sign in sign up
<<
Home , Hubert Humphrey, Tim Russert, Tom Brokaw

Faculty Research Working Papers Series

Diminishing Returns:

A Comparison of the 1968 and 2000 Election Night Broadcasts

Thomas E. Patterson

December 2003 RWP03-050

The views expressed in the KSG Faculty Research Working Paper Series are those of the author(s) and do not necessarily reflect those of the John F. Kennedy School of Government or Harvard University. Copyright belongs to the author(s). Papers may be downloaded for personal use only.

Diminishing Returns: A Comparison of the 1968 and 2000 Election Night Broadcasts*

by

Thomas E. Patterson Joan Shorenstein Center on the Press, Politics and Public Policy John F. Kennedy School of Government Harvard University Cambridge, Massachusetts 02138 www.shorensteincenter.org

Research sponsored by a grant from the John S. and James L. Knight Foundation.

December 2003

Shortly before 8 p.m., the television call led to a falloff in turnout on the networks projected as the West Coast that cost them at least two winner of the vote. Two hours House seats and perhaps a Senate seat later, they retracted the call. Then, just or two. after 2 a.m., the networks claimed George W. Bush had won in Florida and However, exit‐poll projections are a was thereby the president‐elect. Upon bigger threat to broadcasters’ hearing this news, Gore called Bush to reputations than to the integrity of concede defeat. Meanwhile, the elections. There is no firm evidence to networks were trying to get statements support the claim that network from the candidates. “We haven’t heard projections influence voter participation yet from either Al Gore or the in any systematic way.2 Although triumphant Governor Bush,” said CBS’s scholars are divided in their opinions, . “We do expect to hear from nearly all agree that the impact of exit them in the forthcoming minutes.” polls is small. Some scholars have Forty‐five minutes later, the networks concluded that exit polls might even reported that Gore’s concession had boost turnout slightly in a close election. been withdrawn. “Nobody knows for a The 1980 presidential election is the one fact who has won Florida,” Rather told instance when West Coast turnout his audience. At 4 a.m., the networks clearly sagged after the networks named retracted the claim that Bush had won a winner. However, the culprit in this the Florida vote.1 case appears to have been a presidential candidate. Inexplicably, The networks’ performance—“We went on the air to concede the election a donʹt just have egg on our face, we have full two hours before polls had closed an omelet all over our suits,” said NBC’s on the West Coast.3 —contributed to the post‐ election confusion. It also renewed the Should broadcasters be more longstanding complaint that the responsible in their use of exit polls? Of networks’ exit‐poll projections dampen course, they should. Often, the misuse turnout in states where the polls are still of exit polls has stemmed from the rush open. The networks’ first projections to declare a winner. No network wants aired in 1964, and Republicans to withhold a call that others have made complained loudly when Lyndon and every network likes to boast “you Johnson was declared the winner heard it here first.” After NBC called shortly after 9 p.m. EST. However, the Florida for Gore an hour after the first of GOP’s response was mild compared the state’s polls had closed, the other with how Democrats reacted in 1980 networks felt pressured to duplicate the when, shortly after 8 p.m. EST, nearly call. three hours before West Coast polls had closed, the networks declared Reagan This type of pressure, however, is the winner. Democrats claimed the early largely self‐generated. Viewers

Diminishing Returns 3

apparently could not care less whether and contrast the 7:00‐11:00 pm (EST) ABC, CBS, or NBC makes the first call. coverage of the 1968 and 2000 Election There is also no indication that viewers Night broadcasts. These broadcasts punish a network for withholding a call. were selected for analysis in part In the words of the blue‐ribbon team because journalists each time faced that evaluated CNN’s use of exit polls in nearly the same challenge. Each election 2000, “early calls serve no particular was decided by a razor‐thin margin. public or journalistic purpose.”4 If Each time, Americans turned off their broadcasters would exercise a bit more television sets and went to bed without restraint and would show a healthier knowing for sure who their next respect for the statistical error inherent president would be. in polling, much of the controversy surrounding exit‐poll projections might Yet, broadcast television was itself a disappear. quite different enterprise in 2000 than it had been in 1968. The 1968 election took Analysts might then find time to place in a period when exit polls and study other features of Election Night remote feeds were in their infancy and broadcasts. Unlike the debate and when the broadcast networks had a convention broadcasts, those on Election monopoly on the viewing audience. By Night have received little scrutiny apart 2000, broadcast equipment was highly from their use of exit polls.5 Scholars mobile, exit polling had been advanced, have paid so little attention to these and broadcasters were competing with broadcasts that entire books have been cable outlets for viewers’ attention. written on the networks’ presidential campaign coverage without so much as even a footnote about the Election Night The Content Analysis broadcasts. Yet, these broadcasts are an undeniably important part of our public The analysis in this paper is based on life. They mark the end of the campaign the broadcast “segment”—the and the start of the transition toward continuous portion of an Election Night new leaders and policies. Election Night broadcast in which the same general is one of those increasingly rare topic is discussed. In most cases, a moments when an uncommonly large broadcast segment began with a verbal number of citizens eagerly gather in signal from the network anchor (for front of their television sets to hear example, “We turn now to . . .”). A brief about politics. digression from the main topic of a segment was not considered to mark a What information do these new segment. By this definition, a total broadcasts provide to the public? What of 870 segments were identified in the and who do they emphasize, and what 1968 and 2000 broadcasts. interpretations of the election do they offer? In a preliminary effort to answer Each segment was coded to identify such questions, this paper will compare such things as its main and secondary

