Overview

The news media are offering the American public a fine education in campaign tactics but telling them little about matters that actually will affect them as citizens, a new study of presidential campaign coverage finds.

Leading up to Iowa and New Hampshire, the press has provided only scant reporting on the candidates’ backgrounds, records, or ideas, according to the study by the Project for Excellence in Journalism, a journalist-run group in Washington D.C.

And in all, only 13% of the stories produced were about things that would actually impact the American public if the candidates were elected, such as their ideas, their honesty or how their constituents in the past have been affected.

More than 80% of the stories, in contrast, focused on matters that impact the politicians or parties, such as changes in tactics, who has more money, or internal organizational problems.

Contrary to conventional criticism, the reporting is not particularly framed around the horse race. Nor is it tilted in favor of one candidate over another, including so-called media darling John McCain.

Instead, the reporting is overwhelmingly focused on the internal tactics and strategies of the campaigns-concerns that research suggests people do not care much about and that even the study researchers found numbing to read.

Remarkably less than one percent of the stories-or just two out of 430 examined–explored the candidates’ past records in office in more than a passing reference.

The study examined 430 stories published or aired over two weeks leading up to the Iowa and New Hampshire contests, in five major newspapers and nine television programs on five networks. This was the period when voters were beginning to more seriously focus in on the presidential contest.

Some journalists might counter that the kind of background reporting that is missing was done earlier. While we cannot quantify or confirm this, most evidence suggests that, even if it were true, the public was not yet paying attention. However, the study did capture a time when some papers were doing their big background stories-including three major takeouts in the Washington Post on Bill Bradley’s formative years. Yet this reflects what a small percentage of the mix these pieces represent.

At times, some in the press even sounded resentful of the campaign. Listen to Bryant Gumbel interviewing Hotline editor Chris Crawford January 17 on CBS’s The Early Show:

Gumbel: “I stumbled upon Saturday’s (debate) and it seemed a rather sad show. I mean, here were all the Republican candidates sitting there on a Saturday afternoon answering questions from people in Iowa, and it seemed like, you know, it was just going through the motions.”

Crawford: “Yeah…these debates are sort of like phantom pain…It’s sort of-its gone away, but we still feel it.”

Looking just at the topics with which stories were predominantly concerned, the majority of the stories, 54%, were about political topics, such as fundraising or tactics. A sizable minority, 24%, were ostensibly policy related. Another 11% related to the candidates’ leadership style or health.

But when we looked at how these stories were put together-or framed-around these topics, the coverage takes on a more tactical cast. Only 4% of stories were developed to clearly explore the candidates’ ideas. Only another 3% were developed around the broader theme of their core convictions.

And for all the talk about the importance of character, just 5% of stories were framed around the candidates’ personality or character. Just 4% looked at the candidate’s leadership style.

Overall, the coverage paints a picture not of a contest of ideas between men but of a massive chess game of calculation and calibration in which little seems spontaneous or genuine. And occasionally, the camera turns to the audience for a shot of its reaction.

This comes even though many reporters have called this an unusually issue-oriented campaign. It also comes relatively early in the process, a time when it is still possible to explore the candidates in a larger sense before the dizzying pace of the primaries following Iowa and New Hampshire has begun.

The findings are also striking, given research this year that suggests people do not care to read about internal tactical matters. Rather they say this year they want to know most about the public character of these candidates, including their records, their honesty, and how they connect with people, according to data from the Pew Research Center for the People and the Press.* What’s more, the focus on so-called inside baseball is hardly new, and the political press has vowed in years past to seek better ways of connecting with voters, and making the campaign more relevant. Apparently, even in the early days of the campaign, the press has had difficulty keeping sight of that goal.

The study, which was designed and written by the Project and executed by researchers at Princeton Survey Research Associates, examined stories produced during the week ending January 19 and the week ending December 15. The newspapers studied were the Cleveland Plain Dealer, the Orlando Sentinel, the New York Times, USA Today, and the Washington Post. In broadcast, the study looked at ABC World News Tonight, ABC Good Morning America, CBS Evening News, CBS Early Show, NBC Nightly News, NBC Today, CNN The World and the PBS News Hour and Larry King Live.

