive a Bronze or Silver award (there is no Gold tier, unlike Athena SWAN) tied to delivering that action plan.

Continuing Review (IRB)

Continuing review is the periodic re-review an IRB conducts of previously approved, ongoing research to decide whether approval should continue. Under 45 CFR 46.109(e), an IRB must conduct continuing review of research requiring review by the convened IRB at intervals appropriate to the degree of risk, and not less than once per year, except for the limited categories the revised Common Rule exempts from continuing review entirely under 46.109(f) (research eligible for expedited review, research receiving limited IRB review under certain exemptions, and research that has closed to enrollment and moved solely into data analysis or long-term follow-up using standard-of-care procedures). FDA's parallel provision, 21 CFR 56.109(f), requires continuing review of FDA-regulated research at intervals appropriate to the degree of risk and not less than annually, and does not carry the same 2018 Common Rule exceptions. The operative distinction for a study posing greater than minimal risk: continuing review of that research cannot be conducted through the expedited procedure, because expedited review (whether at initial approval or at a continuing-review interval) is only available for research that is no more than minimal risk. A greater-than-minimal-risk study must therefore go before the full convened IRB, with a quorum present and a majority vote, at every continuing-review interval for as long as the study remains open to enrollment or active intervention -- not just at initial approval.

IACUC Semiannual Program Review and Facility Inspection

The PHS-Policy-mandated recurring self-governance duties every Institutional Animal Care and Use Committee (IACUC) must perform on its own institution: reviewing the entire animal care and use program at least once every six months against the Guide for the Care and Use of Laboratory Animals, inspecting all of the institution's animal facilities on that same six-month cycle, and preparing and submitting a signed written report of the review and inspection to the Institutional Official that classifies any deficiencies as significant or minor and sets a correction plan and schedule.

Generative AI use and disclosure· Contribution→

Fake Citation

A fake citation (also called a fabricated citation or, when the source is a paper never written at all, a phantom reference) is a reference that does not correspond to a real, findable publication -- an invented author, title, journal, DOI, or page range, or some combination of these, presented as if it points to an actual source. It also covers the narrower case of a citation to a real work that is misrepresented: the cited paper exists, but it does not say, show, or support what the citing text claims it does. A citation is operationally 'fake' if a reader who tries to locate and check it cannot verify that the source exists as described, or finds that the source exists but does not support the claim attached to it -- as opposed to an honest error (a typo in a volume number, a wrong year) that still resolves to the intended real work.

Ethical Violation

An ethical violation in research is any act or omission that breaches an accepted professional, institutional, or regulatory research-ethics norm, whether or not it meets the narrower threshold for formal 'research misconduct.' In the US federal system, research misconduct is legally defined by 42 CFR Part 93 as fabrication, falsification, or plagiarism (FFP) only, and explicitly excludes honest error, good-faith disagreement, and behaviors 'not unique to research' such as harassment or grant-management violations. 'Ethical violation' is the broader umbrella: it covers FFP, but also questionable research practices (QRPs), authorship abuse (ghost, guest, or gift authorship; denying deserved credit), undisclosed conflicts of interest, human- or animal-subjects protection breaches (e.g., proceeding without IRB/IACUC approval, exceeding approved protocol scope), peer-review misconduct (manipulating or leaking a manuscript under review), duplicate or salami-slicing publication, and mentorship or supervisory abuse. Whether a specific act rises to the level of an actionable 'ethical violation' is determined by the applicable code of conduct: an institution's own policy, a funder's terms and conditions, a journal's editorial policy (e.g. via COPE), or a national/international integrity code (e.g. ALLEA's European Code of Conduct for Research Integrity, or the UK's UKRIO framework), each of which sets its own scope and remedies.

