The news media reacted to the terrorist attacks of September 11 with great care about not getting ahead of the facts, but over time the press is inching back toward pre-September 11th norms of behavior, according to a new study of press coverage of the war on terrorism.

In the beginning, solid sourcing and factualness dominated the coverage of bombings and their aftermath, according to the study, conducted by the Project for Excellence in Journalism with Princeton Survey Research Associates. A full 75% of what the press reported was a straightforward accounting of events—here is what happened.

As the story moved to the war in Afghanistan, however, analysis and opinion swelled—so much so that the level of factualness declined to levels lower than those seen in the middle of the Clinton-Lewinsky scandal.

The early coverage may help account for why we saw the first measurable upturn in public approval of the press in 15 years. But the changes in coverage offer a caution about why that approval has started to fall again.

Has the news media become jingoistic in covering the war amid intense Pentagon restrictions? Or is there a liberal or negative tilt to the coverage?

The study found that during the periods examined the press heavily favored pro-Administration and official U.S. viewpoints—as high as 71% early on. Over time the balance of viewpoints has broadened somewhat. Even then, what might be considered criticism remained minimal—below 10%.

Coverage of the Crisis1
Fact 75% 63% 63% 69%
Analysis 14 21 22 18
Opinion 9 11 10 10
Speculation 2 4 4 3
Total 100 100 100 100
1 For all charts, totals may not equal 100 due to rounding.

The study, conducted by the Project for Excellence in Journalism with Princeton Survey Research Associates, involved a detailed examination of 2,496 stories contained on television, magazines and newspapers in three key periods in mid-September, mid-November and mid-December.

After the press earned high approval marks from the public and praise from critics for its coverage, the study set out to probe why. To that end, it looked at a cross section of the news media to examine the sourcing, verification, and range of viewpoints in the coverage.

Among the findings:

  • In the earliest days, the news media tended to avoid interpretation. Just 25% of the coverage was analysis, opinion and speculation-including even the talk shows and the opinion pages.
  • By December, that percentage had swelled to close to four-in-ten of all the reportage (36%).
  • The number of sources cited as evidence in stories also declined over time, though it is still relatively high. The level of on-the-record sources has remained consistently high-three quarters of all sources.

On talk shows, journalists often seemed to luxuriate in sounding not like knowledgeable experts on TV stages, but like anyone else standing in a barroom.

The death of Osama Bin Laden’s third in command for CNN’s Margaret Carlson on December 17 was “another reason to be cheerful.”

“Having Osama bin Laden on trial in the United States of America is a nightmare,” Cokie Roberts declared on ABC’s This Week November 18. “With any luck, you know, he is—he is found dead.”

The study examined a snapshot of the news media culture during three different phases of the crisis. We looked at four newspapers (The New York Times, Washington Post, Cleveland Plain Dealer and the Fresno Bee), two news magazines (Time and Newsweek) four nightly news broadcasts (ABC, CBS, NBC and PBS), the three main network morning shows, the Sunday talk shows, three weeknight talk shows (Larry King, Charlie Rose and Hardball with Chris Matthews), Nightline, and relevant segments of three prime time network news magazines (Dateline, 20/20 and 60 Minutes II). The study also included an examination of two cable nightly newscasts (Fox Special Report with Brit Hume1 and CNN’s NewsNight with Aaron Brown). The study focused on three phases of the crisis, September 13-15, November 13-15 and December 10-12 as well as the closest weekend Sunday shows and news magazines.

The Project for Excellence in Journalism is a journalism think tank affiliated with Columbia University Graduate School of Journalism and is funded by The Pew Charitable Trusts.

One reason for the decline in sourcing and factualness and the rise in interpretation over time may be the restrictions the government is imposing on journalists’ access to information. “The restrictions are unprecedented and they are successful,” ABC National Security correspondent John McWethy told a panel at Columbia University Graduate School of Journalism last week.

The evidence strongly suggests that coverage is more factual when journalists have more information and becomes more interpretative, perhaps ironically, when they have less.

It is an oversimplification to suggest that since the U.S. has won the war against the Taliban, the press has returned to a so-called “normal” diet of softer news. Even on programs that often have less in the way of traditional hard news, such as morning television, the coverage of the war on terrorism actually increased from the November to December periods after a significant decline from September to November.

Even if news of the war was easy to find, what Americans know about it varies drastically depending on what medium they get their information from. Television news, for instance, is measurably less likely to include criticism of the Administration than the print media.

Contrary to the suggestions of Fox News executives, there is no evidence that CNN is less “pro American” than Fox or has some liberal tilt. To the contrary, there is no appreciable difference in the likelihood of CNN to air viewpoints that dissent from American policy than there is Fox. This may not be anything to boast about. Both channels tended to favor pro-Administration viewpoints more than most other newscasts—even most talk shows.

Just as the Project found in the Clinton-Lewinsky scandal, those who know the least may be the most prone to offer their opinions. As a rule, the weaker a story’s sourcing, the more likely it is to be interpretive. The better the sourcing, the less likely it is to interpret.

The three phases of the crisis examined in the study each offered distinct story lines.

The first phase, September 13 to 15, began with the day the television media returned to regular news programming. The press focused on a nation in shock. Airlines remained grounded. President Bush was still five days from addressing a joint session of Congress to outline the U.S. response. The coverage focused on four themes—the potential war on terrorism, the September attacks and rescue efforts, personal connections stories, and citizen, community and state response.

The second phase of the study examined two months later, November 13 – 15. The Northern Alliance was making major gains in the north. The Taliban was fleeing the Afghan capitol of Kabul, but it was still unclear whether this was a collapse or a strategic regrouping. The press coverage focused on the action in Afghanistan, the war on terrorism in general and the international response.

The third phase of the study examined December 10 through 12. The Taliban had fled Kandahar to the mountains around Tora Bora. The U.S. military focus had turned to the hunt for Bin Laden, Taliban leader Mullah Omar and other al Qaida leaders. The coverage focused on action in Afghanistan, the fight against terrorism generally, and continuing community and civic response to the September attacks.

Initially, public reaction to the coverage was extremely positive. By November, indeed, The Pew Research Center for the People and the Press found the first upturn in broad public support for the press in 15 years. More Americans suddenly considered the press accurate, professional, moral, caring about people and patriotic—after years of steady decline. Newspaper circulation and TV audience numbers spiked.2

What was it people liked? “Timeliness,” “comprehensiveness” and “informativeness,” were the reasons survey respondents most often volunteered. Few people complained of bias and sensationalism. They liked the coverage, even though they found it tiring and depressing. In short, researchers concluded, people craved the information and felt the media provided it.

With time, that has begun to change. The percentage of Americans who think the press has done an “excellent” job covering the crisis has declined steadily, from 56% in September to 30% by mid-November, the last data to date.3

What accounts for the declining public approval? For years, surveys, focus groups and other research have found consistent patterns in what people say they don’t want from the press. People dislike anonymous sourcing. They want information more than interpretation. They resent journalists offering what they think rather than what they know. They dislike hype and the sense that the media is manufacturing and sensationalizing stories.

As the war on terrorism progressed, the press, for a variety of reasons, began to rely more on the methods and habits disliked by the public. In the months ahead, as the war broadens beyond Afghanistan and becomes harder to see, the pressures on journalists to resort to these means of presentation are likely to only increase.


1 Overall totals for the study do not include Fox’s Special Report with Brit Hume. (See Methodology.)

2 “Terror Coverage Boosts News Media’s Image,” The Pew Research Center for the People and the Press, November 28, 2001.

3 Despite this, the percentage of Americans who rate the coverage “good” or “excellent” remains comparatively high, above 70%.