Free scoping read. A written assessment of your manuscript in two business days, at no cost.

Start one
Contents

Academy · Research Publication · Trust & Retraction Governance

The Fragile Balance of Trust in Scientific Publishing: Retraction Trends and Governance

Retraction rates have increased approximately tenfold since the early 2000s, and surveyed researchers report concern not only about the underlying misconduct but also about whether the retraction process itself is governed fairly and consistently. This article examines the trend, its interpretation and the governance standards that have been proposed.

Pubrica AcademyWritten by a publication ethics writer Reviewed bya research integrity editor Published10 October 2026 Last checked10 October 2026 Reading time8 minutes Sources10 peer-reviewed

Scientific publishing depends on a layered system of trust: readers trust that published findings reflect the research as it was conducted, editors and reviewers trust that submitted data are genuine, and institutions and funders trust that the literature on which they base decisions has been vetted. Retraction is the formal mechanism for restoring that trust when it fails: a public notice that a specific published paper should no longer be relied upon. Because retraction is rare, visible and carries lasting reputational consequences for the authors involved, its governance (who decides, under what criteria and with what oversight) matters as much for sustaining trust as the underlying rate of misconduct it is intended to correct.

Learning objectives

  • The extent to which retraction rates have changed since the early 2000s, and what that trend does and does not indicate
  • The documented distinction between researcher-driven and institution- or publisher-driven retractions
  • Why the governance of the retraction process, and not only the detection of misconduct, is a documented concern

Common misconceptions

  • “A rising number of retractions is in itself proof that science is becoming less trustworthy.” Methodological research on retraction trends attributes much of the increase to greater detection effort and improved tools (digital screening, image-analysis software and dedicated watchdog reporting) rather than to a proportional increase in misconduct; a higher number of retractions may therefore coexist with, or even indicate, a self-correcting system that functions more effectively than before.

The extent of the increase in retraction rates

R. Grant Steen, Arturo Casadevall, and Ferric Fang's widely cited 2013 analysis found that the number of retracted papers indexed in PubMed rose substantially faster than the overall growth of the published literature over the preceding decade and attributed a meaningful share of that increase specifically to misconduct, while also noting that greater institutional and editorial scrutiny contributed to the detection of cases that would previously have gone unreported1. Separately, the companion analysis by Ferric Fang, Steen, and Casadevall of the underlying reasons for retraction found that misconduct (fraud, suspected fraud, duplicate publication, and plagiarism combined) accounted for the majority of retractions in their dataset, a larger share than honest error, which challenged an earlier assumption that most retractions stemmed from unintentional mistakes6. Melissa Grieneisen and Minghua Zhang's broader survey of retracted literature across multiple databases similarly documented a sharp rise in retraction counts over time, while cautioning that improved detection tools and the founding of the public tracking site Retraction Watch in 2010 likely account for part of the apparent increase, separate from any underlying change in researcher behavior410.

The asymmetry between loss and restoration of trust

John Ioannidis's influential 2005 paper modeled why a substantial share of published research findings across many fields are likely to be false under plausible assumptions about study power, bias, and the number of research teams testing similar hypotheses, a structural argument about why trust in any single published finding should be calibrated rather than automatic, independent of any specific case of misconduct2. Daniele Fanelli's systematic review and meta-analysis of surveys asking scientists directly about misconduct found that, while only a small percentage admitted to fabricating or falsifying data themselves (about 2% across pooled studies), a much larger share (roughly one in three respondents in some surveys) admitted to other questionable research practices, and an even higher share reported knowing of colleagues who had engaged in misconduct3. This gap between self-reported behavior and the behavior perceived in others is itself a marker of eroded trust: researchers broadly suspect that the problem is more common than they are willing to acknowledge in relation to themselves.

Documented concerns about the governance of retraction decisions

The Committee on Publication Ethics (COPE) publishes retraction guidelines specifically because editorial decision-making about when and how to retract has historically varied inconsistently across journals and publishers, and the guidelines set out criteria, including clear and reliable evidence of findings' unreliability, plagiarism, unethical research, or redundant publication, meant to standardize that decision-making process rather than leave it to each journal's discretion5. The recommendations of the International Committee of Medical Journal Editors (ICMJE) make a parallel point with respect to the conduct of authors and editors, setting expectations for how allegations of misconduct should be investigated and how corrections and retractions should be communicated once a decision has been made7. A 2017 consensus report on research integrity by the National Academies of Sciences, Engineering, and Medicine similarly called for clearer and more consistently applied institutional processes for investigating and adjudicating allegations of research misconduct, concluding that inconsistent process, and not only the underlying rate of misconduct, was itself undermining confidence in how the scientific community corrects its own record8.

