Using Facebook data to predict hate crimes against refugees in Germany

Summary

This project investigated the use of Facebook to predict crime against refugees in Germany. It analysed the content on the Facebook page of the far-right party “Alternative for Germany” (AfD) – specifically the posts from the party and users, and their locations –, and compared this to 3,300 incidents of crime against refugees in Germany. Additionally, the team analysed 300 major internet disruptions that cut off users and the AfD from Facebook to test the relationship between social media usage and incidents committed against refugees in the corresponding municipalities.

Results

The results of this project show that there is a clear link between social media and hate crime. The team found that with spikes in anti-refugee sentiment on Facebook, attacks on refugees are disproportionately more likely in areas with high social media usage. Simultaneously, disruptions of Facebook – such as local internet outages – reduces local hate crimes, particularly in areas with many AfD users. However, the project does not argue that social media causes hate crime; yet, it produced considerable evidence that social media plays a propagating role in translating malicious online content into offline violence. In sum, this project demonstrated that social media is a valuable new data source for a better understanding of migration-related phenomena and informing migration policymaking. The research approach can be applied with other national contexts as well, depending on the penetration rate of Facebook in the local society.

(Image: © IOM, Muse Mohammed)