Abstract
Research indicates that, when engaging their opponents, strategic candidates will draw (and redraw) lines of conflict, pulling attention to their advantaged topics. But do these expectations hold up in debates, where candidates are at the mercy of those asking the questions? And do strategic debate behaviors matter? This study draws on past literature to hypothesize the specific types of agenda-control behaviors we should see in debates. These hypotheses are tested in the 2008 presidential debates, using quantitative content analysis to examine candidate agenda setting, issue framing, and tone. The results show that both Obama and McCain used all three means of agenda control to continually displace the line of conflict in their favor. These findings offer empirical support for theories of strategic agenda control and heighten our understanding of agenda setting, framing, and tone as agenda-control mechanisms. Additionally, media and public opinion data suggest these debate agenda-control behaviors had real effects during the 2008 election.
As Schattschneider (1960) argued, a political conflict is defined by the line of cleavage along which it is drawn. Politicians and political candidates—savvy to Schattschneider’s insight if not his work—thus continually seek to draw or redraw (i.e., displace) each line of conflict in their favor. Consistent with Riker’s (1996) theory of heresthetics, political opponents draw these lines of conflict using the elusive but powerful mechanism of rhetoric, and in particular, agenda setting and issue framing (Chong & Druckman, 2007; Jacoby, 2000; Shah, Watts, Domke, & Fan, 2002). Researchers have applied these concepts of conflict displacement and heresthetics to political campaigns, showing how candidates should—and, indeed, tend to—draw the line of conflict in an election around the topics on which they have a political advantage (Druckman, 2004; Druckman, Jacobs, & Ostermeier, 2004). Candidates’ advantaged topics include those topics they or their party “own” (Miller & Krosnik, 2000; Petrocik, 1996; Petrocik, Benoit, & Hansen, 2003), with the economy serving as an important conditioning variable: “Clarifying” candidates should (and do) focus on the economy, while “insurgent” candidates should (and do) focus on other advantaged topics (Vavreck, 2009). In short, the literature has made significant theoretical and empirical strides in understanding agenda control in the general context of presidential elections.
Presidential debates, however, represent a very different kind of context. Unlike campaign rallies, television ads, and fundraising dinners, debates are rare instances in which candidates are not completely in control of the message. Moderators (and sometimes audience members) set the agenda, and candidates are incentivized and conditioned by social norms to respond to the moderator and to their opponent, including staying “on topic” of the question at hand (Brown & Levinson, 1987; Dailey, Hinck, & Hinck, 2008; Jamieson & Birdsell, 1990). These agenda-control constraints make debates a prime opportunity to test theories of rhetorical engagement in a high-stakes but also highly controlled environment.
Here, we ask: To what extent are predictions about candidates’ agenda-control strategies borne out by presidential debates? And do agenda-control behaviors in debates matter? Debates offer what may be a candidate’s greatest opportunity for “face time” with the American public. Data from the National Annenberg Election Survey indicate that, on average across all three debates in 2008, nearly two thirds of Americans viewed at least some of each debate (Kenski & Jamieson, 2011). As Benoit and colleagues write, presidential debates “may have become the single most important influence on vote decisions” (Benoit, McKinney, & Holbert, 2001). And although the lasting effects of debates have received mixed evidence in the literature (Holbrook, 1996; Stimson, 2004), at a minimum debates allow audiences to learn about candidate priorities—and potentially become drawn in to the conflict—by observing which policy topics the candidates discuss most often, which frames they use to discuss each topic, and the tone they take throughout (Benoit, Hansen, & Verser, 2003; Geer, 1988; Holbrook, 1999; Lemert, 1993). These cues prime voters to consider some topics over others (and from some perspectives over others), in turn influencing voters’ perceptions of and support for candidates (Druckman & Holmes, 2004; Iyengar & Kinder, 1987; McCombs & Shaw, 1972; Nicholson, 2005). This article tests how theories of agenda control map on to the face-to-face context of debates, in turn contributing to our broader understanding of political engagement.
We expect that, despite the agenda constraints of debates, candidates should nevertheless exhibit discernible agenda-control behaviors—from which behaviors we might infer agenda-control strategies, recognizing that candidates may stray from their strategies when off-script. Although candidates may be socially sanctioned when they redraw the line of conflict set by the moderator (i.e., when they go “off topic” via agenda setting or framing), candidates’ incentives to focus on their advantaged topics outweigh these constraints. Indeed, studies suggest that citizens tend to penalize candidates less for deviating from the question at hand when candidates do so in order to bring up a topic of high perceived importance (Boydstun, Glazier, & Pietryka, In Press)—including, for clarifying candidates, the central topic of the economy (Vavreck, 2009).
In looking at the 2008 debates specifically, we expect to see the candidates focus on their advantaged topics. For Obama, as the clarifying candidate in that election, the advantaged topic was the economy. For McCain, as the insurgent candidate, advantaged topics included his “owned” topics of defense and government spending. We expect that the candidates used both agenda setting and issue framing—both in answering the moderator’s questions and in dodging them—to draw the line of conflict around their respective advantaged topics.
We test these expectations using quantitative content analysis to examine the precise agenda-control behaviors that John McCain and Barack Obama used during the 2008 debates, reaching into the debate transcripts to tell a rich descriptive story about how the candidates drew and redrew the line of conflict. Most important, we test whether these candidates, forced to engage each other directly, behaved as past studies would suggest—but no studies have tested empirically. In the case of the 2008 debates, we find that the candidates indeed exhibited strong agenda-control behaviors, despite the constraints of having the questions set by moderators. Obama’s behaviors across the board communicated a message centered on the economy, while McCain directed the focus to government operations. Additionally, we examine news coverage and public opinion surrounding the debates, offering suggestive evidence that the candidates’ agenda-control behaviors had a meaningful, if short-lived, impact on the political system.
Agenda Control in Presidential Debates
Candidates and their campaigns work tirelessly to get their message out to the American public and are hesitant to give up agenda control. Debates, then—with their structure of unknown questions—would appear to be a campaign manager’s worst nightmare. How can a candidate stay on message when the topic for discussion is externally determined?
