According to data from Mendeley, 91% of retracted articles in the study were read, while 84% were read and cited. This suggests that there is a correlation between readership and citations. 

Persistence of Misinformation from Retracted Scientific Articles

Articles are often cited after being retracted. When the retraction is due to faulty data or other issues that impact the conclusion of the study, this leads to the propagation of misinformation in the scientific community. 


University of Michigan - Ann Arbor

The study presented in this paper looked at articles in the biomedical and medical fields that were published between 2010 and 2014 that were retracted prior to 2017. The study looked at readership of the articles, citations before and after the article was retracted, and social media shares. The purpose of the study was to see how the misinformation in retracted articles spreads and continues to spread after it has been removed from the literature. 

Study on Retracted Articles

Bar-Ilan, J., & Halevi, G. (2020). Retracted articles: The scientific version of fake news. In R. Greifeneder, M. Jaffé, E. J. Newman, & N. Schwarz (Eds.), The psychology of fake news: Accepting, sharing, and correcting misinformation (pp. 47–70). London: Routledge. 

Retracted articles: The scientific version of fake news

1. Contributing to the spread of misinformation, articles that have been retracted often become free 
- Access becomes easier to the public
- Articles continue to be cited after retraction
2. Time gap between publication and retraction has been shorted due to technology 
- Little over a year (now) versus estimated 10 years (past)
- However, information spreads faster due to social media 
- A year is still enough time to influence medical and biomedical scientific process on many levels
3. Misinformation is not just incorrect information
- Selective inclusion or manipulation of the data affirms the author’s claims 
4. Debates continue to happen about retracted articles 
- Even after the science has been proven fault, people are more likely to believe information that supports their opinions
- Information that has been proven false is not always supported 

Key Takeaways from the Retracted Articles Study

The authors recommended that readers read each article critically and to look for sponsored content around the articles, as those can indicate conflicts of interest. They advocated for obviously labeling retracted papers and not making them freely available in online databases. 

Tips for Reading Scientific Articles

Post-Retraction Citations

- Scopus
- Altmetric.com
- PlumX
- Mendeley

Databases to Find Scientific Article Metrics

The amount of time between publication and retraction has decreased from 2.67 years to 1.1 years. This indicates that the methods of fact-checking and reviewing published articles have gotten more efficient. This is a result of access to plagiarism-detecting technology and other technological advances. 

Timeline for Retracting a Scientific Article

Data from PlumX and Altmetric.com shows that Twitter is the dominant social media platform for sharing the retracted articles. It is also the platform where the most meaningful discussion of them takes place. Facebook is the second most popular platform for sharing the articles. 

Where are Retracted Articles Being Shared?

A notable example of retracted research is the false claim that routine childhood vaccines cause autism. Although several peer-reviewed studies initially supported this link, large-scale epidemiological research demonstrated no connection and affirmed vaccine safety. The original studies were retracted after it was revealed that the lead researcher had a significant financial conflict of interest in establishing the link.

Learn Before

Related

Learn After