Source material: new presidents and network news: covering the first year in office of Ronald Reagan, Bill Clinton, and George W. Bush.

Farnsworth, Stephen J. ; Lichter, S. Robert

Introduction

New presidents and their executive branch teams endeavor to make a good first impression. To that end, incoming presidential administrations carefully organize their public affairs staff, select issues they wish to emphasize, and try to present the president in ways that will maximize public support and approval of the new chief executive (Han 2001; Kumar 2002, 2003; Lowi 1985). While some media stories develop as a result of administration efforts to set the political agenda, including the presidential tax plans of 1981, 1993, and 2001, there are limits to what can be accomplished through executive branch media management initiatives. Often-skeptical reporters may or may not accept the framing of a story proposed by the executive branch (Cook 1998; Iyengar 1991; Kurtz 1994). More importantly, unanticipated events, including the assassination attempt on Ronald Reagan in 1981 and the terrorist attacks of September 11, 2001, dramatically affect presidential popularity and can come to dominate media coverage of a significant part of a presidency (cf. Gregg and Rozell 2003; Iyengar 1991; Kernell 1997; Woodward 2002).

The content of news reports can expand or limit the ability of presidents to pursue their policy agendas (Abramson, Aldrich, and Rohde 1982, 2002; Booth 2002; Campbell and Rockman 1996; Ceaser and Busch 2001; Kumar 2003; Sabato 2002; Sabato, Stencel, and Lichter 2000; Woodward 2002). Presidents whose administrations are viewed as effective are likely to enjoy the high public approval ratings that help with passing legislation (Gregg and Rozell 2003; Kernell 1997).

For most people most of the time, public perspectives on the presidency are formed largely through a news media led by the "big three" broadcast television networks of ABC, CBS, and NBC (Farnsworth and Lichter 2003; Graber 2002). Although their influence has declined over the past quarter century, the networks, along with the many news organizations around the country that follow their lead, are the lenses through which most citizens view their government (Cook 1998; Graber 2002; Sparrow 1999). Newer media sources, including 24-hour cable news, the Internet, and a revived talk radio, have not displaced the old; instead, they offer a wider range of choices for news consumers, many of whom continue to rely heavily on the evening network newscasts (Davis 1999; Davis and Owen 1998; Farnsworth and Lichter 2003; Seib 2001).

This study examined network television news coverage of the federal government during the first calendar year of the presidencies of Ronald Reagan, Bill Clinton, and George W. Bush. For Bush, we also divided the results into coverage before and after the terrorist attacks of September 11, 2001. (We excluded the military services, whose role in government is distinctive enough to require its own analysis.) The three presidencies examined here comprise the three most recent partisan transfers of power involving the executive branch. Partisan changes of power involve substantial policy and personnel shifts, and therefore represent more comparable periods for comparison than a same-party change of power. They are also periods of heightened attention to the White House, as partisans look to the future with either optimism or dread.

Although presidents do not take office until January 20, presidents-elect become the focus of governmental coverage well before the inauguration ceremony. We therefore included stories about the incoming administration starting January 1 (but excluded the coverage of the outgoing administration). This approach enabled us to examine the status of proposed agenda items and the reception given to Cabinet nominees and other appointments as a new administration made its transition into power.

A comparative study of first-year coverage of the three most recent partisan changes in the executive branch enables us to chart changing trends in the volume and tone of network news reports during three different decades. We also compare network news coverage of the executive branch with that of the legislative and judicial branches to examine the shares of coverage devoted to the three branches of government in 1981, 1993, and 2001.

Network Television News and the Presidency: Past Research

Debates over media content start from a point of general agreement: today's television news reports of government are more negative than in the past (cf. Farnsworth and Lichter 2003; Kerbel 1995; Patterson 1994; Sabato 1993). Print coverage of the president from 1953 to 1978 and network news coverage from 1968 to 1978 showed a consistently favorable portrayal of the president (Grossman and Kumar 1981). At CBS, for example, the amount of positive coverage approached double that of negative, with a 45 percent to 24 percent advantage for more positive coverage during the 1974-1978 period (Grossman and Kumar 1981, 265). In the New York Times, the comparable percentages for that period were 38 percent positive and 23 percent negative (Grossman and Kumar 1981, 265). Another research project involving the CBS Evening News over a somewhat different time frame, from 1968 to 1985, found a trend of rising negative coverage for presidents during those years (Smoller 1986).

Experimental evidence strongly supports the proposition that less substantive news coverage increases public cynicism. "When journalists frame political events strategically, they activate existing beliefs and understandings; they do not need to create them" (Cappella and Jamieson 1997, 208). But research in this area has not identified a precise threshold where audience effects become observable in experimental settings.

The declining volume of political news coverage over the course of recent decades is another significant development. In 1977, nearly 72 percent of the stories on ABC's evening newscast focused on political news (including military, foreign policy, and domestic policy matters); in 1997 the percentage of such stories fell to 45 (Graber 2002, 274). Reduced political coverage undermines public interest in and respect for government, according to past research (cf. Edelman 1985; Jamieson 2000; Patterson 2000; Postman 1985; Putnam 2000).

