Issue voting has been one of the major concerns for studies in voting behavior. According to Nie et al. (1979), campaign issues have significantly affected American voting behavior since the 1960s. Specifically, when voters were asked about candidate evaluation in the 1950s, only 50% mentioned campaign issues. Yet, since 1964, more than 75% of voters could identify issue positions of candidates. Furthermore, Nie, Verba, and Petrocik found that in the 1950s, issue positions had the least explanatory power among the three major factors (i.e., along with party identification and candidate evaluation) in analyzing voting behavior. But by the late 1960s, issue positions became the most crucial factor in explaining vote choice. Scholars also indicate that issue voting has also increased in some Western countries such as the United Kingdom and the Netherlands in the second half of the twentieth century (Harrop & Miller, 1987).
As issue voting has become increasingly important, how can candidates utilize political issues to build up their campaign strategies to attract voters? Petrocik (1996) argued that voters tend to hold different perceptions toward how parties can handle different issues. For example, US voters generally believe that the Republican Party is good at dealing with foreign policy as well as national security while the Democratic Party is good at handling social welfare and domestic issues. It is called the concept of issue ownership—that is, if a party can convince voters of its capability of dealing with a certain political issue, from the viewpoint of voters, the party may “own” the issue.
When a party indeed “owns” a certain issue, should it focus on such issue in its campaign? What if voters care more about other issues? This article explores the linkage between the perception of issue ownership and campaign advertising. By assuming that a political party has a strong incentive to gain votes via issue setting as part of its campaign trategy, this study utilizes Web survey data to explore the extent to which three issue-related strategies of campaign advertising—namely, issue ownership, issue convergences, and issue trespassing, may affect voters’ assessments of parties. The article is organized as follows: the next section examines the literature on issue-based campaign strategies; the following section analyzes the data obtained from a Web survey conducted prior to the 2012 presidential election in Taiwan. The final section offers conclusions and discusses avenues for future research.
Issue-Based Campaign Strategies
Based on Petrocik’s (1996) concept of issue ownership, scholars have utilized large volumes of empirical data to elaborate the perception of issue ownership from a comparative perspective. For example, voters in the United Kingdom tend to believe that the Labor Party is good at healthcare while the Conservative Party is good at national security (Norris et al., 1999). In Canada, Bélanger (2003) analyzed a 50-year data set and discovered that voters generally perceive that the federal Liberal Party excels in handling international affairs and national unity whereas the Reform Party is good at public finance management.
In Taiwan, Chang (2009) conducted a systematic analysis on issue ownership perceptions among voters. By comparing the two major parties – namely, the Kuomintang (KMT) and the Democratic Progressive Party (DPP) – as well as analyzing two opinion polls before and after the 2008 presidential election, Chang found that prior to the election, Taiwanese voters tend to believe that the KMT is perceived as being better at handling such issues as “cross-Strait relations,” “public safety,” and “local infrastructure” while the DPP has a better reputation with regards to handling such issues as “gender inequality,”, “economy,” and “social welfare.” Yet, after the election, voters considered that the KMT was good at handling “economy,” “education,” and “cross-Strait relations” whereas the DPP was perceived as good at “gender inequality,” “social welfare,” and “local construction.” While issue ownership perception should be regarded as a type of long-term attitude among voters, it still could fluctuate simply due to electoral campaigns. Thus, as indicted by Chang’s analysis, we are not sure whether Taiwanese voters have stable issue ownership perceptions toward the two major parties.
In short, previous research suggests two major findings regarding issue ownership perception: first, voters recognize political parties’ capacities or reputations to competently deal with certain issues. Thus, linkages between issue and party competence do exist. Second, issue ownership perception can be regarded as a long-term attitude. Yet, it is not necessarily invariable – that is, it may change due to a party’s governing performance or emergence of new parties (Bélanger, 2003). Additionally, voters may re-define the linkage between parties and certain issues, depending on what types of campaign messages they may receive (Chang, 2009).
If issue ownership perception does exist, to what extent does a party take advantage of such perceptions when making its campaign strategy? Petrocik (1996) and Petrocik et al. (2003) posited that candidates should build up their campaign strategies in correspondence to voters’ perceptions of issue ownership. Specifically, in order to maximize campaign effects on voters, it would be beneficial for a candidate to reach/persuade voters through advocating for its advantageous issues rather than focusing on those issues that have been “owned” by its rivals. Bélanger (2003) also argued that parties tend to emphasize their advantageous issues during campaign as their positions on such issues rarely change. As different parties (or candidates) focus on different issues owned by themselves, respectively, they would be able to differentiate themselves, and the electoral outcome would be determined by the salience of issues as perceived by the voters. Thus, from the viewpoint of issue ownership perception, the most effective campaign strategy for a party is to advocate its “owned” issues and to make those issues salient among voters (Petrocik, 1996; Abbe, 1998; Sellers, 1998; Bélanger, 2003). Recent empirical analyses on US elections indeed found that candidates in different levels of elections tend to emphasize their campaign messages with respect to their “owned” issues in television spots including advertisements and debates (Benoit, 2003, 2007, 2018; Benoit et al., 2011).
