Issue ownership trends and tensions in 2008: Obama, the transformative Democrat?
ISSUE OWNERSHIP THEORY
The theory of issue ownership, as it applies to presidential campaigns, was initially developed by Petrocik's (1996) content analysis of the 1980 election and has been tested in a variety of locations, such as campaign advertisements and media reports (Benoit & Hansen, 2002; Benoit & Stein, 2005; Benoit, 2004, 2007; Damore, 2004; Hayes, 2008; Petrocik, Benoit, & Hansen, 2003; Petrocik, 1996), primary debates (Benoit & Hansen, 2002, 2004a; Benoit, 2004), nomination acceptance speeches (Benoit, 2004; Petrocik et al., 2003), and general election debates (Benoit & Hansen, 2002, 2004b; Benoit, 2004). This study continues issue ownership scholarship through its application to the nomination acceptance speeches and general debates in the 2008 presidential election, thereby extending the conversation into a new time frame and political context but maintaining the textual stage set by past research.
Issue ownership theory posits that Democrats and Republicans "own" certain issues (Petrocik, 1996). According to this theory, voters perceive that the two main political parties have a certain set of issues that they handle well (Benoit & Hansen, 2004a). The issue sets are based on the strength of handling certain issues over time, is reinforced by politician ideology, and is relatively stable (Benoit & Hansen, 2004a). As an election strategy, this theory states that candidates emphasize their owned strengths and make the argument that handling those issues will be better suited to resolve the problems within the current political environment (Petrocik, 1996). One goal of issue ownership as a strategy is to set the issue agenda for citizens and the media so that it favors the candidate's owned issues (Petrocik et al., 2003). It should be noted that, in a strategic context, issue ownership is helpful at arguing for swing voters and also for targeting the base of each particular party, because candidates talk about the issues that they and their party are deemed better at handling (Petrocik, 1996). Petrocik (1996) asserts that Democrats are perceived as owning and resolving the environment and social welfare issues best, such as: health care, education, poverty, and the elderly. On the other hand, the public perceives Republicans as owning foreign policy, defense, the economy, and issues related to government spending, inflation, and taxation (Petrocik, 1996).
Petrocik's (1996) case study of the 1980 election shows that candidates and political parties own certain issues and that ownership of issues is built on reputation over time. This dedication to certain issues has further developed the public's perception of each party's, and by association each candidate's, competency in handling particular issues. These expectations allow for a sense of comfortable predictability, so much so, that some people cast their vote in elections based on the party's issue focus regardless of the candidate (Jackson, 1975). In short, issue ownership provides a heuristic device to voters, which allows them to make decisions based on party lines without knowing a candidates' exact stance on issues. However, such a heuristic is only valid in the case that issue ownership is relatively stable.
Issue ownership is based on the social cleavage theory of political parties, which conceives of political parties as demographically and socially distinct groups with competing political needs and agendas (e.g. Lipset & Rokkan, 1967). Following this theoretical perspective, political parties naturally focus on specific issues that are important to the groups they represent. After time, the public views each political party as handling certain issues well which leads to the perception of issue ownership. Each party's long-term reputation, as well as recent performance concerning specific issues, remains prominent in voters' minds when they go to the polls (Petrocik, 1996). A party's recent performance on an issue affects issue ownership if the party has recently had little success with a particular issue and are often owned by challenging candidates (Petrocik, 1996).
Past scholarship has indicated that there is inter-election variability of issue ownership findings (Petrocik et al., 2003). When an issue does change from one to the other party, this switch tends to represent a lease on that particular issue rather than a permanent ownership change. Such temporary ownership changes usually revert back to the original party in the next election (Petrocik, 1996). We argue that this variability is more than minute differences in individual elections, but instead embodies a trend that forces scholars to reconsider the assumptions of issue ownership theory and reexamine the issues owned by each political party. Further, we argue that changing campaign strategies, namely the microtargeting of swing voters, influences this change.
HISTORICAL CONTEXT OF THE 2008 ELECTION
The 2008 election was unlike any other. But, was this election also different due to the ownership of issues by the candidates and their parties? This project seeks to analyze this question. According to issue ownership theory, it should have been predictable that the Republican candidate, Senator John McCain, a prisoner of war (POW) who worked closely with the Bush Administration, would engage the issue of national security more than any other issue. This is because the Republican Party tends to invest in this issue, especially since the events of September 11, 2001. Additionally, Campbell and Chollet (2007) stated that "perhaps more than any presidential contest, since 1980 or even as far back as 1968, 2008 will be a national security election" (p. 192). Given the circumstances, the Republican candidate should also be expected to engage in Republican owned issues more than the Democratic candidate because the Democratic candidate will have to own Democrat issues to appeal to the Democratic Party base. Additionally, issue ownership theory predicts that the Democratic candidate, Barack Obama, would engage Democrat issues as well as Republican issues.
However, in 2008 the political environment changed with the economic collapse. The 2008 polling data suggested inconsistency between actual issue ownership trends in 2008 and Campbell and Chollet's (2007) prediction that the 2008 election would be a national security election. During the general election of 2008 the economy was the most important issue to voters (Rasmussen Polling Report, 2007-2008, August-October; Saad, 2008; Washington Post-ABC News, 2008), which suggested that there was a major shift away from the national security political environment of post 9/11 elections toward an economic issue political environment. Adding to the uniqueness of the 2008 campaign, polling also showed that voters perceived Democrats to better handle some of the typically Republican owned issues (Rasmussen Polling Report, 2007-2008, August-October; Washington Post-ABC News, 2008). Research on this shift suggests that the national security issue may have switched to a Democrat owned issue in 2008 because of how President Bush handled the Iraq War (Goble & Holm, 2008). These historic factors, along with the beginning of a shift in campaign strategy due to microtargetting, position the 2008 election as an excellent site for exploring general partisan trends and the assumptions of issue ownership theory by testing the theory in a rather novel political environment.
ISSUE OWNERSHIP RESEARCH
The most common finding in the issue ownership literature is that Democrats addressed Democratic issues more than Republicans and that Republicans addressed GOP issues more than Democrats (Benoit & Hansen, 2004a; Petrocik et al., 2003). This is one of the most consistent findings within the issue ownership literature. Therefore we expect that such a relationship is likely to occur within the context of the 2008 election:
Hypothesis 1: The Democratic candidate engages Democratic issues more than the Republican candidate and the Republican candidate engages Republican issues more than the Democratic candidate.