The sample, while not exhaustive, is an attempt to be representative of the media universe from which the largest number of Americans get their news about the campaign.

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

As for the candidates, the study finds:

• Attacking a candidate is not the surest way to get coverage. Only 6% of the stories were triggered by a candidate or campaign attacking a rival.
• Getting more national coverage does not necessarily translate into better electoral results, at least early on. Steve Forbes and Alan Keyes, for example, received almost no individual coverage leading up to Iowa.
• If politics in part is a battle for control of message, Al Gore fared better than Bill Bradley. The coverage of Bradley focused more on his personality and health than on his much-touted “big ideas.” The coverage of Al Gore, in contrast, was arguably closer to the campaign he wanted to run, paying scant attention to his supposed weak spot-his personality-and emphasizing matters over which he wanted to challenge Bradley-his ideas.
• Among Republican candidates, John McCain arguably had the most control over his message-in the sense that more of his coverage was candidate driven rather than driven by the press or others.

As for the press, the study finds:

• Coverage is not predominantly triggered by what the candidates say or do on the stump. It is almost twice as likely to be initiated by decisions in the newsroom to do an analysis or other enterprise piece.
• In print, the more local the newspaper, the more it covered policy topics. The Orlando Sentinel was the most likely to cover policy, 37% of stories. USA Today was least likely, 18% of stories.
• The New York Times tends to cover what the candidates say and do more as straight news and then write political analysis stories alongside. The Washington Post, in contrast, tends to initiate its own stories, but focus more of them around the candidates’ characters and policies.

* “Bradley and McCain Bios Count More: Campaign Incidents Have Little Punch, December 16, 1999, The Pew Research Center for the People and the Press. “Candidate Qualities May Trump Issues in 2000,” The Pew Research Center for the People and the Press, October 18, 1999.

The study, the first of several the Project will provide through the course of the campaign, examined two weeks of coverage, enough to be sizable and still allow the results to be timely. Future studies will focus on other areas as well, such as the Internet and on additional cable outlets, though these reach far fewer people in general.

The goal of this study was to identify what was covered, how, and to whom it related.

To do so, the study broke each story down three ways. First, it identified what each story was about, topic. Then it noted how each story was put together (Was it a straight news account, or was it framed around political concerns like tactics, around policy, personality, etc.?)

Third, who was affected by what the story was about, or who did it impact? Was it citizens? Politicians? Interest groups? Or a combination?

In addition to these measurements, the study also noted two other features for each story.

The first was what initiated the story, its trigger: Was it something a candidate said or did? Something from his campaign surrogates? An outsider? Or was it press enterprise?

Finally, the study measured the tone of each story. Within its frame, was the story predominantly positive, negative or neutral? In order to fall into the positive or negative category, 50% more of the stories had to fall clearly on one side of that line or the other.

Topic

Even just looking at what subjects were covered, the majority of stories (54%) concerned strictly political matters-polls, tactics, fundraising, etc.

Tactical maneuvering was the most common political topic, accounting for 21% of all stories, followed by candidate performance (9%). Polls and momentum was the next most common political topic (7%), followed by stories about the political calendar (4%) and advertisements (4%).

On the surface, about a quarter of the coverage (24%) was nominally about policy. Another 11% concerned the personal background of the candidates. Fewer than one in ten (7%) concerned voters. And just a fraction (1%), concerned the candidates records.

When it came to policy alone, social issues were the most common (7% of all stories or 31 in all), followed by taxes (5% or 20 stories), health care (3% or 11 stories) and both campaign finance reform (2%) and military issues (2%)

Frame

But when we looked to see how these topics were treated-or framed–we found that in the writing or production process many of these stories were refocused so that they became predominantly about something else.

Consider how, on December 14th’s Good Morning America, George Stephanopoulos frames John McCain’s opposition to federal subsidies for the alternative fuel called ethanol. “It’s what a friend of mine called a ‘candor pander,’ and what he’s doing here is hoping that this straight talk, even though it would end up sacrificing the state of Iowa, will appeal to the rest of the country where it fits in with his point that special interests have too much influence in Washington.”

