Polling and the Media |
Can a Poll Affect Perception of Candidate Traits?
BRUCE W. HARDY is a doctoral student in the Annenberg School for Communication and a senior research analyst at the Annenberg Public Policy Center at the University of Pennsylvania. KATHLEEN HALL JAMIESON is the Elizabeth Ware Packard Professor of Communication in the Annenberg School for Communication and the Walter and Leonore Annenberg Director of the Annenberg Public Policy Center at the University of Pennsylvania.
Address correspondence to Bruce W. Hardy; e-mail: bhardy{at}asc.penn.edu.
 |
Abstract |
|---|
 TOP Abstract Theoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
During presidential elections, poll results frequently are presented in the news. Reporters use these polls to tell the public what it thinks about the presidential candidates. We argue that polling results tell the public what it should think about the presidential candidates as well. This study outlines how a character trait that is not usually used to assess presidential candidates was put into play during the 2004 presidential campaign. By repeatedly ascribing "stubbornness" to incumbent president George W. Bush, Democratic challenger John Kerry may have prompted this trait’s inclusion in a Los Angeles Times summer 2004 survey. The poll’s evidence that the public saw Bush as more stubborn than Kerry then produced an attribute agenda-setting effect that strengthened the link between that term and Bush. Using data from the National Annenberg Election Survey, we argue that the news coverage of this Los Angeles Times poll increased the salience of the trait "stubborn" in assessing President George W. Bush during June of the 2004 presidential campaign.
Both scholars (e.g., Atkins and Gaudino 1984
; de Vreese and Semetko 2004; Lang and Lang 1984; Lavrakas and Traugott 2000; Paletz et al. 1980) and journalists (e.g., Kovach 1980; Von Hoffman 1980) have posited effects from the media’s coverage of polls. As early as 1984, Lang and Lang suggested that poll results can reinforce majority opinion, a process akin to Noelle-Neumann’s "spiral of silence" (Noelle-Neumann 1993). In this tradition, we explore the possibility that by repeatedly ascribing "stubbornness" to incumbent president George W. Bush, Democratic challenger John Kerry may have prompted the trait’s inclusion in a Los Angeles Times summer 2004 survey. The poll’s evidence that the public saw Bush as more stubborn than Kerry, we argue, then had an agenda-setting effect in June 2004 that strengthened the link between that term and Bush. |
Theoretical Background: Agenda-Setting and Priming |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
Twenty-five years ago, in a special issue of Public Opinion Quarterly entitled "Polls and the News Media," Bill Kovach, then Washington editor of the New York Times, suggested that what may have been "noticed because a poll happened to have been conducted at that time, begins to take a life of its own and grow in size and importance" (Kovach 1980
, p. 571). The issue’s guest editor, Albert Gollin, made the tie to agenda-setting explicit by adding, "over time, people’s political beliefs and behavior have been affected by evidence of polls presented by the press—a special case of the larger claim of the mass media’s agenda-setting functions" (Gollin 1980, p. 450).
The relationship between the news media agenda as an independent variable and the public agenda as a dependent one is known as agenda-setting. During the 1968 presidential campaign, McCombs and Shaw (1972) found a "near perfect correspondence between the ranking of major issues on the press and public agendas" (McCombs 2005, p. 157). Further work by these researchers, employing cross-lagged correlations, suggested that media coverage sets the public agenda by making certain issues and not others salient (for a discussion of the evolution of this theory, see McCombs 2005; McCombs and Shaw 1993). Because agenda-setting works by increasing salience, it is related to another cognitive media effect: priming (Scheufele 2000).
The priming hypothesis states that (much like agenda-setting) mass media make some issues more salient than others, and that this heightened salience influences the public’s judgments of public policy, public officials, and candidates for public office (Iyengar and Kinder 1987). Work by Iyengar and his associates (e.g., Iyengar 1990; Iyengar and Kinder 1987; Iyengar, Peters, and Kinder 1982) found that salience of certain issues, primed by mass media, influenced people’s judgment of presidents when the issues the media featured were used as standards for evaluation.
The integration of priming and classical agenda-setting has led researchers to develop a two-level theory of agenda-setting. The first, the attention level, deals exclusively with objects such as a public figure, a public issue, an institution, or "anything else that is the focus of attention" (McCombs 2005, p. 160). The second, the attribute level, deals with that object’s characteristics and traits. For example, in the 2004 campaign, President Bush was a first-level agenda object, while his image and character traits made up the attribute agenda. As McCombs explains, "Attribute agenda-setting is the influence of the descriptions in the press on the public’s image of the candidate" (McCombs 2005, p. 161). In sum, the media both set the agenda by increasing the salience of certain issues and political actors and provide the basis for the assessment of them (Jacobs and Shapiro 1994; Pan and Kosicki 1997; Weaver et al. 1981).
|
Attribute Agenda-Setting by Polls |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
The notion that the perception of candidate attributes is influenced by media reports is a long-lived one. Weaver et al. (1981,
p. 162) observed that "the media provide an agenda of attributes from which voters’ images of the candidates are formed." As evidence, they reported that image attributes stressed in the Chicago Tribune ran in "parallel lines" with Illinois voters’ descriptions of leading candidates (Weaver et al. 1981, p. 201). Additionally, Becker and McCombs (1978) found a correspondence between attributes found in Newsweek and the agenda of attributes in upstate New York Democrats’ descriptions of their party’s candidates during the 1976 presidential primaries. These authors noted that "candidate attributes may be ordered from most to least prominent" (Becker and McCombs 1978, p. 302).