In a third major finding in past research, the executive branch is far more frequently the subject of news coverage than the legislative branch, with the judicial branch functioning as little more than an afterthought for television reporters (Grossman and Kumar 1981; Hess 1981; Kaid and Foote 1985; Lichter and Amundson 1994). This should not be surprising, as a modern president is commander-in-chief of the world's sole remaining superpower, legislator-in-chief when it comes to setting the national agenda, and the head of a largely unified political institution (particularly when compared to a Congress divided by partisan conflict). Presidents and their administrations possess a near-monopoly of foreign policy information and frequently can be decisive actors in international and domestic matters (Lowi 1985; Waterman, Wright, and St. Clair 1999). Given these structural advantages, presidents and their administrations possess essential information for news-hungry reporters (Cook 1998; Kurtz 1994). Much of what Congress does, in contrast, is tentative, and therefore less newsworthy (Hibbing and Theiss-Morse 1995). A presidential pronouncement about a policy preference is Far more definitive, for the executive branch, at least, than a lawmaker talking about a bill that may not pass in both chambers, or that may not survive a conference committee, or that may be vetoed by the president (Graber 2002).

Quantitative analysis has demonstrated that a president has a far greater opportunity to be seen and heard than do his adversaries. A study of the year ending June 30, 2000, for example, found that President Clinton received an average of 60 percent of the news coverage of the national government on the three networks, while Congress received 31 percent and the Supreme Court received roughly 9 percent of the coverage (Graber 2002, 275).

Presidents are thought to enjoy a powerful incentive to use the media to build public support for their political agendas. "Going public" can be seen as a way to break the legislative logjam that may tie up a president's initiatives on Capitol Hill, and it can be a way of helping convince uncertain lawmakers to support the White House (Kernell 1997). An individual lawmaker, even a party leader on Capitol Hill, is far less able to attract sufficient press attention, and therefore public interest, to compete with a president who uses his domination of the nation's airwaves and news columns to advance his agenda (Alger 1996; Han 2001; Lowi 1985).

Conservatives have long complained about allegedly biased coverage by the so-called "liberal media," particularly during presidential campaigns (cf. Goldberg 2002). In 2000 some Democrats perceived an anti-Gore bias in the media, either as a result of personal antipathy or "bending over backwards" to be fair to Bush (cf. Hershey 2001).

The quantitative evidence on media bias is mixed. Over the past two decades, scholarly content analyses of election news have found that Democratic presidential candidates sometimes (in 1984, 1992, and 1996) received more favorable treatment than their GOP opponents; at other times (1980, 1988, and 2000), coverage of the major party nominees was roughly balanced (Clancey and Robinson 1985; Farnsworth and Lichter 2003; Just et al. 1996; Kerbel 1998). The data here address the question of whether the media favor one party or the other after the winners turn from running for office to running the government.

The Data

The 16,078 network news stories analyzed here represent a total of 422 hours of coverage of government. Coverage of the executive branch represented 88 percent of the news devoted to the national government in 1993, 82 percent of the 2001 coverage, and 76 percent of the 1981 network news coverage of government. A network news story was included if at least one third of its length was devoted to discussions of the federal government--the president, any Cabinet member, any agency of the administration, the U.S. Congress, the Supreme Court, and other federal courts. In addition to the volume of coverage, we analyzed each story's focus, be it on job performance, personal character and ethics, political effectiveness, or other qualities of these administrations. Because allegations of partisan media bias are a staple of the political debate, we compared coverage of the sole Democratic administration in this study with its Republican successor and a recent predecessor. To put the executive branch data into perspective, we also examine the treatment of the other two branches during these three years.

The leading topics of first-year network news coverage varied considerably. The war on terrorism, including the handling of both 9/11 and the anthrax attacks, was the dominant story of 2001, representing 24 percent of all coverage. If one includes the categories of general defense, the war in Afghanistan, and homeland security issues, the figure rises to 36 percent of news coverage of the executive branch. In 1993, the top policy issues were defense, which represented 17 percent of all stories, along with health care and the federal budget, each of which represented 13 percent of coverage. The budget deficit dominated network news coverage of the executive branch in 1981 with 10 percent of the coverage, followed by the USSR, the focus of 6 percent of coverage, and the controversial sale of AWACs planes to Saudi Arabia, the subject of 6 percent of the coverage during Ronald Reagan's first year in the White House.

Our analysis is based primarily on individual statements or sound bites within each story. Although time-consuming and labor-intensive, this allowed us to analyze the building blocks of each story separately, rather than making summary judgments of entire stories. Instead of coding an entire story as "positive" or "negative" toward an individual or institution, we coded each evaluation within the story for its source, topic, object, and tone. A single story might contain several evaluations of various actors; our system captured each one individually. This procedure produces a very detailed picture of the news media's treatment of government. Video images are more subject to differing interpretations than the text of news reports, and the content analysis conducted here, like other research in this area, uses the words spoken rather than the pictures shown on television news to maximize inter-coder reliability (Farnsworth and Lichter 2003; Just et al. 1996). Further information regarding the research methodology used here is found in the Appendix.

