Retraction Data Can Bring More Insights and Implications for Not Just Authors and Their Institutions, but Funders, Policy-Makers, and Editors

Although retractions are commonly considered as negative outcomes, the fact remains that they play a positive role in the academic community, for instance, help scientific enterprise perform its self-correcting function; become lessons learned for future researchers; represent social responsibilities or let open review communities offer better "monitoring services" in keeping problematic studies in check. This study provides retraction data, which is believed to give useful insights into retraction and its powerful function. By using RetractionWatch data, a database built based on SQL Server 2016, and some home-made AI, a data set of 18,603 retractions from 1753 to 2019 February covering 127 research field was contributed. The results show a long way of retraction with the popularity of retraction practices since 1999, and the burden in 2010; IEEE, Elsevier, and Springer account for nearly 60% of all retracted papers globally with the record belongs to IEEE even though it is not the organization that publishes most journals; a paper could be retracted for diverse reasons but "fake peer review" becomes a major one