4 Thomas E. Patterson

topics, the people and graphic material They developed a primitive computer it featured, and the style of reporting it model that predicted the national two‐ employed. These measures revealed few party vote on the basis of selected local differences either in 1968 or in 2000 returns furnished by the wire services. between ABC, CBS, and NBC. The model accurately predicted the 1952 Accordingly, the three networks are and 1956 elections, but, given the size of lumped together in the analysis that Dwight Eisenhower’s victories, this was follows. The illustrative examples used hardly an extraordinary feat. In 1960, in the paper should be seen in the same the model led the networks to announce light. Although the examples identify early in the evening that particular networks and journalists, they appeared to be the likely winner, an were chosen in each case because they embarrassment that contributed to their represent general tendencies. decision in 1964 to rely on exit polls.6

The broadcast segments in 1968 and Returns vs. Results. Although the 2000 were nearly equal in length—68 1968 Election Night broadcasts were the seconds on average in 1968 and 72 second ones to use exit polls, they bore a seconds on average in 2000. The resemblance to previous broadcasts. For variation in segment length was also one thing, they opened with general nearly identical in 2000 to what it had commentary. CBS anchor Walter been in 1968. Thus, differences between Cronkite began his nework’s broadcast the 1968 and 2000 broadcasts would not with a quick review of a few early appear to be an artifact of segment returns and then turned to the length. commentator Eric Sevareid and the author Theodore H. White for their observations on the campaign. They Votes and Projections exchanged views on the past and the future of America’s political parties, In the era before exit polling, with Sevareid evoking the scholar broadcasters went on the air on Election Clinton Rossiter’s theory of third Night before they had results to report. parties, while White speculated on the They used their opening minutes to impact of ’s and Spiro prepare the audience for the evening Agnew’s candidacies on the border ahead. They reviewed the campaign and states. speculated on the outcome of key races. Only when vote returns started to come The 1968 broadcasts also made in from the states did the networks extensive use of actual vote returns. A begin to base their judgments on actual full third of the results reported in 1968 numbers. were based on actual returns, as opposed to exit‐poll projections. In fact, Even in this early period, the actual returns were the featured display networks were not fully content to let as, throughout the evening, the the actual returns speak for themselves. networks turned to big boards that

Diminishing Returns 5

showed the running national vote totals were dominated by coverage of election in the presidential race. results. Upwards of 85 percent of the segments each year dealt wholly or in In contrast, CBS began its 2000 significant part with vote results. Election Night broadcast by diving Expectedly, the overriding story on straight into the numbers. In his Election Night is a story of victory and opening words, Dan Rather said: defeat.

Bush gets , Gore gets However, the 1968 and 2000 Vermont—part of our CBS News broadcasts did differ substantially in the Election Night headlines of the hour. source of their voting numbers. Actual Bush picks up his first state in the vote returns, which accounted for a South; Gore gets his first win in New third of the 1968 figures, were nearly a England. But no call yet in what both footnote in 2000. More than 95 percent campaigns say may be the key to this election—Florida. It’s 7 PM in the East, of the numbers shown to viewers in and this [ELECTORAL VOTE 2000 were based on exit polls. During TOTALS ON SCREEN] is how the all‐ the first two hours of the 2000 important electoral vote count shapes broadcasts, almost no real vote counts up at this moment; 270 needed to be were presented. elected. Remember it’s still very early. And this is the national popular vote So complete was the networks’ count at this hour. Florida could turn reliance on exit polls in 2000 that out to be the decisive battleground state broadcasters failed at times to tell tonight. The polls just closed in six viewers the basis of what they were states with sixty‐six electoral votes, including Florida’s big 25. Let’s look seeing. Virtually every presentation of them over. South Carolina, the exit‐poll results in 1968 included a palmetto state, was expected to send its reminder that the numbers were eight electoral votes for George Bush projections rather than actual returns. In and it has done so. Bush wins South 2000, a fifth of the exit‐poll projections Carolina. Vermont, up in the green were presented without an advisory mountain state, three electoral votes statement. Moreover, many of the dropped there for Al Gore, his first advisories did not extend beyond electoral votes of the night. phrases such as “we project the winner

to be . . . .“ Viewers were seldom told in From there, Rather went on to more full and clear terms that statistical numbers, and throughout the opening models and exit‐poll samples were the stage of the broadcast, he and his CBS basis for the reports. colleagues rarely strayed very far, or for very long, from the reporting of election Presidency vs. Congress. In both results. 1968 and 2000, the presidential race

dominated the coverage. Nevertheless, Though the 2000 broadcasts got to the 1968 and 2000 broadcasts differed in the vote more quickly, the Election the amount of attention the presidential Night broadcasts in both 1968 and 2000

6 Thomas E. Patterson

contest received (see Table 1). It was The presidency, a truly national office featured in 81 percent of the 2000 that is embodied in a single individual, segments as opposed to 70 percent of was a natural fit for the networks the 1968 segments. Congressional races because of their national audience and received less attention in 2000 than they their tendency to tell the news through had in 1968, despite the fact that the actions of individuals rather than majority control of Congress was at institutions. By the 1970s, the presidency issue in 2000, whereas the Democrats was getting substantially more coverage were virtually assured of retaining on the evening newscasts than was control in 1968. Segments that focused Congress.7 Since then, except for brief on Senate races fell from 18 percent in periods, such as Newt Gingrich’s first 1968 to 15 percent in 2000 while those few months as House Speaker, directed at House races declined from 3 television news has concentrated most percent to 2 percent. Other races, mainly of its attention on the presidency. those for governor, slipped from 8 percent of the segments in 1968 to 2 The Modernist Influence percent in 2000. The sociologist Kiku Adatto notes that Table 1. Playing Up the Presidency television news has been shaped by a In comparison with 1968, the presidential modernist influence. “The very idea of race received more attention in 2000 news, especially the visual, fast‐paced, episodic style of television news, is Race for the: 1968 2000 inconceivable without the culture of modernism,” Adatto says. “For Presidency 69.5% 80.7% modernism, in contrast to the cultural Senate 17.5 14.5 movements that preceded it, prizes House 3.1 2.4 novelty and speed as values in Presidency & themselves.”8 Congress 2.0 0.7 Other Offices 7.9 1.7 Modernist values did not instantly change the form of television news. Total 100% 100% Roughly three decades went by after the first 30‐minute newscasts in 1963 before In this respect, the Election Night the fast‐paced formula of today was broadcasts are part of a more general firmly in place. The average sound bite trend in news. Until 1963, when the on the evening news was more than 40 television networks launched their 30‐ seconds in the and exceeded 20 minute national newscasts, daily seconds in 1980. Not until the 1990s did coverage of America’s elected it fall below 10 seconds, where it institutions was split evenly between remains. news about Congress and news about the presidency. The networks, however, Long vs. Short. The sound bites on had a preference for presidential news. Election Night broadcasts have also