Amount of Coverage

Both 2001 and 1981 saw far more stories about government than did 1993, the first year of the Clinton administration. As shown in Table 1, there were 16,078 network evening news stories on government during those three years, which represents over 422 hours of coverage. During Ronald Reagan's first year as president, the three networks provided 7,216 stories on the national government. The coverage shrank to just over half that level, 3,848 stories, when Clinton took office in 1993. It then rose to 5,014 stories during George W. Bush's first year in office.

Of course, the increase in governmental news in 2001 reflected special--indeed unprecedented--circumstances. The rebound was entirely due to the sharp increase in coverage that followed the 9/11 terrorist attacks on the World Trade Center and the Pentagon. After the attacks, governmental coverage increased almost 50 percent, from a combined average of 20 minutes a night from January 1 through September 10, to 29 minutes a night for the rest of the year. That made the post-attack coverage in 2001 about as high as the average nightly coverage for all of 1981. (Our analysis includes only the regularly rescheduled evening news time slot during the unprecedented four days of continuous coverage that followed the attacks.)

Both the 1981 totals and the post-attack portion of 2001 were well above not only the 18 minutes per night average of news coverage of government during Clinton's first year as president, but also the 20 minutes per night average during George W. Bush's first eight months as president. If we projected the 2001 annual total from the pre-9/11 coverage, the number of stories would have been only slightly higher than in 1993.

Executive branch coverage makes up so much of the media's government news agenda that any general trend in the coverage would necessarily show up in its largest component. As Table 2 shows, the networks broadcast a total of 15,307 discussions of the executive branch during the three years examined here. The number of discussions stood at 6,512 in 1981, before falling first to 4,871 in 1993 and then to 3,924 in 2001, an overall drop of roughly 40 percent.

The continuing downward trend in 2001 is notable in view of the events that took place during that year. The Bush administration received the least coverage, despite the deadliest terrorist attack against civilians in the country's history. Even the subsequent military intervention in Afghanistan was insufficient to reverse the trend of a decreasing volume of network news coverage of the executive branch.

Another pattern in coverage of the executive branch was the increasing focus of network news on the president. In 1981, coverage of Ronald Reagan was exceeded by coverage of the White House and Cabinet staffs, as well as the coverage afforded other parts of the executive branch. In 1993, Bill Clinton personally received more coverage than his White House and Cabinet staffs. But the remaining parts of the executive branch, including federal agencies, received more than twice his coverage. By contrast, President George W. Bush received more coverage than his White House and Cabinet staffs, and the gap between coverage of the president and the other parts of the executive branch narrowed in 2001.

In comparing the legislative to the executive branch, the 3,004 discussions related to Congress during the three years represented less than one fifth of the executive branch's coverage. In fact, it represented a smaller amount than any of the three executive branch categories. The best year for Congress in terms of the proportion of federal government coverage was 1981, when 20 percent of the stories featured the legislative branch. That year had the greatest number of discussions of the legislative branch. The 1,739 network news discussions that featured Congress during 1981 were more than three times the 531 discussions about Congress in 1993 and twice that of the 734 legislative branch stories in 2001.

The judiciary is sometimes called the invisible branch of government. This proved to be literally true on network television, where the courts received only 3 percent of the network news coverage of the federal government in each of these years. The 1993 and 2001 totals represent an average of roughly 50 stories apiece for the three networks annually. Thus, a viewer who faithfully watched a network newscast every night would have seen a discussion related to the federal courts less than once a week.

The competition for media coverage among the branches of the federal government became even more one-sided in the weeks that followed the 9/11 terrorist strikes, as Table 3 demonstrates. The networks' attention focused almost entirely on the executive branch in the wake of the attacks--88 percent of coverage, compared to 79 percent earlier that year. During times of crisis, leadership typically becomes more personalized, as the effects of 9/11 on former New York Mayor Rudy Giuliani's public image attest. But the proportion of President Bush's personal coverage actually dropped by almost 30 percent, from 28 to 20 percent of all government news. This anomaly likely reflects a leadership style in which the president frequently called upon his subordinates to give a public face to the administration's response (Woodward 2002). Mr. Bush's managerial style of delegating authority and visibility to his Cabinet, the White House staff, and other appointees was reflected in increased airtime for Defense Secretary Donald Rumsfeld and Attorney General John Ashcroft, among others. The administration's media management strategy helped secure for Bush an extraordinarily large and lengthy "rally around the flag effect" in the two years that followed those attacks (Gregg and Rozell 2003; Kumar 2002, 2003; Nincic 1997).

The executive branch was the focus of 2,597 stories before the attack, compared to 584 stories that featured the legislative branch during this period--a margin of about 4 to 1. In fact, there was considerably more coverage of President Bush personally than there was of the entire legislative branch. After the attacks, Congress received proportionately even less coverage. The 1,327 stories that featured the executive after 9/11 represented nearly nine times the coverage provided to the legislative branch. Indeed, the 298 stories that featured President Bush alone were nearly twice the 150 stories on all members of Congress combined during this period. Finally, coverage of the judicial branch almost disappeared. In 2001 there were 118 stories on the federal courts before the terrorist attacks (about four per week), but only 28 afterward (less than two per week).