The inverted pyramid, or straight news account, remains the most common way of telling a story. But it is hardly the dominant one. This is a story in which the news is presented not in any thematic way, but as a traditional description of what happened, offering who, what, when, where, why, and how in rough order of their importance. In all, 38% of stories were written as straight news.

Interestingly, straight news accounts are the primary way in which the press writes about candidates’ ideas. Four out of ten straight news accounts were about policy topics.

On the other hand, reporters apparently believe that policy stories are a turn off. When they develop stories about a policy oriented event into something other than a straight news account of the facts, they rarely choose to explicitly explore policy. While policy made up 24% of the topics covered, only 4% of stories were framed as explorations of those ideas.

After straight news accounts, the next most common way of telling a story was to build it around strategy and tactics. Fully 22%, nearly a quarter of all stories, were told within a tactical frame.

Another 9% of stories were told as horse race stories, who was moving up or down.

A significant number of stories, 12%, were crafted in a way that they told us more about larger issues involving the political system, such as the concerns of voters, or the changing role of primaries.

Yet relatively few stories were developed in a way that delved into the candidate himself. For all the talk about character this year, just 5% were framed around a candidates’ personality and temperament. Four percent looked at a candidate’s leadership style. And 3% considered the health of a candidate. In a way, the character of each of the candidates is lost in the focus on tactics and strategy. Tactics becomes the motive for everything. Even the candidates’ beliefs take on an air of insincerity and calculation.

Consider this Washington Post story December 15 about George Bush differing with John McCain on finance reform and taxes. Bush’s differences, it said, are “a sort of political judo….By highlighting points on which McCain strays from party orthodoxy, Bush is trying to build a firewall around New Hampshire-in the Iowa caucuses a week before the Granite State primary, for example, and in the South Carolina primary soon after.”

Tone For all that the press is often charged with negativity and cynicism, there is no proof that the coverage is biased toward one party over another, one candidate over another, or is relentlessly negative. Overall, fully 44% of the coverage was neutral in tone toward the dominant figure in the story. An evenly balanced 24% was positive, and 24% negative. The numbers were virtually identical when it came to coverage of Republicans or of Democrats.

Speculativeness

The study also tried to find out how much of the coverage involved journalists speculating on the future. To do so, we looked at each story to see if it was mostly about things had already happened or things yet to come which the journalist could not know for sure. For example, did a story about tactics mostly discuss the latest move by a candidate or did it speculate as to whether the journalist thought those tactics would lead the candidate to a victory in New Hampshire?

We found that the press was not particularly speculative. Less than two in ten stories were mostly speculative in nature. A full 84% were not. What’s more, this was true for both print and broadcast.

Frame and Tone The tone of stories tended to be influenced by how the reporter framed the story. When reporters develop stories around policy, they apparently take pains to avoid making judgments. Policy stories were much more likely than others-including even straight news accounts-to have a neutral tone, 63%. Straight news accounts were neutral 56% of the time.

When reporters developed stories around political matters, however, they apparently feel more confident making or seeking out normative judgments. Only 30% of the tactics and strategy stories were deemed neutral, as were 26% of the horse race stories.

Impact

The study also tried to isolate whether the coverage was relevant or not to citizens. One way of doing that is to note who is primarily affected or impacted by the concerns that the story is talking about. We called this measurement impact. Did the story affect citizens? Was it talking about things that only impacted the candidates and their parties? Did it affect specific interest groups? Or did the story touch on how several of these constituencies might be affected?

The topic of the story does not necessarily determine its impact. Even a story about internal political matters could be written in a way that impacts citizens, if the story made clear that how a campaign is run reveals how a candidate would govern. Few stories we saw attempted to make this link between tactics and their relevance. An overwhelming amount of the coverage (82%) dealt with things that mainly affected only the candidates, their campaigns and campaign workers. This involved such matters as who was winning or losing, their strategies, fundraising, etc.

Only a little more than one in ten stories (13%) dealt with matters that affected mainly citizens, subgroups of citizens, or even subgroups in a given state or county. These are such things as a candidates record, his honesty, his policy ideas, his ethical background, etc.

A small percentage of the coverage (4%) mainly affected specific interest groups, such as the National Rifle Association or the National Association for the Advancement of Colored People.