One of the more interesting findings within the issue ownership literature is the tendency for Democrats to engage in more Republican owned issues during the general election than GOP party members engage in Democrat issues (Benoit & Hansen, 2004a; Petrocik et al., 2003). Petrocik, Benoit, and Hansen (2003) explain this result:
Democratic issues are consistently, prominent in American politics, but they are relatively few in number. As a result, most of the problems that a President is expected to handle are GOP issue-handling specialties. Democrats' campaigns emphasize GOP issues more than they might prefer because they cannot ignore taxation, government spending levels, and national security (p. 609-610).
Immediately after the primary stage of the campaign, in the nomination acceptance speeches, Democrats begin addressing more Republican issues (Petrocik et al., 2003). Benoit and Hansen (2004a) found that issues raised by Democratic candidates shifted to traditionally Republican owned issues during the general election debates compared to the primary debates. Since our study examines nomination acceptance speeches and general election debates, both non-primary events, we expect this relationship will be present again in the 2008 election:
Hypothesis 2: The Democratic candidate discusses more Republican issues than the Republican candidate discusses Democratic issues.
Further, Petrocik (1996) identified performance issues that tend to be leased by candidates from the challenging party. While there is some evidence to suggest that national security was shifting into a performance issue based on Bush's handling of the Iraq War (Goble & Holm, 2008), the most prominent performance issue was the economy given the 2008 economic recession. Further, the economy is often a performance issue because of its volatility over time (Petrocik, 1996). Therefore we predict:
Hypothesis 3: The candidate from the challenging party, the Democratic candidate, discusses the performance issue, or those issues related to the economy, more than the Republican candidate.
Research also shows that issue ownership communication has a strong and interesting relationship with public opinion polling. First, Petrocik and colleagues (2003) showed that prior to the 2000 election the issue rhetoric of the candidates shaped the issue concerns of the electorate. This means that in many cases the issue agenda of the public is set by candidate rhetoric, showing that issue ownership can be a good strategy to raise the prominence of owned issues within the electorate. It is likely that in the 2008 election issue rhetoric during the convention speeches and debates will influence public polling: Hypothesis 4: Candidate issue rhetoric will influence what issues the public recognizes as important through public opinion polling.
Ultimately, it is the candidates' goal to attract not only their core constituency, but also to garner votes from people outside their party. Candidates' campaigns are turning to microtargeting to determine how to best persuade voter segments to vote for them (Levy, 2008). Microtargeting allows campaigns to determine what issues need to be owned in order to best persuade those voter segments. Issue ownership provides campaigns a tool to demonstrate competence handling the issues of most importance to critical constituencies. Hacker and Pierson (2005) reiterate this point by noting that political norms assume our political leaders obey the results of surveys and focus groups, and that our electoral structure encourages major political parties to compete for the center (i.e. swing voters). This suggests that polling and public opinion will also influence the issue rhetoric of the candidates:
Hypothesis 5: Candidate issue rhetoric will be influenced by what issues the public recognizes as important through public opinion polling.
In addition to the significance of presidential campaign communication on the electorate's concerns, the issues of concern to citizens also influence President elections:, "Just as the issue content of the campaign appears to predict the issue agenda of the electorate [...], the issue agenda of the electorate predicts the vote" (Petrocik et al., 2003, p. 617). Petrocik, therefore, suggests that the issue agenda of voters combined with the issue rhetoric of candidates is directly related to which candidate wins the election. Therefore, we predict that whichever candidate owns the issues of most importance to the public in the 2008 election will be the candidate that wins the election:
Hypothesis 6: The candidate who engages the issue that is most important to the public wins the 2008 election.
Prior research has examined public opinion and campaign rhetoric to evaluate issue ownership assumptions (Benoit & Hansen, 2004a; Petrocik et al, 2003; Petrocik, 1996), but did these assumptions hold true in the 2008 general presidential election? What this study seeks to understand from the 2008 election is: Did the candidates' rhetoric parallel the assumptions of issue ownership theory? In order to test issue ownership in the 2008 election, an examination of public polling and a content analysis on selected presidential campaign communication events, specifically the nomination acceptance speeches and the three presidential debates, were performed.
Nomination acceptance speeches and debates from the 2008 general election were major events that typified the presidential campaign rhetoric, and are therefore the best sites to test issue ownership theory. Nomination acceptance speeches, in particular, are helpful in learning "how the candidate relates to the party's tradition and identity" (Jamieson & Birdsell, 1988, p. 200). In addition, nomination acceptance speeches are presented at party conventions and lay out the platforms from which the candidates will position themselves throughout the general election (Trent & Friedenberg, 2007). Every word in the nomination acceptance speech is purposeful and strategically planned out. Therefore, the candidates are committed to the words in their nomination acceptance speech, and these speeches are excellent sources of information regarding the candidates and how they plan to position themselves within the issues.
Debates are also important sites to test issue ownership theory because they offer the candidates the opportunity to engage the issues and each other on the issues. One of the major reasons debates are set up, in addition to facilitating the democratic process, are so the electorate can get a sense of the candidates and where they stand on the issues in comparison to their competitor. Debates are structured events with questions dealing with certain issues, but there is room within each question so that candidates can emphasize different issues. Benoit and Hansen (2004a) elaborate on this idea when they state that, "Issue ownership is not the only factor which influences statements in debates. . . however, if there are differences in the emphasis on issues between candidates from different parties, those differences should reflect issue ownership" (p. 146). Moreover, debates are the focal points of campaigns and offer a summary of the major arguments for each candidate's election (Carlin, 1992).
The general election debates and nomination acceptance speeches in 2008 had record-breaking television audiences (Wenger & MacManus, 2009); therefore, these are excellent events to examine the relationships between public opinion polling and candidate rhetoric in a macro manner. This will allow us to determine if reciprocal relationship between issue ownership by the candidates and the issue agenda of the electorate at large exists or if a different process is occurring. Consequently, in order to test issue ownership trends in the 2008 general election, it is imperative that nomination acceptance speeches and debates be the sites for inquiry.