Tone of Coverage

As Table 4 shows, coverage of the executive branch was quite similar in tone for all three administrations, ranging from a low of 34 percent positive in tone in 1981 to a high of 38 percent positive in 1993. Personal coverage of the presidents was nearly identical despite very different media management styles: 39 percent positive evaluations of George W. Bush, versus 38 percent for Bill Clinton and 36 percent for Ronald Reagan. Coverage of Congress was most favorable in 2001, rising to 35 percent positive from 28 percent in 1993 and 13 percent in 1981.

The House and the Senate received roughly similar treatment in 2001--37 percent positive for the House and 34 percent positive for the Senate. This was a much closer margin than the 20 percentage-point advantage that favored the House in 1993. In 1981, the tone of Senate coverage was more favorable than that of the House by 6 percentage points.

The Supreme Court's more favorable coverage reflects the praise accorded two recent Supreme Court nominees: Sandra Day O'Connor in 1981 and Ruth Bader Ginsburg in 1993. The federal judiciary endured by far its most negative treatment during 2001, in the wake of the Supreme Court's controversial 5-4 decision to stop the 2000 presidential election ballot recounts in Florida. Even so, it received the least negative coverage in all three years examined here.

Table 4 provides a comparison of the different presidents and their administrations. The personal coverage of Ronald Reagan was 36 percent positive, compared to 38 percent positive for Bill Clinton and 39 percent for George W. Bush. For the executive branch as a whole, the comparable figures were 34 percent positive in 1981, 38 percent in 1993, and 36 percent in 2001. The most obvious interpretation of these findings is that all three administrations received mainly (and almost equally) negative coverage.

Television news coverage during 2001 was a tale of two presidencies, as shown in Table 5. Before September 11, coverage of the new administration was only 35 percent positive. After the terrorist attacks, coverage of the executive branch overall improved somewhat, rising to 42 percent positive. But this slight overall change in tone masks movements in opposite directions for the president personally and the rest of his administration.

President Bush enjoyed a dramatic increase in favorable coverage after 9/11, rising from 36 percent positive evaluations before the tragedy to 63 percent positive afterward. Conversely, comments directed toward presidential appointees and executive agencies became more negative after the attacks, dropping to 31 percent positive after 9/11 from 48 percent positive previously. The latter finding is not the paradox it may seem. The period after the attacks included pointed questioning as to why the nation's intelligence services had not done more to protect the public, as well as the Justice Department's secretive and controversial detention procedures relating to possible terrorists. The administration's strategy of limiting the president's exposure on network news allowed Bush to concentrate on executing his commander-in-chief functions, and in doing so in circumstances that emphasize the president as leader, including televised speeches to the nation, formal addresses to Congress, and informal comments made during visits of other heads of state (Gregg and Rozell 2003).

Table 5 demonstrates that 2001 was also a year of two Congresses on network news, but in the opposite direction. Coverage of the legislative branch before the attacks was 37 percent positive in tone, falling to only 13 percent positive in their wake. The House and Senate had roughly similar ratings during both periods. But heavy criticism of Congress after 9/11 reversed a long-term trend toward more favorable coverage; the legislative branch received 28 percent positive coverage in 1993, compared to only 13 percent in 1981. This abrupt change in tone after 9/11 may reflect a redirection of partisan politics away from a newly popular president and his policies and toward the congressional realm, where criticism was less likely to be branded as unpatriotic. Such one-sided coverage makes it difficult, if not impossible, for the legislature to compete with a president for the public affections on anything approaching a level playing field.

Trivial Coverage?

Critics often charge that political news is too oriented toward scandal, political strategy and tactics, and various forms of trivia. Our data tell a different story, with the important caveat that the data only cover the first year of these three recent administrations. First years are a time when incoming presidents lay out their new policy agendas, and these three years also happened to exclude most of the political and personal scandals that drew heavy coverage later in these administrations. Table 6 shows that all three administrations were overwhelmingly evaluated on their performance in office during their first year. The most substantive orientation was toward the Bush administration; evaluations of the Bush team were based on policy and performance 83 percent of the time. For both the Clinton and Reagan administrations, job performance also represented more than two thirds of coverage.

Job performance likewise was an important part of the legislative and judicial branches' evaluations. Ethics was more a matter for congressional than executive branch evaluations, at least in the first year of each administration. Even in the Clinton administration, such assessments constituted a mere 5 percent of all executive branch evaluations during 1993. (The Clinton-Lewinsky matter became public in January 1998; Edwards 2000; Sabato et al. 2000).

Although the focus on job performance is to the networks' credit, a consideration of the tone of that coverage highlights the extent of network news negativity. For the job performance of the executive branch, the tone of coverage was about two to one negative for all three administrations. As Table 7 shows, 35 percent of evaluations were positive in 2001 and 33 percent were positive in both 1993 and 1981.

Many more plaudits were given to the Reagan administration's political effectiveness (63 percent positive) and to George W. Bush's team in 2001 (56 percent positive). Both presidents successfully pushed Congress to adopt major tax cut programs in their first year in office. The Clinton administration attracted too few evaluations of its political effectiveness to make a meaningful comparison, even though it also passed a key tax program in its first year (Woodward 1994).

In all three years, the performance of the executive branch received more positive assessments than that of the legislative branch. This finding is consistent with the argument of scholars that the complicated procedures of Congress make it easy to portray that institution in a negative light (Hibbing and Theiss-Morse 1995). The image of the legislative branch did improve over time; we recorded the most positive ratings for Capitol Hill in 2001, when the tone of legislative branch performance was 32 percent positive.