Yet, scholars have argued that the use of issue ownership perception should not be regarded as the dominant factor. As one of the competing theories, Kaplan et al. (2006) argued that each election tends to have its own salient issue at the time and all contenders are likely to pay more attention on those issues during the campaign period. Thus, in order to garner the most support, parties/candidates may focus on the same issues but offer different policy alternatives. When all the parties/candidates’ efforts converge on the same issues, the issue ownership strategy seems of no use. Instead, parties (or candidates) may adopt issue convergence strategy and pay attention to the most important issue specified by voters.
The third campaign strategy with respect to the use of issue setting is called the issue trespassing strategy. Damore (2004) argued that under certain circumstances, parties (or candidates) would like to focus on the issues that have been owned by rivals and try to alter voters’ issue ownership perceptions. Following a logic similar to that of issue convergence strategy, parties (or candidates) tend to focus on the issues that voters are most concerned with and aggressively adopt an issue trespassing strategy, even if those issues are not their advantageous issues. Brazeal & Benoit (2008) also found that in US congressional elections, although candidates from both Democratic and Republican parties discuss their own issues more, winners tend to discuss issues from the other party more than losers.
The three issue-based campaign strategies – issue ownership, issue convergence, and issue trespassing – are based on different assumptions about voters’ perceptions toward the most important issues in the election. The issue ownership strategy assumes that voters do not have any preference on the importance of issues, and each party is capable of persuading voters that the issue it owns should be the most important one. In order words, the importance of an issue is endogenous to the making of a party’s campaign strategy. In contrast, both the issue convergence and issue trespassing strategies assume that the importance of issue is exogenous to a party’s campaign strategy and is determined by the political environment prior to the election, when parties would adopt either of the two issue-based campaign strategies to meet the demands of the general public.
One of the prominent manifestations of campaign strategy is campaign advertising. Shaw (1999) specified two types of campaign advertising effects on voters, namely mobilization and persuasion. Specifically, campaign advertising contains three elements to mobilize voters, including raising voters’ party identifications, stimulating voters’ perceptions of the external environment, and increasing people’s concern with certain issues. The latter two elements may also persuade independent voters who do not affiliate with any party. Shaw (1999) argued that mobilization and persuasion effects can be used as the two criteria to measure the success of campaign advertising. Nevertheless, scholars have difficulties truly measuring advertising effects on voting behavior due to the fact that a multitude of factors may contribute to explaining voting behavior and electoral outcomes. Indeed, the formation of political attitudes such as party identification is a long-term process. The role of family or peer pressure can be more influential than any advertisement in shaping one’s political attitudes. It is difficult to specify what kinds of factors would change voters’ attitudes during the campaign period even though a number of studies have shown that the effect of campaign advertising on voting behavior does exist (Ansolabehere & Iyengar, 1994, 1997; Holbrook, 1996).
This analysis does not aim to directly measure the effect of campaign advertising on voting behavior. Rather, from the three perspectives of issue-based campaign strategies mentioned above, this paper explores whether different issue-based campaign advertisements may affect voters’ assessments of parties’ capacities to deal with various issues. In other words, we focus on whether different types of campaign advertising strategies may reinforce or damage parties’ images on issue-specific capacities. Empirically, we hypothesize that different campaign advertisements may reinforce party supporters’ preferences, persuade those voters in the middle, and even alter the preferences of the opposite party supporters. The findings have several implications for the set-up of campaign advertising that can be assumed to exert certain impacts on voting behavior.
In this section, we use Web survey data to test the hypotheses mentioned above. The data set was collected two weeks prior to the 2012 Taiwan presidential election through PollcracyLab.com, which is an experimental Web survey platform hosted by the Election Study Center at National Chengchi University, Taipei, Taiwan. The panelists of PollcracyLab have been recruited via random digit dialing (RDD) telephone surveys so that the list of panelists could be regarded as a semi-random sample. Yet, it still has certain degrees of sample selection bias vis-a-vis the general population (Yu, 2013). As the major purpose of this study is not to make inferences about the population from the sample data, this Web survey platform suffices to show advertisements and test the effect of advertisement on issue assessment.
We first explored the importance of issue dimensions among voters. In the first part of the survey, we asked the respondents to specify the most important issue in Taiwan from among seven issue dimensions, namely “economy” (ECO), “education” (EDU), “public safety” (SAFE), “cross-Strait relations” (STRAIT), “local infrastructure” (LOCAL), “social welfare” (WEL), and “gender inequality” (GENDER)4. Among these seven issue dimensions, 69% of the respondents chose “economy” as the most important issue. Thirteen percent chose “Education.” “Public safety,” “cross-Strait relations,” or “social welfare” were each ranked as the most important issue by approximately 5% of respondents. Only about 1.7% ranked “local infrastructure” as the number one issue. the least important issue seems to be “gender inequality” – that is, less than 1% of the respondents said it is the most important issue.