Sole Authorship

The condition of a scholarly or scientific work having exactly one author: a single named individual who alone satisfies the full authorship criteria a publisher or field applies (most commonly the ICMJE's four-part test -- substantial contribution to conception or design, drafting or critically revising the work, final approval of the published version, and accountability for its accuracy and integrity), with no co-authors to share credit, order, or accountability with. Because there is only one contributor, none of the coordination questions that dominate multi-author authorship management -- author order, corresponding-author designation, CRediT role-splitting, gift or ghost authorship, or authorship disputes -- apply: the sole author is, by definition, first author, last author, and corresponding author simultaneously, and carries full responsibility for the entire work rather than a delineated share of it.

Declaration of Helsinki

The Declaration of Helsinki is the World Medical Association's statement of ethical principles for medical research involving human participants, first adopted in 1964 (Helsinki) and revised at WMA General Assemblies through the most recent revision, adopted October 19, 2024. It is the operative ethical reference a physician-investigator, sponsor, or ethics committee cites specifically for physician-conducted medical research -- distinct from the Nuremberg Code (a fixed 1947 tribunal judgment) and the Belmont Report (a US-specific 1979 national-commission document underlying the Common Rule) -- and it is the ethical source ICH E6 Good Clinical Practice cites as the origin of its own principles. A study or protocol is judged against the Declaration when the question is whether independent ethics-committee review occurred, informed consent was genuinely obtained, any placebo/no-treatment control arm was ethically justified, and results-transparency obligations were met.

Conflict of Interest in Human Subjects Research

A conflict of interest in human subjects research exists when an investigator's, institution's, or IRB member's financial or personal interest could compromise -- or reasonably appear to compromise -- the design, conduct, oversight, or reporting of a study involving human participants. Unlike the general research-COI concept, review here is tied directly to the IRB approval decision: federal guidance and institutional policy require that a significant financial conflict be evaluated and, where warranted, managed under a documented plan before the IRB may approve the protocol, and require any IRB member with a conflicting interest in a specific protocol to recuse from reviewing or voting on it.

Clinical Equipoise

Clinical equipoise is the state of genuine uncertainty, held within the expert medical community as a whole (not necessarily by any one investigator), about the comparative therapeutic merit of a randomized controlled trial's arms. It is the ethical precondition that justifies randomizing participants to a control or comparator arm rather than assigning known-superior treatment to everyone; it is disturbed, and randomization becomes unethical, once credible evidence resolves that community-wide uncertainty toward one arm.

Reproducibility and computational research· Data & methods→

Data Dredging

Searching a dataset for statistically significant relationships across many variables, subgroups, or model specifications without a hypothesis specified in advance, then reporting whichever pattern crosses a significance threshold as if it had been the study’s pre-specified object of investigation, without correcting for the number of comparisons actually performed.

Human Research Participant (Human Subject)

A human research participant (the regulatory term is "human subject") is a living individual about whom an investigator conducting research either obtains information or biospecimens through intervention or interaction, or obtains, uses, studies, analyzes, or generates identifiable private information or identifiable biospecimens (45 CFR 46.102(e)). "Subject" is the term used in the Common Rule and IRB paperwork; "participant" is the term preferred by APA style and many journals for manuscript prose. Both refer to the same regulatory category of person.

Funding lifecycle and financial vocabulary· Assessment→

National Health and Medical Research Council (NHMRC)

The National Health and Medical Research Council (NHMRC) is the Australian Government's independent statutory authority for health and medical research funding, research translation, and health/medical research ethics guidance, established under the National Health and Medical Research Council Act 1992. A grant, ethics guideline, or research-translation activity falls within NHMRC's remit when it concerns health or medical research — the causes, treatment, diagnosis, monitoring, management, or prevention of human disease — administered through NHMRC's own National Competitive Grants Program (via its Sapphire platform) or through NHMRC's ethics-guidance function (e.g. the National Statement on Ethical Conduct in Human Research), rather than through the Australian Research Council's (ARC) National Competitive Grants Program, which covers all other fields of research.