Petrocik and colleagues (2003) argue that a content analysis of presidential campaign rhetoric allows for the measurement of issues raised by the candidates; and systematic content analysis studies continue to be supported as a valid method to study politicians' speeches and debates (Neuendorf, 2002). Also, content analysis has been used on similar studies in the past that have tested issue ownership in presidential nomination acceptance speeches and debates (Benoit & Hansen, 2004a; Petrocik et al., 2003; Petrocik, 1996). Finally, McKinney (2005) suggests, "a count of the total words devoted to discussion of each issue provides a more accurate measure of each debate's issue agenda," compared to the questions asked of candidates in the debate (p. 206). This type of methodological approach is especially helpful because of its applicability to the different formats being considered (McKinney, 2005). Therefore, a concordance computer program that utilized the keyword in context method produced the analysis of this content by counting issue words uttered by the candidates.
The units of analysis were presidential campaign rhetoric from the two main Presidential candidates in the 2008 general election: Senator Barack Obama and Senator John McCain. Content analyses were performed on five data sets: the 2008 Democratic and Republican nomination acceptance speeches and the three televised presidential debates. The specific unit of data was the issue mentions by both presidential candidates. The study focused on perceived owned issues, and therefore the content analysis examined words and utterances that were relevant to the perceived owned issues. A total of eleven issues were considered. The Democratic issues considered were education, environment, elderly, health care, and poverty. The Republican issues considered were national defense, foreign policy, government spending/deficit (shortened to "government spending"), taxes, and crime/illegal drugs (shortened to "crime"). Mentions of the economy/inflation/jobs (shortened to "economy") were classified as a performance issue.
In order to conduct the content analysis, a wordlist was needed with a set of words that referenced each issue. To ensure validity of this study, we used a wordlist taken from a prior content analysis study testing issue ownership assumptions (Benoit & Hansen, 2004a; Petrocik et al., 2003). The wordlist was slightly modified by unpacking words that were concatenated together into two separate words and by adding new words that were pertinent given the evolving political context (e.g., adding Al-Qaeda, Bin Laden, and climate change). Both the original, and the modified, wordlist can be obtained from the first author upon request.
The candidate utterances were divided up into separate text files from transcripts of each debate and nomination acceptance speech. These files were then searched for each keyword by the concordance software to find out what issues the candidates referenced. The researchers recorded the utterance frequencies of each keyword. The keyword in context method utilized within the concordance program allowed the researchers to assure that each keyword was utilized within the correct issue type context. For example, the word "class" can refer to an education issue or when combined with the word "middle", an economic issue. The surrounding words were displayed with each word and the researchers together came to an agreement on what issue was being referenced in cases that were unclear.
In his nomination acceptance speech, Obama used 4,834 words, of which, 207 were issue related words. McCain used 3,976 words in his nomination acceptance speech, of which 167 were issue related words. In the first debate event, Obama used 7,799 words, including 426 issue words, and McCain used 7,305 words, including 355 issue words. The second debate contained 7,106 words stated by Obama, including 370 issue words, and 6,548 words stated by McCain, including 295 issue words. Finally, in the third debate, Obama used 7,465 words, including 326 issue words, while McCain used 6,662 words, including 267 issue words.
Public polling is taken from several locations and used throughout the analysis to compare issue mentions with the issues that were important to the public at the time. Published press releases from polling organizations are cited where they were used. The 2008 panel data from the National Annenberg Election Survey (NAES) telephone surveys were used to test certain hypotheses as well (The Annenberg Public Policy Center, 2010). The NAES survey randomly selected participants and conducted phone interviews every day starting from December 2007 leading up to Election Day. This enabled us to locate particular issues deemed most important to voters surveyed on those days. The specific question of interest was the verbatim answer to the following question: "In your opinion, what is the most important problem facing the country today?" (Item number: CA01rlv; The Annenberg Public Policy Center, 2010). The answers to this question are verbatim responses to that question resulting in many different permutations of the same issue (e.g., "economy" is different from "the economy" in the NAES data). For this analysis only the top two permutations were combined because the two top issues in each case were "economy" and "the economy."
The first hypothesis stated that the Democratic candidate would engage Democratic issues more than the Republican candidate, and that the Republican would engage Republican issues more than the Democrat. The results from all the data sets show that Obama mentioned Democratic issues 269 times (20.2% of his total issue mentions), while McCain mentioned Democratic issues 172 times (15.9% of his total issue mentions). This difference was significant, z = 2.77, p < .01, and indicates that the Democrat candidate mentioned more Democratic issues than the Republican candidate. Also, Obama mentioned Republican issues 736 times (55.4% of the issues he mentioned), while McCain mentioned Republican issues 686 times (63.3% of the issues he mentioned). This difference was also significant, z = --3.93, p < .001, indicating that the Republican candidate mentioned Republican issues more frequently than the Democrat candidate. Hypothesis one was supported by the data as both Obama and McCain referenced issues owned by their own party more than the other candidate. Additionally, these results support previous research reporting that during the general election the Democratic candidate tends to move to the right and mention more Republican issues. The number of total issue mentions in each category can be found in Table 1.
Interestingly, the predicted relationship was observed in the different communication events overall, but differences observed in the nomination acceptance speeches were not statistically significant whereas the differences were significant in the debates. During the nomination acceptance speeches, Obama mentioned Democrat issues 55 times (26.4% of his issue mentions in the acceptance speech) and McCain mentioned Democrat issues 34 times (20.4% of his issue mentions in the acceptance speech), which was not a significant difference, z = 1.38, p > .05. Also, in the nomination acceptance speeches, Obama mentioned Republican issues 89 times (43.0% of his issue mentions in the acceptance speech) whereas McCain mentioned Republican issues 76 times (45.5% of his issue mentions the acceptance speech), but this was not a significant difference, z = -.49, p > .05. The total issue mentions in each category from the nomination acceptance speeches can be found in Table 2.
In the debates, Obama mentioned Democrat issues 214 times (19.1% of his issue mentions in the debates) and McCain mentioned Democrat issues 138 times (15.0% of his issue mentions in the debates), which was a significant difference, z = 2.39, p < .05. Obama mentioned 647 Republican issues (57.7% of his issue mentions) and McCain mentioned 609 Republican issues (66.4% of his issue mentions), which was a significant difference, z = -4.04, p < .001. The total issue mentions across debates can be found in Table 3.
The second hypothesis predicted that, in the 2008 election, the Democratic candidate would discuss Republican issues more than the Republican candidate would discuss Democratic issues. Overall, Obama mentioned significantly more Republican issues 736 times (55.4% of the issues mentioned) than McCain mentioned Democratic issues 172 times (15.9% of the issues mentioned), z = 19.93, p < .001. This data supports the second hypothesis.
