Retraction Watch Database: A Global Resource for Monitoring Research Integrity and Retractions in Scholarly Publishing

The Retraction Watch dataset provides a comprehensive, regularly updated record of retracted, corrected, and scrutinized scholarly publications across multiple disciplines. Maintained by Retraction Watch and hosted by Crossref, this open dataset includes metadata on retractions since the 1970s and is updated daily. Here, we describe the dataset structure, scope, and potential applications for research in scientometrics, artificial intelligence ethics, and research integrity. The dataset enables large-scale analyses of trends in publication corrections, misconduct detection, and the evolution of editorial practices. By making this data openly available, Retraction Watch facilitates transparency, reproducibility, and accountability in the global research ecosystem.