Perverse Incentives (in Research)

In research integrity, a perverse incentive is a feature of an evaluation, funding, or reward system that motivates behavior contrary to that system's stated goal. Research is organized around proxy metrics for quality — publication count, journal impact factor, citation counts, grant dollars secured — that are easier to measure than the underlying goals they stand in for (rigor, originality, real-world value). When careers, tenure, promotion, and funding decisions are tied tightly to these proxies, researchers face pressure to optimize the measurable proxy rather than the underlying goal, even when doing so degrades the research itself. A situation exhibits a perverse incentive when three conditions hold together: (1) a metric or reward is used as a stand-in for research quality, (2) career or funding consequences are attached to that metric, and (3) the metric can be increased through behavior that does not improve, or actively harms, the actual science.

National Research Act of 1974

The National Research Act (Public Law 93-348) is the U.S. federal statute, signed into law by President Nixon on July 12, 1974, that created the modern legal framework for protecting human subjects in research. It established the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research and charged it with identifying the basic ethical principles that should govern human-subjects research; that mandate is what produced the Belmont Report in 1979. Separately, the Act amended the Public Health Service Act to require that any institution receiving federal funding for biomedical or behavioral research involving human subjects establish an Institutional Review Board (IRB) to review and approve such research before it begins -- the statutory origin of mandatory IRB review, later codified in regulation at 45 CFR 46 (the Common Rule). A statute qualifies as "the National Research Act" in this sense only if it is Public Law 93-348 itself (or a provision directly traceable to it); the Belmont Report and the Common Rule are downstream products the Act required or made possible, not the Act itself.

Clinical Study

A clinical study is the umbrella term ClinicalTrials.gov and the U.S. National Library of Medicine (NLM, part of NIH) use for any research study involving human volunteers (participants) that is intended to add to medical knowledge. Under this definition, every clinical study is one of two subtypes: an interventional study — commonly called a clinical trial — in which participants are prospectively assigned, according to a protocol, to receive a specific intervention (a drug, device, procedure, or behavioral change) so its effect can be evaluated; or an observational study, in which participants are assessed for health-related outcomes or characteristics as they occur naturally, without the investigator assigning any intervention. "Clinical study" is therefore broader than "clinical trial": every clinical trial is a clinical study, but not every clinical study is a clinical trial. This is a separate axis of classification from NIH's own narrower, funding-and-policy-consequential test for whether a specific study is an "NIH-defined clinical trial" (prospective assignment to an intervention to evaluate its effect on a health-related biomedical or behavioral outcome) — a study can be a clinical trial in the general ClinicalTrials.gov sense while the NIH four-question test is what actually determines registration, reporting, and funding-opportunity obligations for NIH-funded work.

Vulnerable Populations in Research

A research participant or class of participants that 45 CFR 46 (the Common Rule) subjects to additional, IRB-enforced protections beyond ordinary informed-consent review: either because they fall into one of three subparts with dedicated, codified requirements (Subpart B — pregnant women, human fetuses, and neonates; Subpart C — prisoners; Subpart D — children), or because the IRB determines under §46.111(b) that they are otherwise likely to be vulnerable to coercion or undue influence (e.g., individuals with impaired decision-making capacity, or economically or educationally disadvantaged persons) and documents the additional safeguards included to protect their rights and welfare.

Reproducibility and computational research· Data & methods→

SciScore

SciScore is an automated text-mining tool from SciCrunch that scores a manuscript's methods section (0-10) against named rigor and resource-transparency criteria -- randomization, blinding, sample-size justification, sex as a biological variable, and RRID-identified research resources -- drawn from NIH rigor guidelines, MDAR, ARRIVE, CONSORT, and STAR Methods, integrated into journal submission workflows at publishers such as the American Heart Association, Rockefeller University Press, and FASEB.