Diminishing Returns 7

shrunk. In 1968, many of the segments 1960. At this early stage, we can say contained a lengthy statement by a that this is just one indicator that single speaker. When CBS’s Walter Illinois may yet go Republican. Cronkite turned to Dan Rather for a Now let’s look at Ohio, also with 26 rundown on the Midwest, Rather talked electoral votes and with 36 percent of without interruption for the next 173 the vote counted in Ohio. Very close seconds: there. Nixon leading in Ohio, razor‐thin margin over Humphrey. George Walter, in the Middle West, Hubert Wallace, as expected, running well in Humphrey, according to CBS News Ohio. It was assumed before the polls estimates, has won in Michigan and closed that any votes that Wallace got . All the rest of the Midwest in Ohio would be taking votes away belongs, as expected, to Richard Nixon. from Hubert Humphrey. Again though, No surprises anywhere so far tonight in Ohio, a substantial number of our except that Illinois, Ohio, and Missouri CBS News sample precincts are in, and are still out, and there’s every it’s simply too close for us to spot a indication that they are very close. trend, or indicate that anyone is even leading, and those are the two big First, let’s go to the board in Illinois. electoral giants in our area, Illinois and With 19 percent of the vote counted in Ohio, and nowhere is the see‐saw Illinois, Hubert Humphrey leads nature of this race any more apparent Richard Nixon with George Wallace than it is in those two big states. running third. Our CBS News estimates it’s simply too close at the Now, in Missouri, which has 12 moment to come out with anybody even electoral votes and is the only other leading in Illinois and that, in itself, is state still not called as far as our CBS somewhat of a surprise. And, I might News estimates are concerned. This is point out that Chicago, as usual, the actual vote total [VOTE TOTALS appears to be the key in this close race, PROJECTED ON SCREEN]. this presidential race. And the CBS Humphrey leading in Missouri with 45 News analysis of returns from sample percent of the actual vote in. George precincts shows that Humphrey, even Wallace running third and not running though he is winning 6 out of every 10 nearly as well in Missouri as many Chicago votes, is running at a slower Wallace supporters had expected. Now, pace in Chicago than John Kennedy did once again our CBS News sample in 1960. Now why is that important? precincts—and we have a good many of Simply because Kennedy won Chicago them in—but things are too close in by more than 450,000 votes in 1960 and Missouri to spot a trend or even call managed to carry Illinois by less than anyone leading. 9,000 votes. And that, in the suburbs, small towns, and rural areas, Nixon is And that’s the general picture in the winning by about the same ratio as Midwest, Walter, as far as the Humphrey is in Chicago. That could be presidential race is concerned. Illinois important as the night goes along. and Ohio, the two important ones, are Hubert Humphrey on the basis of our still out. And if we may add here in sample precincts is not running as well , Gaylord Nelson has won the in Chicago as John Kennedy did in

8 Thomas E. Patterson

Wisconsin senatorial race, in South Florida and Bush can run the table. Dakota Ferrar has won the Much more likely. I think it’s—right, gubernatorial race there. Hearns has now it’s—I would say that Florida is been elected in Missouri. more of a must win for Gore than for Bush. In 2000, such reports were rare. Although the 2000 segments, as Jennings: And so Mr. Bush, in general terms, has slightly less of a indicated previously, were about the struggle than Mr. Gore. same length as the 1968 segments, they were split into more pieces, as Stephanopoulos: Slightly less. But, exemplified by this ABC segment, boy, it’s awfully hard to say. which has 21 sound bites, most of which are less than 10 seconds in length: Jennings: I quite agree. Mark Halperin, our political director, your : Bring us up to date thoughts on the same question. on what you’re thinking at the moment. Mark Halperin: Florida is the key. George Stephanopoulos: Getting a And it’s been key in these two little easier to follow now. campaigns’ minds for weeks. So they put in a fair fight. The winner of Florida Jennings: That’s true. would not have surprised the other gentleman. Some of the other states Stephanopoulos: Both candidates weren’t quite as closely contested. are in exactly the same situation now. Florida they both know would be It’s simple. Whoever wins Florida, only important. has to win one other state. Whoever loses Florida has to run the table. That’s Jennings: And do you agree that Mr. the ballgame now. Gore could win Florida and Mr. Bush could run the other states? Jennings: OK, so if Mr. Gore wins in Florida, he only has to win one of the Halperin: Absolutely. If you look at other battleground states—Iowa or those states, it’s possible that Gore Wisconsin. could end up winning Florida, but then lose in the others. Stephanopoulos: Or or Oregon. Any one of the four gets him Jennings: Most of what we are over 270 [electoral votes]. getting now is anecdotal, incomplete, numerical information. At this point, Jennings: And looking at it before the any hints, any clues? polls actually closed tonight, I don’t think you would have wanted to take a Halperin: Well, we’ve got the raw guess at how those states would go, or votes and we’ve got the confidence of even at this late date. both campaigns that when the final vote comes in they’ll be the victor. Stephanopoulos: Can’t tell. I think it’s very plausible that Gore can take

Diminishing Returns 9

Jennings: What does the raw vote broadcasts served no purpose other show anyway that catches your eye at than to ease the transition to the next all? speaker.