Remarkably, only 1% of all the stories were written in such a way that they addressed how multiple stakeholders would be impacted. A story need not just impact one group, citizens versus politicians. It can be written in ways that show how different groups, or what some call “stakeholders,” are affected. A story about Bill Bradley’s health care plan, for instance, could explore both how he was being attacked for it and who it covered versus his rival’s plan. Again, however, few stories were written in a way to make clear their relevance to more than one constituency.

Overall, the findings here suggest that journalists may want to be far more conscious of crafting stories in ways that, regardless of the general topic, make their relevance clear and address the concerns of voters, not just insiders. In that sense, this idea of writing a story with a mind toward its impact may be a way of helping journalists cover matters they consider newsworthy and making sure that their coverage remains relevant to the largest possible audience.

Frame and Impact

One way of doing this is to consider how stories are framed.

The most popular story frames identified in this study tend to leave citizens out.

For instance, 95% of the stories framed around tactics and strategy impacted politicians. This was also true of every single story framed around horse race.

Stories framed around the political system as a whole or the nature of politics, however, were much more likely to be about things that affected voters. Fully 35% of these stories had citizen impact, though 54% still related overwhelmingly to politicians.

Journalists were most likely to write stories that impacted citizens when they developed a topic by exploring the policy issue involved, which, as mentioned above, they did little of in this study. Still, 50% of stories framed around policy were written in such a way that they made clear the impact on citizens. A quarter of these stories impacted specific interest groups and another quarter impacted candidates.

The Watchdog Role

Playing watchdog–by scrutinizing the veracity of rhetoric or advertisements or conducting investigations–has not made up a large percentage of the press’ role, at least in the period studied. All told, only nine of the 430 stories were of a watchdog nature. There were no investigative reports. Most of the rest were ad watches, more than half of which were published by one news organization, the Washington Post.

The study reveals some clear differences in the way candidates were covered, not so much positively or negatively, but the nature of the coverage, what got covered and who controlled the coverage. Bill Bradley

If the former Senator from New Jersey wanted to be the candidate of big ideas, he did not succeed in doing so in the press in the middle of December and January.

To begin with, he was less successful than rival Al Gore in projecting himself in the press as talking about ideas. Only 12% of the coverage of Bradley focused on the topic of his ideas.

Arguably, this may be because Bradley rolled out some of his major policy positions in November, the month before the study began. Yet the numbers here reveal how hard it can be to sustain press coverage of one’s ideas for very long. Remember, too, that while the campaign started early, many voters even in Iowa and New Hampshire-let alone elsewhere-may not have been paying close attention before December.

Rather than focus on Bradley’s ideas, the press had its gaze on his health or fitness for office. Fully 36% of the stories in which Bradley was the dominant figure focused on the topic of his health, compared with 6% for candidates overall. Bradley’s first incident of a heart murmur occurred during the first week the study examined. Interestingly, this event was a minor story in print, accounting for just 3% of coverage. Yet it was a major event on television, the second biggest story after the then-pending Iowa caucuses themselves.

One reason, perhaps, is that the networks are now heavily invested in health reporting, with doctor-reporters on staff, ready to mull over the implications of any medical event. In this case, the networks may have focused heavily on an incident that the medical community has clearly established poses no meaningful risk whatsoever to Bradley’s fitness for office. “The disorder amounts to little more than a nuisance, according to the American Heart Association,” as the Atlanta Journal and Constitution, and many other newspapers reported.

Al Gore

The Vice President, in contrast, managed to get coverage that matched the kind of campaign he wanted to project.

As an example, 45% of the stories about Gore were candidate driven (that is, they were triggered by things the candidates themselves said or did). By comparison, only 33% of Bradley stories were candidate driven. Gore, in fact, had the most candidate-driven coverage of any candidate in either party. To the extent that political candidates see their relationship as a struggle for control with the press over the story that is told about them, Gore was the most successful candidate in dictating his coverage.

In a similar vein, Gore had more success in controlling his coverage than Bradley, or any other candidate, in the sense that more of the stories were written as straight news accounts rather than framed around some more analytical or thematic story angle of the news outlet’s choosing. Fully 47% of the coverage of Gore was written as straight news compared with 34% for Bradley.

