The world of online knowledge-sharing has an intriguing yet concerning issue: the persistence of retracted scientific papers on Wikipedia. A recent study, led by Haohan Shi, a Ph.D. candidate at Northwestern University, has shed light on this problem. The findings reveal a slow and uneven response from the Wikipedia community when it comes to addressing post-publication knowledge failures, despite Wikipedia's reputation for agility in other areas.
One key insight is the significant time lag between a paper's retraction and the correction of its citation on Wikipedia. With a median correction time of 3.68 years, it's clear that the platform struggles to keep up with the ever-evolving nature of scientific knowledge. This lag can have serious implications, as it allows misinformation to spread and persist, potentially influencing millions of readers.
The Impact of Retracted Citations
Retracted papers, once cited on Wikipedia, can have a lasting impact. Many of these papers are found on specialized pages, but some also appear on general-interest pages. This means that ordinary readers, unaware of the retraction, may accept the paper's claims as factual. The study's authors emphasize the potential for these citations to "quietly misinform" a vast audience.
Variation Across Academic Fields
Interestingly, the time it takes to correct a citation varies depending on the academic field. Physical science citations are corrected relatively quickly, with a median time of 228 days. In contrast, life sciences and health sciences citations take much longer, with median correction times of 2,720 and 1,846 days, respectively. This variation highlights the challenges Wikipedia editors face in keeping up with the diverse and rapidly changing nature of scientific research.
Addressing the Problem
The study's authors suggest that an automated approach, using tools like RetractionBot, could help accelerate the correction process. These tools can continuously monitor citations and flag retracted papers, aiding Wikipedia editors in their task. Additionally, browser extensions like LibKey Nomad can provide users with warnings when they encounter retracted papers.
While the number of retracted papers on Wikipedia is relatively small (less than 0.1% of all scientific papers cited), the impact of these citations should not be underestimated. As Laura Kurek, a Ph.D. candidate at the University of Michigan, notes, "People keep tending to [Wikipedia] like a garden, and over time things do get caught." However, the study's authors argue that the overall rate of retractions doesn't indicate Wikipedia's success in avoiding bad science, but rather reflects the rarity of retractions in the scientific world.
A Call for Action
This study serves as a call to action for the Wikipedia community and scientific researchers alike. It highlights the need for improved systems to identify and address retracted papers, ensuring that Wikipedia remains a reliable source of information. As Haohan Shi points out, expecting Wikipedia editors to manually monitor every citation is unrealistic, making automated tools a necessary solution.
In conclusion, the persistence of retracted scientific papers on Wikipedia is a complex issue that requires a multi-faceted approach. By combining human vigilance with technological advancements, we can work towards a more accurate and trustworthy online knowledge base.