Surprisingly, the agenda-setting function of media reports of polled candidate attributes has not been documented in a campaign using survey research as evidence. Vatz and Weinberg’s (1987) discussion of the impact of a Newsweek cover story that detailed the results of a Newsweek-commissioned Gallup poll relied, for example, on qualitative data. The cover story reported that "more than 50 percent of the public believe [George H. W. Bush’s] ‘wimp’ image will be a serious problem" (Vatz and Weinberg, 1987, p. 25). Vatz and Weinberg (1987, p. 27) concluded, "Before the Newsweek poll there was no firm evidence that the label ‘wimp’ was a salient association with Bush for the general public. . . . Indisputably, however, Newsweek gave it tremendous impetus, with the poll’s findings seeming to provide a scientific confirmation of public perception with tremendous persuasiveness of a major newsweekly cover story."
Conscious of the limitations of their method, these authors (1987, p. 26) noted that the impact of the Newsweek cover story is "difficult to gauge with precision." In this essay we hope to increase the precision with which we describe the possible effect of reports of a poll by employing data from the National Annenberg Election Survey to create a case study of the attribute agenda-setting effect of a June 2004 Los Angeles Times poll that found that Bush was perceived as more stubborn than Kerry.
|
A Case Study |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
To demonstrate this possible function of polls we examine the impact of media coverage of that June 2004 Los Angeles Times poll, which asked a sample of national adults if "too ideological or stubborn" applied more to George W. Bush or John Kerry. We examine the character trait "stubborn" because it is not one that scholars commonly use to evaluate presidential candidates, and it is an attribute that is not typically in play in the rhetoric of campaigns. An influential study by Kinder et al. (1980)
outlined the character traits that are generally used in the assessment of presidential candidates. Participants responded to open-ended questions that asked them to name things that would describe an "ideal" and then an "anti-ideal" president. From the responses, Kinder and his colleagues developed two lists of traits, each containing 16 items. Next, respondents were asked to select the six most important attributes from these lists. From these results a "profile of an ideal president" (Kinder et al. 1980, p. 319) was developed. This profile contained eight positive and eight negative traits. The trait "stubborn" is not a component of this profile. The Los Angeles Times poll was released on June 10, 2004. This presented an opportunity for this research because the poll was released when there were no other major events happening in the campaign. The closest events that might have had an impact on the campaign were the death of former president Ronald Reagan (on June 5, 2004) and the release of Michael Moore’s film Fahrenheit 9/11 (June 25, 2004). A search of major newspapers’ headlines in the Lexis-Nexis database, using the search terms "Bush" and "Kerry," from June 9 through June 24, 2004, did not reveal any other major events that might have affected perceptions of President Bush’s "stubbornness." Therefore, we are able to examine the influence of the media coverage on the Los Angeles Times poll without the daunting task of separating the poll’s influence from the influence of some major event.
|
Setting the Media Agenda: The Kerry Campaign’s Influence |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
Setting the news media agenda is a preoccupation of political campaigns. Their efforts pay off. Analyzing the parallels between candidates’ agendas and the news media agenda during the 1996 presidential campaign, Jamieson (2000)
concluded that major-party presidential candidates influenced what the New York Times reported.
In the contest over control of the ascription of traits in the 2004 election, the Democrats translated the perceived Bush asset embodied in his slogan "steady leadership" into "stubborn leadership." On March 10, 2004, the New York Times captured the tactic in the headline "Kerry Accuses Bush of ‘Stubborn,’ Not Steady Leadership." "I think his stubborn leadership has led America steadily in the wrong direction," said Kerry (Wilgoren and Sanger 2004). A subsequent (March 15, 2004) New York Times article explained that campaign advisers said that the stubborn leadership idea "worked on several levels—not least because it evokes immaturity. ‘It’s a basic truth about [Bush’s] approach to the presidency, and it’s a basic truth about how his policies are affecting the country,’ one senior Kerry adviser said. ‘It’ll be a case that’ll be made throughout the entire campaign, from candidate to surrogates, to potentially advertising, to every other facet of the campaign’" (Halbfinger 2004).
