PROVING THE POLLS WRONG: 'The researchers' analysis of the data--two terabytes worth containing 70,000,000 websites--shows exactly when and where voters made decisions that led to the election of Donald Trump.'.

• Author: Niksa, Matt
• Position: POLITICAL LANDSCAPE

AS THE PREDICTIONS for the 2016 presidential election remind us, polling the electorate is an imperfect science. Most polls claimed that Hillary Clinton would be our next president --it seemed a foregone conclusion--and most polls were wrong, although many forecasts for the popular vote were very close, off by less than one percentage point. Election polling always has been inexact. It also has been time-consuming, expensive, and lacking the ability to measure the influence of shortlived events, like a candidate's speech, or to read the electorate of small geographic areas.

Now, a pair of Boston University researchers believe they have found an alternative, one that not only is accurate but has the potential to be faster and less expensive, can target areas as small as towns, and can measure the people's response to specific issues and events. The methodology, which correlates web browsing patterns with public opinion from polls, was developed by two professors at the College of Arts & Sciences: computer scientist Mark Crovella and political scientist Dino Christenson.

The pair worked with Giovanni Comarela --formerly a Ph.D. student at BU under Crovella --assistant professor of computer science from the University of Vicosa; Ramalcrishnan Durairajan, assistant professor in the Department of Computer and Information Science at the University of Oregon; and Paul Barford, professor in the Department of Computer Sciences at the University of Wisconsin.

Barford, who also works for comScore, Inc., a kind of Nielsen ratings of the Internet, negotiated an arrangement by which comScore provided the researchers with the web browsing histories of more than 100,000 U.S. residents over the 56-day period preceding the 2016 election. The researchers' analysis of the data--two terabytes worth containing 70,000,000 websites--shows exactly when and where voters made decisions that led to the election of Donald Trump.

It also suggests that, contrary to popular opinion, a last-minute dip in support for Clinton was not precipitated by a letter to Congress in which FBI Director James Comey revealed that the Bureau had found a new batch of relevant emails on the former Secretary of State's server. Crovella and Christenson's analysis clearly indicates that support for Clinton began to decline Oct. 25, three days before the letter was sent. "The slippage could have just been a coincidence," he says. "It may have been a small dip that would have rebounded had it...