The longevity of retracted scientific papers on Wikipedia is a cause for concern, according to a recent study. The research, conducted by Haohan Shi, a Ph.D. candidate at Northwestern University, reveals that retracted citations on Wikipedia persist for years, often without warning readers. This is particularly problematic given the vast reach of Wikipedia, which boasts over 66 million articles and receives approximately 508 million views daily. The study's findings highlight a disconnect between the retraction data and Wikipedia's editorial workflow, with a median correction time of 3.68 years for citations added before retraction. This delay could potentially misinform millions of people, as retracted papers may still be cited on general-interest pages. The study also notes that retractions in specialized fields, such as life sciences and health sciences, take significantly longer to correct compared to physical sciences. To address this issue, Shi suggests implementing automated tools like RetractionBot to expedite the correction process. However, the overall rate of retracted citations on Wikipedia is relatively low, at less than 0.1% of the 1.6 million scientific papers cited. This rate, according to Shi, reflects the rarity of retractions in the scientific community rather than Wikipedia's ability to avoid bad science. The study underscores the importance of timely corrections to retracted citations to maintain the integrity of Wikipedia as a reliable source of information.