The cultural resonance of the concept "strong leader" made undercutting its link to Bush a key objective of the Kerry campaign. An influential study by Bem (1974) isolated adjectives describing character traits that both men and women found desirable in each sex. Attributes such as "act as a leader," "willing to take a stand," "defends own belief," and "strong personality" all were considered positive for males. By contrast, as Asch’s classic study (1946) showed, "stubborn" is generally an undesirable characteristic. Asch described an individual with traits in this order: intelligent, industrious, impulsive, critical, stubborn, and envious. He then described an individual with the same traits but in reverse order. Participants were more likely to rate the individual whose traits were listed with "envious" and "stubborn" first as significantly less happy and less sociable than were other participants who heard exactly the same personality traits read in the opposite order. In short, turning "strong" into "stubborn" resonated with culturally defined negative male traits. The move also assumed that the meaning of "stubborn" was stable and not subject to reframing as a positive trait synonymous with strong.
Bush campaign consultants saw the perception that Bush was a strong leader as their ace in the hole in both the 2000 and 2004 elections (Jamieson and Waldman 2001; Jamieson, in press). In 2000 Bush consistently beat Democratic nominee Al Gore on this trait (Johnston, Hagen, and Jamieson 2004). Despite Kerry’s attempt to turn it into a negative attribute, Bush did the same against Kerry in 2004. Table 1 reports the results of Gallup polls, conducted toward the end of the 2004 campaign, that asked respondents if the term "strong leader" applied more to John Kerry or to George W. Bush. Bush consistently was seen as more of a "strong leader" than Kerry, sometimes by a margin of 20 percentage points or more. Therefore, a logical move by the Kerry campaign was to recast strong as stubborn.
|
Did the Kerry camp prompt the inclusion of the "too ideological and stubborn" item on the Los Angeles Times poll? Recent research has shown that campaigns use private internal polling to direct image priming strategies (Druckman, Jacobs, and Ostermeier 2004; Jacobs and Shapiro 1994). Presumably, Kerry campaign polls found that the trait "stubborn" could be plausibly affixed to George W. Bush and, therefore, primed this trait both in public and in behind-the-scenes contact with reporters for the Los Angeles Times. So, for example, Tad Devine, a Kerry strategist, was quoted in the Washington Post as saying Bush’s negative traits included that he was "stubborn and ideological" (Allen 2004a). Moreover, during the Annenberg Public Policy Center Election Debriefing (Allen 2004b) on December 4, Kerry campaign pollster Mark Mellman noted that the campaign identified weaknesses of Bush, including that he was "too stubborn."
Jacobs and Shapiro (1995–96) outlined ways in which campaigns use back-channel communication to encourage the placement of questions that may benefit them on public opinion polls. In their review of the relationships between the Nixon administration and Gallup and Harris polls, these authors found that the White House was often successful in its efforts. Reporters who are assigned to a campaign may provide the link to explain how a trait of interest to a campaign finds its way into a news organization’s poll. When asked about the origin of the "too ideological and stubborn" question, for example, Jill Darling Richardson, associate director of the Los Angeles Times Poll, reported that "we work with the reporters and editors who cover the campaigns to be sure we’re drafting accurate questions on the most timely issues. They read what we’ve written, suggest topics they are interested in that we might have missed, then we collaborate on what we can cut out of our wish list to keep the questionnaire within a reasonable length" (e-mail response to authors’ question, October 5, 2005).
|
Media Coverage of the Los Angeles Times Poll |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
The potential of the Los Angeles Times poll to influence the public was magnified by the fact that, in addition to being featured in the newspaper, it was carried in an Associated Press story and on CNN. The author of the June 10, 2004, Los Angeles Times story, staff writer and CNN political analyst Ronald Brownstein, cast the poll’s results in this way in the paper’s report: "Perhaps most troubling for the Democrat, nearly half said Kerry ‘flip-flops on the issues,’ while just a quarter applied that description to Bush. But for Bush, the flip side of the flip-flop charge is a deepening perception that he is too rigid: By a resounding 58% to 16%, poll respondents said the phrase ‘too ideological and stubborn’ applied more to Bush than to Kerry" (Brownstein 2004
).
Most of the media coverage on the June 10, 2004, Los Angeles Times poll led with the fact that the poll showed Kerry ahead of Bush. The Associate Press (AP) wire headline read, "L.A. Times Poll: Kerry Leads Bush by 7 Percentage Points." Since the report that Senator Kerry was ahead was followed by the "stubborn" finding, a reader might reasonably assume that one reason for the Kerry lead was the president’s vulnerability on this trait. One might also surmise that "flip-flop" was a smaller vulnerability for Kerry than "stubborn" was for Bush. The second section of the wire read: "More than a third of those surveyed said they don’t know enough about Kerry to decide whether he will make a better president than Bush. Asked who is more likely to flip-flop on issues, they chose Kerry by 2 to 1. But by 56 percent to 16 percent, voters felt that Bush was ‘too ideological and stubborn’" (Associated Press 2004).1 Although national television nightly news did not pick up the story, the 24-hour cable news channels and newspapers did.2
The day the poll was released, Ron Brownstein appeared on the CNN television show American Morning: "In our poll, voters were much more likely to say John Kerry is a flip-flopper but by more than 3–1 they were more likely to describe George Bush as ideological and stubborn. I think those are the comparisons that are actually relevant to Americans." That same day Mr. Brownstein guested on the CNN show Judy Woodruff’s Inside Politics, saying, "The flip side of the flip-flop charge, though, is that people are more likely to see Bush as arrogant or stubborn. That may be a central point of argument between them through the fall. What is the proper mode of leadership? Is it adjusting or is it setting a course and sticking to it?"