Australian Code for the Responsible Conduct of Research

Australia's foundational research-integrity governance document (the "2018 Code"), jointly issued by the National Health and Medical Research Council (NHMRC), the Australian Research Council (ARC), and Universities Australia in June 2018, replacing the more prescriptive, rules-based 2007 Code. It is a principles-based document setting out eight principles of responsible research conduct (honesty, rigour, transparency, fairness, respect, recognition of the rights of Indigenous peoples to be engaged in research that affects or is of significance to them, accountability, and promotion of responsible research practices) plus 29 specific responsibilities — 13 for institutions (R1–R13) and 16 for researchers (R14–R29). Unlike the National Statement on Ethical Conduct in Human Research, which governs the ethical review of studies involving human participants specifically, the Code covers research-integrity conduct broadly across all disciplines: authorship, data management and retention, supervision, publication and dissemination, peer review, conflicts of interest, and how institutions must manage and investigate potential breaches.

Animal Welfare Act (AWA)

The primary US federal law (7 U.S.C. §2131 et seq., originally enacted 1966 as the Laboratory Animal Welfare Act, Pub. L. 89-544) that sets minimum standards for the humane care, handling, and treatment of certain warm-blooded animals used in research, teaching, testing, exhibition, and transport, enforced by USDA APHIS under the Animal Welfare Regulations (9 CFR Parts 1-3). It excludes rats and mice bred for research, birds bred for research, and cold-blooded animals from its definition of "animal," and — together with PHS Policy — is one of two legal bases requiring a research institution to convene an IACUC.

Originality Report

An automated text-matching report generated by a plagiarism-detection service (such as Turnitin or iThenticate) that compares a submitted manuscript, thesis, or dissertation against prior publications, previously submitted papers, and web content, expressing the proportion matched as a similarity index — a screening aid for human reviewers, not an automated finding of plagiarism.

Pseudonymous and Anonymous Authorship

Pseudonymous authorship is publication under a consistent, invented author name that is not the author's legal name but that the journal or publisher can trace back to a real, accountable person even though that identity is withheld from the public byline -- the same pseudonym is typically used across multiple works, building a durable scholarly identity distinct from the legal one. Anonymous authorship is narrower: the published work carries no attributable name or durable alias at all, though COPE's guidance expects the editor to still hold the real author's identity on file even when no name is published. Both are treated as narrow, conditional exceptions to the default expectation of a named, accountable author -- legitimate chiefly where disclosure would compromise the author's safety, or (for pseudonyms specifically) by durable convention such as a shared or house pen name -- and not where the goal is simply to distance a paper from an employer, avoid a workplace dispute, or preserve general privacy.

PubPeer

PubPeer is an independent, web-based platform where anyone can post public comments — anonymously or under their own name — on papers already published in the scholarly literature, identified by DOI, PubMed ID, or a title/author search. It is a venue for open, informal post-publication commentary, not a peer-review, editorial, or investigative body: it has no authority to correct, retract, or adjudicate a paper, and the mere existence of a PubPeer comment thread on a paper is not, by itself, evidence of research misconduct.

Conflict of Interest (COI) in Research

A conflict of interest (COI) exists whenever a researcher's, reviewer's, official's, or institution's outside financial or personal interests could reasonably be seen to compromise, or actually do compromise, objective and unbiased conduct, review, or oversight of research. The test is whether a reasonable person could conclude the secondary interest created a risk of undue influence over the primary interest -- not whether bias in fact occurred. COI is distinct from research misconduct: an undisclosed or unmanaged conflict is treated as a violation, but the underlying interest itself is not, by default, evidence of wrongdoing.

Belmont Report

The Belmont Report is the 1979 statement titled "Ethical Principles and Guidelines for the Protection of Human Subjects of Research," issued by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research (created by the National Research Act of 1974 in response to the Tuskegee Syphilis Study). It establishes three basic ethical principles for human-subjects research -- Respect for Persons, Beneficence, and Justice -- each linked to a specific application (informed consent, risk/benefit assessment, and equitable selection of subjects, respectively). It is not itself a regulation, but its principles were subsequently codified into binding federal regulation at 45 CFR 46 (the Common Rule) and remain the ethical basis every U.S. IRB applies when reviewing human-subjects research.

