A vast 74 percent of Americans plan to watch the debate. And while eight in 10 say it won’t change their minds, that leaves more than enough to shift the balance in an increasingly closely fought contest, with unprecedented levels of qualms about both major-party candidates.
Trump, in particular, is running competitively despite persistent doubts. Around six in 10 Americans continue to see him as unqualified, untrustworthy, temperamentally unsuited or insufficiently knowledgeable of world affairs to serve effectively as president. Yet he’s capitalizing on strength in his core support groups and on Clinton’s own weaknesses, including concerns about her health.
In all, 44 percent of likely voters say they’d vote for Trump if the election were today, numerically his best since spring. Forty-six percent prefer Clinton, unchanged from an ABC/Post poll early this month and virtually unchanged since June. The 2-point gap between them is not significant, given the survey’s margin of sampling error. The race has closed from an 8-point Clinton lead in early August.
Watch the first presidential debate at 9 p.m. ET. Full live coverage and analysis of the debate will begin on ABCNews.com/Live at 7 p.m. ET.
Support for third-party candidate Gary Johnson slipped to a new low, 5 percent in this poll, produced for ABC by Langer Research Associates, with the biggest departure among Republicans and GOP-leaning independents -- thus mainly benefiting Trump.
Clinton, for her part, has advanced to a 25-point lead among college-educated white women, 57-32 percent, extending her advantage in this group from 10 points early this month. Romney won them by 6 points in 2012.
The division between non-college white men and college-educated white women is its widest by far in exit poll data back to 1980. The outcome among these two groups, in turnout and preference alike, may well be a decisive factor in the election results.
Further, Clinton now leads by 19 points among women, Trump by 19 among men -- a yawning 38-point gender gap, triple the average gender gap in exit polls dating to 1976.
Other contours of the race also are without precedent, including both candidates’ personal unpopularity. Fifty-nine percent of Americans see Trump unfavorably; 55 percent say the same about Clinton. That said, Clinton’s unfavorable rating is a point from its record high, set in late August, while Trump’s is well down from its peak this year, 70 percent in mid-June.
Further, while neither candidate is seen as particularly honest and trustworthy, Trump has gained an advantage on this gauge among likely voters. Fewer than half, 45 percent, see him as honest and trustworthy; but it goes lower, to 36 percent, for Clinton.
Clinton pushes back on other key factors. Fifty-five percent of likely voters say she has the personality and temperament to serve effectively, 57 percent say she’s qualified for the office and 68 percent say she knows enough about world affairs. These drop to 41, 47 and 42 percent, respectively, for Trump -- albeit his best scores on these qualities to date.
Much of the race reflects two fundamentals: Partisanship and turnout. Among likely voters who are Republicans or GOP-leaning independents, 87 percent back Trump; among Democrats and Democratic leaners, 84 percent favor Clinton. The strong pull of partisanship is one factor that boosts Trump despite compunctions about his suitability.
Trump also may benefit from greater enthusiasm among his backers than for Clinton among hers. The upshot, in any case, is that Trump’s weakness on his qualifications, while still present, eases somewhat as we move from all adults to likely voters. He’s seen as unqualified for office by 57 percent of all adults, vs. 51 percent of those likely to vote.
Among other groups, independents also may prove key; their preference was +2 points for Clinton a few weeks ago vs. +5 for Trump in this poll. While that’s within the margin of error, independents customarily are more movable than other likely voters, given that they’re less rooted in partisan predispositions. They haven’t been swing voters (i.e., supporting the winner) in the last two elections -- but where they land in six weeks may matter.
Health, Taxes and Prejudice
While questions about Clinton’s health are a comparative negative, she largely gets a bye on transparency in this regard. Sixty-four percent of Americans think Clinton was justified in keeping her recent diagnosis of pneumonia private until she fell ill in public. Trump gets much more criticism, by contrast, for not releasing his tax returns; 63 percent call this unjustified.
Fifty-nine percent think Trump is trying to win support by appealing to people’s prejudices about groups other than their own. That’s mitigated, somewhat, by the fact that 45 percent say Clinton is doing the same. Moreover, 64 percent think it’s unfair to describe a large portion of Trump’s supporters as prejudiced against women and minorities, as Clinton did in remarks at a fundraiser (and for which she subsequently apologized).
Among other results:
• Despite reports of a stronger Democratic get-out-the-vote organization, voter contacts at this point look closely matched. Among likely voters, 30 percent of Clinton supporters say they’ve been contacted by her campaign, while 32 percent of Trump supporters say they’ve heard from his side.
• President Barack Obama has 55 percent approval, a continued help to Clinton. That said, Americans divide on whether or not they’ve gotten better off financially under Obama’s presidency -- 29 percent say yes, but 25 percent say they’re less well off, and a plurality, 45 percent say they’re in the same shape financially as when he took office. These views cut strongly to vote: Better-off likely voters overwhelmingly for Clinton; worse-off for Trump; the rest much more divided.
Among five issues offered in this survey, the economy leads as most important, cited by 32 percent, followed by terrorism, up a scant 4 points to 23 percent after the New York/New Jersey and St. Cloud attacks. Corruption in government, immigration and “law and order” follow. Economy voters favor Clinton by 62-27 percent; those focused on terrorism, by contrast, back Trump by 58-38.
Separately, 78 percent are worried about a major terrorist attack, 37 percent “very” worried, similar to a reading last year but high by historical standards.
The survey finds a substantial lack of differentiation between candidates on issues; they score evenly among likely voters in trust to handle terrorism, “looking out for the middle class,” health care and ethics in government alike. Trump is +5 on the economy, Clinton +6 on immigration issues, neither statistically significant.
Clinton is +9 in trust to handle an international crisis, much less of a margin than she’s enjoyed recently. That jumps to a 22-point lead in trust to handle social issues such as abortion and same-sex marriage.
The candidates’ close ratings on the issues reflects the contest in vote preference. More voters say they’re focused on issues than on the candidates’ personal qualities, and in this group the race stands at 46-45 percent. Among those more concerned with candidate attributes, it widens to 52-39 percent. (Trump leads among those concerned equally with both factors.)
With the first debate upon us, Trump has an opportunity to boost poor perceptions of his qualifications, temperament and world knowledge. Clinton can try to capitalize on her advantages on these, as well as to address the health issue. Both candidates may seek to address their relatability and bolster their perceived trustworthiness. Both also may try to reinforce their opponent’s perceived negatives, as well as to advance their appeal to their sharply differing core support groups in an effort to drive turnout.
Clinton has more than a diminished advantage in vote preference to worry about. She also labors under greater expectations that she’ll win the debate, 47-33 percent. And much is at stake: Seventy-eight percent of Americans say that who wins the 2016 election matters a great deal to the country’s future. Among likely voters, that rises -- to a near-unanimous 93 percent.
A table of vote-preference results among key groups follows.
This ABC News/Washington Post poll was conducted by landline and cellular telephone Sept. 19-22, 2016, in English and Spanish, among a random national sample of 1,001 adults, including 651 likely voters. Results have a margin of sampling error of 3.5 points, including the design effect, for the full sample, and 4.5 points for likely voters. Partisan divisions are 33-23-36 percent, Democrats-Republicans-independents, in the full sample, and 37-27-28 among likely voters.
The survey was produced for ABC News by Langer Research Associates of New York, N.Y., with sampling, data collection and tabulation by Abt-SRBI of New York, N.Y. See details on the survey’s methodology here.