Halperin: Well, Governor Bush in Anchors vs. Correspondents. The the raw vote has a bit of a lead and the conversational style of the 2000 Gore campaign says that that will be overtaken here when the rest of the vote broadcasts brought the network anchors comes in. regularly into view. Peter Jennings spoke 11 times during his discussion of Jennings: Because if you look at the Florida vote with Mark Halperin every state, as you have told me a and George Stephanopoulos. Anchors hundred times, you have to divide it up were rarely out of the picture for long into many, many pieces in order to during the 2000 broadcasts. They were understand where the real strengths the primary source of information in 47 are. percent of the segments, the facilitator

(as in the Jennings example) in 26 Halperin: There’s going to be a small gap. No matter who wins Florida, the percent of the segments, and had an vote will be agonizingly close for both of ancillary role in 16 percent. In only 1 these guys. percent of the segments was the anchor completely out of the picture or was on Jennings: Thank you, Mark Halperin camera only for as long as it took to and George Stephanopoulos. We’re introduce the next speaker or topic. going to go now to . . . In contrast, the anchor was out of Occasionally, presentations of this the picture or an incidental part of 30 type brought out insights that might not percent of the 1968 segments. In four of have come out in the more structured every five of these segments, presentations that typified the 1968 correspondents virtually had the air to segments. Typically, however, the 2000 themselves, as illustrated by Dan presentations merely skimmed the Rather’s report on the Midwestern surface, as the participants jumped from states. Correspondents were also one point to the next. involved in more segments in 1968 than they were in 2000. They had a leading Looking back at the 1968 coverage, role in half of the 1968 segments, one is struck by just how much compared with only two in five information was packed into a segment. segments in 2000. And rarely did a To be sure, if the words had been put correspondent speak for longer than 30 into a newspaper story, they would seconds at a time in 2000. Nearly all the have filled no more than a couple of lengthy sound bites in 2000 were paragraphs. Nevertheless, the segments delivered by the anchors. were efficient in the sense that few words were wasted. In contrast, many Indeed, the larger presence of the of the words spoken on the 2000 anchors in the 2000 broadcasts, as

10 Thomas E. Patterson

compared with 1968 segments, came and self‐referential statements were entirely at the expense of network virtually nonexistent on the 1968 correspondents (see Table 2). All other broadcasts. However, they were participants, including expert commonplace on the 2000 broadcasts. consultants and political figures, had as Shortly after Al Gore was declared the much or more airtime in 2000 as they winner of the Florida vote, for example, had in 1968. Correspondents were the NBC correspondent David Gregory only speakers who were a smaller reported: presence in 2000 than they had been in 1968. Well Tom [Brokaw], you and I have talked about just how chilly the Table 2. The Diminishing Role of the Thanksgiving meal might be between Network Correspondent Governor and Governor W. Bush. Well, ironically enough tonight, Anchor‐centered, fast‐paced broadcasting just a mile from where I stand at the has reduced the correspondent’s role even Four Seasons Hotel, the Bush family when he or she is part of a segment.. was at just such a private dinner when the results came in from Florida. Correspondents’ Certainly not good news. We know now Speaking Role: 1968 2000 that this family and all the Bush campaign is focusing very closely on Solo or nearly solo 59.6% 1.1% . In fact, the entire Primary but with entourage has moved from the Four Seasons Hotel back to the Governor’s anchor 35.1 36.4 mansion. That’s a change in plans. It’s Equal or a fairly festive atmosphere at the hotel, secondary to but everybody knows in the top echelons anchor 5.4 62.5 of this campaign that there’s a lot of watching to be done here in the course Total 100% 100% of very many hours. Pennsylvania, Tom, there’s probably not any state that “Me” vs. “Them.” Anchors and I’ve come to know more in the course of correspondents alike were more self‐ this campaign. I’ve probably memorized referential in 2000 than they were in the siding of all the airport hangers throughout the state. Governor Bush 1968. An aspect of modernism is the has worked it hard, as has Governor shrinking of distance, social as well as Ridge. They’re counting on that 9 temporal. Proximity and intimacy are tonight, and also, as [correspondent] valued, which on television news has is doing, doing all of the meant among other things, that math on some of the other combinations journalists have found ways to work in terms of how they get to victory. themselves into their stories. They are They’ve been encouraged tonight and no longer mere reporters: they are “part throughout the day, but gone is that of the message.”10 sort of boundless confidence that we’ve This conception was not a large part seen in the past few days. As you’ve been talking about, it’s been very tight. of television journalism in the 1960s,