Journal Reviewer

A journal reviewer (also called a peer reviewer or referee) is a subject-matter expert an editor invites, from outside the journal's own staff, to independently evaluate a submitted manuscript's validity, originality, significance, and clarity, then report a written assessment and a recommendation — accept, revise, or reject — back to the editor who requested it. Someone is a journal reviewer only while performing this evaluator role for a specific editor and a specific submission: it is a per-manuscript assignment, not a standing staff position, an authorship credit, or publication-decision authority in its own right.

Retraction Watch

An independent, non-CASRAI nonprofit project — a blog plus the Retraction Watch Database — that identifies, documents, and publicly reports individual retractions, expressions of concern, and corrections across the scholarly literature. It aggregates and investigates after a publisher's own retraction decision; it has no editorial or regulatory authority to retract anything itself.

CRediT extensions and adjacent contribution vocabularies· Contribution→

ANSI/NISO Z39.106-2023 (Peer Review Terminology Standard)

The ANSI/NISO standard (Z39.106-2023, Standard Terminology for Peer Review, published July 5, 2023; DOI 10.3789/ansi.niso.z39.106-2023; ISBN 978-1-950980-27-7) that gives journals and publishers a common, structured vocabulary for describing how a manuscript's peer review was actually conducted. Developed from a taxonomy first built by an STM Association working group (convened 2019, brought to NISO in 2021) and refined through public comment and publisher piloting, it organizes description around four dimensions: (1) identity transparency — e.g. single anonymized, double anonymized, all identities visible; (2) who the reviewer interacts with — e.g. editor only, author and editor; (3) what review information is published alongside the article — e.g. none, reviewer identities, review reports, decision letters; and (4) whether post-publication commenting is enabled. A journal's peer-review policy is described using Z39.106 terminology when it states a position on all four dimensions with the standard's controlled terms, rather than a single unqualified label like "blind" or "peer reviewed."

Clinical Trial Registration

The act of submitting a clinical trial's identifying and design information — sponsor, condition studied, intervention, key eligibility criteria, and primary/secondary outcome measures — to a recognized public clinical-trial registry (ClinicalTrials.gov or any other WHO ICTRP primary registry) so the trial's existence and design are publicly discoverable independent of whether or how its results are ultimately published. Registration identifies the trial before or during its conduct; posting the trial's results to the same registry after completion is a separate, later obligation (results reporting), not part of registration itself.

Double-anonymized peer review

A manuscript review model in which neither party can identify the other during review: the reviewer does not know who the author is, and the author does not know who the reviewer is, up to and including the editorial decision. A given review counts as double-anonymized only if both conditions hold - the manuscript file sent to reviewers has been stripped of author names, affiliations, and other directly identifying material, and the editorial system withholds reviewer identity from the author throughout. Withholding only one identity (author-blind but reviewer known, or vice versa) is single-anonymized review, not double.

ICMJE (International Committee of Medical Journal Editors)

Working group of medical journal editors whose Recommendations set de facto authorship, disclosure, and trial-registration standards, not a formal regulator.

Single-anonymized peer review

The peer-review model in which reviewers are told the identity of a manuscript's authors, but authors are not told the identity of the reviewers assessing their work. Traditionally called "single-blind" review — the STM Association's 2020 peer-review terminology recommends "single-anonymized" instead, since "blind"/"blinded" have an ableist origin, though "single-blind" remains in very wide use in journal policy pages and search queries for the identical practice. It is the long-standing default across most STEM and biomedical publishing, including the majority of Elsevier, Springer Nature, and Wiley journals in those fields, and is distinguished from double-anonymized review (both parties' identities withheld from each other) and open review (some part of the process — identities, reports, or both — made public, before or after publication).