This relationship held across the different communication events. During the nomination acceptance speeches, Obama mentioned Republican issues (n= 89, 43.0% of the issues) more often than McCain mentioned Democratic issues (n = 34, 20.4% of the issues), which was a significant difference, z = 4.63, p < .001. In the debates, Obama more frequently mentioned Republican issues (n = 647 times, 57.7% of the issues) than McCain mentioned Democratic issues (n = 138, 15.0% of the issues), z = 19.67, p < .001.
The fourth hypothesis stated that candidate communication in convention acceptance speeches and debates would influence which issues the public found to be important, and the fifth hypothesis stated that candidate communication in these events would be influenced by what issues the public found to be important. These hypotheses predict that the nomination acceptance speeches, which occurred on August 28, 2008 for the Democratic candidate and on September 4, 2008 for the Republican candidate, along with the three presidential debates, which occurred on September 26, October 7, and October 15, 2008, both influence, and are influenced by, the issues that the public finds most important in public opinion polls. To test these hypotheses, telephone survey panel data from the National Annenberg Election Survey (NAES) from the two days prior to, and two days following, each event were used to find the issue of most importance to the population (The Annenberg Public Policy Center, 2010). The most mentioned issue on all dates was "economy" followed by "the economy" (recall that in the NAES data used in this project verbatim responses fall into different categories if they are phrased or spelled a different way, even if they mention the same issue), and Table 4 shows the proportion of voters who indicated that the economy in either of these permutations was the most important issue (The Annenberg Public Policy Center, 2010). On most dates "economy" was mentioned significantly more than "the economy" (The Annenberg Public Policy Center, 2010), implying that the economy was, by far, the most important single issue to the citizens.
In terms of the fourth hypothesis, if the issue rhetoric in the speeches influenced the issue agenda of the population then the economy issue should have become more important to the public as indicated by the frequency of mentions in the polls following the speeches. Looking at the two most important issue mentions together, given they both deal with the economy, there was a significant increase in mentions of the economy issue in the NAES poll after Obama's nomination acceptance speech, z = -2.49, p < .05. A similar trend emerged for McCain's nomination acceptance speech which indicated a slight increase in the proportion of mentions of the economy in the NAES polling data, z = -.01, p > .05. This suggests that Obama's nomination acceptance speech may have influenced the issue agenda of the public, but not McCain's nomination acceptance speech, lending partial support for hypothesis four.
Pertaining to hypothesis 5, as indicated in Table 4, the economy was the most important issue to the population both before and after both nomination acceptance speeches (The Annenberg Public Policy Center, 2010). In his nomination acceptance speech, Obama mentioned the economy more frequently than any other issue (n = 63). However, Obama's economy issue mention frequency was not significantly different from the next highest mentioned issue, national defense (n = 48), z = 1.66, p > .05. In McCain's nomination acceptance speech, the most mentioned issue was also the economy (n = 57), which was significantly different from the next highest mentioned issue, national defense (n = 40), z = 2.05, p < .05. If the issue rhetoric in the speeches were responsive to public opinion then it would be expected that the economy should have been the most emphasized issue by a significant margin, which it was in one case, though only marginally. These findings partially supports hypothesis five.
In the first debate, the issue with the most mentions by both candidates was national defense (353 combined mentions) and that was significantly greater than the next highest mentioned issue, foreign policy (138 combined mentions), z = 10.35, p < .001. As previously mentioned both prior to, and after, the debate the most important issue to the public was the economy (The Annenberg Public Policy Center, 2010). After the debate, there was a significant increase in mentions of the economy in the NAES polls, z = -1.61, p < .05. This data does not support hypothesis four or five. However, the topic of the first debate was national security and foreign policy, so the opportunity for discussing the economy was limited, making this debate less relevant to hypothesis testing. Interestingly, outside of defense and foreign policy, the economy was the most mentioned topic (18.2%) followed by taxes (6.2%), a topic related to the economy. Together, these topics made up a combined 24% of the mentions in a debate that was supposed to focus on national defense. These numbers seem to indicate the pervasive influence of the economy as a talking point in the election.
The issue mentioned the most in the second debate by both candidates was also national defense (206 combined mentions) which was mentioned significantly more often than the economy (153 combined mentions, the next most often mentioned issue), z = 2.93, p < .01. Again, the most important issue to the public was the economy both before and after the debate (The Annenberg Public Policy Center, 2010). Mentions of the economy on the poll slightly decreased following the debate, z = .19, p > .05. Although these results do not support hypothesis four or five, the format of the debate strongly influences the topics candidates discuss. In this case, the questions were submitted by undecided voters in a town hall style debate and were about equally divided between foreign policy and the economy.
The topic of the third debate was domestic issues and the economy which was the most mentioned issue in this debate (163 combined mentions) and was mentioned significantly more than the next most mentioned issue, education (101 combined mentions), z = 3.95, p < .001. As with previous debates, the most important issue to the public was the economy both prior to and after the debate (The Annenberg Public Policy Center, 2010). Since the economy was the most important issue prior to the debate and the most discussed issue during the debate, this data provides support for hypothesis five. Following the debate, there was a non-significant increase in mentions of the economy in the polls, z = -.68, p > .05. The data do not support hypothesis four.
Hypothesis six predicted that the candidate who more frequently discussed the most important issue to voters would ultimately win the election. Because Obama won the election and because the economy was the most important issue to voters, this hypothesis predicts that Obama mentioned the economy significantly more than McCain. Also, hypothesis three predicts that the candidate from the Democrat party, Obama, would engage the issue of the economy more than the Republican candidate. Overall, Obama mentioned the economy 324 times (24.4% of his total issue mentions) and McCain mentioned the economy 226 times (20.8% of his total issue mentions). This difference was significant, z = 2.06, p < .05. Therefore, hypotheses three and six are supported by the data.
An examination of the events individually shows that, although Obama mentioned the economy more frequently than McCain overall, these differences are not consistent across events. In the nomination acceptance speeches, Obama mentioned the economy 63 times (30.4% of his total issue mentions) and McCain mentioned the economy 57 times (34.1% of his total issue mentions in that event). This difference was not significant, z = -.76, p = .45. During the debates, Obama mentioned the economy 261 times (23.3% of his issue mentions) and McCain mentioned the economy 169 times (18.4% of his issue). This difference was significant, z = 2.67, p < .01. The data imply that hypotheses three and six were supported during the debates but were not supported during the nomination acceptance speeches.
