The War in Gaza May Be Skewing Polling in the U.S. Election
Polls in the U.S. have shown the presidential race between Democratic nominee Vice President Kamala Harris and her Republican rival, former President Donald Trump, in a dead heat for several weeks. Yet over the weekend, a narrative developed that voters were breaking for Harris . A poll out of Iowa by the respected pollster Ann Selzer suggested the shift was being driven by older female voters concerned about reproductive rights. This prompted a bevy of speculation as to whether it is possible pollsters might be wrong , or at least dramatically underestimating support for Harris.
As Dan Drezner summarized here , several commentators over the past few days are beginning to think that is indeed the case. One reason given by Nate Silver for the appearance of a dead heat is that, after several election cycles of embarrassing failures, polling firms may be "herding"—that is, hewing to the general consensus as to where things lie by hiding or distorting poll results that buck those trends.
Additionally, Justin Brown argues that pollsters may not have had enough time after Harris replaced President Joe Biden as the Democratic nominee to figure out how to measure the opinions of her likely voters. On top of that, over the past eight years, pollsters have compensated for failing to predict Trump's victory in 2016 and underestimating his support in his losing bid in 2020, leaving their tools and models more sensitive to Trump voters. Combined, all of that could mean that Harris' support is being undermeasured relative to Trump's.
The view from Honesdale, Pennsylvania, where I spent the weekend knocking on doors as an election canvasser for the Harris campaign, provides some insight into these arguments. My team and I observed two other reasons polls may have made the race appear closer than it really is until now, and why some polls may now be showing Harris gaining ground as the election nears.
First, pollsters may underweight their own polls' recursive influence on political views. People are sensitive to the opinions of others around them, and polls reporting those opinions can change attitudes . For example, dead-heat polls themselves can influence election preparedness by mobilizing Democratic fence-sitters and over-scheduled working people, many of whom stayed home in 2016 when experts widely reported that Clinton would easily win against Trump. Also, fewer people are responding to polls anymore, even as they are awash in reported information about what their fellow citizens—usually the angriest and most vocal on both sides who take the time to answer surveys—are presumably saying. Political science research shows this creates the appearance of a more polarized society than is actually the case.
But that perception of polarization can also drive low responses to polls and a decline in voters' willingness to publicly express their views. For example, talking to Pennsylvania voters, the widespread perception that things are a dead heat has some Harris supporters in battleground states keeping their heads down to avoid openly voicing their true intentions, out of fear of harassment from their Trump-supporting neighbors. Among the doors we knocked on, the vast majority told us they were definitely voting for Harris, and all had a well-conceived plan to do so. They knew where their polling stations were, and they had carefully read the information on their mail-in ballots. These voters were ready and willing to turn out. But even among strong and unequivocal Harris supporters, very few were willing to accept a yard sign to place on their front lawn or porch. "I don't want to advertise to the folks down the block," one told us. Indeed, we passed one house whose Harris-Walz sign appeared to have been vandalized, with a red X duct-taped over the candidates' names.
The other thing we noticed is that the main issue causing Democratic voters to equivocate on their support for Harris, among those we spoke to, is Israel's war in Gaza . Some Democratic-leaning voters are painfully ambivalent about Harris on this issue and feel that trumpeting their willingness to cast a protest vote is the best and perhaps only way to express their horror over ongoing U.S. military support for a country they sincerely believe is committing atrocities, if not genocide. As The Intercept's Jonah Valdez reports , these voters are divided between the "anguished un-decideds," "strategically anti-Trump" and "third-party / opt-outers."
"Dead heat" polls have complicated some Americans' willingness to show their cards ahead of the election, and those that do may be communicating strategically in ways that will not reflect their actual behavior at the ballot box.
We noticed two things in our conversations on doorsteps and front porches. First, even the "anguished undecided" voters we spoke to grudgingly admitted that at the ballot box, they'll likely vote strategically, meaning anti-Trump. This could be idiosyncratic: Social desirability bias—that is, the tendency for people to answer questions in the way they believe will be viewed most favorably by others—is as real in door-to-door conversational settings as it is in public opinion polling. But these citizens' views may also reflect a genuine civic "strategic ambiguity" that some progressives have been playing with in this election: using their political voice in the run-up to the election to mobilize and signal to both domestic and foreign audiences they will hold Harris accountable for the Biden-Harris administration's abetting of war crimes , while reserving their anti-Trump vote for the privacy of the ballot box.