Perhaps this explains something else. If Gore wanted to project himself as a candidate of ideas-not just Bill Clinton’s Vice President–and to downplay his reputation as a dull or stiff personality, he succeeded in the press. The Vice President was nearly three times as likely as Bradley to get coverage of his policy ideas (30% versus 12% for Bradley). And he was ten times less likely to have stories focused on his personality or personal fitness (4% versus 44% for Bradley).

For Gore, as for others, most of the coverage focused on political matters like tactics. This, too, may have helped him somewhat, for it came at a time when he was righting his campaign before Iowa and gaining momentum. This is part of the inevitable challenge for political reporters. How are they to write about candidates without amplifying the momentum up or down? In this case, the overwhelming focus of the coverage on tactics and strategy tended to benefit the candidate on the rise and frustrate the one sinking.

John McCain

The Arizona Senator was the most successful of the Republicans at controlling his coverage in the sense that stories about him were candidate driven. Fully 40% of the stories about McCain were candidate driven compared with 26% for Bush, and 17% for all other Republicans.

McCain also got more coverage of his ideas than any other Republican. Fully 31% of the McCain stories were about his policy ideas, compared with 25% for Bush and 17% for the others.

This may be a signal that accessibility pays. McCain offers reporters constant access, unlike any candidate in years. The result is not so much that the coverage is more positive, but that the candidate is a direct link to the reporters covering him. The campaign is not filtered through aides and spin doctors. The candidate is the story, and so can dictate to a greater degree what he wants the coverage to be about.

George W. Bush

The Texas Governor stood out in one significant feature. He won the race for the most coverage. A full 18% of all the stories studied were predominantly about Bush, compared with 14% about Bradley, 14% about McCain, and 11% about Gore.

One feature of the Bush coverage is that nothing stood out. The press, in other words, did not fix on a particular feature or question about his candidacy, as it did in the case of Bradley’s health. Bush was the only candidate to receive any coverage explicitly focused on how smart he might be, five stories in all, or roughly 7% of the coverage in which he was the dominant figure. Another 13% of the coverage of Bush examined his performance as a candidate. During the debates leading up to Iowa and New Hampshire, some critics questioned whether Bush had handled himself well, especially given the high expectations about his skills on the stump. These may even in some people’s minds have connected to whether he was up to the job of president intellectually. Yet the amount of coverage, while noticeable, does not rise to the level of becoming a major story at this point, as Bradley’s health clearly did.

Roughly half of the Bush stories (54%) dealt with political topics, less than was true of McCain or the other GOP contenders.

Other Republicans

Outside of the two main contenders, Bush and McCain, Republicans had little luck getting coverage of their ideas at all. In total, only one story was produced that focused solely on the policy positions of Steve Forbes, Alan Keyes, Orrin Hatch, or Gary Bauer. Two more did look at their core convictions. The vast majority concerned political matters. The debates clearly helped these candidates in that they were events at which they stood as equals to Bush and McCain. They also provided them with TV exposure that they would not have otherwise gotten in the national press.

Steve Forbes

For all the money he spent and his success in Iowa, the business magazine publisher generated only marginal coverage, accounting for just 3% of the stories. Only two of these stories were candidate driven. Of all the candidates in either party, Forbes stood out for having the largest percentage of his coverage be about his tactics and strategy, roughly six out of ten stories.

The study also suggests, perhaps, two models for candidates to drive coverage of their campaigns. One is the Gore model, which is tightly focused and controlled, and aided, of course, by the fact that the Vice President has the trappings of his office, the experience of White House staffers and the logistics of the Secret Service and the executive branch at his disposal. The other might be called the McCain model, in which openness and access to the candidate leads to coverage of the ideas he’s running on and the policies he might implement.

What makes something a story in campaign coverage? Apparently, the decisions of reporters, editors, correspondents and producers are the main answer. Fully 54% of all stories were initiated not by events outside but within the newsroom in the form of analytical or enterprise stories. The most common of these was to do analysis stories, which made up 42% of all the press driven coverage. Another quarter of these press driven stories were enterprise pieces that probed the candidate’s history, the status of the campaign, etc. Media polls made up another 10% of the media driven stories.

