An Experimental Study of the Effects of Government Terror Warnings on Political Attitudes

Download data and study materials from OSF

Principal investigator:

Robb Willer

Stanford University

Email: willer@stanford.edu

Homepage: http://robbwiller.org/


Sample size: 1282

Field period: 5/22/2008-6/09/2008

Abstract

How do concerns about terrorism affect the way Americans view the 2008 presidential candidates? How would an event that increases the prominence of terrorism, like a threat or attack, affect the 2008 election? Past theory and research are ambivalent on these questions. In this paper we empirically investigate the effects of exposure to a journalistic account of an imminent terror threat in a national (N = 1,282), internet-based field experiment. Overall, we find that exposure to terror threats increased concerns about "homeland security" without affecting candidate preferences. However, analysis of politically moderate respondents - a substantial subset of the total sample (40%) with a high rate of undecided, likely voters - showed that this group expressed significantly lower support for Senator John McCain when exposed to the terror threat than in the control condition. These findings converge with past research suggesting that Americans' views of the war on terror have changed significantly (Davis and Silver 2004) and that terror threats may serve as "anti-rally" events for candidates with unpopular foreign policies, especially among moderate or undecided voters (Bali 2007). We conclude by proposing a more general model of the effects of external threats on leader preferences suggested by past and present research.

Hypotheses

We originally anticipated that exposure to a media account of a terror alert would increase support for (1) then President George W. Bush, and (2) the Republican presidential nominee (who eventually turned out to be Sen. John McCain).

Experimental Manipulations

Salience of the threat of terrorism (manipulated via presentation of a terror-related news article)

Outcomes

Presidential approval
Ranked issue importance
Preferred candidate in the 2008 presidential election
Several questions taken from larger political psychology batteries, including measures of tolerance for ambiguity, openness to uncertainty, social dominance orientation, and traditionalism

Summary of Results

Overall, we find that exposure to terror threats increased concerns about "homeland security" without affecting candidate preferences. However, analysis of politically moderate respondents - a substantial subset of the total sample (40%) with a high rate of undecided, likely voters - showed that this group expressed significantly lower support for Senator John McCain when exposed to the terror threat than in the control condition.

References

Robb Willer and Nick Adams. 2008. "The Threat of Terrorism and Support for the 2008 Presidential Candidates: Results of a National Field Experiment." Current Research in Social Psychology. 14(1):1-22