Principles of Transparency and Best Practice in Scholarly Publishing

A joint framework of 16 disclosure and practice criteria — organized around journal content (name, website, publishing schedule, archiving), journal practices (publication ethics and editorial policies, peer review, access, copyright and licensing), organization (ownership and management, an advisory body, a named editorial team with contact information), and business practices (author fees, other revenue sources, advertising, direct marketing) — jointly developed and maintained by the Committee on Publication Ethics (COPE), the Directory of Open Access Journals (DOAJ), the Open Access Scholarly Publishers Association (OASPA), and the World Association of Medical Editors (WAME). First published in December 2013; subsequent versions were released in 2015, January 2018 (v3), and September 2022 (v4, the current version). A journal or publisher is an instance of 'meeting the Principles' when each of the 16 items is disclosed, verifiable, and consistent with what the framework specifies — not when a publisher merely asserts general trustworthiness. COPE, DOAJ, and OASPA each use the Principles as part of their own membership/indexing assessment criteria, and WAME applies them as part of its membership evaluation — meaning the framework functions as an operational, checkable definition of what a legitimate journal must show, not an aspirational statement. Compliance with the Principles is also cited as one of the criteria major indexing services (e.g. Scopus) weigh when assessing a journal for inclusion, though each index applies its own separate, additional criteria on top of it.

Responsible Conduct of Research (RCR) training

Funder-mandated instruction, required of specific categories of NIH- and NSF-supported researchers and trainees, covering a defined set of research-ethics and integrity topics (authorship, data management, mentoring, peer review, research misconduct, conflicts of interest, and related areas) — distinct from a general institutional code-of-conduct orientation because its topic list, format, and recurrence are set by the funding agency, not the institution alone.

Whistleblower (research)

A person who, in good faith, raises a concern about possible research misconduct, breach of research integrity, or other serious wrongdoing in a research context. The label applies when the concern is reported through a recognised channel and the reporter would benefit from protections against retaliation.

Expression of concern

A formal post-publication notice issued by a journal to alert readers to substantive concerns about a published article when the journal cannot yet conclude that retraction or correction is the appropriate outcome. An expression of concern is appropriate when concerns are credible but unresolved, often pending an institutional investigation.

Correction (corrigendum)

A formal post-publication notice that amends an error in a published article without invalidating its overall conclusions. A correction is appropriate when the error is bounded and the central findings remain reliable; a retraction is appropriate when the conclusions are unreliable.

Retraction

The formal withdrawal of a published article from the literature by the journal, with a public retraction notice explaining the reason. An article is retracted when the journal determines that its findings are unreliable due to misconduct, honest error, or other invalidating factors.

Hong Kong Principles (2020)

Five principles for assessing researchers, articulated by Moher and colleagues at the 6th World Conference on Research Integrity in Hong Kong (2019) and published in 2020, designed to incentivise responsible research practices through hiring, promotion, and funding decisions. An institution adheres to the principles if its assessment processes embody all five.

Singapore Statement on Research Integrity (2010)

A short framework statement of four principles and fourteen responsibilities for the responsible conduct of research, adopted at the 2nd World Conference on Research Integrity in Singapore in 2010. It is referenced as a foundational global statement rather than a regulatory instrument.

ENRIO

The European Network of Research Integrity Offices, a network of national research-integrity bodies and offices across European countries that share practices, case experience, and policy positions. An organisation joins ENRIO by application from a national body with a research-integrity remit.

UKRIO

The UK Research Integrity Office, an independent advisory body providing confidential, expert advice on research integrity matters to UK research organisations and researchers. A query falls within UKRIO's remit if it concerns the conduct or governance of research in a UK context.

ORI (US Office of Research Integrity)

The US federal office within the Department of Health and Human Services that oversees and monitors investigations of research misconduct involving Public Health Service-funded research. A case is in ORI's jurisdiction if PHS funds (NIH, CDC, FDA, etc.) supported the research at issue.

COPE (Committee on Publication Ethics)

An international membership organisation providing standards, guidance, and case-by-case advice on publication ethics to editors and publishers. A journal is a COPE member if it adheres to the COPE Core Practices and pays the membership fee.