When faced with any question in a debate, candidates must weigh the costs and benefits of redrawing the line of conflict. Past research suggests key expectations for agenda-control strategies and, thus, agenda-control behaviors we should observe in debates. Foremost is Schattschneider’s (1960) work on how placing and displacing the line of conflict can build winning coalitions, in part by engaging additional voters on topics that matter to them (i.e., expanding the scope of the conflict). Following Schattschneider’s work, debate participants have incentives to change the topic or the frame to one that viewers care more about, compared to the topic or frame presented by the moderator. A closely related agenda-control strategy comes from Riker’s work on the art of heresthetics—where candidates attempt to prime the American people to think about the topics that favor the candidate when casting their votes (e.g., Jacoby, 1998; Petrocik et al., 2003; Riker, 1996; Sellers, 1998). Also by using heresthetics, candidates are able to get voters on their side by changing the considerations that voters bring to bear on a topic without actually changing their underlying attitudes. Instead of changing the topic then, candidates can change the way the topic is discussed—often through framing—to make the conversation more favorable to their position.
As this literature indicates, candidates are incentivized to make their advantaged topic the topic of discussion in a debate. But on what topics are candidates advantaged? It very much depends on the context of the campaign (Kingdon, 1995; Vavreck, 2009) and on the specific candidate (Petrocik, 1996; Petrocik et al., 2003).
Candidates have strong incentives to set the agenda to topics on which they have established expertise—either personally or via party affiliation (Kernell, 1993; Petrocik, 1996; Petrocik et al., 2003). Research shows that party reputations for being associated with and/or successful on certain topics (e.g., Democrats on social welfare and Republicans on taxes and “law and order”) have held fairly strongly over time (Pope & Woon, 2009; Walgrave & De Swert, 2007; Walgrave, Lefevere, & Tresch, 2012), allowing candidates to benefit from the reputational advantages of their parties. Of course, just because a candidate’s party “owns” a topic does not mean it is useful to bring up the topic in a particular debate. Other topics may be more pressing to the electorate or to the moderator. 1
Candidate strategies of topic ownership, however, are strongly conditioned by the economic climate. As Vavreck (2009) argues, the state of the economy determines the usefulness of campaign strategies. A candidate of the incumbent party benefits from a good economy, while a candidate not of the incumbent party benefits from a bad economy—Vavreck refers to both types of candidates as “clarifying” candidates. Strategically, it is in the best interest of these clarifying candidates to talk about the economy in the campaign. By contrast, “insurgent” candidates—incumbent-party candidates in a bad economy and challengers in a good one—are not helped by the state of the economy and so would be well advised to direct attention elsewhere. In good economic times, insurgent challengers should find another means of attack against the party in power, while in bad economic times, insurgent incumbents should find a governing success on which to concentrate. Specifically, an insurgent candidate should define the election around an issue on which the clarifying candidate holds a constrained position that is less popular than the insurgent’s position.
Here, we take these concepts and the theoretical expectations that follow and test them in the context of the 2008 debates. Although in debates candidates are necessarily in a reactive position, we expect that they will still attempt to control the agenda in line with theoretical expectations from the literature, specifically by drawing the line of conflict to their advantage. Of course, candidates face more constraints in a debate than in other campaign platforms and violating debate norms can carry serious consequences. Straying from the topic of a moderator’s question in order to set the agenda to a more advantageous topic can be a risky strategy, given the same social norms that govern human interactions in other contexts. If the moderator asks a candidate a direct question, the candidate would be ill-advised to ignore the topic altogether, no matter how disadvantaged he or she is on the topic, because doing so might be perceived as evasive or dodgy (see also research on the social norm of reciprocity, for instance: Berg, Dickhaut, & McCabe, 1995; Dailey et al., 2008; Fehr & Simon, 2000; Gouldner, 1960). The same logic applies to other mechanisms of agenda control, namely, issue framing. Candidates who use too much political “spin” also risk coming across as suspect.
These concepts apply also to tone as a mechanism of agenda control. The popular strategy of “going negative” can backfire in the context of a debate. While research indicates that going negative by criticizing one’s opponent may prove the best strategy in some situations (Lau & Pomper, 2002; Skaperdas & Grofman, 1995), doing so entails particular risk in a face-to-face debate where social norms may curb the number and veracity of acceptable negative attacks, compared to the impersonal medium of television ads. Although a pithy quip can also make for a good sound bite—think Senator Lloyd Bentsen’s “You’re no Jack Kennedy” comment to Senator Dan Quayle in the 1988 vice-presidential debate—negative statements may end up making a candidate look snarky or just plain rude. Research indicates that candidates who directly attack their counterparts in a debate, as opposed to indirectly attacking them by blaming them for important social and political problems, are less likely to be elected (Dailey et al., 2008). Social norms matter in debates, as in everyday life (Brown & Levinson, 1987; Jamieson & Birdsell, 1990).
Within the context of the particular constraints of live debates, we examine here three main mechanisms of strategic agenda control: agenda setting, issue framing, and tone. In a debate, candidates must choose how to handle each new question that is asked: to reinforce or to displace the line of conflict (Schattschneider, 1960). The literature indicates that candidates have incentives to use agenda setting, framing, and tone as tools of agenda control when the potential votes gained by doing so outnumber the potential votes lost by slighting social protocol (Damore, 2005). In the sections that follow, we examine all three tools in greater depth and develop general hypotheses regarding candidate agenda-control behavior in debates and specific hypotheses for the 2008 presidential debates.
Agenda Setting
Agenda setting is the most overt way that candidates can control the agenda. In a very broad sense, agenda setting refers to the process by which those problems that receive attention become topics of political discussion at the necessary expense of other topics (Erbring, Goldenberg, & Miller, 1980; McCombs & Shaw, 1972). In other words, which topics are talked about? Through agenda setting, candidates can define “what politics is about” (Schattschneider, 1960), a powerful tool for building coalitions and gaining votes (e.g., Baumgartner & Jones, 2009; Jones & Baumgartner, 2005; Kingdon, 1995; McCombs, 2004; McCombs & Shaw, 1972).
When should candidates use agenda setting in the context of a debate? If the topic of the moderator’s question is one that fits the candidate’s position relative to the economy and/or is an issue that she or he owns, the best choice is likely to stay on topic. If not, the candidate is likely better served by straying to other topics that might be more advantageous—a behavior analogous to Schattschneider’s concept of conflict displacement. President Bush in the 2004 debates provides a quintessential example of going off topic in order to stay on message. In this campaign, Bush consistently emphasized the national security message, responding to a question about partisanship in Washington in the third debate with the statement, “My opponent has got a plan of retreat and defeat in Iraq” (October 13, 2004).
Displacing the conflict, especially displacing it to a topic the audience is more interested in (Boydstun et al., In Press), has a clear upside. We thus expect that candidates in debates will try to set the agenda by reinforcing topics on which they are advantaged and attempting to displace topics on which they are disadvantaged.
