Hillary Clinton is favored to become the next president of the United States. Yet many in Donald J. Trump’s camp remain undaunted, arguing that hidden supporters will carry him to the White House. As the polls have narrowed in the last few days, the question of whether pollsters could be missing Trump voters has become paramount. Do these unseen supporters exist?
Working with the students in our undergraduate public opinion course at Cornell, we recently commissioned a nationally representative survey of 1,461 adults (carried out by the market research company GfK from Oct. 5 to 25). To our surprise, we found previously undiscovered Trump supporters. But we also found that they are likely to stay home on Election Day.
Like many surveys, ours asked, “If the presidential election were being held today, would you vote for Hillary Clinton, the Democratic candidate; Donald Trump, the Republican candidate; other; or do not intend to vote?” Consistent with most polls during this period, Mrs. Clinton had considerably more support than Mr. Trump. However, more than 20 percent of respondents indicated that they did not intend to vote, supported another candidate or did not want to answer the question. In other words, just weeks before the election, a substantial portion of the electorate was unwilling to commit to one of the leading-party candidates. If hidden Trump supporters existed, they would have been among this uncommitted group.
At first glance, the hidden Trump supporters are indeed hidden. The proportion of uncommitted voters that identified as or leaned Democrat was slightly above the proportion that identified as or leaned Republican (49 percent versus 47 percent). These numbers contradicted the “hidden Trump supporters” argument. According to party affiliation, if the currently uncommitted decided to vote for a leading-party candidate, if anything we would expect a slight benefit to Mrs. Clinton.
But this wasn’t the whole story. It turned out that despite the prevalence of Democrats among the uncommitted, this group appeared to favor Mr. Trump.
One question on our survey, posed by our student Brianne Adler, provided evidence of hidden Trump support. That question asked whether respondents would support a proposal to increase penalties for undocumented immigrants who overstay their visas. But for half of the respondents, the question referred to “a proposal.” For the other half, it referred to “Donald Trump’s proposal.” Because chance determined the version respondents received, any differences could be attributed to the mention of Mr. Trump’s name.
And this mattered. Among those intending to vote for either Mrs. Clinton or Mr. Trump, support for the proposal dropped to 57 percent from 70 percent when Mr. Trump’s name was attached to it. This makes sense, because (as we saw above) the majority of these respondents indicated they would vote for Mrs. Clinton. However, the opposite pattern emerged among uncommitted voters. Support for the proposal increased to 66 percent from 62 percent when it was described as a Trump proposal. This is what we would expect if a substantial proportion of uncommitted voters harbor support for Mr. Trump.
A similar pattern emerged for a question posed by another student, Alexander Rauter: “If you had to choose, which presidential candidate do you find to be more truthful: Donald Trump or Hillary Clinton?” Among those who expressed support for either Mrs. Clinton or Mr. Trump, a greater portion deemed Mrs. Clinton as more truthful (53 percent to 46 percent).
Again, this is not a surprise. Clinton supporters outnumbered Trump supporters, and nearly everyone who planned to vote for Mrs. Clinton viewed her as more truthful and nearly everyone who planned to vote for Mr. Trump viewed him as more truthful. However, those who remained uncommitted to either major candidate (or to voting) were more likely to view Mr. Trump as more truthful, offering further evidence that he has more support than Mrs. Clinton among the uncommitted (46 percent to 40 percent, with 14 percent refusing to respond).
A substantial portion of the public remains uncommitted to voting for either candidate. Consistent with the notion of hidden Trump supporters, this group views Mr. Trump more favorably than Mrs. Clinton. The real question, though, is will enough of these hidden supporters emerge on Election Day to usher him into the White House? Such an outcome would surprise many and send shock waves through the polling industry. A final question posed by our student Hilary Lemonick suggests that this outcome is unlikely.
The question read, “If a person believes that both of the leading candidates have flaws, is it better to vote for the least flawed candidate or not to vote?” Not surprisingly, very few (just 6 percent) of those intending to vote for either Mrs. Clinton or Mr. Trump thought it better not to vote. But among uncommitted voters, a striking 47 percent selected the better “not to vote” option.
Granted, we might expect this uncommitted group to have relaxed attitudes toward showing up on Election Day. But this is precisely the group that Mr. Trump needs to turn out if he is to have any chance of victory. Unfortunately for Mr. Trump, the respondents who appear to favor him — but not enough to say they will vote for him — are also more likely to believe that it may be better to just stay home. This result also suggests that instead of tipping the uncommitted one way or the other, late-breaking negative headlines may simply further decrease turnout among this group.
And so on Election Day, don’t be surprised if most hidden Trump supporters remain hidden.
Continue reading the main story