In June 2013, Thomson Reuters stopped publishing impact factors for four Brazilian journals. Since 2009, editors had been publishing articles carrying hundreds of references to each other's journals, each article avoiding its own journal so that no self-citation check would notice. One of them, Mauricio Rocha-e-Silva, was fired as editor of Clinics that July, and told Nature: "We've been caught wrong-footed" (Van Noorden, 2013a).
That is a citation cartel, one of several documented ways of inflating a citation count. This guide, published by EdCitation, covers only cases and responses read at their source and dated: what COPE counts as citation manipulation, the cartels, coercive citation, extreme self-citation, bought citations, and what Clarivate does each June when it finds them. Our guide on whether citation counts matter covers what an honest count means.
EdCitation helps at two points, free and with no account: Find sources shows each result's "Cited by" figure and names the index it came from, and Verify references checks a list against the publisher's record and flags retracted papers.
What is citation manipulation?
Citation manipulation is citing work to raise a citation count rather than because the text needs it. The Committee on Publication Ethics (COPE) lists three forms in its 2019 discussion document: excessive self-citation by authors, excessive citation of the journal an author is publishing in, and excessive "honorary" citation of an editor, a well-known scholar or another journal, which it also calls citation stacking (COPE Council, 2019).
COPE defines two further terms. Manipulation becomes coercive citation when the references are made a condition of acceptance, whether a reviewer, an editor or the journal's own policy demands them. A citation manipulation cartel is more than one journal collaborating to raise each other's impact factors with citations chosen regardless of relevance.
Self-citation is not the offence
COPE is careful on this point. Citing your own earlier work is often right: a paper may be one step in a long programme, and leaving out your relevant papers can itself raise questions of redundant publication. An editor or reviewer may also know that their own work belongs in a paper. The line COPE draws is motive: requests that are "merely self promotional" breach publication ethics, and an editor's request for citations to their own work should be independent of the decision to accept (COPE Council, 2019). It recommends that journals set policies on acceptable self-citation and keep clear procedures for when a problem arises.
What is a citation cartel?
A citation cartel is a group of journals, editors or authors citing each other to lift their numbers. Two cases show how one works.
Cell Transplantation, 2012
In April 2012 Phil Davis described two 2010 review articles on The Scholarly Kitchen, one in Medical Science Monitor and one in The Scientific World Journal. The first cited 490 papers, 445 of them published in Cell Transplantation in 2008 and 2009; the second cited 124, 96 of them from the same journal and years. Those were exactly the two years the journal's 2010 impact factor counts. Davis calculated that without the two reviews that impact factor would fall from 6.204 to 4.082, and noted that three of the four authors of one review sat on Cell Transplantation's editorial board (Davis, 2012). Nature later reported that the articles had been retracted, and that the tip-off led Thomson Reuters to build a program to detect the pattern (Van Noorden, 2013a).
The Brazilian journals, 2013
That program found the Brazilian arrangement. One of the editors told Nature that by 2009 the editors of eight Brazilian journals had decided to act together. In its June 2013 release, Thomson Reuters suspended the impact factors of 14 journals for citation stacking, four of them Brazilian, among them Clinics and the Revista da Associação Médica Brasileira (Van Noorden, 2013a).
Rocha-e-Silva blamed CAPES, the education ministry agency that judged graduate programmes partly by the impact factors of the journals their students published in, which kept new Brazilian journals at the bottom. Other editors involved disputed that the aim was only a higher impact factor, saying it was also to show off Brazilian research. CAPES ruled that articles in the four journals from 2010 to 2012 would not count in its October 2013 evaluation, and the appeal for Clinics was refused. Marie McVeigh, who led the Journal Citation Reports, said the firm does not assume intent: "Until we can algorithmically measure motive, we are just going to measure citations."
What is coercive citation?
Coercive citation is an editor, or a reviewer backed by an editor, requiring references to the journal or to themselves as a condition of publication. Allen Wilhite and Eric Fong surveyed academics about it for Science in 2012. That paper is paywalled, so we give its findings as the authors summarised them five years later: roughly 20% of respondents had been coerced, and more than half said they would add superfluous citations when submitting to a journal known to coerce (Fong & Wilhite, 2017).