Issue Framing
Topics are inherently multidimensional. In addition to choosing whether to stay on topic or not (and, if not, what topic to address instead), candidates must also choose how to frame each topic. Framing then, also known as second-level agenda setting, is the process of emphasizing a particular aspect of a topic over competing aspects (Chong & Druckman, 2007; Nelson, Clawson, & Oxley, 1997). As with agenda setting, candidates have strong incentives to use framing to draw and redraw the line of conflict in a debate to their advantage.
Whether the topic is favorable or not, candidates and their teams of “spin doctors” have preferred ways of framing it to make it more favorable. Through framing, candidates are able to control the agenda in a more nuanced way—even when discussing a disadvantaged topic, they can use framing to prime citizens to think of the topic in terms of a more advantaged topic. For instance, in response to a question about the conflict in Iraq, an insurgent GOP candidate may choose to use a safety frame—describing the deployment of troops to Iraq as critical to preventing future terrorist attacks on the United States. A clarifying Democratic candidate, on the other hand, may respond to the same question with an economic frame—describing the U.S. involvement in Iraq as a massive burden on U.S. taxpayers. In the 1992 presidential debates, for instance, Bill Clinton took a question about defense and framed it in economic terms saying, “I want to take every dollar by which we reduce defense and reinvest it in technologies for the 21st century—in new transportation, in communication, in environmental clean-up technologies.” Through Clinton’s emphasis on the economic frame, other aspects of the defense topic, such as national security, the ability of the United States to project power, and the safety of nuclear weapons, were absent from (at least that portion of) the discussion.
Thus, even when candidates are staying “on topic” by responding directly to a question, they do have discretion over how they talk about the topic and, thereby, over which dimensions of consideration citizens will use in evaluating the topics and candidates. Indeed, research indicates that candidates are more willing than previously thought to trespass on topics owned by their opponents, specifically by using an alternative framing of the topic (Sides, 2006).
Framing, then, is another way of controlling the agenda—another way that candidates can draw or redraw the lines of conflict. We expect that candidates will use framing to draw attention to their advantaged topics. For instance, clarifying candidates will use economic frames, while insurgent candidates will instead use frames that emphasize their owned topics.
Tone
Finally, candidates are also able to exercise some control in a debate through the tone of their remarks. Although tone is rarely discussed in the literature on agenda control (but see Sheafer, 2007, and Schoenbach & Semetko, 1992, for notable exceptions), it is a critical mechanism candidates can use to shape the discussion in their favor. Specifically, candidates may choose to use a negative, positive, or neutral tone when talking about the substance of a topic as well as when talking about their opponent.
A change in tone can have a significant effect on the meaning and interpretation of a statement. For instance, in the first 1992 debate, the moderator posed a question about racial division. Ross Perot responded with a positive statement about how “We ought to love one another because united teams win and divided teams lose.” Bill Clinton responded with a negative statement about growing up in the segregated South: “I saw the winds of hatred divide people and keep the people of my state poorer than they would have been, spiritually and economically.” George Bush responded by criticizing the state of Arkansas for not having any civil rights legislation. In these ways, each candidate spoke about race, but they communicated very different messages. Perot was hopeful, Clinton used a negative tone on the substance of race to express empathy (similar to his famous “I feel your pain” commiseration), and Bush used a negative tone to make a personal critique of Arkansas and, by extension, Clinton (Davis & Womack, 2001; Stuckey & Antczak, 1994). Even though the candidates stuck to the topic of the moderator’s question, they used tone (and framing) in attempts to draw the line of conflict in their favor.
We expect that candidates will use the tone of their substantive remarks to reinforce their positions on advantaged topics. For instance, clarifying candidates in poor economies are incentivized to criticize the economic performance of the party in power—and, by association if not by actual record, the economic credentials of their opponent. We also see personal attacks as a tactic of a disadvantaged candidate. If a candidate is behind in the polls coming in to a debate, he or she will be more likely to reach for the risky tool of negative personal statements.
The 2008 Presidential Campaign
Given candidates’ incentives to use agenda-control strategies to shift the debate discussion to their advantage, what can we expect in the 2008 presidential debates? Political campaigns don’t take place in a vacuum, and the messages McCain and Obama communicated in the 2008 presidential debates were certainly influenced by the campaign environment. Given the candidates’ divergent positions with respect to the economy in particular, we expect them to exhibit very different debate behaviors (Vavreck, 2009).
In the 2008 campaign, Obama represented the clarifying candidate with a strong incentive to draw attention to the poor economy. The housing crisis and economic downturn burst on to the political scene in late 2007, drawing attention toward the economy. Obama’s campaign was well positioned to take advantage of the increased salience of the economy. As his campaign strategists said, they “wanted the whole fall to be about the economy” (Jamieson, 2009, p. 71). The collapse of Lehman Brothers on September 15, 2008, just 11 days before the first presidential debate, along with the broader economic troubles, fits well with the Obama campaign’s strategy of focusing on the economy. Indeed, a look at Obama’s campaign ads confirms this focus on the economy. Quantitative content analysis of the Stanford Political Communication Lab archive of presidential election television ads shows that 56% of the 88 archived ads for Obama during the 2008 general election were either primarily or secondarily on the topic of the economy, compared to only 44% of McCain’s 105 archived ads. 2 Vavreck’s theory, plus the context of the 2008 campaign, leads us to expect Obama to use agenda setting, framing, and tone in the debates to draw attention to the economy in particular.
In contrast, McCain represented the insurgent candidate in 2008. The bad economy disadvantaged his campaign, making it in his interest to draw attention to other topics. The issue ownership literature (Damore, 2004; Petrocik, 1996; Petrocik et al., 2003) suggests candidates should shift attention toward topics that are “owned” by either the candidate or the party. This theory is not mutually exclusive from Vavreck’s suggestion that insurgent candidates should shift the attention to topics on which they are closer to the median voter than their opponent. In line with these theories, McCain’s campaign put early and strong emphasis on foreign policy and defense, traditionally advantaged topics for the GOP and ones McCain personally “owned” through his combat experience in the Vietnam War. McCain’s political team used the slogan “Country First” and emphasized his willingness to make tough but correct choices in defense, as he did in backing the surge in Iraq (Jamieson, 2009). Looking again at the television ads for McCain in the general election, we see that defense was a major focus—16% of ads for McCain were either primarily or secondarily about defense, compared to only 4% of Obama’s ads. This strategy of focusing on defense appeared to be an effective one during the primary campaign (Jacobson, 2009), but as economic conditions worsened and became more salient to the public, McCain began to emphasize another topic he owned in addition to defense: government operations (specifically, spending reform). Not only does the topic of government size and inefficiency traditionally belong to the GOP, but McCain’s record of speaking out against wasteful spending and earmarks gave him particular ownership of the topic. His campaign ads reflect this focus, with 16% of his ads primarily or secondarily about government operations, compared to less than 6% for Obama. 3
Thus, drawing on both Vavreck (2009) and the issue ownership literature, we see that the candidates were advantaged by different topics in the 2008 election: Obama by the topic of the economy, and McCain by the topics of defense and government operations. Combining these divergent agendas with our theoretically derived expectations regarding strategic agenda control, we predict very different behaviors by these two candidates in the 2008 debates.
