A Researcher’s Suspiciously High H-Index Revealed a Vast Citation Ring | The ScientistSkip to main content

Register

Article<br>A Researcher&rsquo;s Suspiciously High H-Index Revealed a Vast Citation Ring<br>An interconnected global network of computer scientists aggressively boosted each other&rsquo;s metrics through citation manipulation and compromised peer review.<br>Written byRetraction WatchRetraction Watch<br>Retraction Watch is an outlet that reports on scientific paper retractions and research ethics.<br>View Full Profile<br>Learn about our Editorial Policies.

| 7 min read

Save for Later

A group of scientists around the world who cite each other&rsquo;s work has both raised their profiles and research integrity questions.<br>Image credit:&copy; iStock.com, Alllex

PDF versionRegister for free to listen to this article<br>Listen with Speechify<br>0:00

7:00

Share

With an h-index of 75, computer scientist Thippa Reddy Gadekallu ranks among the world’s most highly cited researchers. But the speed and means of his ascent to those lofty heights of scholarship has been as remarkable as the achievement itself.<br>In less than a decade, Gadekallu, a professor at Zhejiang A&F University in China, has managed to bootstrap himself from scientific obscurity by collaborating with colleagues around the world who cite each other’s work in ways that have raised questions. In some years, Gadekallu received more citations than Yoshua Bengio, a pioneer in artificial intelligence and the top-rated computer scientist on Google Scholar.<br>Earlier work uncovered a network of reviewers on papers Gadekallu edited who frequently suggested adding citations to his work. A closer look by Retraction Watch shows the impact of that strategy on Gadekallu’s h-index, and reveals additional possible collaborators in the network.<br>“Man, this is crazy,” said Vincent Larivière, an information scientist at the University of Montreal, whom we asked to review the metrics. “These numbers are definitely suspect.”<br>The h-index is based on citation rates, and growth becomes more difficult with each new paper. Extraordinarily prolific researchers, such as John Ioannidis, have experienced single year jumps as high as 16, but yearly increases in the single digits are more typical.<br>Continue reading below...<br>Like this story? Sign up for FREE Newsletter updates:<br>Latest science news storiesTopic-tailored resources and eventsCustomized newsletter content<br>Subscribe

David Robert Grimes, a Retraction Watch Sleuth in Residence, examined the citation records of Gadekallu and his closest collaborators, and found that, in 2019, Gadekallu’s h-index jumped 17 points, from 10 to 27. The following year, it jumped 23 points, and then 19. Matching Ioannidis’s career high, Grimes said, “is kind of like matching Usain Bolt’s 100-meter time.”<br>Repeating the feat three times in a row? “You either have a truly exceptional scholar or something else is going on,” Grimes said.<br>Year Number of publications Number of citations H-Index 2014 12512016 86342017 7280102018 257102019 16102020 284057272021 494906502022 837531692023 933987742024 821883752025 8723975<br>Thippa Reddy Gadekallu’s publications, citations and h-index by yearGrimes’s work has revealed that his network includes dozens of coauthors spread around the world. Like Gadekallu, Grimes found, his most frequent collaborators have experienced remarkably steep increases in their h-indices, raising the question of whether such metrics should continue to be relied upon.1<br>Over the course of several emails with us, Gadekallu admitted that he typically leaned on collaborators to conduct peer reviews, but insisted that he has never participated in any intentional manipulation.<br>“I completely understand why it might look concerning from the outside, but I categorically deny any behind the scenes coordination, pressure, or exchange regarding citations,” he wrote.<br>Hung-Wen Chiu’s experiences tell a different story.<br>In April 2021, Chiu, a biomedical engineer at Taipei Medical University in Taiwan, submitted a new paper to PLOS One. The paper described a method to automatically classify medical images of different body parts that had been captured on different types of machines.<br>PLOS One’s editorial board is made up of thousands of volunteer editors from around the world. Chiu’s paper was assigned to Gadekallu, then at the Vellore Institute of Technology in India, who delegated it to two peer reviewers.<br>The peer review comments Chiu received several weeks later were brief and generic.2 “Discuss the limitations of the current work,” the first reviewer suggested. “Typo’s [sic] in the whole paper,” the second reviewer said.<br>But in one area they offered more precision: citations. Both reviewers felt Chiu’s paper should reference two of Gadekallu’s own articles and two others from one of Gadekallu’s collaborators. One provided a link to the Google Scholar profile for the collaborator: “Visit this profile and cite related papers.”<br>The papers were broad, descriptive...