1. Startsiden
  2. sentimenttest_ai_2019

Fake news about fake news

The researchers thought that falsehoods spread just as easily as quality information in social media, but they boomed in their study. The study received a lot of attention when it was published in June 2017.

In a time when fake news is rife, many people have confirmed what they thought - that social media is as easy to spread as low-quality and high-quality information.



Error information that gets viral spread and threatens our democracy, as the researchers put it in the scientific article in Nature Human Behavior .


But the findings turned out not to agree.



Withdraws article

Researchers are now withdrawing the article on social media on their own initiative. According to the website Retraction Watch.


Scientists from China, the United States and Yahoo company created a model for information flow and used Facebook as an example.


They argued that although we may prefer good quality of information, it is too much to deal with in the information flow. We can't distinguish good from bad.


Now, in a statement in Nature Human Behavior , they write that they got errors in one of the analyzes due to a software error.


Used incorrect data

In another analysis, they used the wrong research data as a starting point.


When they had corrected the error and calculated correctly, they realized that the original conclusion was not correct.


On the contrary, the model predicts that people disseminate quality information to a much greater extent. Low quality information is less popular.


While other researchers have found that fake news on Twitter is easier to follow than true news.


- Do double check

Gordon Pennycook at the University of Regina, Canada, has been researching the phenomenon of fake news. He tells Retraction Watch that this reminds researchers that they need to double-check - and triple-check - their analyzes.


- The pressure to publish is very strong, but no researcher wants to be in this situation.


He praises the researchers for having taken the affair himself and pulled his article from the journal.