Threatening Information: An Experiment on the Effect of Terrorism News on Foreign Policy Attitudes

By Shana Kushner Gadarian

More than five years after the attacks on the World Trade Center and the Pentagon, terrorism still occupies a sizeable proportion of the nightly news. In 2006 alone, more than 700 news stories on terrorism aired on national network news, adding to the more than 4,300 stories related to terrorism that aired in the previous four years. Compare those 5,000 terrorism stories to 138 stories on poverty, 592 stories on education, and 724 stories on crime over the same time period, and the dominance of terrorism news is clear.

As part of a larger study on how sensationalistic media coverage affects foreign policy attitudes, I designed and ran a media experiment using a nationally representative sample of 1,229 adults through Polimetrix.8. The experiment was designed to uncover the mechanism by which television influences attitudes—threatening message content or content paired with threatening imagery—and to determine the causal impact of media exposure on attitudes in a way that secondary survey data cannot.

Subjects of the experiment received an email invitation to participate in the study and were randomly assigned to watch one of three videos showing stories that had aired on the national news in early fall 2005. The subjects, who completed the experiment online in December 2006, were randomly assigned to either the control group or one of two treatment groups. Those in the treatment conditions watched a story about the potential for a new wave of terrorism. All of these respondents received exactly the same information about terrorism from the videos: that government authorities were troubled by the increase in smaller-scale terrorist attacks such as the ones in London, Sharm el-Sheik, Egypt, and Madrid, and were concerned that the attacks might signal a new wave of terrorism. The story emphasized the potential for new terrorist attacks in the near future.

What differed between the treatment groups was the visual imagery of the videos they were shown. The “scary visuals” group viewed the story with evocative imagery of terrorism, while the “neutral visuals” group viewed the same story with neutral imagery instead. In the scary visuals video, the terrorism news story was edited to enhance the threatening nature of the visual imagery, adding, for example, such images as the burning World Trade Center and bloodied victims of the London subway bombings. In the neutral visuals condition, the visuals were replaced by less violent imagery. For example, in place of the London victims, the neutral condition included maps of the London subway system. Respondents in the neutral visuals condition received threatening information only through the verbal content, while those in the scary visuals condition received threatening information through both the verbal and visual story components. Control subjects watched an unrelated story on India’s economy. Table 1 summarizes the design of the experiment.

Click table to go to videos

To measure the effects of exposure to these different stories on foreign policy attitudes, respondents in each group answered a series of questions after watching the videos. I used a statistical procedure called ordinary least squares (OLS) to analyze the effect of exposure to the videos and respondents’ concern about terrorism on two measures of foreign policy attitudes—a foreign policy index and a measure of militarism. I created the foreign policy index using six questions found on the 2002 survey for the American National Election Studies (NES). An additive index, it contained five questions on, respectively, federal spending on foreign aid, homeland security, approval of the president’s handling of terrorism, approval of the government’s handling of terrorism, and whether the war in Iraq was worth the cost. The militarism question asked respondents to place themselves on a seven-point scale indicating how they thought the government should “handle international problems,” with diplomacy/international pressure at one end and military force at the other.

The OLS procedure controlled for the effect of respondent’s prior level of hawkishness (that is, preference for military action over diplomacy), measured before exposure to the videos, as well as respondent’s pretest level of threat (that is, how likely he or she believed there would be a terrorist attack in the next year).

Results of the analysis indicated that perception of the threat of terrorism—how vulnerable respondents believe the country is to a terrorist attack—significantly increases preferences for militarism over diplomacy in international affairs. Additionally, and most relevantly, the effect of threat on attitudes was greatest for respondents in the scary visuals condition.

The partisan political environment adds another dimension to the way that media influence the public. To evaluate whether partisans exposed to the same information reacted similarly to it, I ran the militarism analyses separately for Democrats, Independents, and Republicans. The quite different effects that the media treatments had on the partisan groups suggest that, in the area of foreign affairs, the adage that “politics stops at the water’s edge” no longer applies. Not only does politics influence the type of foreign policy that citizens desire from government; it also shapes how citizens interpret media messages and apply that information to their preferences.