Citation cartel

A group of authors, editors, or journals that systematically cite one another's work beyond what the science would warrant, in order to inflate citation-based metrics. A pattern is cartel-like when the citation flows between members are dramatically denser than flows to comparable non-members.

Tortured phrases

Unusual paraphrases of established technical terminology that result from automated synonym substitution, often used to evade plagiarism detection. A phrase is tortured when a domain reader recognises it as a mangled rendition of a standard term.

Paper mill

A commercial operation that produces fabricated or low-effort manuscripts and sells authorship slots, or that brokers acceptance of such manuscripts into journals. A submission originates from a paper mill if the underlying research was not conducted by the named authors or was fabricated to order.

Image duplication

The undisclosed reuse of an image, or portion of an image, to represent a different experimental condition, time point, or sample. Duplication is problematic when the figure legend implies the images are distinct sources.

Image manipulation

Adjustment of a research image (gel, blot, microscopy, etc.) in a way that misrepresents the underlying data or violates journal image-integrity policies. Manipulation is problematic when it changes the conclusion a reader would draw or when it would not survive disclosure.

Coercive citation

The practice by editors or reviewers of pressuring authors to add citations that are not scientifically warranted, typically to the journal or to the reviewer's own work, as a condition of acceptance. A request is coercive when no substantive scientific reason is given and the citations would not otherwise be added.

Gift authorship

The inclusion as an author of a person who did not meet the substantive authorship criteria, typically as a courtesy, in exchange for resources, or for prestige. An authorship is a gift when removing the named person would not require any change to the manuscript.

Ghost authorship

The undisclosed substantial contribution to a manuscript by a person not listed as an author or named in the acknowledgements. A contribution is ghost-written when, by ICMJE criteria, it would have warranted authorship had it been disclosed.

Authorship dispute

A disagreement between contributors over who should be listed as an author, in what order, or with what role designation. A dispute is formally recognised when raised through an institutional process or with the journal.

Salami slicing

The practice of dividing a single coherent body of research into the smallest publishable units to maximise paper count. Outputs are salami-sliced when separating them obscures rather than clarifies the research and when readers must reassemble multiple papers to understand the whole.

Duplicate publication

Publishing the same study, or substantially the same study, in more than one venue without cross-reference and editorial permission. Two articles are duplicates if they share the same hypothesis, sample, methods, and core results to a degree that a meta-analyst would not treat them as independent.

Self-plagiarism

Reusing substantial portions of one's own previously published work in a new publication without disclosure or appropriate citation, such that a reader is led to believe the material is new. Not generally illegal or copyright infringement (absent a prior copyright transfer to a publisher) and, in the US, excluded from the federal 42 CFR Part 93 research-misconduct definition of plagiarism — but treated as a publication-ethics and institutional-policy violation by COPE, journals, and most universities.

Plagiarism

The appropriation of another person's ideas, processes, results, or words without giving appropriate credit. A passage or idea counts as plagiarised if a reasonable reader would attribute it to the new author when the substance originated with someone else.

Falsification

Manipulating research materials, equipment, processes, or data such that the research record does not accurately represent the actual results. A modification qualifies if it is undisclosed and changes the conclusions a reader would draw.

Fabrication

Making up data or results and recording or reporting them as if real (42 CFR § 93.211). A finding is fabrication if no underlying experiment, observation, or measurement actually produced the reported result — distinct from falsification, where genuine data are altered.

Research misconduct

Fabrication, falsification, or plagiarism (FFP) in proposing, performing, or reviewing research, or in reporting research results. An act qualifies if it is a significant departure from accepted practices, committed intentionally, knowingly, or recklessly, and proven by a preponderance of evidence.

Research integrity

The adherence to professional values and practices (honesty, rigour, transparency, accountability, fairness) such that research can be trusted by other researchers and by society. A practice exhibits research integrity if it could be openly described to peers without embarrassment or sanction.