Working Paper: NBER ID: w24631
Authors: Yuriy Gorodnichenko; Tho Pham; Oleksandr Talavera
Abstract: This paper studies information diffusion in social media and the role of bots in shaping public opinions. Using Twitter data on the 2016 E.U. Referendum (“Brexit”) and the 2016 U.S. Presidential Election, we find that diffusion of information on Twitter is largely complete within 1-2 hours. Stronger interactions across agents with similar beliefs are consistent with the “echo chambers” view of social media. Bots have a tangible effect on the tweeting activity of humans but the degree of bots’ influence depends on whether bots provide information consistent with humans’ priors. Overall, our results suggest that the aggressive use of Twitter bots, coupled with the fragmentation of social media and the role of sentiment, could contribute to the vote outcomes.
Keywords: Social Media; Public Opinion; Bots; Brexit; US Election
JEL Codes: D72; D83; D84
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Stronger interactions among agents with similar beliefs (C92) | Echo chamber effect (C92) |
| Pro-Leave users react faster to pro-Leave messages (C92) | Pro-Remain messages (F24) |
| Information diffusion on Twitter (D85) | Information rigidity (L15) |
| Social media sentiment (Z13) | Public opinion (D72) |
| Bots (Y60) | Human tweeting activity (Z13) |
| Bots' messages with a given sentiment (D90) | Human messages with the same sentiment (C92) |