Policing political advertising and rhetoric did not constitute a sizable amount of the coverage, as mentioned above under the watchdog role.

The second most common way to make news is through what the candidates say themselves, but it was much less likely to generate a story than was the enterprise or analysis of journalists. Roughly 15% of the stories were initiated by an individual candidate speaking on the stump.

Looking deeper, it is not true that the candidate needs to talk trash to make news. Candidates generated slightly more stories (9%) with non-accusatory rhetoric than with attack statements, which accounted for just 6% of all stories. This does not, however, account for nasty things a candidate might have said in a debate. Another 9% of the coverage was triggered by debates, a place where such vitriol seems generally safer, presumably because the opponent is there and can defend himself.

Trigger and Topic

Candidates apparently want to talk about policy more than do others in the campaign community, including the press. When stories were triggered by something the candidate said or did, roughly half (48%) were policy related. In contrast, when the press initiates a story, it seems to be most interested in politics. A notable 58% of all press-initiated stories were about internal political topics. The press divided the rest of its stories among policy topics (14%), the personal background of the candidates (13%) and the electorate at large (11%).

When others trigger stories-be they outside observers, campaign surrogates or advertisements, the topic was also largely political in nature (61% and 62% respectively).

Trigger and Frame

Even when candidates triggered stories about policy topics, journalists did not always develop those into policy stories. Stories triggered by candidates were usually either a straight account of what the candidate said or did (72%) or were developed around tactics and strategy (16%).

Only 3% of stories triggered by the candidates were written or produced by journalists as explorations of their policy positions.

Stories the press initiated, on the other hand, were mostly about tactics and strategy (28%) or the electorate and the nature of politics (22%). Another 14% developed the theme of who was winning or losing.

There is a good deal of social science research now suggesting that people are not interested in learning about the inside baseball of politics, or being educated in the black arts of how to run campaigns. The survey data from this year tends to reinforce that. In October Americans told pollsters from the Pew Research Center for the People and the Press that what they considered most important to learn about a future president was first, his honesty, second, how well he connected with people, and third his record. Subsequent data has tended to corroborate that people are not just saying what they think is socially acceptable. The discovery that Al Gore had feminist author Naomi Wolfe advising him on how to reach women voters had no impact on how voters felt about him, Pew research in December found. Conversely, people’s view of John McCain was influenced by how he handled his experience in Vietnam and stories about his temper.

The Medium clearly makes a Difference

Of the 106 television broadcasts studied on nine programs, television averaged less than one story per show each day. That compares with roughly five stories a day on average in the newspaper.

In general, television covered more about the personalities of the candidates, 16% versus 10% for print. It was less likely to cover policy topics, 16% of stories versus 27% for print, and the electorate, 6% versus 10% for print.

It covered more stories about how the candidates were performing with the media. That comprised a marked 20% of TV stories, compared with 7% for print. It did fewer stories about tactical topics, 15% versus 24% for print.

And as mentioned earlier, it did a good deal more about candidate health, 15% versus 3% for print.

TV also covered Bradley a good deal more than Gore, again heavily on his health, 19% of the stories were about Bradley, 5% about Gore.

TV coverage was less candidate driven than print, (22% versus 30%) and more driven by decisions in the newsroom (62% versus 51%).

The TV coverage was also more subjective or analytical. Fully 40% of TV stories were press analysis stories, versus 18% in print. In a similar vein, TV was less likely to do a straight news account of an event, only 28% of stories were framed that way versus 40% for print. And more of the stories in TV were framed around political matters, rather than policy or personality, 42% versus 32% for print.

Finally, more stories were framed around tactics than in print, 13% versus 8%.

The numbers for individual programs are too small to offer any refined comparisons, but one set of differences are worth mentioning. One program, ABC’s Good Morning America, took a noticeably different approach than other programs.

The program was twice as likely to cover policy topics as any other program. It did not do a single story about the candidates’ personal qualities. And it was more likely to do stories based on what candidates said or did and less likely to do stories based on newsroom decisions than any other program.

Larry King Live, in contrast, did no segments about the presidential campaign during the two weeks examined.