The poll also was highlighted by major regional papers such as the Boston Globe (Johnson 2004), the Chicago Tribune ("Key States" 2004), the Miami Herald ("Campaign Briefs" 2004), and the Washington Times (Pierce 2004), as well as smaller regional papers such as the Columbia (SC) State ("Poll: Kerry" 2004) and Manchester’s New Hampshire Union Leader (DiStaso 2004). Most of these stories ran the AP wire. We are unable to assess the total pickup of the story, but we do know that since it ran both on the Los Angeles Times and the AP wires, the story was available for use by papers, radio stations, and local broadcast news throughout the country. The newsworthiness of the story should have been heightened by the fact that it showed Kerry in the lead. Additionally, the lack of other campaign events during this time also should have increased coverage. A Web search of the Los Angeles Times poll, using the search phrase "bush + ‘ideological and stubborn’ + times + poll," suggests that the poll was picked up by both "The Hotline" (http://nationaljournal.com/pubs/hotline) and a large number of Web logs, or blogs. (Although beyond the scope of this article, it would be an interesting to see the influence, if any, that blogs played in the agenda-setting process that we are examining here.) In the following section we examine the influence of the media coverage of the Los Angeles Times poll on the public.
|
Data and Methodology |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
The data for this study come from the 2004 National Annenberg Election Survey (NAES), a rolling cross-sectional survey (Johnston and Brady 2001
; Romer et al. 2004) that interviewed by phone 81,422 adults from October 7, 2003 through November 16, 2004. The rolling cross-sectional design is a special type of a repeated cross-sectional survey design in that interviews are conducted on a set schedule across a specific time period. The NAES was conducted on a daily schedule with an average of 207.2 completed interviews per day. It is important to note that, with the rolling cross-sectional survey design, the day that the respondent happens to be interviewed is a product of random selection exactly like the random selection of the respondent to be included in the sample. Therefore, "overtime comparison is possible with few or no controls, and the sample can be partitioned pretty much at will" (Johnston, Hagen, and Jamieson 2004, p. 16). In this study we examine a segment of the NAES that spans from April 19, 2004, through September 17, 2004, and includes 35,742 respondents. During this time, respondents were asked, "I am going to read you some phrases. For each one, please tell me how well that phrase applies to George W. Bush. Please use a scale from 0 to 10, where ‘0’ means it does not apply at all and ‘10’ means it applies extremely well . . . ‘Stubborn.’" Respondents who did not provide an answer or "did not know" were coded as missing data (N = 22,392; M = 6.68; SD = 3.04).3
One major advantage of the rolling cross-sectional design is that it allows researchers to examine variations in public opinion over time. To assess respondents’ identification of George W. Bush as "stubborn," and how it changed over time, we first aggregate the data by calculating a mean of the individual responses of each day (Kenski 2004). This allows us to create a data point for each day and allows us to graphically visualize the data across time. The segment of the NAES that we are analyzing does include July 4, 2004, and no interviews were conducted on this national holiday. Following Kenski’s (2004) advice, the missing data were imputed by taking the average of the days surrounding the missing date. Because daily cross-sections are subject to random sampling variation, real swifts are "scarcely detectable through the uninteresting day-to-day fluctuations induced by sampling error" (Johnston et al. 1992, p. 26). Therefore, we pooled the data across days to "smooth" out the random variation. Throughout this study, we will employ a 5-day prior moving average, that is, any particular day’s value is an average of values of that day and the preceding 4 days.
|
The Public’s Assessment of George W. Bush as Stubborn |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
Figure 1 displays the 5-day prior moving average of the level of agreement with the statement that the term "stubborn" applies to George W. Bush. As figure 1 shows, the greatest variation occurs in the month of June. Recall that the release date of the Los Angeles Times poll that showed that 58 percent of respondents agreed that George W. Bush is "too ideological and stubborn" was June 10, 2004, directly prior to the increase in Bush’s stubbornness rating in mid-June.
|
Other polls that asked respondents if they thought Bush was stubborn were released by Pew on May 12, 2004, and September 16, 2004. However, the items on these polls did not receive as much media attention as the Los Angeles Times poll. Searches of the Lexis-Nexis database the week following the release of the May Pew poll using the search phrase "Pew and Poll and Stubborn," produced only one hit: a Washington Post article on May 16, 2004. The Associated Press did not pick up the results. As figure 2 shows, the release of the May Pew poll, which did not receive media attention, does not correspond to any shift in the NAES data. Unfortunately, because the question tapping respondents’ perception of Bush as stubborn was only included on the NAES from April 19 to September 17, 2004, we do not have the NAES data available to determine whether there was any influence from the September Pew poll. Searches of the Lexis-Nexis database the week following the release of the September Pew poll, using the same search phrase, did not produce any hits.