Diminishing Returns 11

The tendency of today’s television criticisms of George Wallace’s 1968 journalists to conflate themselves and campaign, the broadcasts were nearly events is perhaps nothing more than a devoid of negative statements about the small conceit. Maybe it even draws candidates. Although the effect of this viewers more fully into the material. posture on the Election Night audience Whether the material provides insights has not been studied, it may help or information worthy of viewers’ citizens to put aside some of the attention is an entirely different issue. partisan divisions created by the campaign. Good vs. Bad. The modernist influence has not affected the Election Night broadcasts in all respects. Lionel Context and Analysis Trilling identified adversarial posturing as a hallmark of modernism.11 This Daily television news coverage changed posturing is clearly evident in daily markedly between 1968 and 2000. The election coverage. In the 1960s, 1960s were dominated by a descriptive presidential candidates received mostly style of journalism that for decades had favorable press coverage. By the late characterized newspaper reporting. The 1980s, partly as a consequence of journalist’s task was to describe events, Vietnam and Watergate, their coverage which often meant telling the audience was mostly unfavorable, and has what newsmakers had said and done. remained so.12 On evening newscasts during the 2000 general election The descriptive style, however, was campaign, George W. Bush’s coverage poorly suited to television. Viewers did was 63 percent negative while Al Gore’s not have to be told what they could see was 60 percent negative. A good deal of with their own eyes. Moreover, the Bush’s coverage suggested that he was descriptive style seemed dull when the not too smart. There were nine such words were spoken to a viewing claims on the evening news for every audience. The networks preferred a contrary claim. Gore’s coverage was livelier, more story‐like style of dotted with suggestions that he was not reporting. , an executive all that truthful. Such claims outpaced producer of NBC’s nightly news in the rebuttals by seventeen to one.13 1960s, told his correspondents: “Every news story should, without any sacrifice But if journalists are quick to fault of probity or responsibility, display the the candidates during the campaign, attributes of fiction, of drama. It should they bring a different perspective to the have structure and conflict, problem Election Night broadcasts. In both 1968 and denouement, rising action and and 2000, journalists embraced the falling action, a beginning, a middle, winners and the losers alike. The and an end.”14 elections were described as “hard fought” and candidates were said to An interpretative style of reporting have earned “respect.” Aside from that was explanatory as well as

12 Thomas E. Patterson

descriptive gradually emerged. to provide anything resembling Reporters found it difficult to develop a insightful commentary. Take, for dramatic story line without engaging in example, CBS correspondent Bob synthesis and interpretation. Schieffer’s response to a graphic on the Accordingly, television journalists Florida Senate vote: began to think like analysts as well as reporters, telling their audiences not just Well, look at this. About a third of the the “what” of events but the “why.” By voters down there said using the the 1980s, interpretive reporting had surplus to take care and keep Social displaced descriptive reporting as the Security solvent was an important thing to them. And 60 percent of those dominant form of television journalism. people voted for the Democrat, Nelson.

So that may tell you that Al Gore may Reports v. Analyses. The change is have a bit of an edge in Florida. We’ll evident in Election Night broadcasting, have to wait and see what happens. though less so than in daily television news. At peak moments on Election Superficial statements of this nature Night, broadcasters are hard pressed to dotted the 2000 broadcasts. In the whole do anything more than deliver of these telecasts, there were only a straightforward accounts of the results. handful of presentations where a trend Nevertheless, interpretation and or development was examined in analysis were a larger part of the 2000 substantial depth. broadcasts than of the 1968 broadcasts. Of course, elementary forms of analysis Strategy vs. Policy. In both 1968 and crept into otherwise straightforward 2000, candidates’ campaign strategies 1968 reports, as when CBS’s Dan Rather were the major focus of the analysis. compared Hubert Humphrey’s showing However, because analytical content in Illinois to John F. Kennedy’s made up a larger share of the 2000 performance there eight years earlier. broadcasts, campaign strategy was also Typically, however, analysis in 1968 was a larger theme of these broadcasts. Of set off from reporting and assigned to the broadcast segments in which designated commentators, such as CBS’s election results were discussed, nearly Eric Sevareid. Only one in eight of the 40 percent in 2000, compared with less 1968 segments were analytical in nature. than 15 percent in 1968, included The rest were fully or primarily statements about campaign strategy. descriptive. In contrast, one in three of the 2000 segments were analytical in In their Election Night analysis, nature, and many of the others journalists looked backward, toward the contained some analysis. strategies the candidates had employed The greater frequency of analysis in in the campaign, rather than forward, 2000, however, did not translate into toward the policies the candidates higher‐quality analysis. The fast pace of promised to pursue if elected. Only 1 the 2000 broadcasts meant that most of percent of the broadcast segments in the analytical statements were too short 1968 and in 2000 focused on the

Diminishing Returns 13

election’s policy consequences. Fewer difference of 10 points. Now let’s take a than 5 percent contained even a passing look at the women’s vote. Here we go. reference to the election’s policy Just flip the numbers and you will see implications. that Mr. Gore is getting 53 percent of the vote, Mr. Bush only 43 percent of

the vote. That adds up to a 20‐point

gender gap. It could be the largest since Exit‐Poll Journalism we started tracking all of this back in 1980. And Peter, the reason is that men Exit polls offer a means of discovering consistently want smaller government. what voters are thinking when they cast Fewer women are so sure about it.” their ballots. By itself, the ballot reveals only the decision that a voter made. Exit The quality of the exit poll analysis polls collect this information while also was somewhat higher in 1968 because gathering information about a the longer sound bites allowed for more respondent’s personal background and substantial commentary. But in truth, political opinions. even the 1968 exit‐poll analysis was superficial in most cases. Seldom did Broadcasters in 1968 and 2000 used correspondents say much beyond what this feature of exit polls to help explain the numbers themselves revealed. voters’ decisions. The most striking Perhaps network correspondents are feature of most of this analysis was its inadequately trained in survey analysis superficiality. Frequently, the or insufficiently practiced in the correspondent did little more than reporting of percentages and correlates. summarize a set of numbers. During Whatever, rather than unleashing their ABC’s 2000 telecast, for example, Peter reportorial skills, exit polls seem almost Jennings asked correspondent Lynn to suppress them. Sherr to analyze “what the demographics are showing us in Certainly, exit polls do lend a degree Florida.” As the breakdown of the male of precision to statements about voting and female vote in Florida appeared on patterns, as the following report from the screen, Sherr said: CBS’s 2000 broadcast illustrates:

Peter, we’ve been talking all year long Anthony Mason: Dan, our exit polls about the gender gap. It is here again show us today that 11 percent of the big time. The gender gap, as you know voters made up their minds in the last 3 of course, is the difference between days. And they swung to Gore, but by a the—between the way men vote and narrow margin—as you can see, 48 to women vote. Let’s take a look at the 44 percent. For many of those numbers, and see what’s happening undecided voters, it was not an easy here today. Let’s start with the male decision. When asked about the strength vote: 59 of that—52 percent, excuse of their support, nearly three‐quarters of me—of that vote went to George Bush those undecideds said they had today—is going for Mr. Bush, only 42 reservations about the candidate they percent for Mr. Gore. That’s a ultimately voted for. So they may have

