Volunteer watchdogs pushed a small country up the rankings

10/31/2018 Science


Romania has among the highest retraction rates, thanks partly to oversight by volunteers.

Drumroll, please: The countries that top our rankings of most retractions by nation are … Iran and Romania.

Why? They’re not among the world’s leaders in the absolute number of retractions—that dubious honor goes to the United States and China.

But ranking countries that way can be misleading. The United States and China fund many researchers who together publish many papers, which in turn can increase the number of papers that must be retracted.

Instead, Science and Retraction Watch created two measures that allow consistent comparisons across countries. The first is retractions per dollar of national research funding from 2003 to 2016, which is a proxy for the size of a nation’s scientific establishment. The second is retractions per paper published.

CREDITS: (GRAPHIC) J. YOU/SCIENCE; (DATA) RETRACTION WATCH AND NSF

By the funding measure, Romania takes the top spot. (The United States falls to 34th, and China to 14th.) But the story doesn’t end there: Romania’s leading rate of retractions per research dollar probably reflects the outsize effect of some dogged watchdogs—a small band of researchers who have been politely but firmly contacting journals to point out suspected plagiarism by Romanian authors. That activism has led to dozens of retracted papers.

The effort was launched in 2013 by Stefan Hobai of the University of Medicine and Pharmacy in Târgu Mureş, Romania, who dubbed it, with no intentional irony, the Project dedicated to arrest of the name decline of the Romanian achievement (PANDORA) in biomedical publishing. Hobai tells Science and Retraction Watch that he acted because the editors of Acta Medica Marisiensis—published by the same university Hobai works for—ignored 17 messages in which he reported articles suspected of plagiarism.

Since then, PANDORA’s few members, who other than Hobai have remained anonymous, have posted allegations on two blogs, including side-by-side comparisons of similar text. The editor of Acta Medica Marisiensis has questioned Hobai’s motives and called his allegations “vicious.” Editors at three other publications have acted on PANDORA’s allegations, but not always in ideal fashion: They often simply removed plagiarized papers from their websites with no notification or reason, leaving no trace that the paper ever existed. That practice runs contrary to guidelines issued by the Committee on Publication Ethics, an international group that advises journal editors.

PANDORA is just one example of bands of researchers taking it upon themselves to clean up the literature. Some are subject specific, and many, such as PubPeer, are international. Others—such as VroniPlag, which began as a way to crowdsource suspected plagiarism in theses in Germany, and PANDORA—are country specific.

By the second measure—retractions per paper published—Iran tops the leaderboard and Romania drops to second (among countries that published at least 100,000 papers from 2003 to 2016; see graphic, left). Iran’s position may reflect several high-profile scandals involving fake peer review. But this analysis may overstate Iran’s retraction rate. That’s because the incidence was calculated using a tally of published papers developed by the U.S. National Science Foundation. That count includes only papers published in English. If it also included papers published in Farsi—Iran’s national language—the rate could change.