|
To further assess the influence of the Los Angeles Times poll we examine data from the month of June 2004. Figure 3 shows the average rating of Bush as "stubborn" for this month. After the death of President Reagan (June 5), there was a dip in the average rating of George W. Bush as stubborn. At the bottom of this dip the Los Angeles Times poll was released, and the subsequent media coverage of the poll took place. This pattern suggests to us that the media coverage of the poll may have primed the salience of the attribute stubborn.
|
By imposing the negative cultural meaning of stubborn on positive traits such as "strong leader," the Kerry camp was successful at creating not only a new character trait to assess George W. Bush but also a reassessment of Bush’s trump positive trait. Figure 4 details evaluations of Bush as a "strong leader" alongside the evaluations of Bush as "stubborn" for the month of June 2004 (see the appendix for exact question wording). We see almost a mirror image between the two traits (r = –.775, p 
.001).4 This suggests that the priming of stubborn produced a negative association with the trait "strong leader," causing individuals to reassess George W. Bush regarding this trait.
|
To test empirically the hypothesis that the fluctuations we see here are, indeed, a media effect we compared the mean ratings of Bush as stubborn before and after the release of the poll by levels of media use. Specifically, we compared the means of heavy newspaper readers versus light or nonreaders and then compared the means of heavy 24-hour cable news viewers versus light or nonviewers.5 We partitioned the data into two independent samples. The first, representing the "before the poll" time period, consisted of 3,052 respondents who were surveyed from May 27 to June 9, 2004. The second sample, representing the "after the poll" time period, consisted of 2,105 respondents who were surveyed from June 13 to June 27, 2004. Notice that the second sample starts 3 days after the release of Los Angeles Times poll, therefore, allowing for 3 days for media coverage. This serves two purposes. First, it compensates for the lag created by the 5-day moving average. Second, it adjusts for the fact that the influence of the media on issue or trait salience is not instantaneous, especially for issues that are not of high consequence (McCombs 2005). At the same time, it draws on Johnston, Hagen, and Jamieson’s (2004) suggestion that campaign news stories have a maximum life of roughly 3 days.
Table 2 details the results of the independent sample t-tests. First, the differences between the means of heavy and light media users within the "before" sample are not statistically significant. There is no difference in ratings of Bush as stubborn among media users before the release of the Los Angeles Times poll. However, in the sample after the release of the poll, we see that the differences in heavy and light media users are significant. Furthermore, the difference in means between heavy newspaper readers in the two independent samples is significant, with the sample after the release of the poll showing a higher mean. The difference in mean ratings of Bush as stubborn between the light newspaper readers across the two samples is not significant. These differences hold for the 24-hour cable news viewers as well, with heavy viewers showing significant differences in the mean rating of Bush as stubborn, while there is no difference between light 24-hour cable news viewers across the two samples. These results support our hypothesis that the increase in the mean rating of Bush as stubborn is a media effect. First, the heavy and light media users in the sample before the release of the poll do not differ with regard to their rating of Bush as stubborn. Second, heavy and light media users differ in the sample after the poll. Third, the only variation across the two samples is found among heavy media users. Together, these results suggest that media coverage of the Los Angeles Times poll was the causal agent of the increase in the rating of Bush as stubborn.
|
One potential limitation to this approach is that we do not know if the individuals polled in NAES actually viewed news coverage of the Los Angeles Times poll. There is the possibility that the effect is actually attributable to some other content in the news media. Recall that we located no major events during this time that could have had an influence on that attribute. However, it must be noted that during the coverage of President Reagan’s passing (which occurred 5 days before release of the poll), the news media often recited Reagan’s misquotation of John Adams during his farewell address at the 1998 Republican convention in New Orleans. Instead of saying "facts are stubborn things," Reagan said, "facts are stupid things." This coverage took place directly after his death and right before the release of the Los Angeles Times poll. Therefore, there is a slim possibility that an effect of the coverage shows up in our analyses. Albeit possible, it seems to us unlikely that such coverage would shape evaluation of George W. Bush’s stubbornness. Indeed, in the days following Reagan’s death the perception that Bush is stubborn dropped rather than increased (see figure 3). Additionally, the priming of stubborn is not likely from the references to the Reagan quotation because he was attributing stubbornness to facts not a person.
To exclude other possible media influences we examine the possibility that the June 25, 2004, opening of Michael Moore’s film Fahrenheit 9/11 might have had an effect. Although the release date is relatively far from the Los Angeles Times poll and the effect of the poll appears to have substantially dissipated by then, this period is included in our "after the poll" sample, which ends on the June 28, 2004. Of the 2,105 respondents in this sample only 11 individuals reported having seen the movie. Such a small number of respondents should not influence the results of our independent t-tests.