The results from this study reveal several interesting factors regarding issue mentions during the 2008 election. First, overall, the Democratic candidate engaged more Democratic issues than the Republican candidate and the Republican engaged more Republican issues than the Democratic candidate. Second, the Democratic candidate mentioned Republican issues more than the Republican candidate mentioned Democratic issues. Third, the Democratic candidate mentioned the economy more than the Republican candidate. Also, the results suggest the relationship between public polling and issue rhetoric by candidates. We cannot conclude that there was a significant agenda setting effect by the candidates onto the issue agenda of the public, nor can we conclude that there was a significant relationship between the issue of most importance to voters and the issue rhetoric of politicians, though we have some limited evidence of a relationship between these two groups. Further, we can suggest a relationship between the candidate's discussion of the issues and the public's issue agenda since the economy was the most important issue to the public and was a highly mentioned issue by the candidates. Finally, Obama mentioned the economy, the most important issue to the public, more than McCain and won the election.
Overall, the data show that both candidates engaged Republican issues more than Democratic issues, which is typical of general election presidential communication because "most of the problems a President is expected to handle are GOP issue-handling specialties" (Petrocik et al., 2003, p. 609). However, it is important to note the variations in the 2008 results compared with previous studies (Benoit & Hansen, 2004a; Petrocik et al., 2003; Petrocik, 1996). Specifically, the frequency of Republican mentions in 2008 by both candidates, especially the Democratic candidate, was much higher than past averages.
Both the Democratic and Republican candidates mentioned Democratic issues less and Republican issues more in 2008 than they ever had in the examined election cycles. Considering only frequency, the Democratic candidate mentioned more Republican issues than the Republican candidate. Past research on issue ownership during the nomination acceptance speeches show that Democratic candidates tend to focus on issues owned by their party in the nomination acceptance speeches and then pivot towards more Republican issues in the general election (Petrocik et al., 2003). However, in the case of 2008, the Democratic candidate focused extensively on Republican owned issues during his nomination acceptance speech. This is important, because, although data indicate that general issue ownership trends were present in 2008, these differences demonstrate a shift in which issues were mentioned most by each candidate toward traditionally Republican owned issues, especially by the Democratic candidate. This result indicates a potential shift from the original assumptions of issue ownership theory as applied to nomination acceptance speeches.
In the general election debates the candidates engaged the issues they owned more than the other candidate. The results from the three debates are similar to and different from the results examining general presidential debates from 1960 through 2000 (Benoit & Hansen, 2004a). Proportionally, in 2008, Obama engaged Republican issues significantly more than Democratic candidates in previous elections, McCain uncharacteristically mentioned Republican issues slightly less often than Republican candidates in past debates. Again, these differences are important and illustrate tensions between issue ownership assumptions and issue ownership as it applies to the 2008 election. This finding aligns with research that indicates issues owned by each party are in flux (e.g. Goble & Holm, 2008).
Further, in the 2008 general debates, Obama and McCain discussed Republican issues more than Democratic issues. For example, the focus of the first presidential debate was foreign policy and national security, which is reflected in the issue frequency data. For both candidates, the most mentioned issues in this debate were traditionally "owned" by Republicans. Obama and McCain both mentioned national defense and foreign policy the most with the economy coming in a close third. During the second debate, which was a town hall style format mixing domestic and foreign policy issues, both Obama and McCain mentioned traditional Republican owned issues such as national defense most often: although the economy was featured prominently. Finally, the focus of the third debate was domestic policy, and the most mentioned issue by both candidates was the economy.
Overall, the data from the debates show that national security and the economy were major rhetorical foci of the candidates. The results from the nomination acceptance speeches mirror the results from the debates, showing that in these speeches, the candidates mentioned national defense and economic issues the most. As referenced in Table 2, Obama and McCain both mentioned national defense and the economy more often than any other issue. Although the proportions may indicate that Obama mentioned these issues less often, Obama mentioned these issues more often than McCain in terms of raw frequencies. Petrocik's (1996) issue ownership theory suggests that national defense is a Republican issue and that the ownership of the economy as an issue depends on a candidate's perceived performance or competence, which in most years has favored the Republican candidate. The decrease in Democratic issue mentions and increased references to Republican issues by the Democratic candidate again indicate that a shift in what issues are owned by each party may be occurring.
Also, the issue of the economy was highly mentioned by both of the candidates. Petrocik (1996) indicates that challenger candidates usually own performance issues when a sitting candidate fails to perform well on that issue. However, neither candidate was an incumbent, so it is interesting to observe both candidates vying for ownership of that issue. Although Obama ultimately mentioned the economy more, McCain also mentioned the economy quite frequently This was likely a feature of the current political situation unfolding with the collapse of the economy in 2008 and the shift in public polling indicating that the economy was the most important issue to voters (e.g. Rasmussen Polling Report, 2007-2008, August-October; Saad, 2008; Washington Post-ABC News, 2008).
The importance of the economy as an ownership issue is reflected in Table 5, which indicates a clear shift in respondents' perceived handling (and therefore ownership) of issues in 2008 compared with results reported in Petrocik's (1996) original study. The 2008 election also depicts a remarkable shift in ownership of national defense, an issue regularly perceived to be better handled by Republicans in the past, was perceived to be better handled by Democrats in 2008 (Rasmussen Polling Report, 2007-2008, August-October). Given that one candidate owned both of the major issues in the election, it should be no surprise that Obama won the election decisively. The economy and defense were so important in the 2008 election that issues such as the elderly and the environment, which are typically perceived to be owned by the Democrats, were not important enough to receive enough mentions to make the issue list.
Overall, the data results are noteworthy variations from Petrocik's (1996) original study. Variations to issue ownership between parties have been shown to occur (Petrocik et al., 2003), but the findings in this study are relatively novel. Whether this is a variation or a shift in owned issues between parties deserves further explanation. Future studies into issue ownership should examine this trend and perhaps consider that owned issues are more fluid than previously thought. This is evidenced in Table 5, suggesting that the public's perception, as indicated by polling, of party issue handling was different in 2008. As presented by Petrocik (1996) typically Republican owned issues, such as taxes and national defense, were perceived to be better handled by Democrats in 2008. Therefore, three typically Republican "owned" issues became Democratic "owned" issues in 2008. Meanwhile, issues that are typically perceived to be owned by Democrats continued to be perceived as Democratic issues in 2008.
