4,500+ Research Papers Were Retracted in 2025: Countries, Journals, and Causes
The year 2025 marked a watershed moment in scholarly publishing, with an unprecedented rise in research paper retractions across disciplines, countries, journals, and publishers. Using data compiled from RetractionDatabase.org, this blog presents a comprehensive analysis of research paper retractions in 2025, highlighting key trends, geographic patterns, publisher-level insights, and the underlying causes driving this global phenomenon.
While retractions are an essential mechanism for correcting the scientific record, the scale and concentration observed in 2025 raise serious concerns about research quality, peer review integrity, and systemic publication pressures worldwide.
Overview: Retractions in 2025 at a Glance
- Total research papers retracted (2025): 4,544
- Most affected subject areas:
- Computer Science and Technology
- Mathematics
- Data Science
- Engineering
- Single journal with highest retractions:
- Journal of Intelligent & Fuzzy Systems — 1,041 retractions (23%)
The data clearly show that retractions in 2025 were not evenly distributed across disciplines or journals. Instead, they were highly concentrated in specific fields and publication outlets, pointing to vulnerabilities in fast-growing, high-output research areas.
Countries with the Highest Number of Retracted Papers (2025)
The following table summarizes the countries that reported the highest number of retracted research papers in 2025:
| Country | Number of Retracted Papers | Percentage of Total (%) |
|---|---|---|
| China | 1,824 | 40.14% |
| India | 922 | 20.29% |
| Iraq | 490 | 10.78% |
| Saudi Arabia | 374 | 8.23% |
| Russia | 371 | 8.16% |
| United States | 297 | 6.54% |
| Pakistan | 252 | 5.55% |
| Uzbekistan | 187 | 4.12% |
| Egypt | 171 | 3.76% |
| United Kingdom | 109 | 2.40% |
| Malaysia | 104 | 2.29% |
| Turkey | 88 | 1.94% |
| South Korea | 86 | 1.89% |
| Nigeria | 63 | 1.39% |
| Vietnam | 62 | 1.36% |
| Australia | 59 | 1.30% |
| Poland | 59 | 1.30% |
| South Africa | 58 | 1.28% |
| France | 52 | 1.14% |
| Taiwan | 52 | 1.14% |
| Germany | 51 | 1.12% |
Key Observations
- China alone accounted for over 40% of all retractions in 2025.
- India followed with more than 20%, meaning that over 60% of global retractions originated from just two countries.
- Several emerging research systems appear disproportionately affected, suggesting rapid publication growth without parallel strengthening of research integrity frameworks.
It is important to note that high retraction counts do not automatically imply poor science at the national level, but they often reflect systemic stress points such as publication pressure, incentive misalignment, and weaknesses in peer review oversight.
Publishers with the Most Retractions
Publisher-level analysis reveals notable clustering of retractions:
| Publisher | Number of Retractions |
|---|---|
| IOS Press | 1,384 |
| Springer – Nature Publishing Group | 962 |
| EDP Sciences | 679 |
| Wiley | 676 |
| Elsevier | 472 |
IOS Press recorded the highest number of retractions, followed by Springer Nature and EDP Sciences. These figures largely reflect specific journals and special issues rather than publisher-wide failures, but they nevertheless underscore the importance of editorial vigilance and robust peer review systems.
Journal-Level Concentration: A Red Flag
One of the most striking findings of the 2025 data is that:
1,041 retractions (23%) occurred in a single journal — the Journal of Intelligent & Fuzzy Systems.
Such concentration strongly suggests systemic editorial breakdowns, including:
- Compromised peer review
- Inadequate reviewer vetting
- Possible manipulation of special issues
- Insufficient screening of submissions
This highlights the need for journal-level accountability, particularly in high-volume, fast-turnaround publication environments.
Subject Areas Most Affected
The majority of retractions in 2025 were concentrated in:
- Computer Science and Technology
- Mathematics
- Data Science
- Engineering
These fields have experienced explosive growth, driven by AI, big data, and interdisciplinary applications. Unfortunately, rapid expansion often brings quality control challenges, especially when combined with aggressive publication incentives.
Common Reasons for Retractions in 2025
Analysis of retraction notices reveals the most frequently cited causes:
- Compromised Peer Review
- Concerns or Issues with Referencing and Attribution
- Concerns Regarding Third-Party Involvement
- Investigations by Journals or Publishers
- Data Fabrication or Falsification
- Plagiarism and Duplicate Publication
- Ethical Approval and Authorship Issues
Notably, many retractions involved multiple overlapping concerns, suggesting coordinated or systematic misconduct rather than isolated errors.
What Does This Mean for Global Research?
The 2025 retraction surge should be seen as a wake-up call, not merely a statistic. It highlights:
- The consequences of publish-or-perish cultures
- The risks of outsourcing manuscript preparation or peer review
- The need for mandatory research integrity training
- The importance of publisher-led audits and transparency
At the same time, increased retractions also reflect improved detection mechanisms, better whistleblowing, and stronger post-publication scrutiny—positive signs for long-term scientific credibility.
Looking Ahead
This blog is part of a broader series examining research integrity trends. Upcoming posts include:
- Scopus Discontinued Journals 2025: 56 Journals Removed from Index
- Scopus Discontinues European Public and Social Innovation Review (2026 Update)
Together, these analyses aim to foster evidence-based discussions on responsible research, ethical publishing, and sustainable academic evaluation systems.
Final Thoughts
Retractions are not failures of science—they are corrections of the scientific record. However, when they occur at scale, they signal deeper structural issues that demand coordinated action from researchers, institutions, publishers, and funding agencies.
In the long run, quality, integrity, and transparency must matter more than counts, rankings, or impact metrics.
0 Comments