Just in the last few months, we put out 37 from Proceedings of the Institution of Mechanical Engineers and 21 from Concurrent Engineering. And there are more to come. While we don’t celebrate this type of action, the news is not all bad. The high numbers of retractions at times reflect a problem of industrialized cheating, but also, as in our case, a belief that rigorous scholarship – robustly reviewed by researchers who are experts in their fields – can and should improve the world.
Retractions must be supported as an essential part of healthy science. Sleuths should be compensated and given access to tools to improve the hunt for errors and fraud — not face ridicule, harassment and legal action. Publishers could create a cash pool to pay them, similar to the ‘bug bounties’ that reward hackers who detect flaws in computer security systems. At the same time, institutions should appropriately assess researchers who honestly aim to correct the record. Retractions should not be career killers — those correcting honest errors should be celebrated.
JACI is the journal with the poorest experience that I ever encountered as an author and as a reviewer. The editors never adequately responded to numerous errors in an earlier paper where I sent a long letter describing all details.
Most of the papers that cite discredited COVID research in The Lancet and The New England Journal of Medicine don’t mention that the studies have been retracted. The infamous studies relied on health-record analyses from a company, Surgisphere, that declined to share its raw data for an audit. Science looked at 200 academic articles that cite the Surgisphere papers and found that 52.5% — including some in prominent journals — failed to mention the retractions.
Having a discussion recently with the editor of a small journal, I think it is worth to repeat that withdrawal of a paper is not always a sign of deliberate fraud as it could be just an honest error. Science had an analysis last year of the retractiondatabase that clearly showed the relationship.