Fong and Wilhite (2017) then sent surveys to more than 110,000 scholars in eighteen disciplines, from medicine and nursing to economics and marketing, over five years, and received about 12,000 replies. The key question asked whether an editor had requested citations to the editor's journal "for reasons that were not based on content". Of all respondents, 14.1% said yes. More than 90% thought the practice inappropriate, 85.3% thought it lowers a journal's prestige, and 73.9% said they were less likely to submit to a journal that does it.
How common is extreme self-citation?
Rare among the most cited scientists, but not negligible. Ioannidis et al. (2019) built a public database of the 100,000 most cited scientists in Scopus, giving every indicator with and without self-citations, which they count as any citation from a paper sharing an author with the cited one. In the career-long data to 2017, the median share of self-citations was 12.7%, ranging from 0% to 93.8%; 1,085 authors had more than 40% and 8,599 more than 25%. In the 2017 single-year data the counts were 1,565 and 8,534.
The authors advise against using any citation metric for someone with very high self-citation, since it may signal other problems, and say removing self-citations may not be enough. They also describe "citation farms", small clusters of authors citing each other massively, which show up as a high ratio of citations to citing papers. Nature's report on the database was titled "Hundreds of extreme self-citing scientists revealed in new database", and noted that researchers warn against policing self-citation (Van Noorden & Singh Chawla, 2019).
Can citations be bought?
On Google Scholar, yes. Ibrahim et al. (2025) examined about 1.6 million Google Scholar profiles and, surveying faculty at highly ranked universities, found Google Scholar was the most used source of citation counts when scientists are evaluated. They then created a fictional researcher and contacted an "H-index & Citations Booster" service, which offered 50 citations for 300 US dollars or 100 for 500. They bought 50, which arrived from five papers.
They also found citations planted through preprints. Some anomalous profiles drew most of their citations from non-peer-reviewed papers on ResearchGate, and citations from preprints later deleted from their server stayed on Google Scholar. The authors note that Scopus and the Web of Science keep preprints out of their citation metrics, which closes that route.
How do the citation indexes respond?
Clarivate, which publishes the Journal Citation Reports (JCR), withholds a journal's impact factor from a year's release when self-citation or citation stacking distorts it. The impact factor began as Eugene Garfield's tool for choosing journals to index; see the first citation index. Clarivate's 2017 explanation says each decision rests on quantified criteria, that "precise thresholds are not released so as to prevent gaming", and that it judges neither intent nor wrongdoing (Clarivate Analytics, 2017).
What a suppression looks like
Clarivate's list for the 2024 release, which we read in an archived copy, suppressed 17 titles: 6 for self-citation, 11 for stacking. Self-citations made up 93% of the impact-factor citations of Exploratory Animal and Medical Research, shifting its category rank by 55%, and 88% of those of SOCAR Proceedings. For stacking the list pairs recipient and donor: 46% of the Ukrainian Journal of Physical Optics' impact-factor citations came from Optik, and 92% of all Optik's citations to it fell in the two years the metric counts. Donors can be suppressed too: Environmental Science and Pollution Research and Resources Policy are on the list. Suppression does not immediately change a journal's Web of Science coverage, though the title may be reviewed, and full metrics return once the pattern is resolved (Clarivate, 2024).
The documented cases side by side
| Case or practice | Year | What happened | Response | Source |
|---|---|---|---|---|
| Stacking into Cell Transplantation | 2012 | Two reviews supplied 541 citations to its 2010 impact factor | Reviews retracted; a detection program built | Davis (2012); Van Noorden (2013a) |
| Brazilian journals' cartel | 2013 | Editors cited each other's journals from 2009 | Four impact factors suspended; an editor fired; CAPES discounted the papers | Van Noorden (2013a) |
| JCR suppressions | 2011 to 2013 | 34, then 51, then 66 journals | Impact factors withheld | Van Noorden (2013b) |
| Coercive citation | 2012, 2017 | About 20%, then 14.1%, of surveyed academics coerced | COPE's 2019 guidance | Fong and Wilhite (2017) |
| Extreme self-citation | 2019 | 1,085 of the top 100,000 over 40% self-citations | Metrics published with and without them | Ioannidis et al. (2019) |
| JCR suppressions | 2023 to 2025 | 4, then 17, then 20 journals | Withheld, eligible again next year; from 2025 citations to retracted papers excluded | Attride and Orrall (2024); Orrall (2025) |
| Bought citations | 2025 | 50 citations bought for a fictional profile | Curated indexes exclude preprints from metrics | Ibrahim et al. (2025) |
| Mandatory citations at a surgery journal | 2026 | Required reporting-guideline citations inflated its impact factor | Six journals put on hold | Joelving (2026) |
The 2025 list, 16 journals for self-citation and 4 for stacking, drew protest. Wiley said an internal investigation of Applied Organometallic Chemistry found no integrity issues, and appealed; the chief executive of the publisher of Chemical Methodologies called the decision "very wrong and dangerous". That year Clarivate also excluded citations to retracted papers from impact factors for the first time, affecting 1% of journals (Orrall, 2025).
