Lessons from Brexit: Could Our Election Polls Be Lying?

Election polls may not be taking every factor into account.

By Michael MarksNovember 3, 2016 9:36 am

A week before the United Kingdom voted whether or not to leave the European Union, it looked like those in the ‘Remain’ camp could relax – the pundits, the press, even the betting markets were sanguine about the status quo. The polls confirmed their confidence.

But upon closer examination, in the harsh light of retrospect, the polls didn’t tell the whole story. Those pointing to a Brexit win were the oddball outliers – until they proved to be correct.

Ahead of the Brexit referendum, polls showed the chance of a Brexit loss at over 85 – and in some cases 90 – percent. But those polls failed to detect the actual temperature of voters.

Fast forward to the U.S. presidential contest, ending with Election Day on Nov. 8. Most national polls show Democrat Hillary Clinton with a comfortable lead. But Republican Donald Trump would like his supporters to think that we’re looking at another Brexit.

Matthew Goodwin, professor of politics and international relations at the University of Kent in Britain, has written about the parallels between the Brexit vote and the presidential election for Politico.

“In the case of Brexit, it was mainly a failure to capture unlikely voters and, in particular, older working-class voters who had not been responding to the polls,” Goodwin says. “So we saw around 2 to 2-and-a-half million extra voters that we didn’t see at the previous general elections of the seven last polls during that campaign. Only one had Brexit ahead.”

There was also a strong enthusiasm gap, Goodwin says.

“Those who were saying they were going to vote for Brexit looked more enthusiastic and they looked more committed to the vote,” he says. “That enthusiasm gap really mattered because it led to differential turnout – it’s a fancy way of saying one side’s voters turned out in greater numbers than another side.”

The two are considerably different contests – a referendum versus an election based on the electoral college – but Goodwin says the U.S. polls could be making the same mistake. Although polls show Trump as the current political underdog, that may not actually be the case.

“You can begin to see actually some evidence of an enthusiasm gap, in particular over the last few days, between supporters of Donald Trump and supporters of Hillary Clinton,” Goodwin says. “When you look at the characteristics of Trump voters, they’re often coming from the very same section of society as Brexit voters. And they are incredibly distrustful of mainstream elites and they are incredibly motivated to have their views heard.”

So Trump finishing strongly doesn’t surprise him. But Goodwin says the big question is whether Trump will finish strong enough to cross the line in essential battleground states like North Carolina.

“That is ultimately where he’s going to be putting all of his hopes on there being a modest polling error,” Goodwin says. “Because if he can get within the margin of error within 2 to 3 points of Clinton and the polls have failed to put together a sample that captures Trump voters, then he’s got a chance. He’s got a realistic chance. The Trump presidency has moved from being implausible to being more than plausible.”

Post by Beth Cortez-Neavel.