Social media is used to amplify harmful content. The Integrity Institute, an advocacy group, is trying to measure exactly how much, and on Thursday it began publishing results that it plans to update each week through the election.
The institute found that a well-crafted lie will get more engagements than usual, and that features of social media sites contribute to the spread of misinformation.
The analysis shows that the great misinformation amplification factor is due to the fact that people are able to re-tweet posts easily. TikTok is a Chinese-owned video site that uses machine-learning models to predict engagement and make recommendations.
Jeff Allen, founder and chief research officer at the Integrity Institute, said that each platform has different mechanisms for virality. There are more ways to spread misinformation on the platform.
The institute used the engagement of previous posts that were not flagged from the same accounts to calculate its findings. In September, it analyzed nearly 600 fact-checked posts on a variety of topics, including the war in Ukraine and the upcoming elections.
According to the sample that the institute has studied so far, Facebook had the most instances of misinformation but amplified them to a lesser degree because it required more steps to share a post. Some of its newer features are more prone to amplify misinformation.
The institute found that the amplification factor of video content was close to TikTok. According to the institute's calculations, the platform's Reels and Facebook Watch both rely heavily on algorithmic content recommendations.
Facebook had the highest amplification rate. The institute said there wasn't enough data to make a statistically significant estimate.
The amplification will be tracked by the institute to see how it changes as the elections approach. The institute said that misinformation is more likely to be shared.
The report said that amplification of misinformation can occur if misinformation narratives take hold. Platforms can fall if they implement design changes around the event that reduce the spread of misinformation.