Using a statistical procedure called multiple regression analysis, we first examined the extent to which the news media’s coverage of terrorism is related to public perceptions of the gravity of the terrorist threat to the United States. One might expect the total volume of mass-mediated threat messages to have an effect on perceptions of terrorism as the most important national problem, but this is not the case. We found that it is not the volume of threat messages that matters, but who conveys such messages.

As we see in Figure 1, all the variables show decreases over time in the same direction, with increases during some of the same short-term periods. While some of the smaller short-term changes and others cited below might not appear significant by themselves, they are compelling in
their consistency with the overall patterns of statistical relationships.

When Bush’s reactions to and comments about terrorist threats are reported in television news and when administration officials make these statements, the public is more likely to perceive terrorism as most important. This happened at several points: In June 2002 the peak in the public’s threat perception followed several terrorism alerts in the previous months, when administration officials initiated a heightened state of alert for railroads and other transit systems and warned of a special threat against the Statue of Liberty and the Brooklyn Bridge. Moreover, in early June, U.S. Attorney General John Ashcroft made the dramatic announcement that Jose Pardilla, a U.S. citizen and al-Qaeda associate, had been arrested before he could acquire and explode a “dirty bomb” in an American city. As television news covered these threat announcements heavily, the public’s view of terrorism as a major problem for the country increased from 22 percent in May to 33 percent in June.

Figure 1

It is noteworthy that there was a dearth of mass-mediated threat messages and actual threat and alert statements in the preceding period. In February 2003, a month before the invasion of Iraq, President Bush’s relevant statements were reported on TV three times, and there were five original statements by administration officials. The following month, we see a slight increase in the public’s perception of terrorism (from 10 percent to 13 percent) after a decrease in the previous months (from 18 percent to 10 percent).

In December 2003, when Saddam Hussein was captured, the same pattern occurred. One might think that this success would lead to a decreased level of threat (no correlation was found between these types of messages with responses to any of our survey questions), and a smaller percentage of people thinking that terrorism is the most important issue facing the country. But two heavily covered domestic events could have affected the public: First, the chairman of the 9/11 Commission, Thomas Kean, said publicly that the 9/11 attacks could have been prevented. Three days later, Secretary of Homeland Security Tom Ridge raised the terror-threat alert for the upcoming holidays.

Finally, in the months before the presidential elections, the pattern occurred again: From June 2004 (or even earlier) the public’s perceptions regarding terrorism as a major problem strengthened steadily, with a peaking in September-October 2004 to the same level it reached in November 2002 (when the American-led coalition had made progress in the fight against al-Qaeda and the Taliban).

We next looked at the public’s concern about more major terrorist attacks occurring in the United States. In this case, we found that the media had the only apparent influence on people’s perceptions, with several concurrent peaks appearing in the trend: When media professionals talked about terrorism threats or increased levels of alert, the public became more concerned. This occurred particularly around February/March 2003, with 27-29 percent of the public very concerned about another major attack occurring (rising from 22 percent in previous opinion polls). In February, network TV mentioned the threat of terrorism twenty-five times and increased levels of official alerts, followed up by six messages in March, when the Iraq invasion was launched. The threat messages reached a peak of fifty-five in August 2004, and, a month later, 25 percent of the public was very concerned about another attack in the United States.