Figure 1 gives a possible explanation for the shift to the right. The top five issues of importance in the Gallup poll are issues that should be Republican owned issues: the economy, federal budget deficit, terrorism, energy/gas prices, and the situation in Iraq (Petrocik, 1996; Saad, 2008). In fact, out of all of the issues listed in the Gallup poll (Saad, 2008), there are only three issues mentioned that are typically perceived as owned by the Democrats: health care, education, and the environment (Petrocik, 1996). This is a move to the right of center regarding issues of importance to the public. Therefore, it was strategic for the candidates to rhetorically move to the right of center so as to address the needs and issues of importance to the people. Perhaps this is a short-term variation that will revert in future election cycles, which would indicate that the perceived issue handling competencies were based on recent past handling of issues. This would imply that Obama may have "leased" these issues based on the handling of those issues by President Bush. However, if Democratic candidates and elected officials continue to emphasize and handle these issues well, ownership of these issues may switch. Future studies should examine how the public views the parties at handling each issue, and investigate whether these findings represent a more long-term trend. Specifically, the results of this study imply that future studies may find that Democratic candidates are increasingly engaging in Republican issues, and perhaps securing ownership on certain issues.
This study found that in most cases, the issue rhetoric of the candidates did not influence and was not influenced by public polling. The only significant findings were that Obama's nomination acceptance speech seemed to reflect public concern indicated in polls and/or had an agenda setting effect on the importance of the economy to the public. In addition, it appears McCain's nomination acceptance speech may have been responsive to the public's issue agenda. The finding that the third debate mirrored the public's concern about the economy should be interpreted carefully because the third debate was solely constrained to domestic issues and likely contributed the frequency with which the economy was mentioned issue in that debate. Overall, the data do not entirely support the hypotheses that both the issue agenda of the public affects the issue rhetoric of politicians and that the issue rhetoric of the politicians influences the issue agenda of the public. However, we argue that even though these hypotheses are not supported, the results suggest that there is a reciprocal relationship between the candidate issue rhetoric and public opinion polling given the very high number of mentions of the economy in events that have constraints on the topic of discussion.
Therefore, the argument that the Democratic candidate moved his issue rhetoric to the right challenges the results from this study. However, the relationships between public opinion and issue rhetoric tested by this study were micro in temporal scope, but macro due to national sampling. It could be that more temporal macro data, which examine the most important issue over the entire election cycle, would have yielded significant results. This temporal macro relationship between polling results and the candidate's rhetorical attention to typically Republican owned issues is also exemplified in Table 5, which compares perceived owned issues in Petrocik's (1996) study with perceived owned issues in 2008. The polling suggests that the public supported the Democrat's ability to handle typically Republican owned issues. This may have encouraged Obama to engage and challenge the Republican candidate on these issues. The potential temporal macro relationship between public opinion and campaign issue rhetoric is one that should be tested by future research.
Further micro sampling of swing voters, rather than our macro national sample, may have also yielded significant results. The campaign communications examined in this study drew large record-breaking audiences (Wenger & MacManus, 2009). The campaigns could have been microtargeting swing voters watching those events and mentioning issues that were important to those voters. Our finding that the macro sample issue agenda was not a significant influence on the issue's owned of the candidates could be interpreted as an indication that microtargeting is occurring. The potential microtargeting of swing voters using issue ownership warrants further examination.
LIMITATIONS AND FUTURE RESEARCH
A major limitation of this study is the absence exploring framing (i.e. how different issues are explained by the candidates). How the issues were discussed in the communication events could influence the salience of these issues in the minds of the electorate. Future research should examine how the issues are framed and probe for differences between the two parties.
A second potential limitation is the wordlist used for content analysis. Although we updated the wordlist by including more words, words that may be considered out of date were not removed. Additionally, the wordlists were disproportionately skewed towards having more words and issues associated with Republicans. Future research should employ survey research to refine and improve the wordlist.
The relationships tested between the issue rhetoric of candidates and the public opinion polling is also limited. We cannot make any strict causal claims that the issue rhetoric of the candidates influenced public opinion or that public opinion directly influenced candidate issue rhetoric. Our data is limited in this way and does not control for any outside influences on the issue agenda of the people or if the people sampled even watched the events considered. Future research should experimentally assess the relationship between issue ownership and issue agendas in a more controlled setting.
This study determined that Obama and McCain addressed the issues of most concern to voters: the economy and national defense. Although the candidates may not have engaged the issue of most importance, the economy, more than any other issue overall, they did, however, rhetorically engage the issues of concern to voters more than other issues. Most notable was their attention to the economy in the third debate. However, this relationship may be because the third debate was a domestic issue debate rather than a strong indicator that the candidates were responding to voters needs. It could also be the case that engagement in the issue topics of concern to voters by the candidates was depressed in the other debates because of the topic focus of those events. We also argue that public opinion perceives traditionally Republican owned issues as more important than Democratic issues and that issue mentions parallel this move. The choice by both candidates to focus more attention on Petrocik's (1996) assumed Republican issues, including national defense, was strategic. For example, had Obama focused on traditional Democratic owned issues, such as the environment, then he very likely would have lost the election. Obama clearly understood his advantages and used them strategically against his opponent.
McCain did engage Republican owned issues as well as the issues of importance to the voters; however, he still lost. Therefore, it may have been more prudent for him to focus on convincing voters of his advantages over Obama by exploiting Obama's weaknesses. Perhaps more importantly, McCain would have been well advised to focus his message on the economy to a greater extent since this was, the issue of most importance to voters. In October 2008, typically Republican owned issues were perceived as better handled by the Democrats which proved extremely problematic for McCain (Rasmussen Polling Report, 2007-2008, August-October).
Overall, since Republican issues were the issues both owned by the Democrats and of most importance to voters in 2008, the candidates needed to shift to the left to align with voter preferences. The polling from 2008 suggested a political environment in which traditional Republican issues were perceived to be owned by Obama. Although the issues of high concern to voters, including national defense, were once Republican issues, McCain's loss in the election suggests that past issue ownership assumptions did not hold true for the 2008 general election. McCain needed to do more to prove to the voters that he owned traditional Republican issues. Instead, the public was convinced that the Obama held the most palatable position on national defense issues.
