| 01 October 2008
Presidential candidates who play up the threat of terrorism to bolster votes may want to rethink their game plan. New research from the University of California, Berkeley, indicates the war on terror has less impact on presidential popularity than it did during President Bush's first term.
Contrary to earlier studies that found that the threat of terrorism favors conservative leaders, a new national field study conducted by UC Berkeley sociologists Robb Willer and Nick Adams shows that terror warnings delivered by such government agencies as the Department of Homeland Security may reduce support for Sen. John McCain among moderates or swing voters.
While the survey shows that terror alerts have little, if any, influence on how self-described conservatives and liberals cast their ballots, politically moderate voters or swing voters are less likely to vote for McCain in the face of an imminent terror threat, according to a report on the survey published this week in the journal Current Research in Social Psychology.
"Most past research led us to expect that terror threats would increase support for conservatives," said Willer, lead author of the study and assistant professor of sociology at UC Berkeley. "But discontent with Bush's approach to the war on terror could be impacting views of McCain."
And, the survey found, while the war in Iraq still ranks as a major concern, the economy is a greater priority than the "war on terror."
The survey is a follow-up to Willer's earlier study at Cornell University in which he tracked 131 Gallup polls between 2001 and 2004 and found that each government-issued terror alert prompted an increase the following week in President Bush's approval rating.
To test the effect of the threat of terrorism on the presidential election, Willer and Adams designed an Internet-based survey experiment funded by a National Science Foundation program for large-scale field experiments. The survey was conducted by the media research firm Knowledge Networks in late May and early June with a nationally representative group of 1,282 Americans. Of the total sample, 36 percent identified themselves as conservative; 40 percent as moderate and 24 percent as liberal.
Respondents first were divided into a control group and a "threat-exposed" group and asked to rate various journalistic accounts based on their newsworthiness and importance. Both groups evaluated two articles about social welfare policy and health technology, but the "threat-exposed" group rated an additional article, adapted from The New York Times, which warned of a possible Al Quaeda attack on the United States.
Respondents were then asked whom they planned to support in the 2008 election and to what degree they favored President Bush.
For political moderates, exposure to the Al Quaeda attack article led to an 11 percent drop in their support for McCain. With just weeks until the Nov. 4 election, recent polls show that the race for the president is very close and that both campaigns are heavily courting moderate swing voters.
"A variety of studies conducted during Bush's first term showed that terror concerns led to greater support for the Bush presidency," Willer said. "Now, however, growing evidence suggests that many, possibly even most, Americans may prefer a new approach to combating terrorism."