Trump’s Hidden Supporters

In the United States, there’s an expression for when the surveys get it wrong: “The Bradley effect.” The idea dates back to 1982 when Los Angeles Mayor Tom Bradley lost the election for governor of California despite a clear lead in the polls, including exit polls conducted on the day of the election. The reason? Bradley was black, and those who didn’t intend to vote for him didn’t wish to appear racist to the pollsters.

Now, 34 years later, the Bradley effect is making a comeback with Donald Trump. Are there large groups of voters hiding their preferences for the Republican candidate, meaning that surveys placing him 1-12 points behind Clinton are wrong?

The experts aren’t convinced. And yet, this isn’t enough to dispel all doubts. Last year, Trump was doing well, but not spectacularly, in telephone polls. However, he achieved noticeably better results in online surveys, like those created by SurveyMonkey, founded by David Goldberg, husband of Sheryl Sandberg, the woman responsible for the daily operations at Facebook.

At the time, America’s most popular survey guru, Nate Silver, disregarded the online polls. And yet, they turned out to be more worthwhile than those conducted over the telephone, the reason being that online polls are anonymous, meaning that people feel greater freedom to give their opinion.

It’s not clear why people might hide their intention to vote for Trump, particularly as his followers attend meetings en masse and are the first to express, through all means possible — including a wide range of insults — their political preferences and opinions about their dissatisfaction with America.

One potential theory is that people won’t admit to voting for Trump for fear of being perceived as racist, or so that they’re not accused of committing America’s worst social offense: being a loser, a poor person.

In a country based on the idea of individual success, this is something that is unforgivable. Especially not within Trump’s principal demographic group: lesser-educated males. Minority groups, particularly African-Americans, often tend to place blame for social injustice on society, racism or political systems. White people, however, do not. Therefore, suicide rates within this demographic group are triple those of black people.

Nevertheless, pollsters have been constantly adjusting their models ever since Trump burst onto the scene, therefore confirming that there is no indication of a hidden voter base. The polls are not bulletproof, however. And even less so with a strategy like Trump’s, which places so little importance on surveys that it’s spent almost double the amount on baseball caps (approximately $3.2 million) than it has on polls (approximately $1.7 million) to date, has invested next to nothing on TV advertising, has fallen out with the entire Republican party and is now following in the footsteps of former campaign manager Corey Lewandowski in mobilizing those who do not wish to vote.

The problem is that people who don’t vote tend to be ignored by the pollsters, who, in turn, struggle to determine the stance of new political parties, as they are unable to project beyond results. It happened in Spain with Podemos. And it’s happening again in America with Trump. It’s as if he is, by himself, representing a new party, one that’s very far removed from Republican orthodoxy.

Last week, this author set out to ask homeless people in Las Vegas if they could vote, and, if they could, whom they were voting for. Although a large number of those surveyed weren’t registered to vote, Trump won by a significant majority.

“Homeless people in Las Vegas aren’t a demographic group that I would include in my surveys,” explained Gonzalo Rivero, pollster for a public opinion studies center in Washington and editor of the popular blog Politikon. And it seems he doesn’t need to. According to official figures, Las Vegas’ homeless community comprises a mere 7,000 people, whose limited votes count for nothing.

There’s just one more possibility. People aren’t lying; instead, they’re refusing to answer the surveys. “More and more people are refusing to respond to surveys,” explained Rivero. And, given the anti-establishment mentality held by a large proportion of Trump’s followers, it’s not impossible to believe that many of them are refusing to respond