Disputed retractions

Most retractions are not controversial once issued: the data of Steen, Casadevall, and Fang, together with the companion analysis by Fang, Steen, and Casadevall, show that a clear majority of retractions are either author-initiated corrections of honest error or confirmed cases of serious misconduct established through institutional investigation, and such cases rarely generate continuing dispute about whether the retraction was warranted16. A smaller number of cases are genuinely contested: an author may dispute the journal's grounds for retraction, or a retraction may appear to follow external pressure (a legal threat, a business interest or a politically sensitive finding) rather than a straightforward evidentiary problem with the data. The guidelines of the Committee on Publication Ethics anticipate this scenario by recommending that journals maintain a documented, evidence-based process with which an author can engage and which the author can, in principle, appeal, rather than leaving a contested retraction decision solely to a single editor acting without independent review5. Disputed cases bear disproportionately on the question of trust raised at the outset of this article, because a retraction that appears to have been driven by considerations other than evidence may make researchers reluctant to publish inconvenient or controversial findings for fear of a similarly non-evidentiary response, thereby undermining the self-correcting function that retraction is intended to serve.

Requirements for durable trust in the retraction process

The 2017 “manifesto for reproducible science” by Marcus Munafò and colleagues argued that structural reforms (pre-registration, open data and materials, replication incentives and transparent reporting) address trust at its source by facilitating the detection of problems before publication rather than relying entirely on retraction after the fact9. Taken together with the COPE and ICMJE guidance on the governance of retraction, this evidence indicates that trust in scientific publishing depends on two separate but related systems: mechanisms that reduce how often correction is needed in the first place, and a fair, transparent and consistently applied process for correcting the record when necessary, the latter requiring as much deliberate institutional attention as the former. Those evaluating a field, a journal or a body of evidence should therefore treat both systems as verifiable rather than assumed. The self-correction capacity of a field can be assessed by examining whether its retraction notices are specific and retrievable, whether its leading journals have adopted COPE-aligned criteria, and whether disputed cases are handled through a documented process rather than unilateral editorial discretion; each of these questions can be investigated directly by a careful reader. Trust is therefore not merely a perception that the research community holds or lacks about itself; it is the cumulative, verifiable product of specific institutional practices, which can be examined, compared across journals and deliberately improved, independently of the current retraction count considered in isolation. For researchers deciding which journals to entrust with their own work, or which findings to build upon, this perspective is more actionable than a general sense of rising or falling confidence in “the literature” as an undifferentiated whole: specific, verifiable questions about the documented practices of a journal or field, applied consistently across candidates, yield a considerably more useful signal than any single aggregate measure of trust. Journals that publish their editorial policies, misconduct-investigation procedures and retraction criteria openly, rather than leaving them implicit or dispersed across several pages that are difficult to locate, make such verification straightforward for the researchers, institutions and funders who most need to carry it out quickly and reliably when deciding where to submit, review or rely on published work. None of the checks described in this article requires specialized access or tools; the published policies of a journal, its COPE membership status and its retraction notices are intended to be publicly verifiable, so the principal barrier to their use is the failure to consult them before trust is extended rather than afterwards.

Key points

Key points

Rising numbers of retractions reflect both more misconduct and better detection, and in either case they indicate that a correction mechanism is functioning. The greater long-term threat to trust in the system is inconsistent and opaque governance of the retraction decision itself, which is why COPE, ICMJE and national research-integrity bodies have each published guidance aimed specifically at standardising that process.

Practical application

Practical application

  1. A rising number of retractions should be interpreted as a signal of detection and not only as a signal of misconduct. Discussions of retraction trends should acknowledge both possible drivers, namely increased misconduct and improved detection tools and scrutiny, rather than defaulting to the more alarming interpretation.
  2. The stated retraction criteria and process of a journal should be checked before considerable reliance is placed on its record. Explicit adoption of the COPE retraction guidelines and of ICMJE-aligned misconduct procedures is an indicator of consistent governance.
  3. Evidence of self-correction should be distinguished from evidence on rates of misconduct in written work. A journal or field with more documented retractions is not necessarily less trustworthy than one with fewer; the higher number may reflect more rigorous post-publication scrutiny.

Frequently asked questions

Not necessarily. Methodological analyses attribute a substantial share of the rise in retraction counts to improved detection tools, greater editorial scrutiny, and increased public tracking since Retraction Watch launched in 2010, alongside any genuine change in underlying misconduct rates; the two explanations are therefore not mutually exclusive and cannot be distinguished on the basis of the raw count alone.

