China ‘marked down’ in scholar’s retraction-adjusted rankings

Chinese universities would be heavily marked down in a proposed new ranking system based on highly-cited researchers, owing to penalties linked to retraction and self-citation.

Under a proposed scoring system led by Stanford University scientist John Ioannidis, some 2,380 universities were classified based on the number of “high-impact scientists” rated in the top 2 per cent for citations in their discipline, or who had published at least 40 Scopus-indexed items in their career.

That shift in emphasis from traditional university rankings, which focus on overall research activity and citations, gives a better insight into whether “high-impact work is produced by a few individuals or spread across many”, argues Ioannidis in a bioRxiv preprint published with researchers from Elsevier and Mines Paris, PSL University.

Of the top 116 institutions covered by the ranking, 27 were from China, behind only the US (38 institutions), and well ahead of France (eight) and the UK (seven).

Harvard University had the most top-cited researchers (1,894, or 7.2 per cent of all researchers with at least five papers), followed by the University of Toronto (1,521, or 5 per cent of all researchers), while the University of Oxford was fifth placed (with 1,293 top-cited researchers, 7 per cent), just ahead of UCL (1,213, or 6.9 per cent)

Overall, China had the most institutions (410) included in the highly-cited author ranking overall, ahead of the US (344), followed by Japan (152), France (137) and the UK (117), according to the preprint.

However China’s showing would be downgraded if adjustments were made to reflect retraction rates, self-citations and publication in journals subsequently delisted from Scopus, the paper explains. Some 75 of the top 100 institutions affected by the retraction adjustment are from China, representing nearly one in five of its “highly-cited” universities.

A further nine universities in China would figure in the top 100 for excessive self-citation – a metric “dominated by Russia”, half of whose highly-cited institutions (26 in total) feature. For citations linked to Scopus-discontinued titles, China again had the most top 100 institutions for this type of adjustment, tied with India (both have 12 institutions within the top 100 on this metric).

“Several mid-size institutions predominantly from Saudi Arabia and some from Egypt, India, and United Arab Emirates appear at the very top ranks in the absence of adjustments, but almost all of them lose their top-ranks or even end up with negative adjusted numbers,” adds the paper which criticises the “common use of Saudi Arabian institutional affiliations by scientists in other countries who are paid heavy bonuses to do this”.

Adjusting the rankings to consider self-citations and retraction would help the proposed new system to consider both “potential gaming and misconduct”, explains the study, noting recent analyses that had revealed “that some institutions, especially in China, Saudi Arabia, India, Pakistan, and Ethiopia are indeed retraction hotspots”.

“Accounting for retractions, and inordinate self-citations and publications in discontinued titles can substantially alter the placement of some institutions in the adjusted rankings,” it adds, stating the “proposed percentile ranking may thus strengthen proper incentives, weaken the drive for publication and citation misconduct, and identify institutions and author groups that require more in-depth scrutiny”.

“Overall, in our baseline analysis, Saudi Arabia, China, Malaysia, Iran, India and Indonesia have cumulative penalties that on average exceed their number of top-cited authors,” states the paper on the penalties for appearing in the top 5 per cent for self-citations, retractions or discontinued papers.

Despite the focus on highly-cited authors, such metrics have also faced criticism, with 2,400 individuals excluded from Clarivate’s 2025 list amid concerns that such lists might fuel “hyper-prolific authorship” or citation gaming.

jack.grove@timeshighereducation.com