14 Thomas E. Patterson

been reluctant, even tortured decisions, broad changes that have taken place in but they gave a narrow edge to Al Gore. television news since the 1960s are a hindrance. They have contributed to a Nevertheless, exit poll analysis is form of broadcasting driven heavily by about as lifeless as election reporting personality and pacing. The changes gets. Real people dissolve into faceless have been so gradual and are now so numbers and facile explanations. What’s deeply ingrained that it may be difficult the difference between men and women to reverse them. Yet, in an era where voters? The answer is 20 percentage there are so few opportunities to points. Why the difference? Because communicate at length with the men want smaller government and American public about politics, it would women might not. be a shame not to try to strengthen the Election Night broadcasts. Remarkably, exit polls are not even the most revealing or interesting Places and Faces. Election Night method of reporting election outcomes. broadcasting is less interesting than it Actual vote returns are superior in this could be. These broadcasts—with their respect. Voting patterns vary widely live reporting from studio settings— within a state, and the first returns from have an old‐fashioned look. Talking a state often diverge from later ones. heads and graphic displays dominated Countless candidates have seen a lead in 2000 as well as in 1968. Each time, shrink and then disappear entirely as they accounted for roughly 90 percent of the late returns come in. Except when what the viewers were presented. The the networks miss the call, exit polls graphics were slicker in 2000 than they hold out no such prospect. With one set were in 1968, but not much use was of numbers, they yield a final verdict on made in either year of television’s what voters in a particular state or capacity to tell a story with pictures. district decided. Given the other Only occasionally did the networks problems associated with exit polling, make use of pre‐shot footage or remote Congress would probably do the cameras, even though they would have networks a favor if it carried through on opened the broadcasts to a wider range its periodic threat to ban exit poll of faces and voices. projections until all polls had closed. Political leaders, for example, did not figure prominently on either the Thoughts and Recommendations 1968 and 2000 broadcasts, speaking in only about 5 percent of the segments. Election Night broadcasting is not They are featured on these telecasts only deeply flawed. It certainly fulfills the in the sense that, as the votes are being major interest of the viewing audience, counted, their fates are being decided. which is to discover election outcomes. Otherwise, they are bit players. By one Nevertheless, the potential of these indicator, they even shrank in telecasts has not been fully realized. The significance in 2000 as compared with

Diminishing Returns 15

1968. When a candidate’s photograph breakdowns of the vote by such appeared on screen in 1968, as it did in 6 variables as race, gender, and issue percent of the segments, it filled the opinions are now part of these screen two thirds of the time. For broadcasts. In truth, however, these example, as ABC’s Tom Jarriell did a snapshots of the electorate are far less rundown of the statewide races in vivid—and in this sense, less real—than and Arkansas, full‐screen the portrayals conveyed in the photographs of and traditional style of election reporting, as popped up. is evident from this 1968 example, in Jarriell could be seen in a small insert in which CBS correspondent Mike Wallace the lower right corner. In 2000, full‐ dissects the Pennsylvania vote: screen shots were reserved for reporters and anchors. The candidates were the The Democrats said they would need a small inserts. Although photographs of quarter of a million votes—250 candidates were used six times more thousand votes. They got a 263,000 vote often in 2000 (37 percent of the segments plurality in the city of . That plurality made possible by a big had a candidate photograph), these turnout for Humphrey in the Negro photographs filled the screen in only 1 and Jewish areas, with the Negro vote percent of the segments—four times less turnout higher than usual. The only often than in 1968. weak spot was in the Italian areas, which might be sensitive to racial issues Voters are even less prominent on where Republicans did somewhat better these broadcasts. Their votes may than normal, and it is conceivable that determine the outcomes, but citizens are that is what is making the difference in represented largely through numbers the race between the incumbent and statistical relationships. At times, Democratic Senator Joseph Clark and Congressman , the the public does not even seem to be on Republican. Let’s take a look at that the minds of the network journalists. Pennsylvania board. With one third of For example, in light of the persistent the vote now tabulated in Pennsylvania, complaint that exit polls depress Schweiker is leading with 792 thousand turnout, broadcasters might be expected to 745 thousand for Clark. to go out of their way to remind viewers to vote. Nevertheless, fewer than 5 Senator Clark has had his difficulty percent of the segments broadcast with the big Italian community in during the hours when West Coast polls Pennsylvania. He has made over the were still open contained even as much period of the past two or three years disparaging remarks about other as a brief statement to that effect. Democrats of Italian origin. And then,

in addition, he was for gun control Exit Polls. Of course, it could be legislation that was a little stiffer than claimed that exit polls bring the voters the gun control legislation that more fully into Election Night Schweiker was for in the state of broadcasts than was previously Pennsylvania. And so one million possible. As a result of exit polls, sportsmen in western Pennsylvania