In conclusion, more research should investigate issue ownership to better explain the relationship between public opinion and campaign issue rhetoric; however, for now the results show that neither the public nor the candidates completely control the issues of the election. In particular, there is not an indicator of public opinion and campaign issue rhetoric influencing, as opposed to dictating, one another on a micro level., However, there is some anecdotal evidence that a macro temporal relationship or micro geographical relationship exists. These results suggest that candidates should engage in the issues of most importance to the voters, since that predicts electoral success. Additionally, it is the relationship between presidential campaign rhetoric and public opinion that gives insight into the political environment within which both the candidates and the public are situated. This relationship, if studied more, can assist in understanding how to run a successful campaign that demonstrates understanding of the electorate's needs and perhaps insight into better serving those needs.
Benoit, W. L. (2004). Political party affiliation and presidential campaign discourse. Communication Quarterly, 52(2), 81-97. doi: 10.1080/01463370409370183
Benoit, W. L. (2007). Own party issue ownership emphasis in presidential television spots. Communication Reports, 20(1), 42-50. doi: 10.1080/08934210601182818
Benoit, W. L., & Hansen, G. (2002). Issue adaptation of presidential television spots and debates to primary and general audiences. Communication Research Reports, 75(2), 138-145. doi:10.1080/08824090209384841
Benoit, W. L., & Hansen, G.J. (2004a). Issue ownership in primary and general presidential debates. Argumentation and Advocacy, 40(3), 143-154.
Benoit, W. L., & Hansen, G.J. (2004b). The changing media environment of presidential campaigns. Communication Research Reports, 21(2), 164-173. doi: 10.1080/08824090409359978
Benoit, W. L., & Stein, K. (2005). A functional analysis of presidential direct mail advertising. Communication Studies, 56(3), 203-225. doi:10.1080/10510970500181181
Campbell, K., & Chollet, D. (2007). The national security election. The Washington Quarterly, 37(1), 191-199. doi: 10.1162/wash.2007.31.1.191
Carlin, D. P. (1992). Presidential debates as focal points for campaign arguments. Political Communication, 5(4), 251-265. doi: 10.1080/10584609.1992.9962949
Damore, D. (2004). The dynamics of issue ownership in presidential campaigns. Political Research Quarterly, 57(3), 391-397. doi: 10.1177/106591290405700304
Goble, H., & Holm, P. M. (2008). Breaking bonds?: The Iraq war and the loss of Republican dominance in national security. Political Research Quarterly, 62(2), 215-229. doi:10.1177/1065912908320669
Hacker, J., & Pierson, P. (2005). Off center: The Republican revolution and the erosion of American democracy. New Haven, CT: Yale University Press.
Hayes, D. (2008). Party reputations, journalistic expectations: How issue ownership influences election news. Political Communication, 25(4), 377-400. doi:10.1080/10584600802426981
Jackson, J. E. (1975). Issues, party choices, and presidential votes. American Journal of Political Science, 19(2), 161-185. doi: 10.2307/2110431
Jamieson, K. H., & Birdsell, D. S. (1988). Presidential debates: The challenge of creating an informed electorate. New York: Oxford.
Levy, S. (2008, April 23). In Every Voter, A "Microtarget". The Washington Post. Retrieved from http://www. washingtonpost.com/wp-dyn/content/article/2008/04/22/AR2008042202471.html
Lipset, S. M., & Rokkan, S. (1967). Cleavage structures, party systems, and voter alignments: An introduction. In S. M. Lipset & S. Rokkan (Eds.), Party Systems and Voter Alignments.
McKinney, M. S. (2005). Let the People Speak: The Public's Agenda and Presidential Town Hall Debates. American Behavioral Scientist, 49(2), 198-212. doi: 10.1177/0002764205279428
Neuendorf, K. (2002). The content analysis guidebook Thousand Oaks, CA: Sage Publications.
Petrocik, J. (1996). Issue ownership in presidential elections, with a 1980 case study. American Journal of Political Science, 40(3), 825-850. doi:10.2307/2111797
Petrocik, J., Benoit, W. L., & Hansen, G. (2003). Issue ownership and presidential campaigning, 1952-2000. Political Science Quarterly, 118(A), 599-626. doi:10.1002/j.1538-165x.2003.tb00407.x
Rasmussen Polling Report. (2007-2008, August-October). Trust & Importance on issues. Retrieved from http://www. rasmussenreports.com/public_content/politics/issues2/trust_importance_on_issues
Saad, L. (2008, October 29). Economy reigns supreme for voters: More than half rate it "extremely important" to their vote for president. Gallup News Service. Retrieved from http://www.gallup.com/poll/111586/Economy-ReignsSupreme- Voters.aspx
The Annenberg Public Policy Center. (2010). The national Annenberg election survey: 2008 phone survey. Retrieved from http://www.annenbergepublicpolicycenter.org
Trent, J. S., & Friedenberg, R. V. (2007). Political campaign communication: Principles & practices (6th ed.). Lanham, MD: Rowman & Littlefield.