Beyond suppression: a journal on hold
Clarivate can go further. In March 2026 Retraction Watch reported that mandatory citation of reporting guidelines had inflated the impact factor of the International Journal of Surgery, 15.3 in 2022 and second in its category. On 19 March, Clarivate put that journal and five sister titles on hold, indexing no new content while it re-evaluated them. Clarivate says about 85% of journals put on hold are eventually delisted; the publisher said it continues to work with Clarivate (Joelving, 2026).
How do I spot citation manipulation as a reader or reviewer?
Look for the patterns, because motive cannot be seen. These checks follow the cases above:
- Check the journal's metric. A journal on the year's JCR suppression list, or on hold, deserves a closer look, and a suppression names the reason.
- Scan the reference list. A block of references to one journal or author, unrelated to the claims beside them, is the Cell Transplantation pattern: 445 of 490.
- Read the author's figures without self-citations. The Ioannidis database gives both; above 40% places an author among about 1% of the 100,000 most cited.
- Compare indexes. A count far higher on Google Scholar than in a curated index may include preprint citations. Find sources names the index behind every "Cited by", and our guide to the most cited papers of all time shows how far indexes differ.
- As an author, ask what each requested citation supports. COPE holds that an editor's request to cite their own work should be independent of acceptance.
- As a reviewer, suggest only what fills a gap, your own work included when it truly does.
- Check for retractions. Verify references flags a retracted paper anywhere in a list.
For journals that fake their review rather than their numbers, see how to avoid predatory journals and hoax papers that fooled journals.
Which EdCitation tools help you read a citation count?
No free tool reads a citation count more honestly than Find sources, because every count it shows belongs to a real record, with the index that supplied it named on the page; a chatbot's list offers no such guarantee. It searches about 300 million published works by topic or by the claim a sentence needs to support, with filters for kind of source, years, open access and times cited, and a Most cited sort for finding a field's landmarks.
On 27 September 2026 we searched "citation manipulation". Best match returned 133,863 results, and the page read "Results from OpenAlex." First came Fong and Wilhite's 2017 study, cited by 217, then COPE's discussion document, cited by 41, then Ibrahim and colleagues' citation mills paper, cited by 25. Sorted by Most cited, OpenAlex answered again with 133,863 results, led by a 2010 paper titled "Detecting h-index manipulation through self-citation analysis", cited by 307, then Fong and Wilhite, 217, and COPE, 41. Those are OpenAlex's counts that day; another index gives others.
Cite a source then turned the DOI 10.1371/journal.pone.0187394 into this APA 7 entry from Crossref's record, finding no retraction notice:
Fong, E. A., & Wilhite, A. W. (2017). Authorship and citation manipulation in academic research. PLOS ONE, 12(12), Article e0187394. https://doi.org/10.1371/journal.pone.0187394
Verify references checks a whole list: each entry comes back verified, "check this" or not found, retracted papers are flagged, and "could not check" is never shown as "not found". All three tools are free. For a full draft, References from a file also ties every in-text citation to its entry, in Pro at $8 a month (pricing).
EdCitation never writes any part of anyone's work. It finds, cites and verifies from the record, and leaves the judgement of whether a count deserves trust to you, sharpened by cases like these.
What could not be confirmed?
Clarivate's 2026 suppression list could not be read: its help pages refused our requests, and its 17 June 2026 release, covering 22,643 journals, gives no suppression count (Clarivate, 2026). Wilhite and Fong's 2012 paper is paywalled, so its figures come from the authors' 2017 summary. The 2024 list was read in an archived copy from September 2024.