Another possibility would suggest that the differences found in the independent t-test were a campaign effect and not a media effect. To test that possibility, we looked at the ratings of Bush as stubborn by "battleground" and "nonbattleground" states. At the time, the battleground states were Colorado, Florida, Iowa, Michigan, Minnesota, Nevada, New Hampshire, New Mexico, Ohio, Oregon, Pennsylvania, Wisconsin, and West Virginia. Our analysis suggests that there was very little difference between the battleground and nonbattleground states in the weeks following the release of the Los Angeles Times poll. If these differences were a campaign effect, we would see differences in the ratings of Bush as stubborn between battleground and nonbattleground states.
To further rule out alternative explanations, we also looked to see if the Democrats released advertisements that primed stubborn. To investigate this possibility, we examined the storyboards of the ads released by the Democrats during the week before the release of the Los Angeles Times poll and the two weeks following the poll. We examined the storyboards of Republican spots as well to rule out the possibility that a Republican message highlighted stubbornness. We drew our storyboards from the TNS Media Intelligence/Campaign Media Analysis Group (CMAG), whose data set consists of advertisements’ airdate, media market, spot count, and estimated cost.
Specifically, we examined the seven major English-language television advertisements that aired during this time. Three of them were produced by the Kerry campaign, three by the Bush camp, and one by the Democratic issue advocacy organization the Media Fund. The closest connection to any content that might have primed "stubborn" was found in one of the Bush campaign ads. This advertisement, referring to the Patriot Act, attacked John Kerry’s voting record by explicitly stating, "John Kerry changed his position." Although this might translate into Bush’s not changing his position and therefore being stubborn, it is unlikely that such an implicit suggestion would produce the effects demonstrated here. The Kerry campaign ads focused on Kerry’s service as a veteran, prosecutor, and senator, his plan for health care, and his "strong at home, respected in the world" message. None the Kerry campaign advertisements mentioned George W. Bush. The Media Fund advertisement focused on the no-bid contract for Halliburton in Iraq. In sum, there was not anything obvious in these ads that would account for the jump in stubborn during this time.
|
Conclusion |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
Although analysts have long posited the function, this study used rolling cross-sectional data to document the attribute agenda-setting effect of a poll in the 2004 campaign. It details a process by which campaigns, polls, and media reports interact with culturally shared social meanings of terms that are used to describe character. This study showed how a character trait that is not usually studied by scholars, or employed by campaigns, was put into play in June in the 2004 presidential campaign. On the presumption that it was culturally a negative trait, the Kerry campaign characterized Bush not as steady but as stubborn. The coverage of the Los Angeles Times poll did not produce overwhelming shifts in the public’s assessment of President Bush’s character. It did, however, produce a small, detectable change in the public’s assessment of both Bush’s stubbornness and his steady leadership.
Any change in perception invites the question, did the change affect vote choice? Although this study focuses simply on the possible attribute agenda-setting function of polls, we conducted some exploratory analyses to determine whether the effect we found affected vote choice. Clearly, uncovering any direct impact of the coverage of the poll’s question on ballots cast five months later is next to impossible. Additionally, NAES data that would make possible such an assessment do not exist. That said, when examining the impact of Bush’s stubborn rating on the proportion of respondents who would vote for Bush if the "election were held today," we found no significant effect.6
The fact that we did not find any direct impact on vote choice in June does not suggest that poll-produced priming effects are necessarily inconsequential. This type of priming of a candidate’s attribute may have a more pronounced impact on vote decision as Election Day nears and as the "undecideds" start deciding. Isolating the priming effect of a poll toward the end of the campaign would, of course, be more difficult because of the rush of messages, heightened news coverage, and presence of debates, among other factors.
A main intention of this essay is to sketch the relationship between poll results and candidate evaluation, specifically attribute assessment. We believe that this relationship has a unique and important role in political campaigns. We suggest that poll results may prime character traits through attribute agenda-setting. Future research might examine if poll results also provide information on the opinion environment that could possibly influence individuals to conform to the majority opinion. An experiment testing this hypothesis might be conducted. Future research might also examine the cultural meaning and stability of the terms each candidate uses to describe him or herself and opponents. Additionally, researchers might focus on how campaigns influence the meaning of these terms once they are in play. In sum, this study opens a number of questions for future research.
|
Appendix |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
NAES EXACT QUESTION WORDING
Traits:
- I am going to read you some phrases. For each one, please tell me how well that phrase applies to George W. Bush. Please use a scale from 0 to 10, where "0" means it does not apply at all and "10" means it applies extremely well . . . "Stubborn."
- I am going to read you some phrases. For each one, please tell me how well that phrase applies to George W. Bush. Please use a scale from 0 to 10, where "0" means it does not apply at all and "10" means it applies extremely well . . . "Strong leader."
News Media Use:
- Now I would like to ask about where you got your news during the past week. Please tell me how many days in the past week you did each of the following. How many days in the past week did you watch a 24-hour cable news channel, such as CNN, Fox News Channel, or MSNBC?