Washington Post-ABC News. (2008, November 30). Washington Post-ABC News tracking poll. Washington Post. Retrieved from http://www.washingtonpost.com/wp-srv/politics/documents/postpoll_110308.html
Wenger, D. H., & MacManus, S. A. (2009). Watching history: TV coverage of the 2008 Campaign. Journalism Studies, 70(3), 427-435. doi:10.1080/14616700902987207
Hayley J. Cole (PhD University of Missouri) is an assistant professor in the Department of Communication at the University of Missouri. Joshua Hawthorne (MA University of Missouri) is a PhD student in the Department of Communication at the University of Missouri. Correspondence concerning this article should be addressed to Hayley J. Cole, Department of Communication, University of Missouri, 108 Switzer Hall, Columbia, Missouri, 65211. Email: Hayley.Cole@gmail.com
TABLE 1. TOTAL CANDIDATE ISSUE MENTIONS DURING 2008 ELECTION NOMINATION SPEECHES AND GENERAL ELECTION DEBATES Democratic Republican Candidate Candidate z-test Issue Education 7.60 (101) 6.37 (69) 1.18 Health Care 9.55 (127) 6.55 (7l) 2.68 ** Poverty .45 (6) .18 (2) 1.13 Environment .60 (8) .92 (10) -.91 Elderly 2.03 (27) 1.85 (20) .33 Democratic Issues 20.24 (269) 15.87 (172) 2.77 ** National Defense 26.64 (354) 31.64 (343) -2.70 ** Foreign Policy 12.57 (167) 12.92 (140) -.26 Government Spending 5.87 (78) 7.20 (78) -1.32 Taxes 7.98 (106) 8.12 (88) -.13 Crime 2.33 (31) 3.41 (37) -1.60 Republican Issues 55.38 (736) 64.28 (686) -3.93 ** Economy 24.38 (324) 20.85 (226) 2.06 * Total (1,329) (1,084) Note: Percentage of issue mentions for each candidate (Frequency). Data results are totaled from all the data sets, the 2008 nomination acceptance speeches and the three 2008 general presidential debates. TABLE 2. ISSUE MENTIONS BY BOTH CANDIDATES IN THEIR 2008 NOMINATION ACCEPTANCE SPEECHES Democratic Republican z-test Score Candidate Candidate Issue Education 11.11 (23) 11.98 (20) -.26 Health Care 8.70 (18) 4.79 (8) 1.48 Poverty 2.42 (5) 1.20 (2) .86 Environment .48 (1) 0 (0) .37 Elderly 3.86 (8) 2.40 (4) .80 Democratic Issues 26.57 (55) 20.36 (34) 1.40 National Defense 23.19 (48) 23.95 (40) -.17 Foreign Policy 7.73 (16) 5.99 (10) .66 Government Spending 1.93 (4) 4.79 (8) -1.56 Taxes 5.80 (12) 4.79 (8) .43 Crime 4.35 (9) 5.99 (10) -.72 Republican Issues 43.00 (89) 45.51 (76) -3.36 *** Economy 30.43 (63) 34.13 (57) -.76 Total (207) (167) Note: Each cell displays percentage of the mentions of that issue compared to the total issues in that event, frequency of occurrences is in parentheses; * p < .05, ** p < .01, *** p < .001. TABLE 3. ISSUE MENTIONS BY BOTH CANDIDATES IN EACH 2008 GENERAL ELECTION DEBATE Democratic Candidate 1st Debate 2nd Debate 3rd Debate Issue Education 2.11 (9) 2.70 (10) 18.10 (59) Health Care 2.58 (11) 13.24 (49) 15.03 (49) Poverty 0 (0) .27 (1) 0 (0) Environment .47 (2) .54 (2) .92 (3) Elderly 1.17 (5) 2.16 (8) 1.84 (6) Democratic Issues 6.34 (27) 18.92 (70) 35.89 (117) National Defense 44.37 (189) 28.65 106) 3.37 (11) Foreign Policy 17.60 (75) 12.70 (47) 8.90 (29) Government Spending 5.40 (23) 7.84 (29) 6.75 (22) Taxes 8.22 (35) 6.49 (24) 10.74 (35) Crime 1.88 (8) 1.08 (4) 3.07 (10) Republican Issues 77.46 (330) 56.76 (210) 32.82 (107) Economy 16.20 (69) 24.32 (90) 31.29 (102) Total (426) (370) (326) Republican Candidate 1st Debate 2nd Debate Issue Education .56 (2) 1.04 (5) Health Care 1.69 (6) 8.81 (26) Poverty 0 (0) 0 (0) Environment .85 (3) 1.69 (5) Elderly 1.13 (4) 2.71 (8) Democratic Issues 4.23 (15) 14.92 (44) National Defense 46.20 (164) 33.90 (100) Foreign Policy 17.75 (63) 13.22 (39) Government Spending 8.73 (31) 5.42 (16) Taxes 5.63 (20) 10.51 (31) Crime 4.79 17) .68 (2) Republican Issues 83.10 (295) 63.39 (187) Economy 12.68 (45) 21.14 (63) Total (355) (295) Republican Candidate 3rd Debate z-test Score Issue Education 15.73 (42) 1.5 Health Care 11.61 (31) 2.40 * Poverty 0 (0) .90 Environment .75 (2) -1.15 Elderly 1.50 (4) -.08 Democratic Issues 29.59 (79) 2.35 National Defense 14.61 (39) -2.82 ** Foreign Policy 10.49 (28) -.46 Government Spending 8.61 (23) -.90 Taxes 10.86 (29) -.27 Crime 3.00 (8) -1.44 Republican Issues 47.57 127) -4.04 *** Economy 22.85 (61) 2.67 ** Total (267) Note: Each cell displays percentage of the mentions of that issue compared to the total issues in that event, frequency of occurrences is in parentheses; * p < .05, ** p < .01, *** p < .001; z-test score is calculated across all debates. TABLE 4. THE PROPORTION OF VOTERS MENTIONING THE ECONOMY AS THE MOST IMPORTANT ISSUE ON SELECTED DATES DURING 2008 ELECTION Event Mentions Prior Mentions Following z-test Obama Acceptance 19.95 [+ or -] 3.57 26.62 [+ or -] 3.78 -2.49 * McCain Acceptance 24.42 [+ or -] 3.57 24.63 [+ or -] 3.67 -.09 First Debate 27.11 [+ or -] 3.90 31.69 [+ or -] 3.97 -.09 Second Debate 30.04 [+ or -] 3.99 24.49 [+ or -] 3.95 -.09 Third Debate 35.09 [+ or -] 4.48 37.25 [+ or -] 4.26 -.09 Note: * p < .05, ** p < .01, *** p < .001; Number indicates the proportion that answered "economy" or "the economy" compared to the total frequency of interviews [+ or -] the 95% margin of error. TABLE 5. PERCEIVED ISSUE HANDLING COMPETENCE OF THE PARTIES: PETROCIK VS 2008 Petrocik Study October 2008 Problem Better Handled By Trust More on Issue Issue Democratic Republican Democratic Republican Economy 32% 49% 51% 38% National Security 17% 68% 47% 44% Taxes 35% 44% 47% 42% Education 44% 28% 53% 34% Health Care 51% 22% 54% 34% Note: Petrocik Study data comes from table 1, p. 832, results from 1988-1991 in Petrocik, 1996. October 2008 data comes from Rasmussen Polling Report, 2007-2008, August-October.
|Printer friendly Cite/link Email Feedback|
|Author:||Cole, Hayley J.; Hawthorne, Joshua|
|Publication:||Argumentation and Advocacy|
|Date:||Sep 22, 2013|
|Previous Article:||Dick Gregory and activist style: identifying attributes of humor necessary for activist advocacy.|
|Next Article:||How much jail time? Returning women to the abortion debate.|