On Feb. 6, 2017, US president Donald Trump stated that media coverage of terrorist attacks under-represented the extent of terrorist actions. To support this assertion, White House staff released a list of 78 attacks they said were “under-reported” by the media. Some were perhaps inappropriately included because they were either committed by deranged individuals without political motivation or were prevented.
Many news organizations responded by enumerating how often they reported on the listed events, while also noting some terrorist attacks they covered that were not included. However, the back and forth between journalists and officials does not shed much light on the overall level of reporting around the world of terror events.
What do the data tell us?
To better understand journalistic reporting of terror attacks over the past several decades and the real risk of terrorism risk for Americans, the Dow Jones Factiva service was queried. This database contains articles from more than 33,000 sources, 74% of which are not available on the free web, reaching as far back as the 1960s.
Four searches were conducted on February 12: The general term “terrorist attacks” yielded 1,280,835 articles, while “terrorist attack” – more likely to refer to a specific terrorist event – produced 581,839, half as many. Combining “terrorist attack” with the term “killed” yielded 197,633 citations, one-sixth as many. “Terrorist attack” combined with “wounded” received the least coverage, with a little over 50,000. (Note that all the search terms were in English which almost certainly under-represents the number of article worldwide reporting terrorist events.)
While articles with the keywords “terrorist attacks” may or may not refer to a single attack, they do indicate an increased focus on terrorism, suggesting both public interest in and exposure to the subject. This may also lead readers to have a heightened sense of insecurity, whatever the reality of the situation. For example, air ticket and hotel bookings to Paris from America significantly declined after terrorist attacks in the French capital.
The distribution of the articles over time, region, reporting sources, and authors offers further information. First, citations containing the words “terrorist attacks” rose dramatically in 2001:
The 9/11 attacks moved terrorism from under-reported (by comparison) news that happened somewhere else to a major concern by a broad collection of media outlets. Assuming that independent media attempt to satisfy their readers’ demands for information, this suggests that the public wanted, or at least consumed, stories about deadly terrorist attacks.
Let’s put the number of these articles into perspective. The combined number of stories found by the terms “Super Bowl”, “World Cup” or “Olympic Games” was 7,528,026. “Marijuana”, “LSD”, “cocaine”, or “heroin” were featured in 3,282,773. Other examples: “right to life” or “abortion”, 1,154,722; “Holocaust”, 755,001; “Academy Awards” or “Oscar”, 303,375; “nuclear war”, just 103,103. It appears that deadly terrorist attacks have an enduring interest for journalists and their readers, falling somewhere between competitive sports or conflictual social issues at the top and entertainment awards or nuclear war at the bottom.
The news organizations that met this demand were not evenly distributed globally or within individual countries:
North America is by far the region most referenced in the articles. The deadly attacks in Asia nudged Europe into third place. Given the large number of deadly attacks in the Middle East, there are relatively few cited reports. This is probably because the search terms were in English, so local language news outlets are undoubtedly under-represented. Curiously, this was not the case of Asia, which is home to many different languages.
Fewer than half of the American states were referenced in the articles on deadly terrorist attacks according to the Factiva database:
States with recent and major terrorist attacks are unsurprisingly referenced in many articles. What is interesting is the large number of articles which mention states with few terrorist events. For example, while the Midwest states of Illinois, Missouri, and Ohio suffered few terrorist events relative to other regions, they were frequently referenced in the articles.
Factiva searches export the top 100 news organizations producing the articles. Those reporting more than 10,000 articles over the period examined are represented below.
The Associated Press is by far the leader as are other English-language organizations. French and Chinese agencies are strong contributors collectively accounting for over 3% (43,000) of the global articles on deadly terrorist attacks. The top 18 reporting news agencies contribute nearly 40% of the world’s total articles about terrorist attacks.
There has been much discussion in the press and among academic researchers about the motives and social consequences of publishing articles about terrorist attacks. Some question whether widely reporting terrorist attacks leads to more of them in the future. (Doward 2015; Jetter, 2014.) Although this is not the question of this report, it is interesting to note that some of the most prolific writers about deadly terrorist attacks are located in countries that have experienced few, if any, terror attacks:
Two of the top three writers are in Canada, while the number-two writer is in New Zealand, a country that has had extremely few terrorist events.
Specific terrorist attacks with mortal effects
Over the past several years, articles about deadly terrorist events surpassed the press reporting of the September 11, 2001 attacks.
