English
Arabic
Chinese (Simplified)
Dutch
Esperanto
French
German
Greek
Italian
Japanese
Korean
Portuguese
Russian
Spanish
UrduThe findings of the study, conducted by researchers at New York University’s Courant Institute of Mathematical Sciences, are in contrast to post-debate polls, which showed Biden winning the exchange by double-digit figures.
In addition, the results from Thursday differ from the team’s analysis of last week’s town hall events, in which Biden outpaced Trump in both Twitter mentions and Google searches; however, they are more consistent with its study of the first presidential debate, in which the president led in Twitter mentions while trailing in Google searches.
Notably, Biden received approximately 6 percent more Twitter mentions leading up to the second presidential debate, but during and after the debate, Trump led by 8 percent—a lead he held through the two hours after the debate’s conclusion.
“While a majority of viewers saw former Vice President Biden winning the debate, related online activity told a much different story,” observes Anasse Bari, a clinical assistant professor in computer science at the Courant Institute and the senior author of the study. “President Trump, who has made Twitter a cornerstone of his communications, received more, and better, attention there than did his opponent.”
In addition, Biden’s statement that he would “transition away from the oil industry” appears to have prompted a flurry of related Twitter activity. Tweets mentioning Biden were led by “wants end oil industry,” followed by “Trump [got] Biden” [to admit he wants to end oil industry] and “energy industry workers”—a reference to the potential impact on jobs as a result of a decline of the petroleum industry.
The team calculated Twitter activity and Google searches in the two hours before, during, and the two hours after the debate, held in Nashville, Tenn.
The researchers also examined Twitter users’ attitudes toward the nominees using Sentiment Analysis algorithms, a commonly used tool that processes natural language and deploys artificial intelligence to interpret and categorize emotions. Their results showed notable contrasts from the first debate and the town halls:
- The first debate: For both Trump and Biden, positive mood (approximately 5 percent), negative mood (approximately 31 percent), and neutral mood (approximately 64 percent) remained nearly constant before, during, and after the first debate, indicating that the encounter did not have much effect on public opinion among Twitter users.
- Town halls: Both candidates received almost the same number of tweets expressing negative sentiment (approximately 30 percent of the study’s tweets), or approximately 4 percent less than from the first debate. The number of tweets expressing positive sentiment after the town hall in Trump tweets rose by approximately 54 percent, as compared to the first debate, while the number of tweets expressing positive sentiment in Biden tweets rose by approximately 37 percent as compared to the first debate.
- The second debate: The number of positive Trump tweets increased by 22 percent before, during, and two hours after the debate while Biden’s percentage of positive tweets remained the same from the period of two hours leading up to the debate, during the debate, and two hours after the debate.
On Google, “Joe Biden” was searched over twice as much as “Donald Trump” in each of the 50 states—a result consistent with the first presidential debate and the town hall events. In fact, this gap grew slightly during the second debate–from 7:35 p.m. to 11:35 p.m. EDT, Joe Biden made up on average 77 percent of search volume including the two candidates’ names, compared to 76 percent during the first debate and 72 percent during last week’s town hall event.
The focus of these searches centered on Biden’s stance on fracking, the president’s criminal justice policies, and the development of a vaccine for COVID-19—a search not linked specifically to either candidate.
The team’s studies this fall used computer programs to collect, in near-real-time, approximately three million tweets related to the two presidential debates and town halls. Its algorithms collected and filtered tweets specifically pertaining to Trump and Biden. It also studied online activity surrounding the vice-presidential debate.
The other authors for these analyses included Courant researchers Alankrith Krishnan, Aashish Khubchandani, Julia Damaris Yang, Daniel Rivera, Vikas Nair, Shailesh Apas Vasandani, and Matthias Heymann.
# # #
