Following up on the previous post about a study from the University of Bern tracking errors in academic papers (linked above), it's worth highlighting a recent article in Nature that offers an in-depth look at Argos, a newly launched science-integrity platform. Argos aims to identify potentially problematic research papers by assigning risk scores based on factors such as the authors' publication history and the extent to which the work cites previously retracted studies. https://www.nature.com/articles/d41586-024-03427-w
However, Argos raises several concerns despite its intention to flag high-risk papers. A high-risk score does not automatically indicate that a paper is of poor quality or fraudulent. Its heavy reliance on retraction history can result in false positives, potentially penalizing authors for past associations rather than reflecting the actual quality of their current work. Furthermore, the use of open data sources, such as Retraction Watch, increases the risk of misidentifications, particularly with common author names. Labeling papers as high-risk without thorough investigation may lead to misguided actions. Additionally, Argos's commercial interests could present a conflict of interest, fostering excessive retractions to protect reputations rather than ensuring genuine integrity.
Declaration of Competing Interests - In 2019, I authored a critique of Retraction Watch in a post entitled 'Public Shaming in Academia or Sharia Law in Academia?' https://pacheco-torgal.blogspot.com/2019/10/public-shaming-in-academia.html