With regard to agenda setting, we expect candidates to dedicate more statements in general to their advantaged topic (H1a) and also to use more off-topic statements to emphasize their advantaged topic (H1b). Substantively, this means that Obama should spend more time and more off-topic time than McCain talking about the economy, while McCain should spend more time and more off-topic time than Obama talking about defense and government operations.
We believe that the influence of the advantaged topic will carry over into issue-framing behaviors as well. Remember, it is in the candidates’ best interest to draw attention to their topic—whether through agenda setting or framing—whenever possible. Thus, we expect that candidates will use framing to emphasize their advantaged topic in their statements in general (H2a) and will also use more of their off-topic statements to emphasize their advantaged topic through framing (H2b). In 2008, we expect that Obama should use more economic frames (to emphasize the economy) than McCain, while McCain should use more safety frames (to emphasize defense) and political frames (to emphasize government operations) than Obama.
Finally, we expect that the tone of candidate remarks will also be influenced by the context of the debate and each candidate’s position relative to the economy. We expect clarifying challengers in a bad economy (in this case, Obama) to make more substantively negative comments, particularly about the economy (H3a). However, we expect “underdogs” in an election (in this case, McCain) to be more personally negative—a somewhat risky strategy in a face-to-face debate, and thus, more likely to be used by the candidate down in the polls (H3b). 4
Thus, building on classic literature that leads us to expect candidates to use agenda-control strategies to direct attention to their advantaged topics, we are able to derive general hypotheses that can apply to future research on other debates, while also specifying hypotheses for the 2008 debate in particular. In summary, we expect:
Hypothesis 1 (H1): Agenda setting:
H1a: Candidates will dedicate more of their statements to their advantaged topic than will their opponent (Obama to the economy and McCain to defense and government operations).
H1b: When speaking off topic, candidates will dedicate more of their statements to their advantaged topic than will their opponent (Obama to the economy, and McCain to defense and government operations).
Hypothesis 2 (H2): Framing:
H2a: Candidates will use more frames to emphasize their advantaged topic than will their opponent (Obama will use more economic frames and McCain will use more safety and political frames).
H2b: When speaking off topic, candidates will use more frames to emphasize their advantaged topic than will their opponent (Obama will use more economic frames, and McCain will use more safety and political frames).
Hypothesis 3 (H3): Tone:
H3a: Clarifying candidates in a bad economy will make more substantively negative statements than will their opponent (Obama will make more substantively negative statements than McCain).
H3b: Candidates down in the polls prior to a debate will make more personally negative statements than will their opponent (McCain will make more personally negative statements than Obama).
Research Design
In order to examine candidate agenda-control behaviors in the 2008 presidential debates, we collected and coded the full transcripts from each of the three presidential debates between John McCain and Barack Obama. Each moderator question and candidate response was hand coded by at least one of the authors, with an overlap of 25% of the statements in order to evaluate reliability. The unit of analysis was the “statement,” defined as an independent clause. 5
We coded several variables in each statement, most notably the topic, the frame, and the tone. 6 The question and response topics were coded based on the Policy Agendas Topics Codebook (Baumgartner & Jones, 2006), containing 19 major policy topic codes (e.g., macroeconomics, health, defense). 7 By cross-referencing the topic of the statement with the topic of the question posed by the moderator (or audience member), we calculated a binary “on-topic/off-topic” variable for each statement. 8 This variable thus allows us to measure how often and to which topics each candidate went “off topic.”
In order to track agenda-control attempts via framing, we also coded each statement according to the frame the candidate used in conveying it. On the basis of our readings of a sample of debate transcripts from 1992 through 2008, we identified and coded for eight different frame dimensions general enough to span policy topics: economic, political, logistical, patriotic, legal, moral, safety, and effectiveness (plus an “other” category). For example, we coded John McCain’s statement in the second debate, “Nuclear power is safe,” as being on the topic of energy and framed in safety terms. By contrast, we coded his subsequent statement, “and it creates hundreds of thousands of jobs,” as being framed in economic terms. Later in the debate, McCain made a statement about an energy bill with benefits for oil companies, saying, “I have fought time after time against these pork barrel—these bills that come to the floor and they have all kinds of goodies.” This statement was coded as on the topic of energy but with a political frame.
The next variable that we measured was tone. We coded whether the tone of the statement was generally positive (e.g., supportive/hopeful), negative (e.g., critical/fearful), or neutral. For example, in the second debate, we coded Obama’s statement, “We have a moral commitment as well as an economic imperative to do something about the health care crisis,” as positive, but we coded his statement, “I can’t tell you how many people I meet who don’t have health insurance,” as negative. In addition to coding the general tone of each candidate’s statements, we also tracked the tone each employed with reference to his opponent. Any statement that included a reference to the opponent (by using the opponent’s proper name, by using a pronoun, or clearly in context) was coded as a candidate reference. Each candidate reference was also coded negative, neutral, or positive, in order to capture whether and when either candidate “went negative” on a personal level. Any time that the opponent is referenced with a negative tone, we consider it a negative “personal” statement. For example, McCain’s statement about Obama, “Again, a little bit of naiveté there,” is coded as personally negative.
After presenting our findings on the candidates’ debate behaviors, we evaluate the potential consequences of these agenda-control behaviors by looking at news coverage before and after the debates as well as public opinion polls.
Findings
Campaign teams have historically turned negotiations over debate structure and format into fierce battles (Carlin & McKinney, 1994). There were no such battles in 2008, as both camps settled on a format with relative ease. There were no opening statements in any of the three 2008 debates, and each candidate was allowed only 2 minutes for a closing statement in the third and final debate. Initial responses were 2 minutes long with an optional 1- to 5-minute follow-up discussion at the discretion of the moderator. This structure led to a nearly even split in speaking time between the two candidates. Although the candidates responded to the same questions and were allotted the same amount of time, we present findings in the following three sections to indicate that both candidates used agenda setting, issue framing, and tone to control the agenda of the debate and, specifically, to continually draw the line of conflict in their favor.