Network news evaluations of the judiciary's performance were consistently more positive than those of the other two branches--52 percent positive in 2001, 41 percent positive in 1993, and 56 percent positive in 1981. However, these percentages were based on very little coverage. For every evaluation of the federal courts' performance, there were nearly a hundred evaluations of the executive branch. In 2001, for example, there were 2,170 evaluations of executive branch performance, 323 evaluations of legislative performance, and a mere 27 evaluations of the judicial performance.

As we have argued throughout this article, any consideration of news coverage during 2001 must take account of the very different media environments that existed before and after 9/11. Table 8 divides the 2001 results for the Bush administration and the Congress into the periods before and after the terrorist attacks. (There are too few post-attack references to the judiciary to allow for meaningful comparisons.) As we have seen before, support rose for the executive and fell for the legislative branch after those attacks. What is striking, however, is the "dog that did not bark." Job performance was virtually the only topic addressed in network television evaluations after the terrorist attacks. For the executive branch, only 4 of the year's 102 assessments of political conduct aired after the attack, as did 4 of the 50 assessments of political effectiveness. Thus, the Bush administration was aided both by favorable coverage of the job it was doing, and by an absence of attention to any other topic. Both the tone of coverage and the media agenda favored the White House after 9/11, as the networks had virtually nothing to say about the more contentious "political" aspects of the president's job.

Conclusion

The breadth and depth of this study permitted us to test some widely held assumptions about network television news coverage of the federal government, and of the presidency in particular. We are not the first to find decreasing television news coverage of government. Our findings, however, challenge the notion that the decline affected all political actors equally. Over these first years of new presidents, the networks increasingly have highlighted the president's role as communicator-in-chief (cf. Kurtz 1994; Waterman et al. 1999).

There was a mild increase in the proportion of discussions about President Clinton above those of President Reagan, but no further increase in the attention paid to President George W. Bush. Taking into account the declining number of discussions, the attention given to the president actually fell by nearly one third (32 percent) from 1981 to 2001, with Bush receiving 577 fewer discussions than Reagan did. But the executive branch continued to dominate throughout the period, making it easy for presidents to "go public" on network newscasts that present government largely from the executive branch's point of view.

In the age of television, the president has the most decisive voice in American government, and that public role is inevitably heightened during times of crisis (Lowi 1985; Woodward 2002). But presidential media management stories can shift at least some of the coverage away from the Oval Office toward other parts of the executive branch. In the wake of the 9/11 terrorist attacks, for example, George W. Bush received proportionately less attention than he had earlier in the year as the administration sought to shift television news coverage toward Bush's executive branch subordinates. Given the sustained presidential approval during this period, media management strategies and leadership styles can be important weapons in the president's ongoing wars with the legislative branch and the press. Bush's media strategies are the latest examples of a trend across all three presidencies to direct television news coverage of executive agencies and other administration appointees outside the traditional power centers of the White House and the president's Cabinet. Given the negative orientation of much of the executive branch coverage seen here, this can be a very effective strategy for maintaining a president's public prestige by deflecting criticisms in the direction of a president's subordinates.

Declining coverage of the president did not mean increasing coverage for other elected officials. Most legislators, even the party leaders, are far less well known than presidents. Stories involving these less visible figures are more difficult and time-consuming to present than a report on the president as commander-in-chief or legislator-in-chief (cf. Iyengar 1991). Given the decentralized nature of Congress, this trend toward a growing legislative disadvantage does not seem likely to be reversed in the years ahead.

The "rally 'round the flag" effect is normally used to explain increases in public support for the president during crises (Adams et al. 1994; Nincic 1997). We found a similar tendency in network news. During the final 16 weeks of 2001 (after September 11), George W. Bush was portrayed far more positively than he was before the terrorist attacks, while coverage of the Cabinet and the White House staff actually became more negative. Favorable congressional coverage fell more sharply than that of the president's staff, further evidence of the great media advantages that Bush and the executive branch had over the legislative branch in late 2001.

In other words, on television the legislative branch is losing the battle for the "hearts and minds" of citizens to the executive branch and losing it badly. While this is not a new finding, this content analysis suggests that the coverage inequity between the two branches is becoming larger. Future legislatures may be even more deferential to presidents in such a media environment, unless news coverage reverses the anti-Congress trends seen in recent decades. Of course, the increasingly large number of gerrymandered "safe seats" for lawmakers may make nearly all House members immune to presidential influences regardless of the magnitude of coverage inequalities. On the other hand, such a development could make lawmakers more immune to citizen influences as well.

Our findings on the tone of television news coverage belie overarching generalizations about partisan bias. The two Republican presidents, Ronald Reagan and George W. Bush, received roughly as much negative coverage personally during their first year in office as their Democratic counterpart Bill Clinton. The same goes for the tone of coverage of their job performances. Overall, the Clinton administration received only slightly better coverage than did its GOP counterparts on some measures, though the difference between the evaluations of the public policies of the Clinton and Reagan administrations was statistically significant (p < .05).