Figure 1 shows the effects for all respondents. It indicates that respondents more concerned about terrorism preferred more hawkish policies than those less concerned. In addition, among those respondents highly concerned about terrorism (the blue line), watching one of the two terrorism stories increased their hawkishness.

As Figure 2 shows, Democrats threatened by terrorism became the most hawkish after viewing the threatening story paired with threatening visuals.

In comparison, Republicans preferred militaristic foreign policy, no matter which video they viewed (Figure 3).

The shape of the graph in Figure 4 indicates that for Independents, the neutral visuals condition moved opinions the most, meaning that the threatening information contained in that story, but not the threatening visuals, mattered for this group.

Being in the scary visuals condition had more than twice the impact on Democrats than it had on Republicans or Independents in pushing them toward militarism. This finding is particularly striking given that, when asked about their emotional reactions after watching the news stories, politically engaged Democrats tended to dismiss the scary visuals condition as manipulative, and to show evidence of “negative persuasion”—that is, reacting in the opposite way to that intended by the communication.

Using the foreign policy index instead of the diplomacy measure in the analysis produced the same results: Democrats were more affected by the news stories than Republicans or Independents, and were most affected by the scary visuals condition. For respondents most concerned about terrorism, the influence of the scary visuals condition was more than twice as large among Democrats as among Republicans or Independents. While threatening information paired with neutral visuals mattered little for Democrats’ attitudes, the neutral visuals condition influenced Independents more than the scary visuals condition did. As with the diplomacy measure, these findings also suggest that exposure to the most threatening types of news stories influences the attitudes of Democrats by moving them toward supporting policies they otherwise would not.

The analysis thus far compares the effect of exposure to two types of terrorism stories to exposure to a nonterrorism story. Yet 26 million Americans a night watch news that consistently features threatening terrorism stories. To the extent that watchers are exposed to threatening stories, we may wonder how presentational features of these terrorism stories matter. The previous analyses found that television news exposure had the largest impact on respondents already concerned about terrorism. To test whether frightening imagery by itself affects foreign policy attitudes, Figure 5 presents the difference in average foreign policy attitudes between high-threat respondents (those whose pretest results showed a level of threat above the mean) in, respectively, the scary visuals condition and the neutral visuals condition. Average scores above zero indicate that the scary visuals condition moved attitudes toward the hawkish end more than the neutral visuals condition, while negative scores indicate that the neutral visuals condition caused more hawkishness among respondents than the scary visuals condition.

The figure demonstrates that exposure to the scary visuals condition made high-threat Democrats prefer significantly more militant policies than did exposure to the neutral visuals condition. This finding is particularly remarkable, given that the Democrats in this sample seemed particularly unlikely to be moved by terrorism cues: Prior to the experiment, only 6 percent said they voted for President George W. Bush in 2004, 89 percent strongly disapproved of the president, and 55 percent wanted to decrease the defense budget. The combination of the strong message and the evocative images of the scary visuals story, however, induced Democrats concerned about terrorism to adopt a more hawkish policy than only the cue in the neutral visuals condition that terrorism was likely in the near future. The effect of the scary visuals condition on Democrats is a testament to the indelible emotional mark left on Americans by the images associated with 9/11.

In contrast to Democrats, Independents and Republicans reacted to the neutral visuals condition and the scary visuals condition by the same magnitude. Republican attitudes were unmoved by any of the experimental stories. As noted before, Independents’ attitudes reacted to the threatening content in the neutral visuals condition more than the threatening imagery offered by the scary visuals condition. Since Independents in this sample were less politically knowledgeable than both Democrats and Republicans, their attitudes were most affected by the information common to the neutral visuals and scary visuals story. Overall, this figure demonstrates that the presentation of threatening news does matter in shaping public opinion on foreign policy, but that it matters by convincing Democrats to support more militant types of foreign policy.

