Haunschild & Bornmann/arXiv.org
While using bibliometric techniques to measure how disruptive research papers are to their field of study, Robin Haunschild and Lutz Bornmann stumbled across a strange phenomenon.
Just under 45,000 academic papers contained citations to themselves, they found. Haunschild and Bornmann — both information scientists at the Max Planck Institute for Solid State Research in Stuttgart, Germany — found these “paper self-citations” in journals indexed by Clarivate’s Web of Science since 1980.
Some 7,943 different journals had at least one self-citing paper, the researchers report in their study, posted on arXiv.org earlier this month. Eight journals alone covered 10% of the sample papers, and 129 publications covered the top third. More than 31,000 of the papers appeared under the ‘article’ category in Web of Science, followed by just over 6,000 listed as ‘corrections’ and just under 2,500 as ‘reviews.’
Stéphane Bonhomme, an economist at the University of Chicago who edits Quantitative Economics — the journal with the highest number of paper self-citations in study, 165 self-citing papers out of 416 studies in total — says he was surprised to see the numbers for his publication.
But after taking a closer look, Bonhomme said, he discovered that many paper self-citations were coming from the authors referencing their own appendices and supplementary materials that were published under different DOIs — a fairly common practice in economics, particularly with papers with long appendices or supplementary material.
“It’s supplementary information which is completely related to the paper, but it’s not part of the published paper,” Bonhomme told us. “My conjecture is that this is most or all of the [paper] self-citations for this particular journal.”
Authors based in the United States had the highest number of paper self-citations at 13,128, followed by those based in China at 5,363, and the United Kingdom at 4,493.
From the study, it seems that around half the papers self-citations in the sample are the result of database errors, said Molly King, a sociologist at Stanford University in California. King published a large-scale analysis finding men cite themselves an average of 56% more than women do.
King said she would like to see a larger sample that explores how many studies contain true paper self-citations versus how many are the result of mistakes in the database. “That’s the remaining question in my mind,” she said. “How much of this is a Web of Science problem, and how much of this is actually happening in the real world?”
Some true self-citations may indicate researchers or journals trying to game the system to boost their citation counts, said Bornmann, who is also a sociologist of science at the Max Planck Society in Munich. Bornmann said he’s not aware of any guidelines for authors or journals on paper self-citations. The next step would be to explore the reasons behind this trend, he said.
“Given the very low percentage of it happening, it’s possible there’s a few scholars out there who have uncovered this as an approach to artificially inflating their citation indices,” King said. “But it does not seem to me like this is anything approaching a widespread gaming of the system.”
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on X or Bluesky, like us on Facebook, follow us on LinkedIn, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].