More than 14,000 articles related to deadly acts of terrorism were published in 2013, a 55% increase over the 9,245 published in 2001. In 2016, 23,548 were published, a 155% rise over 2001. It is evident that there has been a very large increase in articles published in recent years about specific deadly attacks.
Some of the increase could be attributed to the introduction of new publication sources indexed by Factiva. However, restricting the search to four publications founded before Factiva’s creation and continuously included – the Washington Post, New York Times, Wall Street Journal, and Financial Times – shows a similar U-shaped curve.
The natural conclusion is that the pattern of reporting has not been overly influenced by the increase in the number of indexed sources, which was also confirmed by Factiva’s curators.
More reporting or more killing?
There is strong evidence that journalists report more terrorist attacks when there are fatalities. But what is the trend in the number of deaths? In 2015 deaths due to terrorist attacks were nearly eight times greater than in 2003, the low point after 9/11. However, much of the increase from 2012 stems from new reporting methods used in the START Terrorist Incident Database.
Deaths were trending downwards before 2012. After a high point in 2007, deaths due to terrorism declined steadily while the reporting of terrorist attacks by the world’s English-language press generally continued to rise, roughly doubling between 2007 and 2012. Apparently, there was no shortage of journalists writing about terrorism, especially when there were victims.
There does seem to be an upper limit on the number of articles published about terrorist attacks, however. If we examine the number of articles published annually and divide this figure by the number of people killed during these attacks, we find that there were proportionally fewer articles published over the past few years:
In 2003, more than three articles were published for every reported death, while in 2014 there were 0.6. Does this indicate that terrorist attacks are currently under-reported in relation to their impact if not in actual number? This judgment is best left to future researchers but it is a fact that the vast number of deaths related to terrorism over the past decade occurred in countries in the Middle East and Africa, where reporting would most likely be in languages other than English.
Putting it in perspective
Based on the available data, the current controversy in the USA about protecting residents appears unwarranted in relation to the extremely small number of American victims. Compared to the total number of victims around the world, few Americans are killed in terrorist attacks.
Over the past half century the average annual number of American deaths due to terrorism is 195, and if the unusual year of 2001 is excluded, the average is 65. Factually, US residents are more likely to die from “accidental handgun discharge and malfunction” – there were 83 such deaths in 2015 according to the Centers for Disease Control and Prevention – or “unintentional cuts” (109 deaths in 2014). Furthermore, the number of terrorism-related deaths is absolutely miniscule compared to deaths from poisoning (51,966 annually), motor-vehicle crashes (33,804), firearms (33,736), or falling (33,018). (All statistics from the CDC’s National Center for Health Statistics, 2014, 2015.)
Overall, the risk of death by terrorism in North America is extremely low – in fact, it is among the lowest in the world. According to the START Terrorist Incident Database, since 2012 the number of terrorism-related deaths was highest in the Middle East and North Africa (52,105), followed by Sub-Saharan Africa (30,580), South Asia (30,277), Eastern Europe (2,587), and Southeast Asia (2,508). The database lists 173 deaths in North America. Yet the rising number of articles about terrorist attacks may create a climate of fear among Americans, who in reality are exposed to very little risk.
Perception versus reality
American worries, and to some extent those of Europeans, about terrorist attacks are exaggerated. Even the tragic deaths in France and other European countries over the past few years are dwarfed by the deaths in the Middle East, South Asia, and Africa. Terrorism is not just a problem for Americans and Europeans; it is a world problem that often indiscriminately kills people regardless of their religion or nationality.
People certainly have a right to know the reality and risk of terrorism and journalists have a duty to inform them. However, when does reporting cross over into fear-mongering? Disproportionate levels of reporting terrorism can also have emotional, social and economic costs that are poorly examined. The 1995 terrorist bombing in Oklahoma City by Timothy McVeigh and Terry Nichols was deadly, killing 168 people, injuring 680, and causing more than $600 million dollars in property damage. But was it really more dangerous to visit the city afterward? It is worth noting that the American government did not create an “extreme vetting” procedure for rental trucks entering Oklahoma – perhaps because it would have been too expensive, invasive, and ultimately futile.
Whatever the subject, journalists should report the truth supported by fact. Creating alarm to sell ads or promote a political agenda is inappropriate, even dishonest. If further statistical analysis finds that the 78 terrorist attacks listed by the White House were indeed under-reported, perhaps the reason will be that they were under-important for the average American and not reflective of the true risks of living in America.