16 Thomas E. Patterson

particularly organized against Joseph Clark and it begins to look as though did not have the opportunity to display Shweiker may be going down—I mean the knowledge they had acquired in the Clark may be going down before course of covering the campaign. Republican Richard Schweiker in spite

of the fact that Humphrey now has Unleashing their considerable talent taken the state of Pennsylvania and 29 big [electoral] votes and it looks like a would require the networks to lengthen big, big Democratic sweep throughout their sound bites and downplay their the eastern area. anchors, who now carry too much of the reporting burden on Election Night. It is ironic that, for all their emphasis on They are expected to be on top of lively and colorful forms of reporting, virtually every aspect of the election, a the broadcast networks have embraced division of labor that is unworkable if in exit polling a relatively drab way of the content of these broadcasts is a top talking about the electorate. priority. Correspondents. The networks should find ways to bring their Topics and Emphasis. In the U.S. seasoned correspondents more fully into governing system, the executive and their Election Night broadcasts. In legislative branches are theoretically co‐ today’s fast‐paced, anchor‐centered equal. A close evaluation of these broadcasts, correspondents are an branches’ constitutional powers underutilized resource. Their talent is indicates, however, that they are not, in largely squandered in a world where fact, equal. Congress actually has the they are confined to 10‐20 second sound larger powers. Yet, on Election Night, bites. As they spoke on the 2000 the presidency looms far larger than the broadcasts, network correspondents Congress. It is doubtful that these seemed almost to be responding to an quadrennial broadcasts contribute internal timer, knowing their remarks significantly to what the political would have to be short, and therefore scientist Hugh Heclo calls “the illusion nondescript. Perhaps because of this, of presidential government.” they did not display on the 2000 Election Night broadcasts the seasoned Nevertheless, there is no good judgments they exhibited in 1968. The reason why the congressional races talent pool may no longer support the should receive such short shrift on kind of journalism that enabled Walter Election Night. Although local Election Cronkite to sit in the center of the studio Night broadcasts provide some and turn alternately to Mike Wallace, coverage of congressional elections, they Dan Rather, John Hart, , do so from a local rather than a national and other CBS correspondents for perspective. The networks alone are detailed statements. More likely, positioned to inform the public about though, correspondents in 2000 simply the Congress as a whole and its relationship to the presidency.

Diminishing Returns 17

The storyline on election night is understandable is the failure to achieve necessarily one of victory and defeat. this standard on the Election Night But there is no reason why this storyline broadcasts, which are hours long and should be developed almost entirely devoted to a single subject. within the context of campaign strategy. There are lots of other possibilities. Recommendations Comparisons with previous elections are an obvious example. Lessons are Format nearly always learned from comparing the current election with previous ones. ‐‐Increase correspondents’ role and cut Yet, in both 1968 and 2000, fewer than 2 back the anchors’ role. percent of the segments contained a ‐‐Lengthen the sound bites. historical reference worthy of note. ‐‐In explaining the vote, cut back the use of exit polls and rely more heavily on But the most glaring interpretive traditional reporting. omission in the 1968 and 2000 ‐‐Bring the voices and faces of broadcasts was the infrequency with candidates and voters more fully into which journalists stepped back from the the broadcasts. election returns to ask: what does all of ‐‐Increase the display of actual vote this mean for national policy? Fifty returns and decrease the display of exit‐ years ago, the Hutchins Commission on poll results. a Free and Responsible Press concluded that the news media’s main shortcoming Substance was its failure to place political developments in a context that would ‐‐Cut back somewhat on the coverage of help citizens to understand their the presidential race and increase significance. The networks somewhat the coverage of congressional understandably have difficulty races. achieving this standard on their evening ‐‐Cut back sharply on explanations newscasts, which seek to cover the based on campaign strategy and nation and the world in 30 minutes. Less increase significantly the commentary on the political and policy implications of the election.

18 Thomas E. Patterson

Endnotes

* The author would like to thank Alison Kommer and Tami Buhr for their extraordinary contributions to this paper, which was written with support of a grant from the Knight Foundation. With painstaking precision, Alison Kommer conducted the content analysis that is the foundation of this paper. Tami Buhr assisted in that task, prepared the data for analysis, and did the computer analysis.

1 “CBS News Coverage of Election Night 2000: Investigation, Analysis, Recommendations,” CBS News (January, 2001): 16‐22. 2 See, for example, Douglas A. Fuchs, “Election‐Day Radio‐Television and Western Voting,” Public Opinion Quarterly, 30, 2 (1966): 226‐236; Harold Mendelsohn, “Western Voting and Broadcasts of Results on Presidential Election Day,” Public Opinion Quarterly, 30, 2 (1966): 212‐ 225; Seymour Sudman, “Do Exit Polls Influence Voting Behavior?” Public Opinion Quarterly, 30, 3 (1986): 331‐339; Michael W. Traugott, “The Impact of Media Polls on the Public,” in Thomas E. Mann and Gary R. Orren, eds., Media Polls in American Politics (, D.C.: The Brookings Institution, 1992): 125‐149; and Sam Tuchman and Thomas E. Cronin, “The Influence of Election Night Television Broadcasts in a Close Election,” Public Opinion Quarterly, 35, 3 (1971): 315‐346. 3 See, for example, Michael X. Delli Carpini, “Scooping the Voters? The Consequences of the Networks’ Early Call of the 1980 Presidential Race,” Journal of Politics 46 (1984): 866‐885; John E. Jackson, “Election Night Reporting and Voter Turnout,” American Journal of Political Science, 27, 4 (1983): 615‐635; and Raymond Wolfinger and Peter Linquiti, “Tuning In and Turning Out,” Public Opinion, 4 (1981): 56‐60. 4 Joan Konner, James Risser, and , “Television’s Performance on Election Night 2000: A Report for CNN,” unpublished report, January 29, 2001, p. 19. 5 Abraham McLaughlin, “After Election, “Are Promises Kept?” Christian Science Monitor, Sept. 14, 2000, web download. 6 Martin Plissner, “Television and the Making of the President: Lessons for the Next Round.” Public Lecture presented at Princeton University’s School of Public and International Affairs, March 6, 2001, p. 2. 7 See, for example, Richard Davis, “News Media Coverage of National Political Institutions,” Ph.D. dissertation, Syracuse University, 1986. 8 Kiku Adatto, “Sound Bite Democracy: Network Evening News Presidential Campaign Coverage, 1968 and 1988.” The Joan Shorenstein Center, John F. Kennedy School of Government, Harvard University, Research Paper R‐2, June 1990, p. 20. 9 Adatto, “Sound Bite Democracy,” p. 20. 10 Peter Boyle, Who Killed CBS (New York: Random House, 1988), p. 141. Quoted in Adatto, “Sound Bite Democracy,” p. 21. 11 Cited in Adatto, “Sound Bite Democracy,” p. 20. 12 Thomas E. Patterson, Out of Order (New York: Knopf, 1993), p. 20. 13 Robert Lichter, “A Plague on Both Parties: Substance and Fairness in TV Election News,” Harvard International Journal of Press/Politics 6, no. 6 (Summer 2001): 16. 14 Quoted in Michael Robinson and Margaret Sheehan, Over the Wire and on TV (New York: Sage Foundation, 1983), p. 226.