Agenda Setting
Over the course of the three 2008 presidential debates, the candidates were asked a total of 29 questions on 10 topics. In a (nonrealistic) world where candidates do not exert any agenda-setting power in the debate, candidate responses would be limited to those 10 topics. In reality, of the 19 major policy topic codes we employed from the Policy Agendas Project, 18 were mentioned at some point by one candidate or the other (neither mentioned agriculture). The 29 questions posed by the moderators (and audience members; hereafter clustered with moderators) ranged from simple, single-statement questions, to more in-depth promptings, expositions, and follow-ups. All of these questions comprise valuable agenda space in the debate, and we track the use of this agenda space using the statement as the unit of analysis. Assuming that McCain preferred questions on defense and government operations while Obama preferred questions on the economy, we see that the moderators’ choice of questions was quite balanced: The moderators made exactly 100 statements on the economy and a combined 100 statements on defense and government operations (as well as 69 statements on the remaining 7 question topics). That said, the fact that the economy was the most dominant question topic surely advantaged Obama, consistent with his positioning as the clarifying candidate in a weak economy. Figure 1 presents a comparative overview of the topics introduced by the moderators and the topics covered by each candidate.
Percentage distribution of statements by topic, across candidates and moderators.
If the moderator entirely set the agenda, then the three bars representing the moderators, McCain, and Obama in Figure 1 would be identical for each topic. Instead, we see that McCain especially spoke more about defense and government operations and Obama especially spoke more about energy and health care than the moderators’ questions alone would have predicted. For Figure 1 and for subsequent figures, Table 1 provides the results of difference of proportion tests. Note, for example, that each candidate dedicated a smaller proportion of their attention to the economy than the moderators did but that, consistent with our theoretical story, Table 1 shows that this difference is only significant at the p < .05 level in McCain’s case.
Difference of Proportion Tests Comparing Key Candidate Agenda-Control Behaviors.
Note: ns = nonsignificant. Table 1 shows difference of proportion tests based on different sample sizes. In each comparison, the lower and upper 95% confidence intervals are calculated relative to the speaker with more statements in that category. Thus, a lower-bound confidence interval that does not drop below zero indicates that the speaker with more statements exceeded the other speaker’s statements in that category by a statistically significant margin.
In line with our expectations regarding agenda setting (H1a), Obama dedicated a significantly greater proportion of his attention to the economy than McCain. And McCain spent more of his attention on defense and government operations and defense compared to Obama, although Table 1 shows that only the difference in attention to government operations is significant. Overall, however, the two candidates dedicated upwards of 90% of their attention to the 10 moderator-introduced topics, which were obviously very important. In order to understand more about how the candidates responded to the agenda set by the moderators, we have to take a closer look at the amount of time each candidate spent “on topic” and “off topic.”
Again, deviating from the topic of the moderator’s question can be a risky but potentially beneficial strategy. Going off topic sends a moderately costly signal of candidate priorities, making this behavior a solid measure of candidate topic preference. Overall, McCain and Obama were more likely to stick to the topic of the moderator’s question than they were to diverge from it, but they did go off topic. Both candidates were especially likely to go off topic when presented with a question that did not fit their preferred agenda. For instance, McCain made 281 off-topic statements when presented with questions on the economy, compared to Obama’s 185, while Obama made 43 off-topic statements when asked questions on defense, compared to 21 for McCain. The candidates thus appeared to go off topic in order to draw attention away from a topic on which they were not advantaged. Table 1 shows that these differences are significant.
When the candidates did go off topic, what did they talk about? As predicted, the candidates diverged from the moderators’ agenda in order to redraw the lines of conflict in their favor. They used the agenda-setting power of going off topic to communicate their advantaged messages. Figure 2 presents the off-topic remarks for each candidate across the three debates. Overall, McCain went off topic more often than Obama; McCain had a total of 464 off-topic statements to Obama’s 378 (a significant difference again, as shown in Table 1).
Percentage distribution of off-topic attention by candidate.
We hypothesized that McCain would go off topic in order to draw attention to the topics of defense and government operations. Figure 2 shows that McCain spent the lion’s share of his off-topic statements addressing government operations and significantly more so than Obama (Table 1). McCain also dedicated significantly more of his off-topic attention to defense than Obama did (Table 1), going off topic to defense about 10% of the time (Figure 2). These findings are supportive of our expectations regarding agenda setting (H1b) and validate Vavreck’s (2009) argument regarding insurgent candidates.
As for Obama’s agenda-setting choices, we hypothesized that he would go off topic to talk about the economy (H1b). Supportive of this hypothesis, Figure 2 shows that Obama spent almost twice as much of his off-topic time on the economy, compared to McCain. Again, Table 1 shows that this difference is statistically significant. Figure 2 also shows that, while not as much as McCain, Obama too frequently went off topic to address government operations.
Together, the data in Figure 2 show that Obama was by no means single-minded in terms of his off-topic attention, but he instead went off topic to draw attention to a variety of topics, including energy, economics, and health. Obama also spent some of his off-topic statements on education, a traditionally Democrat-owned topic that McCain very rarely addressed with his off-topic comments. These findings suggest that both candidates tried to use agenda setting to redraw the lines of conflict in the debate in their favor. Through both his off-topic comments and his overall attention distribution, McCain attempted to draw attention to the topics of government operations and defense. Obama, on the other hand, tried to direct the discussion toward the economy, in addition to the topics of energy and health care. One might expect a more focused allocation of off-topic attention, but both the economy and defense were the subject of many of the moderators’ questions, as shown in Figure 1, which perhaps freed up the candidates to broaden their agenda-setting behaviors.
At any rate, by going off topic, both candidates drew the lines of conflict in their favor. To understand exactly what these behaviors looked like in practice, we can take a qualitative look at the nature of each candidate’s off-topic statements, demonstrating just how far from the original question the candidates sometimes strayed. For instance, when Obama was presented in the first debate with a question about the likelihood of another 9/11-type attack on the United States, he responded by discussing the financial troubles of the country. After mentioning heavy U.S. borrowing from China, Obama concluded, “We have weakened our capacity to project power around the world because we have viewed everything through this single lens, not to mention, look at our economy. We are now spending $10 billion or more every month.” By going off topic from the question about terrorism to discuss the problems with the economy, Obama employed agenda setting to redirect the debate to a topic on which he was advantaged. McCain, by contrast, chose to emphasize government operations, saying, “after 9/11, Senator Joe Lieberman and I decided that we needed a commission, and that was a commission to investigate 9/11, and find out what happened, and fix it.” He went on to say, “And I’d like to remind you, also, as a result of those recommendations, we’ve probably had the largest reorganization of government since we established the Defense Department.” Qualitatively, we see that both candidates took a question about 9/11 and found a way to connect it to their advantaged topic. In short, both quantitative and qualitative evidence support H1 as a whole: The candidates used agenda setting to draw the line of conflict in their favor.
