Study finds X's algorithm prioritizes engagement, shows more 'ragebait' to Democrats

Algorithm Prioritizes Engagement

A study published in the Proceedings of the National Academy of Sciences (PNAS) titled "Value misalignment of X’s feed algorithm is a reflection of value tensions in engagement" found that X's algorithm prioritizes engagement above other factors when generating a user's 'For You Page'. This optimization for engagement influences the type of content users encounter on the platform.

Increased 'Ragebait' for Democrats

The study also revealed that X's algorithm serves more 'ragebait' content to users who identify as Democrats. The exact reasons for this specific targeting are not yet clear, but it indicates a differential content experience based on political alignment.

Study Methodology

Researchers recruited 715 active X users in September and October 2024, forming a nationally representative sample matched on ethnicity, gender, and partisanship. Participants installed a browser extension to collect data from their 'For You Page' and Following feeds. The study also involved collecting a 'values inventory' from volunteers using the Schwartz Theory of Basic Values, which included 19 points corresponding to features like 'tolerance' and 'dominance', alongside self-reported political alignments.

Implications for Information Consumption

Ziv Epstein, a co-author of the paper and postdoctoral researcher at Stanford University, stated that the study was observational and aimed to prompt questions about social media feed design and its impact. He emphasized the significant power of social media algorithms in shaping information consumption and the lack of transparency regarding their operations and implications for civil society.