These data identify a pervasive negativism in network television news coverage, results consistent with previous studies of political news. Coverage of all three presidencies was nearly two to one negative in tone, and evaluations of Congress were even worse, ranging as low as seven to one negative in 1981. It is no wonder that supporters of each new administration see the media as uniquely biased against them. But it would be more accurate to say that the networks were about equally "biased" against presidents and administrations of both parties.

Television news is often derided for focusing on trivial or ephemeral matters such as personal foibles and gaffes or political strategy and tactics rather than policy issues (cf. Farnsworth and Lichter 2003; Patterson 1994). But we found more coverage of the policy stands and job performance of government officials on network television than one might expect, particularly in light of the networks' performance in covering recent president elections. When the government is the subject, however, the vast majority of evaluations of both the executive and legislative branches dealt with "the issues"--matters of substance--during all three years covered by our study.

Appendix

Content analysis is a technique that allows researchers to classify statements objectively and systematically according to explicit rules and clear criteria. The goal is to produce valid measures of program content, and the hallmark of success lies in reliability. Other investigators who apply similar procedures to the same material should obtain similar results, although their interpretations of those results may differ. Clear rules and standards have to be set for identifying, measuring, and classifying each news story.

Evaluations were coded as positive or negative if they conveyed an unambiguous assessment or judgment about an individual, an institution, or an action. Only explicit evaluations were coded, in which both the target of the evaluation and its direction were clear. A positive example was provided by a voter quoted on ABC, "I think that Bill Clinton represents the best hope that this nation has for a brighter future" (January 17, 1993). A negative example came from NBC reporter Lisa Myers, "The president already had a reputation on the Hill for making contradictory promises and not keeping commitments. Now House members say even if he tells them the truth, they can't trust the president not to cave later under pressure" (June 10, 1993).

A description of events that reflected well or badly or some political actor was not coded for its tone unless it contained an evaluative comment. For example, an account of the passage of a bill supported by the White House would be coded as positive only if a source or reporter explicitly described it as a victory for the White House, a validation of the president's views or efforts on its behalf, etc.

A minimum reliability of 80 percent was achieved for all variables retained in the final analysis.

TABLE 1
Network News: Amount of Coverage

 2001 1993 1981 Total

Number of stories 5,014 3,848 7,216 16,078
Number of hours 138 107 178 422
Ave min per night 23 18 29 23
Number of minutes ABC 2,807 2,128 3,012 7,947
Number of minutes CBS 2,110 2,065 3,931 8,106
Number of minutes NBC 3,372 2,223 3,707 9,302

 Pre- Post-
 9/11 9/11

Number of stories 2,909 2,105
Number of hours 84 54
Ave min per night 20 29

Projected 2001 Totals without 9/11 Coverage Effects

Number of stories 4,197
Number of hours 121
Ave min per night 20

TABLE 2
Network News: Number of Discussions by Branch of Government

 2001 1993 1981 Total

Executive 3,924 4,871 6,512 15,307
President 1,212 1,395 1,789 4,396
White House & Cabinet 850 715 2,483 4,048
Other executive 1,862 2,761 2,240 6,863

Legislative 734 531 1,739 3,004
House 201 122 636 959
Senate 343 273 833 1,449
Other Congress 190 136 270 596

Judicial 146 156 293 595

Note: More than one individual or institution may be featured in one
story.

TABLE 3
Network News: Number of Discussions by Branch of
Government before and after 9/11/2001

 Pre-9/11 Post-9/11 2001 Total

Total Executive 2,597 1,327 3,924
President 914 298 1,212
White House & Cabinet 544 306 850
Other executive 1,139 723 1,862

Total Legislative 584 150 734
House 164 37 201
Senate 280 63 343
Other Congress 140 50 190

Judicial 118 28 146

Note: More than one individual or institution may be featured in one
story.

TABLE 4
Network News: Tone of Coverage (Percent Positive)

 2001 1993

Executive Total 36% n = 2,509 38% n = 4,658
President 39 n = 1,397 38 n = 2,708
White House & Cabinet 44 n = 455 53 n = 739
Other executive 26 n = 657 26 n = 1,211

Congress Total 35% n = 543 28% n = 562
By chamber: House 37 n = 244 42 n = 147
Senate 34 n = 179 22 n = 263
Other Congress 25 n = 120 23 n = 152

Judicial Total 48% n = 31 70% n = 66

Overall Total 36% n = 3,083 37% n = 5,286

 1981 Total

Executive Total 34% n = 1,606 37% n = 8,773
President 36 n = 705 38 n = 4,810
White House & Cabinet 31 n = 607 43 n = 1,801
Other executive 34 n = 294 27 n = 2,162

Congress Total 13% n = 196 29% n = 1,301
By chamber: House 12 n = 106 33 n = 497
Senate 18 n = 60 26 n = 502
Other Congress 3 n = 30 25 n = 302

Judicial Total 60% n = 82 61% n = 179

Overall Total 33% n = 1,884 36% n = 10,253

Note: Based on evaluations made by sources and reporters on the evening
news.