Framing
The candidates also shifted the focus to their advantaged topics through how they framed both on-topic and off-topic statements (which framing, as we know, can not only displace the line of conflict but also potentially influence how the public perceives each topic). Figure 3 presents an overall look at the frames utilized by the two presidential candidates in the 2008 debates.
Percentage distribution of frame use by candidate.
Figure 3 displays the proportion of each candidate’s framing efforts that were dedicated to each of the eight frame categories listed above. We hypothesized that McCain would use political frames to support his advantaged topic of government operations and safety frames to support his advantaged topic of defense. We similarly hypothesized that Obama would use economic frames to draw attention to his advantaged topic of the economy (H2). Figure 3 shows that McCain dedicated about 10% more of his frames to political framing, compared to Obama, while Obama dedicated about 10% more of his frames to economic framing. Recall that Table 1 shows that these are statistically significant differences. The two candidates are actually quite similar in their remaining frame usage, with the only other noticeable gap found in McCain’s greater use of patriotic frames (although this difference, not included in Table 1, is not statistically significant). We do not find that McCain used more safety frames than Obama; even when McCain was talking about defense, he was more likely to use logistical (36.5%) than safety (20.1%) frames, perhaps reflecting an inability to move beyond the details of government operations and a missed opportunity to use framing to draw attention to his advantaged topic of defense.
Because we expect that candidates went off topic strategically, it is also useful to know what frames they used when they were off topic. Figure 4 presents the candidates’ percentage distribution of frames when they were off topic. Overall, McCain used the economic (29%), logistical (25%), and political (26%) frames as his top three frames, but when he went off topic—that is, when we expect he was actively working to redraw the line of conflict—he used economic (41%) and political frames (32%) much more than any other. McCain’s increase in political frames when going off topic is in line with our expectations, but his use of economic frames—instead of, for instance, safety frames, which would have reinforced his advantaged topic of defense—suggests that McCain again was not using framing as effectively as he could have to redraw the line of conflict. By contrast, Obama’s off-topic frames show a significantly greater focus on the economy (see Table 1). When he tried to shift the focus of the debate, he was nudging it in the strategic direction of the economy. In these ways, Obama appears to have been generally more successful than McCain in using framing to his advantage.
Percentage distribution of off-topic frame use by candidate.
To illustrate the framing differences between the candidates, compare the following representative statements. First, in the third debate, Obama stayed on topic for a question about energy independence, but he framed his response in economic terms, specifically job creation:
If we can get that right, then we can move in a direction not only of energy independence, but we can create 5 million new jobs all across America, including in the heartland where we can retool some of these plants to make these highly fuel-efficient cars and also to make wind turbines and solar panels, the kinds of clean energy approaches that should be the driver of our economy for the next century.
As an illustration of Obama’s use of framing while off topic, we can look to a defense question posed in the first debate, in which the moderator asks if the candidates view Russia as an enemy. Obama went off topic for part of his response to this question in order to talk about energy independence and how Russia’s increasing power through petro-dollars is related to “the pain people are feeling at the pump,” thus, employing an economic frame. In trying to redraw the lines of conflict in the debate, Obama was directing attention away from the topic of defense and toward the topics of energy and the economy.
McCain, on the other hand, used framing to communicate a political message about his suitability for the office of president and, often (but again, perhaps not as often as would have benefitted him), to draw attention to the topic of defense. When McCain was asked the same question about energy independence in the third debate, he responded by discussing the national security dangers of relying on unfriendly countries for our energy supplies, arguing, “Canadian oil is fine . . . we can eliminate our dependence on the places in the world that harm our national security.” And when McCain received a question about the economy in the first debate—a topic he preferred to avoid—he went off topic to talk about government operations and how he would use the veto power as president. Here again, we see McCain trying to move the conversation in the direction of government operations through going off topic and through the use of political frames but perhaps missing a chance to connect with voters by using framing to focus less on nitty-gritty details and more on broad resonant themes of safety and government reform. Whether or not the candidates came to the debates with specific agenda-control strategies, their behaviors in the debates indicate that they were using the tools at their disposal to redraw the line of conflict in their favor—Obama arguably more effectively than McCain. The findings on issue framing support our expectations regarding how candidates use framing to shift the discussion in their favor. Additionally, they bolster Vavreck’s (2009) claims that the economic environment determines the best campaign strategy. We show here that the candidates did not communicate their messages through agenda setting alone—McCain and Obama both used issue framing to emphasize their advantaged topics.
Tone
We can gain additional insight into the messages candidates sent to citizens by examining the tone of their statements. Using tone strategically can help candidates communicate their opinions on matters of substance—including those they would prefer not to address—as well as their opinions of their opponents. As the clarifying candidate in a bad economy in 2008, we expected Obama to use a more negative tone on substantive issues than McCain. The data support this hypothesis. In the 2008 debates, Obama used a negative tone 41% of the time to McCain’s 36%—a significant difference (Table 1). This finding may be somewhat surprising to some, given that Obama was known for running a “hopeful” campaign.
Of course, there were important differences in the topics on which each candidate chose to “go negative.” In Figure 5, we see the distribution of each candidate’s negative statements across topics. These data provide a representative overview of how each candidate used his negatively toned policy statements. Figure 5 reinforces some of the common perceptions of the candidates and the parties: McCain was significantly more critical of the government, and Obama was more critical of defense policy (though not significantly so) and significantly more critical of the economy (again, see Table 1 for the results of the difference of proportion tests).
Percentage distribution of negative toned statements.
The two candidates were clearly using tone differently, but what do these differences mean for the nature of the discussion that took place in the debates? How did the candidates use tone to draw the lines of conflict in their favor? Although the moderator set the initial topic, the candidates decided whether the conversation was going to be a hopeful or a fearful one. Looking at their discussion of the economy, both candidates made negative statements, but Obama made many more: 129 to McCain’s 86. Not only was Obama trying to direct attention to his advantaged topic of the economy, he was trying to draw the line of conflict in such a way as to critique the economic policies of the Republican Party and, by extension, McCain. The most significant gap in tone comes on the topic of government operations, where McCain made 124 negative statements to Obama’s 56. Not only was McCain trying to direct attention to an advantaged topic but, as a Republican candidate in an election year with an unpopular GOP president and public dissatisfaction with the direction of the country (Jacobson, 2009), he likely wanted to get out front in critiquing the government. Thus, both candidates used tone to draw attention to their advantaged topics in a way that benefited their position, a finding supportive of H3a.