Quick questions
What is a citation cartel?
A citation cartel is a group of journals or authors citing each other to raise their counts, regardless of relevance. The 2013 Brazilian case cost four journals their impact factors for a year.
What is citation stacking?
Citation stacking is an unusually concentrated flow of citations from one journal into another's impact factor. Clarivate can suppress both the receiving journal and the one supplying the citations.
Why is a journal suppressed from the JCR?
Because self-citation or citation stacking distorted its impact factor by Clarivate's criteria, which judge neither intent nor wrongdoing. Metrics return in a later release once the pattern is resolved.
Can self-citation manipulate an h-index?
Self-citations raise an h-index in any index that counts them, which is why the Ioannidis database reports figures with and without them. Check both before trusting a high h-index.
How can I check a paper's citation count myself?
Search its title in EdCitation's Find sources: each result shows "Cited by", and the page names the index that supplied it. Compare another index before quoting the number.
References
- Attride, D., & Orrall, A. (2024, June 27). Seventeen journals lose impact factors for suspected citation manipulation. Retraction Watch. https://retractionwatch.com/2024/06/27/seventeen-journals-lose-impact-factors-for-suspected-citation-manipulation/
- Clarivate. (2024). Title suppressions: Journals suppressed from 2023 JCR data (2024 release). Journal Citation Reports Help. Archived September 19, 2024, at http://web.archive.org/web/20240919225613/https://jcr.help.clarivate.com/Content/title-suppressions.htm
- Clarivate. (2026, June 17). Clarivate releases Journal Citation Reports 2026 [Press release]. https://clarivate.com/news/clarivate-releases-journal-citation-reports-2026/
- Clarivate Analytics. (2017, June 23). A refresher course on JCR journal suppression policies. Clarivate. https://clarivate.com/academia-government/blog/refresher-course-jcr-journal-suppression-policies/
- COPE Council. (2019). COPE discussion document: Citation manipulation. Committee on Publication Ethics. https://doi.org/10.24318/cope.2019.3.1
- Davis, P. (2012, April 10). The emergence of a citation cartel. The Scholarly Kitchen. https://scholarlykitchen.sspnet.org/2012/04/10/emergence-of-a-citation-cartel/
- Fong, E. A., & Wilhite, A. W. (2017). Authorship and citation manipulation in academic research. PLOS ONE, 12(12), Article e0187394. https://doi.org/10.1371/journal.pone.0187394
- Ibrahim, H., Liu, F., Zaki, Y., & Rahwan, T. (2025). Citation manipulation through citation mills and pre-print servers. Scientific Reports, 15, Article 5480. https://doi.org/10.1038/s41598-025-88709-7
- Ioannidis, J. P. A., Baas, J., Klavans, R., & Boyack, K. W. (2019). A standardized citation metrics author database annotated for scientific field. PLOS Biology, 17(8), Article e3000384. https://doi.org/10.1371/journal.pbio.3000384
- Joelving, F. (2026, June 24). Major citation index put surgery journals on hold following Retraction Watch investigation. Retraction Watch. https://retractionwatch.com/2026/06/24/major-citation-index-put-surgery-journals-on-hold-following-retraction-watch-investigation/
- Orrall, A. (2025, June 18). Citation issues cost these 20 journals their impact factors this year. Retraction Watch. https://retractionwatch.com/2025/06/18/clarivate-impact-factor-suppression-list-2025-self-citation-stacking/
- Van Noorden, R. (2013a). Brazilian citation scheme outed. Nature, 500(7464), 510–511. https://doi.org/10.1038/500510a
- Van Noorden, R. (2013b, June 19). New record: 66 journals banned for boosting impact factor with self-citations. Nature News Blog. http://blogs.nature.com/news/2013/06/new-record-66-journals-banned-for-boosting-impact-factor-with-self-citations.html
- Van Noorden, R., & Singh Chawla, D. (2019). Hundreds of extreme self-citing scientists revealed in new database. Nature, 572(7771), 578–579. https://doi.org/10.1038/d41586-019-02479-7
- Wilhite, A. W., & Fong, E. A. (2012). Coercive citation in academic publishing. Science, 335(6068), 542–543. https://doi.org/10.1126/science.1212540