Terrorism is newsworthy because it is timely, sensational, and novel, and it is covered extensively by the news media. In the years following the 9/11 attacks, television broadcasts increasingly focused on terrorism and war-related stories, and the stories themselves focused on the threatening aspects of events more than the reassuring ones. In combination with findings of the larger study of which it is a part, this experiment demonstrates that sensationalistic news coverage moves the American public in a more hawkish direction—increasing support for aggressive foreign policy, up to and including war.

More than five years after the 9/11 attacks, exposure to a single terrorism story led experiment subjects to prefer more aggressive types of foreign policy when the story presented both a threatening message and threatening visuals. What makes this finding especially striking is that in the years since 9/11, American citizens saw frequent news stories about terrorism and presumably had stored knowledge and beliefs about counterterrorism policy. Equally remarkable is the fact that even though the public is both interested and knowledgeable about terrorism—according to a 2005 Pew Research Center poll, 88 percent claimed to pay close or very close attention to news coverage of the July 7, 2005, London terrorist bombing—watching one more frightening terrorism story significantly influenced attitudes, particularly among Democrats.

Politicians who invoke terrorism or 9/11 to promote foreign policies run the risk of appearing manipulative to the public, particularly to members of the opposing party. During the 2004 presidential campaign, the Bush campaign was criticized for running ads that used visuals of the wreckage of the Twin Towers. Critics objected to the use of the images for political gain, calling them, as the Associated Press did, “unconscionable,” and in poor taste. Yet these findings suggest that there is a political payoff for reminding citizens of terrorism through threatening news stories and frightening visual imagery. These types of stories may successfully persuade citizens, especially Democrats, to adopt foreign policy positions they would not normally prefer.

In Toward Perpetual Peace, Immanuel Kant argues that once elites need the consent of the governed to engage in war, war should be less likely. This thesis does not account for the ability of institutions like the media to create or heighten a sense of vulnerability and thus increase support for war among the citizenry. If the media can activate citizens’ sense of impending danger and thereby increase support for punitive policies, my findings imply that when political leaders use the right images and trumpet the right threatening message, war may be quite likely in a republic, even when a fully informed citizenry might have chosen otherwise.

Control condition

Neutral visuals condition

Scary visuals condition

Survey Questions Used in the Experiment

Measured prior to experimental treatment:

Threat: How likely do you think it is that the US will suffer a terrorist attack sometime in the next 12 months? Would you say very likely, somewhat likely, somewhat unlikely, or very unlikely? [0=Very likely, .25 = Somewhat likely, .75 =Somewhat unlikely, 1=Very unlikely]

Hawkishness (pretest measure): Which of the following statements comes closer to your view: a. The best way to ensure peace is through military strength or b. Good diplomacy is the best way to ensure peace.[0= Good diplomacy is the best way to ensure peace, 1= The best way to ensure peace is through military]

Partisanship: Generally speaking, do you usually think of yourself as a Republican, a Democrat, an Independent, or what? (IF REPUBLICAN OR DEMOCRAT) Would you call yourself a strong [REPUBLICAN/DEMOCRAT] or a not very strong [REPUBLICAN/DEMOCRAT]? (IF INDEPENDENT, OTHER) Do you think of yourself as closer to the Republican or Democratic party? (7-point scale constructed from answers to above questions; 1=Strong Democrat, 2=Weak Democrat, 3= Lean Democrat, 4=Independent, 5=Lean Republican, 6=Weak Republican, 7=Republican)

Measured after exposure to experimental treatment:

Militarism: Some people believe the United States should solve international problems only by using diplomacy and other forms of international pressure and use military force only if absolutely necessary. Suppose we put such people on a scale that goes from 1 to 7, placing them at the end of the scale numbered "1." Other people believe that diplomacy and pressure often fail and the U.S. must be ready to use military force. Suppose these people are at the other end of the scale, at point number "7." And, of course, other people fall into positions in-between, at points 2, 3, 4, 5, and 6. Where would you place YOURSELF on this scale?

