More than one in five chemistry researchers have admitted to deliberately adding information they believed to be incorrect into manuscripts during peer review to get their papers published. The survey of 982 authors—each with at least two papers published in Royal Society of Chemistry or American Chemical Society journals between 2020 and 2023—was published in Accountability in Research, offering a rare look at how chemists respond to errors in their own work and in others’ studies.
When asked if they felt compelled to modify their manuscript with text they thought to be wrong, 22% of respondents said yes. According to Frédérique Bordignon, a bibliometrician and research integrity officer at École des Ponts ParisTech, one motivation was to speed up the review process: “It’s a bit concerning.”
The survey found that 88% of respondents had discovered errors in the papers they read, and nearly 4 in 5 took some form of action. And whilst most agreed that formally correcting the scholarly record is the best practice, Bordignon noted that “they prefer off‑the‑record activities like discussion with other peers.”
Forty‑two per cent said they raised issues in private conversations with colleagues, around a third cited problematic papers in their own publications to highlight the errors, and about 30% mentioned them to students during training. Other responses included 28% contacting authors directly to encourage retraction or correction, 22% choosing to ignore the error and never cite it, 13% writing letters or notes to journals, 4% launching replication projects, 4% publishing formal refutations, and 2% commenting on PubPeer, a site where scientists discuss work.
The survey also found that 56% of researchers believed errors should always be corrected, a figure that rose to 82% when considering their own studies. A third of respondents said corrections were necessary only if the error altered the paper’s conclusions.
According to François‑Xavier Coudert, a computational chemist at France’s National Center for Scientific Research who is not involved in the study, all mistakes should be addressed. “Identifying errors and their cause is often a lengthy and effort‑consuming process, so the results should be made available to all readers. This is the only way to have more reproducibility in research.”
Bordignon said greater transparency is needed in the field. “The problem is, it’s sometimes difficult to face the consequence of being critical of someone else,” she explained. “I think we should encourage researchers to be more open to critiques and be more open to flagging the research of others as well because that’s part of science.” She added that chemists should make more use of PubPeer, a view echoed by Coudert, who said, “PubPeer is one platform where that can happen, but it is not necessarily the only one,” before adding that research should be treated like “living” documents, updated dynamically as new results emerge.
“This is exactly what happens with preprints, where all readers know that the ‘final’ published version of the article will be different from the preprint version.”
For more information: Bordignon, F. (2025). On and off-the-record correction practices: A survey-based study of how chemistry researchers react to errors. Accountability in Research, 1–14. https://doi.org/10.1080/08989621.2025.2564106