In either case, this is not necessarily a group whose actual preferences or likely behavior is easily captured by standard polling questions, and the impact of the Gaza crisis in this election cycle may be confounding the accuracy of reported support for Harris, not just among Arab Americans but also among left-leaning Democrats. The conventional wisdom is that U.S. voters don't elect presidents based on specific foreign policy issues, and polling data suggests that the "conflict in the Middle East" is "the most important issue" to only 2 percent of Americans. Nevertheless, in an election this tight, that 2 percent can matter in a place like Pennsylvania. The Biden administration's unconditional support for Israel despite its actions in Gaza is clearly a wedge issue this year pushing some Democrats away from the party, in ways that interact with the ongoing exodus of some conservatives from the Republican party.
None of this is easily captured by the major polls, partly because of the way questions are asked. Participants can't generally explain their answers, for instance, and often aren't asked to rank-choice them. This means a question that compares a niche issue like "Conflict in the Middle East" to "The Economy"—a huge basket of issues—in terms of voters' single "most important issue" is going to result in numbers that skew away from foreign policy. But the Carnegie Endowment found in early October that one-third of voters actually care greatly about foreign policy in this election, with Israel's war against Hamas the third most important issue after immigration and climate change. In that poll, 69 percent of Americans say neither of the two presidential candidates will do a satisfactory job when it comes to the Gaza conflict. This may explain why Arab-Americans are divided on what to do, and those who are veering away from Harris are breaking not toward Trump so much as toward Green Party nominee Jill Stein.
In fact, an AJP poll in May found one-in-five Americans were reconsidering their support for the Democratic Party due to Biden's handling of the war in Gaza. A much more recent YouGov poll found the Israel-Hamas conflict tops the list of foreign policy issues for registered voters. Perhaps, as suggested by the attitudes of Democratic voters I spoke to in Pennsylvania, some responses on polls represent a kind of political messaging rather than a genuine intention to boycott Harris—more "voice" than "exit." If so, anguished undecideds of the Democratic party may very well vote strategically anti-Trump. The absence of a third-party run by a candidate like Sen. Bernie Sanders who might have offered a meaningful progressive foreign policy alternative on Gaza has likely helped Harris' chances with this group. So has Trump's open denigration of Palestinians . But at least some voters in Pennsylvania still wished for some reassurance from Harris that their concerns would be heard and addressed on this particular issue.
This is anecdotal data from a small, nonrandom sample of left-leaning and/or swing voters. But they are coming from counties in a battleground state with an outsized impact on today's vote. They suggest Democratic turnout is likely to be high and that even the left-leaning voters least likely to turn out don't need much nudging to support Harris over Trump when it comes down to it. But they do want some nudging and have grown angrier the longer they don't hear it. A few may have even been swayed by Trump's efforts to engage on the issue.
In that sense, Harris' remarks at her rally over the weekend were well-timed. "I will do everything in my power," she said, "to end the war in Gaza, to bring home the hostages, end the suffering in Gaza, ensure Israel is secure, and ensure the Palestinian people can realise their right to dignity, freedom, security and self-determination." Her words were aimed at Arab American voters, but Jill Stein voters of all races are also listening.
What we heard on the ground in Pennsylvania also indicates how media coverage and constant "dead heat" polls have complicated some Americans' willingness to show their cards ahead of the election, and those that do may be communicating strategically in ways that will not reflect their actual behavior at the ballot box. The bigger point is that until pollsters find ways to acknowledge and capture these constituencies and trends— and weight polls to take into account how the reported numbers themselves influence attitudes and turnout, and how savvy citizens use survey responses to exercise influence where they feel they have none—it will be impossible to know how much the numbers can be trusted.
Charli Carpenter is a professor of political science and legal studies at University of Massachusetts-Amherst, specializing in human security and international law. She tweets at .
