Many supporters of Hillary Clinton fear—and many fans of Donald Trump hope—that a terror attack in the United States, such as those that struck New York, New Jersey, and Minnesota over the weekend, will decide the election. The assumption on both sides is that Trump, as the anti-Muslim, anti-immigrant, law-and-order candidate, would benefit. But is that true? Three recent attacks—in Paris on Nov. 13, in San Bernardino, California, on Dec. 2, and in Orlando, Florida, on June 12—have tested this assumption. Trump has not gained ground, and in some ways he has suffered, for several reasons. Here’s what the polls show.
1. No freakouts. After Paris, public concern about terrorism rose significantly. In a Quinnipiac University survey taken shortly before the attack, 32 percent of Americans said it was very likely that “in the near future there will be a terrorist attack in the United States causing large numbers of lives to be lost.” When Quinnipiac repeated that question after Paris, the percentage who said “very likely” had climbed to 44 percent. But the subsequent attack in San Bernardino—an actual terror strike in the United States—made little difference. The percentage who said “very likely” barely increased, and the overall percentage who expected future attacks marginally declined. In ABC News/Washington Post surveys, the proportion of respondents who said terrorism was their top presidential voting issue also marginally declined.
You might also expect such attacks to undermine public confidence in government. But they haven’t. In a May 2015 Fox News poll, 65 percent of Americans expressed confidence that intelligence agencies could detect and prevent terror plots against the United States. In December 2015, after Paris and San Bernardino, that number didn’t fall. It rose slightly, to 67 percent. Likewise, in a series of Quinnipiac surveys taken before Paris, after Paris, and after San Bernardino, the percentage of Americans who said we were “losing the fight against ISIS” steadily declined, from 66 percent to 62 percent to 57 percent.
2. No increase in support for sending troops. In the same sequence of Quinnipiac polls, the percentage of Americans who supported “sending ground troops to fight ISIS in Iraq and Syria” held steady and then marginally declined, from 54 percent to 52 percent. In Fox News surveys taken before and after San Bernardino, the percentage who favored sending “a significant number” of troops to Iraq and Syria also declined, from 42 percent to 38 percent.
3. No sudden increase in appetite for surveillance. Since the 9/11 attacks, public support for domestic surveillance has risen or fallen in response to threats, such as the rise of ISIS. From the beginning of 2014 to the days just after Paris, there was a big shift from concern about civil liberties to concern that the government should do more to monitor potential terrorists. But San Bernardino had no effect. In Quinnipiac polls taken before and after San Bernardino, the numbers moved slightly the other way, from worry about insufficient protection to worry about civil liberties.
4. Concern about Islam but rejection of Trump’s response. Americans slightly prefer Trump’s language to President Obama’s. In an ABC/Post poll taken shortly after Orlando, 48 percent said leaders should use the term radical Islam in describing such attacks. Only 40 percent chose the alternative answer: that leaders should avoid that phrase “because it lends legitimacy to terrorists by falsely suggesting that their actions are supported by Islamic teachings.” But Americans don’t support Trump’s anti-Muslim policies. In a Monmouth University poll taken after the San Bernardino attack, 65 percent opposed “banning all Muslims from entering the U.S.” After Orlando, that number rose to 70 percent. When the ban is phrased as temporary, opposition is lower, but attacks still don’t help Trump. In May, Americans opposed a temporary ban on Muslims, 50 percent to 43 percent. In June, after Orlando, they still opposed it, 52 percent to 43 percent.
5. More fear of radicalization than of refugees. After Paris, polls showed a big drop in Americans’ willingness to accept Syrian refugees. San Bernardino had no such effect. In Quinnipiac surveys taken before and after San Bernardino, Americans narrowly shared Trump’s reluctance to take refugees, but most said they were more concerned about homegrown jihadists than about refugees or other visitors. Likewise, after Orlando, a Monmouth survey found a shift toward concern about domestic radicalization rather than foreign infiltration. So when Trump responds to such attacks by talking about refugee vetting and Clinton responds by talking about monitoring online radicalization, it’s not clear that his message resonates more strongly.
6. No enthusiasm for gun control. The deadly use of guns in Paris, San Bernardino, and Orlando didn’t increase support for stricter gun laws. People seem to view these attacks more as terrorism than as crimes involving firearms. In NBC/Journal surveys between October and December 2015—the period of the Paris and San Bernardino attacks—the percentage of Americans who were very worried about being directly hit by a terror attack increased, while the percentage who were very worried about being directly hit by gun violence did not change. In fact, Quinnipiac polls found a loss of support for gun control: Shortly before Paris, Americans favored stricter gun laws by a 7-point margin, but after San Bernardino, they opposed such laws by a 3-point margin. Before Paris, 24 percent said recent mass shootings in the United States made them more likely to purchase a gun. After San Bernardino, that number rose to 33 percent. Economist/YouGov surveys in May and June found no change in support for stricter gun laws after Orlando.
7. Trump is a turnoff. After San Bernardino, Trump proposed his ban on foreign Muslims. After Orlando, he ratcheted up his rhetoric against refugees, immigrants, and American-born Muslims. These responses didn’t go over well. After San Bernardino, a Monmouth survey asked: “Do you think Trump is saying things that need to be said or is he making things worse?” A plurality, 48 percent to 39 percent, said he was making things worse. In a YouGov survey, Americans approved of Obama’s response to the attack in Orlando, and they marginally approved of Clinton’s response, but they disapproved of Trump’s, 47 percent to 32 percent. In an NBC News/Wall Street Journal poll taken after Orlando, Clinton trounced Trump on every question: Which candidate did a better job of responding to the attack, which showed a better temperament, which offered the better proposals, and which earned more confidence that he or she could handle such attacks?
How do all these factors add up? Do they help either candidate? The best evidence comes from three pollsters who took samples in May, shortly before the Orlando attack, and then in June, shortly afterward. In ABC/Post surveys, Clinton gained ground on the question of which candidate would handle terrorism better, increasing her margin from 3 percentage points to 11. In NBC/Journal surveys, Clinton marginally gained ground on standing up for America, uniting the country, and performance as commander in chief. In Quinnipiac surveys, Clinton gained ground on which candidate was better prepared to be president, but Trump gained ground on other questions: Which candidate was a stronger leader, which was better at handling immigration, which was better at responding to an international crisis, and which was more effective against ISIS?
It’s hard to tell whether these shifts, even when they overlap with terror attacks, are caused by the attacks. It’s equally hard to tell whether other factors, such as Trump’s slurs against a Mexican American judge in early June, counteract and mask pro-Trump effects of the attacks. But some combination of factors—unease about war, distaste for Trump’s reactions, the irrelevance of refugees, and the public’s failure to lose its cool—has limited his ability to capitalize on these incidents. In the wake of this week’s attacks, he’s likely to fail again.