We can take a closer look at how the candidates exercised agenda control through the use of tone by examining the tone across debates. The difference between the candidates’ tone on the economy is seen most clearly in the first debate, with Obama making 59 negative statements about the economy to McCain’s 30. Interestingly, Obama’s division of economic statements between positive and negative grew more balanced in the later debates. Whereas in the first debate Obama made nearly 3 times as many negative statements regarding the economy as positive, in the final debate he actually used a few more positively toned statements than negatively toned ones. Having commiserated with the public and established the economic failures of the GOP in the early debates, Obama spent about half of his economic agenda in the final debate proposing hopeful solutions to the problem. In the first debate, Obama lamented, “Because of the economy’s slowing down, I think we can also expect less tax revenue so there’s no doubt that as president I’m going to have to make some tough decisions.” But by the final debate, he spent more time on solution-filled statements like, “Let’s help families right away by providing them a tax cut—a middle-class tax cut for people making less than $200,000.”
This use of tone not only brought the topic of the economy to the forefront but also allowed Obama to present himself as the candidate best able to handle the nation’s economic problems—a strategy that fits well with his position as a challenger clarifying candidate. In 2008, it was unlikely that the public would have supported a candidate who made nothing but positive statements about the economy—even if the candidate’s complete attention was dedicated to this important topic. These statements by Obama offer quintessential examples of how tone serves as an important agenda-control tactic to help draw the line of conflict in a beneficial way.
In addition to deciding what tone to use when discussing the substance of the debate topics, candidates also must decide how negative they want to be when discussing their opponent. The interpersonal interactions of the candidates draw attention away from the topics in a debate and toward the candidates themselves, thus communicating a very different message to the public (Beck, 1996; Dailey et al., 2008; Harris, 2001). We found, unsurprisingly, that both candidates spoke much more negatively than positively about each other. Yet even though Obama was generally more negative than McCain when discussing substance, McCain was significantly more negative than Obama when talking about his opponent (Table 1). McCain spoke about Obama with a negative tone an average of 75% of the time across the three debates, with the percentage of negative statements increasing with every debate. By comparison, approximately 53% of Obama’s statements about McCain were negative. This finding supports the expectation outlined in H3b that McCain played the underdog role in the debates, perhaps hoping to land a “You’re no Jack Kennedy” type of punch. It also supports the idea that negative personal statements are a tactic of a losing candidate. McCain increased his personal attacks with each debate, as his chances of victory appeared increasingly slim.
Although it is hard to know exactly how the public responded to individual statements made by either candidate, it is possible that McCain’s extensive negative comments regarding, and in the presence of, Obama were seen as rude or in poor taste. Whereas it does not appear that candidates suffer in terms of vote share when they “go negative” on substance or in campaign ads (Brader, 2005; Lau, Sigelman, Heldman, & Babbitt, 1999), going negative on an opponent during a debate may be another story (Dailey et al., 2008). 9 Whether on the substance of policy or on personal characteristics, the tone of candidate statements flavors the entire debate discussion. The 2008 debates show how tone can be used to draw the line of conflict in a candidate’s favor or to distract from substance altogether.
Does It Matter? Evaluating the Media and Public Response
In the 2008 debates, both candidates used agenda setting, framing, and tone to shift the debate toward their advantaged topics and to thereby gain some political advantage. But did these behaviors “work”? Is there evidence that they may have influenced news coverage of the election or public perception of the candidates?
Debate Effects on News Coverage of Policy Topics
One way to examine whether candidate agenda-control behavior in debates matter is to look at whether the topics candidates discuss during a debate—that is, candidates’ debate agenda setting—get picked up in news coverage following the debate. The candidates surely hope that their efforts to draw the line of conflict in their favor stick beyond the debate agenda. The literature on campaign agenda setting gives us reason to believe that, under certain conditions, candidates are able to set the media agenda (McCombs & Shaw, 1972; Roberts & McCombs, 1994) and that topics in the news in turn influence the topics to which the public attends (Gitlin, 2003; Hester & Gibson, 2007; Iyengar, 1994; Kiousis & McCombs, 2004; López-Escobar, Llamas, & McCombs, 1998; McCombs & Shaw, 1993). In particular, news coverage can prime topics as criteria for evaluating candidates (Iyengar & Kinder, 1987; McCombs & Shaw, 1993), and candidates are keen to make their advantaged topics the criteria for evaluation.
Although a full causal test of the relationship between candidate debate agendas and news coverage is well beyond the scope of this study, we can take a simple look at how the percentage of election news coverage in relation to key topics before each debate compare with after each debate. Consulting Figure 1, we selected four topics for our analysis, representing two advantaged topics for each candidate. Specifically, we examine the two topics toward which each candidate diverted the most off-topic attention relative to his opponent. Looking at Figure 2 (and also Table 1), we see that Obama spent significantly more of his off-topic statements on the economy and energy, while McCain spent significantly more of his off-topic statements on defense and government operations (again, government spending reform specifically).
We then searched the news stories of three major television news outlets (CNN, Fox, and NBC) and four major newspapers (Washington Post, New York Times, USA Today, and Wall Street Journal [abstracts]) for 3 days before and after each debate, counting the day of each debate as the third day of predebate coverage (but manually sorting those stories in the evening after a debate into postdebate coverage). For each predebate and postdebate 3-day period, we searched, first, for the total number of stories mentioning Obama and/or McCain in order to estimate a total amount of election news coverage. 10 Then we searched separately by topic for stories mentioning Obama and/or McCain within the same paragraph as keywords capturing that topic area. The four sets of topic keywords (listed in the Appendix) were vetted thoroughly; we began with a wide net of potential keywords and then narrowed the keywords systematically until at least 90% of the hits were true positives, with false negatives constituting less than 10% of the number of stories retrieved. 11 Thus, using the story as the unit of analysis, we are able to estimate the percentage of all election coverage that included discussion of each of these four topics, comparing percentages of coverage before versus after each debate.