1. U.S. should solve problems only with diplomacy and international pressure

2.

3.

4.

5.

6.

7. U.S. must be ready to use military force

[Recoded to vary between 0 and 1]

Foreign policy index: Foreign policy index is an additive, 5-item index, consisting of questions on spending on foreign aid, defense, and homeland security/ war on terror, support for the war in Iraq, approval of the president’s handling of terrorism. The index ranges from -1 to 1, with higher values are more hawkish. (Cronbach’s alpha = .68):”

Spending on foreign aid/defense/homeland security and war on terror: Should federal spending on foreign aid [foreign aid, defense, homeland security/war on terror] be increased, decreased, or kept about the same? (-1=decreased, 0=kept the same, 1=increased)

Iraq war support: Taking everything into account, do you disagree or agree with this statement: The U.S. war in Iraq has been worth the cost. [0=strongly disagree, .25=disagree, 0=neither agree or disagree, .75=agree, 1=strongly agree]

Presidential approval of war on terror: All things considered, do you approve or disapprove of the way George W. Bush is handling the war on terrorism? [0=strongly disapprove, .25 = somewhat disapprove, .75 = somewhat approve, 1=strongly approve]

Source: Polimetrix experiment of foreign policy attitudes, December 2006. Shana Kushner Gadarian, principal investigator

Numbers of News Stories

The total numbers of news stories on the different subjects were found by searching for “terror,” “crime,” “poverty,” and “education” in the Vanderbilt television archive for stories that ran on all networks (ABC, CBS, NBC, CNN, PBS, FOX, MSNBC, CSPAN, and CNBC) from September 11, 2001, to December 31, 2006, excluding commercials and special programs.

The Polimetrix Panel

Polimetrix maintains a panel of respondents, which it recruits through a polling website. Once individuals agree to become part of the Polimetrix panel, they provide demographic information and are offered opportunities to participate in surveys. Since recruitment into the panel is voluntary, this means that the larger panel is an opt-in sample that may be unrepresentative of the larger population. Opt-in internet samples tend to be more interested in politics as well as whiter than the general public.

Additional Reading

Associated Press. 2004. Bush ads anger some 9/11 families. March 4.

Baum, M. 2003. Soft news goes to war: Public opinion and American foreign policy in the new media age. Princeton, NJ: Princeton University Press.

Berinsky, A. 2007. Assuming the costs of war: Events, elites, and American public support for military conflict. Journal of Politics 69 (4).

Berrebi, C., and E. Klor. 2007. The impact of terrorism on voters' preferences. RAND Working Paper WR-477. February.

Brader, T. 2006. Campaigning for hearts and minds: How emotional appeals in political ads work. Chicago: University of Chicago Press.

Broder, D., and D. Balz. 2006. How common ground of 9/11 gave way to partisan
split. Washington Post. July 16, A1.

Gadarian, S. Kushner. 2007. Beyond the water’s edge: Polarized images in the war on terror. Paper presented at the Midwest Political Science Association Annual Meeting. Chicago, IL. April 12-15.

Gans, H. 1979. Deciding what’s news: A study of CBS evening news, NBC nightly news, Newsweek, and Time. New York: Pantheon.

Huddy, L., S. Feldman, G. Lahav, and C. Taber. 2003. Fear and terrorism: Psychological reactions to 9/11. In Framing terrorism: The news media, the government, and the public, ed. P. Norris, M. Kern, and M. Just. New York: Routledge.

Nacos, B., Y. Bloch-Elkon, and R. Shapiro. 2006. Post 9/11 threat perception of international terrorism in the United States: Public officials, the media, and the public. Paper presented at the Annual Meeting of the American Political Science Association. Philadelphia, PA. August 31–September 3, 2006.

Zaller, J. 1992. The nature and origin of public opinion. New York: Cambridge University Press.