- Now I would like to ask about where you got your news during the past week. Please tell me how many days in the past week you did each of the following. How many days in the past week did you read a daily newspaper?
|
Footnotes |
|---|
1. The Associated Press misquoted the LA Times poll results. The poll showed 58 percent to 16 percent.
2. Because searches of news transcripts of the Lexis-Nexis database for the month of June 2004, using the search phrase "LA Times and Stubborn and Bush," did not produce any transcripts from national broadcast television network news, we focus only on newspaper and 24-hour cable news channel use.
3. Due to a split questionnaire design used to maximize the number of questions asked on the survey, this particular question was asked to two-thirds of all respondents.
4. This coefficient was calculated by correlating the 5-day moving averages of these two traits for the 30 days of the month of June.
5. Media use was measured by days of use. These items were split based on their distributions (i.e., median-split) to produce heavy and light users. Newspaper use was split at 4 days or more being heavy users, and heavy 24-hour cable news use was 3 days or more.
6. Two logistic regression models were compared to see if there were any shifts before and after the Los Angeles Times poll in coefficients of the relationship between Bush’s stubborn rating and proportion of respondents who would vote for Bush "if the election were held today." In these models—controlling for party identification, news media use, age, education, and income—the coefficients were relatively stable across the before and after the poll samples, consistently producing a negative relationship.
|
References |
|---|
|
TOPAbstractTheoretical Background: Agenda...Attribute Agenda-Setting by...A Case StudySetting the Media Agenda:...Media Coverage of the...Data and MethodologyThe Public’s Assessment of...ConclusionAppendixReferences |
|---|
Allen, Mike. 2004a. "Bush, Cheney Attack Kerry as Indecisive; President Focuses on Intelligence Issues. Washington Post, March 9.
Allen, Mike. 2004b. "Cheney May Be a Mixed Blessing for Bush Team; He Rallies Base But Can Alienate Others." Washington Post, July 5.
Annenberg Public Policy Center Election Debriefing. 2004. "Polling: Decisive Moments and Audiences." December 4.
Asch, Solomon E. 1946. "Forming Impressions of Personality." Journal of Abnormal and Social Psychology 41:1230–40.
Associated Press. 2004. "LA Times Poll: Kerry Leads Bush by 7 Percentage Points." June 10.
Atkins, Charles K., and James Gaudino. 1984. "The Impact of Polling on the Mass Media." Annals of the American Academy of Political and Social Science 472:119–28.[Abstract]
Becker, Lee, and Maxwell McCombs. 1978. "The Role of the Press in Determining Voter Reaction to Presidential Primaries." Human Communication Research 4:301–7.[CrossRef]
Bem, Sandra Lipsitz. 1974. "The Measurement of Psychological Androgyny." Journal of Consulting and Clinical Psychology 42:155–62.[CrossRef][Web of Science][Medline]
Brownstein, Ronald. 2004. "The Nation: Voters Shift in Favor of Kerry; Unhappiness with Bush and the Nation’s Path Exceeds Doubt about the Senator, a Survey Shows; But in Swing States the Balance Is More Tenuous." Los Angeles Times, June 10.
"Campaign Briefs." 2004. Miami Herald, June 11.
de Vreese, Claes H., and Holli A. Semetko. 2004. "Public Perception of Polls and Support for Restrictions on the Publication of Polls: Denmark’s 2000 Euro Referendum." International Journal of Public Opinion Research 14:367–90.
DiStaso, John. 2004. "Poll: Kerry and Bush Are Deadlocked in NH." New Hampshire Union Leader, June 11.
Druckman, James N., Lawrence R. Jacobs, and Eric Ostermeier. 2004. "Candidate Strategies to Prime Issues and Images." Journal of Politics 66:1180–1202.[CrossRef][Web of Science]
Gollin, Albert E. 1980. "Exploring the Liaison between Polling and the Press." Public Opinion Quarterly 44:445–61.
Halbfinger, David M. 2004. "Kerry Setting Out to Make a Good Second Impression." New York Times, March 15.
Iyengar, Shanto. 1990. "The Accessibility Bias in Politics: Television News and Public Opinion." International Journal of Public Opinion Research 2:1–15.
Iyengar, Shanto, and Donald R. Kinder. 1987. News That Matters. Chicago: University of Chicago Press.
Iyengar, Shanto, Mark D. Peters, and Donald R. Kinder. 1982. "Experimental Demonstrations of the "Not So Minimal" Consequences of Television News Programs." American Political Science Review 76:848–58.[CrossRef][Web of Science]
Jacobs, Lawrence R., and Robert Y. Shapiro. 1994. "Issues, Candidate Image, and Priming: The Use of Private Polls in Kennedy’s 1960 Presidential Campaign." American Political Science Review 88:527–40.[CrossRef][Web of Science]
Jacobs, Lawrence R., and Robert Y. Shapiro. 1995–96. "Presidential Manipulation of Polls and Public Opinion: The Nixon Administration and the Pollsters." Political Science Quarterly 110:519–38.[CrossRef]
Jamieson, Kathleen Hall. 2000. Everything You Think You Know about Politics . . . And Why You’re Wrong. New York: Basic Books.