TABLE 5
Network News: Tone of Coverage before and after 9/11 (Percent Positive)

 Before 9/11 After 9/11 2001 Total

Executive Total 35% n = 2,093 42% n = 416 36% n = 2,509
President 36 n = 1,267 63 n = 130 39 n = 1,397
White House &
 Cabinet 48 n = 351 31 n = 104 44 n = 455
Other executive 23 n = 475 34 n = 182 26 n = 657

Congress Total 37% n = 498 13% n = 45 35% n = 543
House 41 n = 169 10 n = 10 37 n = 244
Senate 39 n = 236 0 n = 8 34 n = 179
Other Congress 27 n = 93 19 n = 27 25 n = 120

TABLE 6
Network News: Focus of Evaluations (Percent of All Evaluations)

 2001 1993 1981 Total

Executive Job performance 83 74 68 76
 Ethics/character 2 5 4 4
 Political conduct 5 13 11 10
 Political effectiveness 2 * 4 1
 Other 7 8 14 9

 99% 100% 101% 100%

Congress Job performance 59 62 52 59
 Ethics/character 20 20 7 18
 Political conduct 14 11 28 15
 Political effectiveness 1 1 6 2
 Other 6 6 8 6

 100% 100% 101% 100%

Judicial Job performance 87 41 34 46
 Ethics/character 0 2 1 1
 Political conduct 3 8 0 4
 Political effectiveness 3 0 0 1
 Other 6 50 65 48
 99% 101% 100% 100%

* Too few evaluations for meaningful analysis.

Note: Percentages do not total 100 due to rounding error.

TABLE 7
Network News: Tone by Focus of Evaluation (Percent Positive)

 2001 1993 1981 Total

Executive Job performance 35% 33% 33% 34%
 Ethics/character * 33 34 36
 Political conduct 23 40 23 35
 Political effectiveness 56 * 63 55
 Other 58 69 40 58
Congress Job performance 32% 26 14 27
 Ethics/character 49 21 * 33
 Political conduct 20 38 6 22
 Political effectiveness * * * *
 Other 60 47 * 54
Judicial Job performance 52 41 56 53
 Ethics/character * * * *
 Political conduct * * * *
 Political effectiveness * * * *
 Other * 94 63 75

* Too few evaluations for meaningful analysis.

TABLE 8
Network News: Tone by Focus of Evaluation (Percent Positive)

 2001 Pre-9/11 Port-9/11

Executive Job performance 35% 33% 42%
 Ethics/character * * *
 Political conduct 23 21 *
 Political effectiveness 56 59 *
 Other 58 61 43
Congress Job performance 32 35 14
 Ethics/character 49 49 *
 Political conduct 20 20 *
 Political effectiveness * * *
 Other 60 60 *

* Too few evaluations for meaningful analysis.

References

Abramson, Paul R., John H. Aldrich, and David W. Rohde. 1982. Change and continuity in the 1980 elections. Washington, DC: CQ Press.

--. 2002. Change and continuity in the 2000 elections. Washington, DC: CQ Press.

Adams, William C., Dennis J. Smith, Allison Salzman, Ralph Crossen, Scott Hieber, Tom Naccarato, William Vantine, and Nine Weisbroth. 1994. Before and after "the day after": The unexpected results of a televised drama. In Media power in politics. 3d ed. Edited by Doris Graber. Washington, DC: CQ Press.

Alger, Dean. 1996. The media and politics. 2d ed. Belmont, CA: Wadsworth.

Booth, William. 2002. Economic anxiety worries politicians. Washington Post, July 21, p. A1.

Campbell, Colin, and Bert A. Rockman, eds. 1996. The Clinton presidency: First appraisals. Chatham, NJ: Chatham House.

Cappella, Joseph N., and Kathleen Hall Jamieson. 1997. Spiral of cynicism: The press and the public good. New York: Oxford University Press.

Ceaser, James, and Andrew Busch. 2001. The perfect tie. Lanham, MD: Rowman & Littlefield.

Clancey, Maura, and Michael J. Robinson. 1985. General election coverage. In The mass media in campaign '84, edited by Michael J. Robinson and Austin Ranney. Washington, DC: American Enterprise Institute Press.

Cook, Timothy E. 1998. Governing with the news: The news media as a political institution. Chicago: University of Chicago Press.

Davis, Richard. 1999. The web of politics: The Internet's impact on the American political system. New York: Oxford University Press.

Davis, Richard, and Diana Owen. 1998. New media and American politics. New York: Oxford University Press.

Edelman, Murray. 1985. The symbolic uses of politics. Urbana: University of Illinois Press.

Edwards, George C. III. 2000. Campaigning is not governing: Bill Clinton's rhetorical presidency. In The Clinton legacy, edited by Colin Campbell and Bert A. Rockman. New York: Chatham House.

Farnsworth, Stephen J., and S. Robert Lichter. 2003. The nightly news nightmare: Network news coverage of U.S. presidential elections, 1988-2000. Lanham, MD: Rowman & Littlefield.

Goldberg, Bernard. 2002. Bias: A CBS insider exposes how the media distort the news. Washington, DC: Regnery.

Graber, Doris. 2002. Mass media and American politics. 6th ed. Washington, DC: CQ Press.

Gregg, Gary L., and Mark J. Rozell, eds. 2003. Considering the Bush presidency. New York: Oxford University Press.

