sciencenewsnet.in

Biden Leads Trump—In Online Searches for Yard Signs

Online searches for yard signs showing support for Joe Biden outpace those for Donald Trump yard signs, shows a new analysis by researchers at New York University’s Courant Institute of Mathematical Sciences. 

Their study, which tracks Google searches for queries containing terms related to “Biden yard sign(s)” and “Trump yard sign(s)” from the beginning of March through the end of October, shows consistently more interest for the Democratic nominee than it does for the president. 

“This is another way to look at candidate support in ways that go beyond typical measurements, such as polls or campaign contributions,” explains Anasse Bari, a clinical assistant professor in computer science at the Courant Institute and the study’s senior researcher. “By mining data in innovative ways, we can perhaps provide insights that other metrics overlook.” 

In the U.S. overall, Biden led Trump in yard-sign searches by 28 percent.

Bari and his colleagues note these results stand in contrast to 2016, when Trump’s yard sign search interest topped that for Hillary Clinton, an advantage that stretched across the country.

This year, searches for Biden signs led those for Trump ones in 35 states, as well as in Washington, D.C. Not surprisingly, the best state for Biden sign searches was his home state of Delaware. However, in addition to New Hampshire and New Mexico, his next best state (where he also led the president) was one that historically votes for the Republican candidate: Idaho. By contrast, Trump searches outpaced those for Biden in the following states, all of which typically back the GOP nominee: Kentucky, Nebraska, Indiana, and South Carolina.  

Interestingly, the findings also showed narrow margins for Biden in several swing states—results that align with some public opinion polls—while others showed differences that are bigger than what surveys are reporting. Among the most notable results were the following: 

The other authors for these analyses included Courant researchers Alankrith Krishnan, Aashish Khubchandani, Shailesh Apas Vasandani, Julia Damaris Yang, Daniel Rivera, Vikas Nair, Laurence Bugasu Ininda, and Matthias Heymann. 

The study relied on a sample of Google searches used in Google Trends, which yielded a dataset designed to be representative of all Google searches.

The research was funded, in part, by Amazon AI and Amazon Web Services. 

# # #