A widely cited analysis found that misconduct categories (fraud, suspected fraud, duplicate publication, and plagiarism combined) accounted for a majority of retractions in the dataset studied, a larger share than honest error, though the exact proportion varies across different studies and time periods.

COPE's guidelines set out specific grounds for retraction, including clear evidence of unreliable findings, plagiarism, unethical research, or redundant publication, and recommend that retraction notices be clearly labeled, linked to the original article, and explain the reason for retraction, aiming to standardize a process that previously varied widely by journal.

Documented disputed-retraction cases exist, which is part of why COPE and related bodies recommend an independent, evidence-based process with clear criteria and the possibility of institutional appeal, rather than leaving retraction decisions solely to an individual editor's discretion.

A systematic review and meta-analysis of survey data found that only a small percentage of scientists admit to fabricating or falsifying data themselves, while a much larger share report suspecting or knowing of such behavior in colleagues, a gap that itself reflects how trust in the system functions differently at the personal and systemic level.

References

  1. Steen RG, Casadevall A, Fang FC. Why Has the Number of Scientific Retractions Increased? PLOS ONE. 2013;8(7):e68397. read 10 October 2026. View on PLOS ONE
  2. Ioannidis JPA. Why Most Published Research Findings Are False. PLOS Medicine. 2005;2(8):e124. read 10 October 2026. View on PLOS Medicine
  3. Fanelli D. How many scientists fabricate and falsify research? A systematic review and meta-analysis of survey data. PLOS ONE. 2009;4(5):e5738. read 10 October 2026. View on PLOS ONE
  4. Grieneisen ML, Zhang M. A Comprehensive Survey of Retracted Articles from the Scholarly Literature. PLOS ONE. 2012;7(10):e44118. read 10 October 2026. View on PLOS ONE
  5. Committee on Publication Ethics (COPE). Retraction Guidelines. read 10 October 2026. View COPE's Retraction Guidelines
  6. Fang FC, Steen RG, Casadevall A. Misconduct accounts for the majority of retracted scientific publications. Proceedings of the National Academy of Sciences. 2012;109(42):17028-17033. read 10 October 2026. View on PNAS
  7. International Committee of Medical Journal Editors. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. read 10 October 2026. View the ICMJE Recommendations
  8. National Academies of Sciences, Engineering, and Medicine. Fostering Integrity in Research. The National Academies Press, 2017. read 10 October 2026. View the National Academies report
  9. Munafò MR, Nosek BA, Bishop DVM, et al. A manifesto for reproducible science. Nature Human Behaviour. 2017;1:0021. read 10 October 2026. View on Nature Human Behaviour
  10. Retraction Watch (The Center for Scientific Integrity). Retraction Watch Database. read 10 October 2026. View the Retraction Watch Database

Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.

How to cite this article

Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.

Pubrica Academy. (2026, October 10). The Fragile Balance of Trust in Scientific Publishing: Retraction Trends and Governance. Pubrica. https://pubrica.com/academy/study-guide/research-writing/trust-scientific-publishing/

This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.

Retraction Governance Check

The following items may be used to assess whether the retraction process of a journal is governed consistently.

[ ] Confirmed the journal has publicly stated retraction criteria aligned with COPE guidelines
[ ] Checked whether retraction notices are clearly labeled, linked to the original article, and explain the reason
[ ] Checked whether the journal distinguishes author-initiated from editor/publisher-initiated retractions in its notices
[ ] Looked for evidence of an independent or appealable review process for disputed cases
[ ] Checked the journal's ICMJE-aligned misconduct-investigation procedures where relevant

Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.

How Pubrica can help

Pubrica’s publication ethics advisory service assists authors and institutions in understanding standards for the governance of retraction, responding appropriately to allegations of misconduct, and drafting compliant notices of correction or retraction.

Find something else

Search the Pubrica Academy

Access is free and requires no account. Enter a topic, or open the full library to filter by category, sub-topic or resource type.

Journal retraction processes assessed against COPE governance standards

A scoping assessment is free and takes two business days. It provides a written assessment of what the work requires and which specialist would undertake it.

Get a free quote

Regulations

Check your project against this rule

  1. 1Your work
  2. 2Your details
  3. 3Send
What do you need?

Material you send is held under signed confidentiality, with access limited to the assigned team and encrypted transfer. It is not used to train anything and is not passed to third-party tools. Ask for the agreement before you send anything you would rather cover first.