Diminishing Returns 19 Appendix: Election Night Broadcast Code Categories

Note: Each of the 870 segments (defined on page 4 7. Human interest (e.g., candidate waiting out of paper) in the 1968 and 2000 Election Night election returns) broadcasts was coded according to each of the 8 Other categories below. The resulting data are the basis for the observations in the paper. Empirical Basis for Judgment about Subject (Major Source) Network: 1. Exit polls/network estimates (vote 1. ABC preferences) 2. CBS 2. Exit polls/network estimates (opinion 3. NBC preferences/groups) Year: 3. Actual vote returns (including partial returns 1. 1968 but not estimates from sampling of partial 2. 2000 returns) 4. Pre-Election Day polls Start time: ______5. Historical data 6. Commentary End time: ______7. Other

Focus (Major) Empirical Basis for Judgment about Subject 1. Presidential race (Secondary source but more than incidental) 2. Senate race (s) 1. Exit polls/network estimates (vote 3. House race (s) preferences) 4. House & Senate 2. Exit polls/network estimates (opinion 5. Pres/Congress preferences/groups) 6. Other races 3. Actual vote returns (including partial returns but not estimates from sampling of Subject (Major) partial returns) 1. Winning/losing (who’s ahead/who’s behind) 4. Pre-Election Day polls 2. Why someone is winning/losing (emphasis 5. Historical data on candidate, e.g., candidate strategy) 6. Commentary 3. Why someone is winning/losing (emphasis 7. Other on voters/voting groups/opinions on issues) 4. Implications of winning-losing: political Extent of Analytic Commentary Does discussion of advantage (e.g. control of Senate) the numbers/data… 5. Implications of winning-losing: policy 1. Stay confined to the numbers (including consequences mathematical combinations using numbers 6. Election hoopla (e.g.., revelers at from more than one state) party/candidate headquarters) 2. Go slightly beyond the numbers 7. Human interest (e.g., candidate waiting out 3. Go significantly beyond the numbers election returns) (numbers serve only as a takeoff for 8 Other extended commentary—e.g., going into the . history of a state’s voting pattern or Subject (Secondary but more than incidental) explaining how the numbers suggest a major 1. Winning/losing (who’s ahead/who’s policy or demographic shift) behind) 2. Why someone is winning/losing (emphasis on candidate, e.g., candidate strategy) Nature of vote projections 3. Why someone is winning/losing (emphasis (Code only if segment includes exit polls in the on voters/voting groups/opinions on issues) context of winning/losing statements. Not in context 4. Implications of winning-losing: political of demographic comparisons. Any noticeable advantage (e.g. control of Senate) reference to the contingent nature of the exit 5. Implications of winning-losing: policy polls/sample precincts is sufficient for a code of “1”.) consequences 1. Clear reference that results are based on 6. Election hoopla (e.g.., revelers at estimates (e.g., “we estimate . . .”) party/candidate headquarters) 2. No mention of estimates (e.g., “___ wins)

Diminishing Returns 21 Anchor’s Role Graphic (Number of separate graphics) __ __ 1. Primary source of information 2. Facilitator (e.g.,, interacting with (FOR FIRST GRAPHIC ADDRESSED IN correspondent, conducting interview) SEGMENT) Was the analysis/commentary 3. Incidental (e.g., simply introducing a primarily in context of graphic(s) or did it go correspondent) substantially beyond the information in the 4. Not part of segment graphic? 1. Simply described graphic Leading Non-Anchor Participant(s) [One entry- 2. Supplied additional information to the dominant category code. If several participants are graphic in the same role, such as correspondents, treat them 3. Used graphic as a take off for a larger as one actor for purpose of determining “leading” point participant. If correspondent is interviewing someone in the 2-8 categories, code the actor who is not the Photo correspondent.] 1. Yes (prominent) 1. Network correspondent 2. Yes (but only as an icon) 2. Expert, part of broadcast team 3. No 3. Outside expert 4. Candidate Live Shot (outside anchor studio) 5. Candidate family member/personal friend 1. Yes 6. Campaign Staff/Party Leader 2. No 7. Group representative 8. Person-in-the-Street Pre-shot video footage 1. Yes Other Non-Anchor Participant [Same code as 2. No above; if correspondent is interviewing someone, this is where the correspondent should be coded] On-screen (dominance, in terms of what the viewer 1. Network correspondent is seeing on the screen during the segment) 2. Expert, part of broadcast team 1. Mostly talking head(s) 3. Outside expert 2. Mostly visuals 4. Candidate 3. About evenly talking heads and visuals 5. Candidate family member/personal friend 6. Campaign Staff/Party Leader Was there a statement that the polls were still 7. Group representative open and viewers should get out and vote? 8. Person-in-the-Street 1. Yes 2. No Setting 1. Studio 2. Studio (other than anchor studio) 3. Field 4. Studio and field (both should be a substantial part of segment. If anchor merely introduces a correspondent who is reporting from the field, it should be coded as field)

22 Thomas E. Patterson