Jamieson, Kathleen Hall, ed. In press. Election Debriefing 2004. Philadelphia: University of Pennsylvania Press.
Jamieson, Kathleen Hall, and Paul Waldman, eds. 2001. Electing the President, 2000: The Insider’s View. Philadelphia: University of Pennsylvania Press.
Johnson, Glen. 2004. "Bush’s Moves on War, Economy Fall Short, Kerry Campaign Says." Boston Globe, June 15.
Johnston, Richard, André Blais, Henry E. Brady, and Jean Crête. 1992. Letting the People Decide: Dynamics of a Canadian Election. Stanford, CA: Stanford University Press.
Johnston, Richard, and Henry E. Brady. 2001 "The Rolling-Cross Sectional Design." Electoral Studies 21:283–95.[CrossRef]
Johnston, Richard, Michael G. Hagen, and Kathleen Hall Jamieson. 2004. The 2000 Presidential Election and the Foundation of Party Politics. Cambridge, UK: Cambridge University Press.
Kenski, Kate. 2004. "Visualizing Data across the Campaign." In Capturing Campaign Dynamics: The National Annenberg Election Survey, by Daniel Romer, Kate Kenski, Paul Waldman, Christopher Adasiewicz, and Kathleen Hall Jamieson, pp. 66–79. New York: Oxford University Press.
"Key States Temper Kerry Lead in Poll." 2004. Chicago Tribune, June 10.
Kinder, Donald R., Mark D. Peters, Robert P. Abelson, and Susan T. Fiske. 1980. "Presidential Prototypes." Political Behavior 2:315–37.[CrossRef]
Kovach, Bill. 1980. "A User’s View of the Polls." Public Opinion Quarterly 44:567–71.
Lang, Kurt, and Gladys Engel Lang. 1984. "The Impact of Polls on Public Opinion." Annals of the American Academy of Political and Social Science 472:129–42.[Abstract]
Lavrakas, Paul J., and Michael W. Traugott, eds. 2000. Election Polls, the News Media, and Democracy. New York: Chatham House.
McCombs, Maxwell E. 2005. "The Agenda-Setting Function of the Press." In The Press, ed. Geneva Overholser and Kathleen Hall Jamieson, pp. 156–68. New York: Oxford University Press.
McCombs, Maxwell E., and Donald L. Shaw. 1972. "The Agenda-Setting Function of Mass Media." Public Opinion Quarterly 36:176–87.
McCombs, Maxwell E., and Donald L. Shaw. 1993. "The Evolution of Agenda-Setting Research: Twenty-Five Years in the Marketplace of Ideas." Journal of Communication 43:58–67.[CrossRef][Web of Science]
Noelle-Neumann, Elisabeth. 1993. The Spiral of Silence: Public Opinion—Our Social Skin. 2d ed. Chicago: University of Chicago Press.
Paletz, David L., Jonathan Y. Short, Helen Baker, Barbara Cookman Campbell, Richard J. Cooper, and Rochelle M. Oeslander. 1980. "Polls in the Media: Content, Credibility, and Consequences." Public Opinion Quarterly 44:495–513.
Pan, Zhongdang, and Gerald M. Kosicki. 1997. "Priming and Media Impact on the Evaluations of the President’s Performance." Communication Research 24:3–30.[Abstract]
Pierce, Greg. 2004. "Nation: Inside Politics." Washington Times, June 11.
"Poll: Kerry Leads Bush by 7 Percentage Points." 2004. Columbia (SC) State, June 11.
Romer, Daniel, Kate Kenski, Paul Waldman, Christopher Adasiewicz, and Kathleen Hall Jamieson. 2004. Capturing Campaign Dynamics: The National Annenberg Election Survey. New York: Oxford University Press.
Scheufele, Dietram A. 2000. "Agenda-Setting, Priming, and Framing Revisited: Another Look at Cognitive Effects of Political Communication." Mass Communication and Society 3:297–316.[CrossRef]
Vatz, Richard E., and Lee S. Weinberg. 1987. "Media and Polling: Measuring and Creating the Salience of George Bush as ‘Wimp.’" Mass Communication Research 14:23–29.
Von Hoffman, Nicholas. 1980. "Public Opinion Polls: Newspapers Making Their Own News?" Public Opinion Quarterly 44:572–73.
Weaver, David H., Doris A. Graber, Maxwell E. McCombs, and Chaim H. Eyal. 1981. Media Agenda-Setting in a Presidential Election: Issues, Images, and Interest. New York: Praeger.
Wilgoren, Jodi, and David E. Sanger. 2004. "Kerry Accuses Bush of ‘Stubborn,’ Not Steady, Leadership." New York Times, March 10.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||