Grossman, Michael B., and Martha Joynt Kumar. 1981. Portraying the president. Baltimore: Johns Hopkins University Press.

Han, Lori Cox. 2001. Governing from center stage: White House communication strategies during the television age of politics. Cresskill, NJ: Hampton Press.

Hershey, Marjorie Randon. 2001. The campaign and the media. In The election of 2000, edited by Gerald M. Pomper. New York: Chatham House.

Hess, Stephen. 1981. The Washington reporters. Washington, DC: Brookings Institution.

Hibbing, John R., and Elizabeth Theiss-Morse. 1995. Congress as public enemy: Public attitudes towards American political institutions. Cambridge, UK: Cambridge University Press.

Iyengar, Shanto, and Donald R. Kinder. 1987. News that matters. Chicago: University of Chicago Press.

Jamieson, Kathleen Hall. 2000. Everything you think you know about politics and why you're wrong. New York: Basic Books.

Just, Marion, Ann Crigler, Dean Alger, Timothy Cook, Montague Kern, and Darrell West. 1996. Crosstalk. Chicago: University of Chicago Press.

Kaid, Lynda Lee, and Joe Foote. 1985. How network television coverage of the president and Congress compare. Journalism Quarterly 62 (1): 59-65.

Kerbel, Matthew Robert. 1995. Remote and controlled: Media politics in a cynical age. Boulder: Westview.

--. 1998. Edited for television: CNN, ABC and American presidential elections. 2d ed. Boulder: Westview.

Kernell, Samuel. 1997. Going public: New strategies of presidential leadership. 3d ed. Washington, DC: CQ Press.

Kumar, Martha Joynt. 2002. Recruiting and organizing the White House staff. PS: Political Science & Politics 35 (1): 35-40.

--. 2003. The contemporary presidency: Communications operations in the White House of President George W. Bush: Making news on his terms. Presidential Studies Quarterly 33 (2): 366-93.

Kurtz, Howard. 1994. Media circus: The trouble with America's newspapers. New York: Times Books.

Lichter, S. Robert, and Daniel Amundson. 1994. Less news is worse news: Television news coverage of Congress. In Congress, the press and the public, edited by Thomas E. Mann and Norman Ornstein. Washington, DC: American Enterprise Institute and Brookings Institution.

Lowi, Theodore. 1985. The personal president: Power invested, promise unfulfilled. Ithaca, NY: Cornell University Press.

Nincic, Miroslav. 1997. Loss aversion and the domestic context of military intervention. Political Research Quarterly 50 (1): 97-120.

Patterson, Thomas E. 1994. Out of order. New York: Vintage.

--. 2000. Doing well and doing good. Research paper, Kennedy School Press Politics Center, Harvard University, December.

Postman, Neil. 1985. Amusing ourselves to death. New York: Penguin.

Putnam, Robert D. 2000. Bowling alone. New York: Simon & Schuster.

Sabato, Larry. 1993. Feeding frenzy: How attack journalism has transformed American politics. New York: Free Press.

--. 2002. The perfect storm: The election of the century. In Overtime: The election 2000 thriller, edited by Larry Sabato. New York: Longman.

Sabato, Larry, Mark Stencel, and S. Robert Lichter. 2000. Peepshow: Media and politics in an age of scandal. Lanham, MD: Rowman & Littlefield.

Seib, Philip. 2001. Going live: Getting the news right in a real-time, online world. Lanham, MD: Rowman & Littlefield.

Smoller, Fred. 1986. The six o'clock presidency: Patterns of network news coverage of the president. Presidential Studies Quarterly 16 (1): 31-49.

Sparrow, Bartholomew. 1999. Uncertain guardians: The news media as a political institution. Baltimore: Johns Hopkins University Press.

Waterman, Richard W., Robert Wright, and Gilbert St. Clair. 1999. The image-is-everything presidency. Boulder: Westview.

Woodward, Bob. 1994. The agenda: Inside the Clinton White House. New York: Simon & Schuster.

--. 2002. Bush at war. New York: Simon & Schuster.

STEPHEN J. FARNSWORTH

Mary Washington College

S. ROBERT LICHTER

Center for Media and Public Affairs

Stephen J. Farnsworth is associate professor of political science at Mary Washington College in Fredericksburg, VA. He is coauthor of The Nightly News Nightmare: Network Television's Coverage of U.S. Presidential Elections, 1988-2000 and author of Political Support in a Frustrated America.

S. Robert Lichter is president of communications at George Mason University where he directs the Center for Media and Public Affairs. He is coauthor of The Nightly News Nightmare: Network Television's Coverage of U.S. Presidential Elections, 1988-2000.

AUTHORS' NOTE: Thanks to the Council for Excellence in Government and the Pew Charitable Trusts for the support of this study and to the staff of the Center for Media and Public Affairs, including Dan Amundson, Mary Carroll Willi, Raymond Shank, Matthew Curry, Trevor Buttorworth, Amy Shank. and Irina Abarinova. Thanks also to George Edwards, Robert Spitzer, and the referees for their helpful suggestions and to Mary Washington College for its financial support. An earlier version of this article was delivered at the 2003 American Political Science Association Annual Meeting. All interpretations and errors remain the authors' responsibility.