Our findings suggest that the candidates’ debate agenda-setting behaviors did indeed influence news coverage of the election, particularly in the case of those topics that were not high on the media’s agenda before the debates. We examine these media data by averaging the percentage of coverage each topic received in each predebate 3-day period—out of the total amount of news coverage (from our general Obama and/or McCain search)—to the percentage of coverage it received in each postdebate 3-day period. We find that coverage of energy doubled, from an average of 4% of election coverage in the predebate periods to 8% in the postdebate periods. Coverage of government reform increased from an average of 29% of predebate election coverage to an average of 36% of postdebate election coverage. The topics of the economy and of defense were more prominent in the news—as well as on the public’s radar—before the debates began, and so the increase in coverage of these topics is more modest though still present. Coverage of the economy increased from an average of 63% of predebate coverage to 64% of postdebate coverage, and coverage of defense increased from an average of 30% of predebate coverage to 33% of postdebate coverage. Note also that these two highly salient topics—economy and defense—were topics that the debate moderators emphasized, while the topics with a greater seeming impact on news coverage—energy and government reform—were topics of more interest to the candidates than the moderators. These findings, while suggestive in nature only, indicate that the candidates’ debate agenda-setting behaviors had at least a short-term effect on news coverage of the election. That said, the candidates’ agenda-setting behaviors were, of course, working at cross purposes. While each candidate seems to have drawn significantly heightened media attention to at least one advantaged topic (energy for Obama and government reform for McCain), each candidate was simultaneously disadvantaged by having his opponent’s topics in the news (including, especially, the negative effects McCain suffered by having the economy stay in the news).
Public Perceptions of Who Won the Debates
Another way to evaluate the effectiveness of the candidates’ agenda-control behaviors is to look at public opinion. According to most political pundits and public opinion polls, Barack Obama won each of the three debates over opponent John McCain. The average perceived margin of victory for Obama (i.e., the gap between the percentage of respondents saying that Obama won and saying that McCain won) across the CNN, Annenberg, and CBS “uncommitted voters” polls was 11.7% for the first debate, 28.3% for the second debate, and 34.7% for the third debate. 12
We argue that Obama’s perceived debate victory is at least in part attributable to his efforts to draw attention to his primary advantaged topic: the economy. The economic crisis in 2008 brought Obama’s advantaged topic to the forefront of the public mind. Gallup’s “most important problem” series shows that, a year prior to the debates, in August, September, and October of 2007, the percentages of adults surveyed who cited the economy as “the most important problem facing this country today” were 8%, 11%, and 9%, respectively. But by the lead-up to the presidential debates 1 year later, things had changed dramatically. In August, September, and October of 2008, the percentages of respondents who cited the economy as the most important problem were 38%, 36%, and 42%, respectively, by far exceeding the secondary topic of concern, generally the situation in Iraq (registering at 19%, 5%, and 5%, respectively). 13
Beyond who won the overall debates, public opinion data suggest that Obama’s efforts to direct the agenda toward the economy earned him particular favor on that key topic in the eyes of voters. Of course, Obama’s campaign had been hitting the topic of the economy hard for months, and surveys show that most Americans already had more trust in Obama to handle the economy by the time of the first debate. For example, two separate CNN polls in September 2008, before the first debate showed that 52% and 53% of respondents, respectively, thought Obama would better handle the economy, compared to 44% and 43% who thought McCain would be better, for a margin of 8% to 10% in Obama’s favor. 14 These surveys speak to the strength of the Obama campaign; earlier versions of this same question by CNN began with a February 2008 survey showing a 15% margin in McCain’s favor, followed by fluctuating results but generally improving for Obama.
Still, the debates appear to have considerably strengthened Obama’s economic edge. A CNN poll (in this same survey series) following the first debate showed that Obama’s edge on being perceived as the best to handle the economy had increased to 20%—the highest margin in the series. The last survey in the series, conducted immediately following the third debate, showed Obama’s margin holding at 19%. Of course, the debates were not the only medium of campaign messaging during the fall leading up to Election Day; they were simply the most widely televised and most widely viewed 4.5 hours of campaign messaging during this crucial time. The point is that Obama effectively used the agenda-control tactics at his disposal during the debates to draw attention to the economy. Arguably then, we can attribute at least part of the upward shift in public trust in Obama’s ability to handle the economy to how he handled the debates. 15
We cannot say whether this public opinion data identifies effects specific to the debates or simply reflects Obama’s growing advantage as the clarifying candidate in light of a struggling economy; perhaps Obama’s advantage on the economy coupled with the dominance of the economy as a topic of concern was enough to win him all three debates—and trust in handling the economy—in the public’s eye. Yet, at a minimum, these findings suggest that the debates provided the electorate with a sharply salient microcosm of the dueling campaigns, transmitting signals that reinforced the candidates’ broader agenda-control strategies despite the highly controlled and unscripted debate context.
Conclusions
This article tested existing theories of agenda-control behaviors in the unique (and electorally significant) face-to-face context of presidential debates. Despite being forced by the format of debates into the role of respondents, candidates in the 2008 debates repeatedly redrew the line of conflict in their favor by going off topic from the moderators’ questions, by framing their responses, and by adding valance (tone) to their answers. As the clarifying challenger candidate in 2008, Obama focused his off-topic attention on the economy, framed topics in economic terms, and used a strongly negative tone about the economy. McCain, on the other hand, went off topic to talk about government operations and defense, used a greater proportion of political frames (though, interestingly, not a greater proportion of safety frames), and spent more of his time attacking Obama personally. In doing so, it seems that McCain, as the insurgent candidate in 2008, was attempting to draw attention away from the economy and toward two topics that issue ownership put at his advantage. In this way, McCain helps to illustrate the disadvantage of insurgent candidates: picking the right noneconomic issue on which to define the election, while still giving obligatory treatment to the economy when absolutely necessary, namely, in response to point-blank questions in a face-to-face debate.
The wisdom of the choices each candidate made in 2008 could be debated; what we have emphasized here is that they used the agenda-control behaviors predicted (and prescribed) by the rhetorical engagement literature (e.g., Schattschenider, Riker, & Vavreck) in the tricky context of the debates. Obama, however, was arguably more successful in using these agenda-control behaviors. Moreover, it appears that these behaviors had real consequences. Both the news media and the public appear to have been influenced by the debates. And although speculating on causal relationships between debates and election outcomes raises more questions that it resolves in this case, it appears that Obama’s focus on the economy may have heightened the edge in terms of public opinion, specifically surrounding the economy.
Further research on additional debates is necessary to determine whether and to what extent these agenda-control behaviors might have electoral consequences, but the evidence in this case is suggestive. What we do know for sure is that the candidates’ debate behaviors in 2008 pointed to tried-and-true political strategies of conflict displacement, issue ownership, and awareness of the economic climate. These findings indicate that even in the most controlled setting, candidates engage each other by engaging specific agenda-control strategies.
Footnotes
Appendix
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.