To see whether voters have issue ownership perceptions between the two major parties, we then asked our respondents to rate KMT’s and DPP’s performances on a scale of 0 to 10 on the seven issue dimensions. The exact wording of the questionnaire is as follows:
If 0 represents highly incapable of handling (issue dimension) and 10 represents highly capable of handling (issue dimension), how would you rate KMT/DPP on a scale of 0-10?
Table 1 suggests that the respondents held the perception that the KMT performs well on “economy,” “education,” and “cross-Strait relations” while the DPP has advantages in “social welfare” and “gender inequality.” Although our sample is not a fully random sample, the results correspond to the past findings shown by Chang (2009). Respondents do not perceive any significant differences between the two parties regarding their performances on “public safety” and “local infrastructure.”
As “economy” is the most important issue for voters, from a viewpoint of issue convergence, each of the two major parties may be forced to pay attention to economic issues. From the perspective of issue trespassing, the DPP may have strong incentives to address the economy even though that issue is “owned” by the KMT. “Cross-Strait relations” is clearly a KMT issue while “social welfare” belongs to the DPP.
Although there were three candidates in the 2012 Taiwan presidential election, the main focus was still on KMT nominee Ma Ying-jeou, who was seeking reelection, and DPP chairperson Tsai Ing-wen. During the campaign period, the KMT seemed to always focus on economic development and Taiwan’s relationship with China, while the DPP was mainly concerned about the uneven distribution of wealth and social inequality. According to statistics collected by Hsu (2012), there were two advertisements concerning cross-Strait relations in KMT’s 60 TV campaign advertisements. In fact, only KMT had TV advertisements specifically on cross-Strait relations. In contrast, there was no advertisement concerning cross-Strait relations among Tsai Ing-wen’s 29 TV campaign advertisements. Yet, Tsai had two TV advertisements concerning the issue of housing prices while Ma had none. At a glance, there seemed to be some differences between the two major parties as they allocated their resources to different issue areas.
In the second part of the survey, we showed six campaign advertisements that had been broadcast on TV – specifically, three of them were made by the KMT and the other three were made by the DPP. All of them are related to party’s issue positions. We chose two advertisements (one from each party) dealing with the economy issue, three advertisements dealing with the social welfare issue, and one KMT advertisement about the cross-Strait relations issue. We assumed that the KMT’s advantageous issue area is “cross-Strait relations” while the DPP’s advantageous issue area is “social welfare.” “Economy” was the most important issue identified by voters. Because the DPP had no advertisement on “cross-Strait relations,” we chose two advertisements concerning “social welfare.”
After viewing those ads, did the respondents change their attitudes with respect to their assessments on issue handling capability? Table 2 shows that the respondents did not change their attitudes dramatically. Specifically, the respondents indeed rated the KMT’s ability to handle cross-Strait relations better than they did before viewing the advertisements. On the other hand, the respondents also rated the DPP’s ability to handle welfare issues better than before. Yet, regarding other issue dimensions, the respondents did not significantly increase or decrease their ratings toward each party’s capabilities after viewing the advertisements. It seems that the advertisements reinforce perceptions of issue ownership among the respondents.
Our next question is: did campaign advertising reinforce party supporters’ preferences, persuade voters in the middle, or even alter the preferences of supporters of the opposite party? Table 3 shows how supporters of different parties (including independent voters) assess KMT’s capability of handling issues before and after viewing the advertisements. After viewing the advertisements, the KMT supporters increased their ratings for their party’s capability of handling both “social welfare” and “cross-Strait relations.” So did independent voters (IND). In contrast, the DPP supporters decreased their opponent’s ratings on “economy” and “social welfare.” Thus, the six campaign advertisements we showed here not only reinforced KMT’s existing image (i.e., capable of dealing with “cross-Strait relations”) but also improved its image on a disadvantageous issue dimension (i.e., social welfare), at least among KMT supporters and independent voters.
Table 4 shows how supporters of different parties assess the DPP’s capability of handling issues before and after viewing the advertisements. There was not much difference except that the KMT supporters decreased their ratings for the DPP’s capability of handling “cross-Strait relations” after viewing the advertisements. It seems that the campaign advertisements we showed here did not reinforce DPP’s existing image (i.e., capable of dealing with “social welfare”) among all types of voters. In addition, maybe because the DPP did not have any advertisements on “cross-Strait relations,” it is not surprising that the KMT supporters downgraded the DPP’s capability on dealing with that issue dimension.
By analyzing experimental Web survey data collected prior to the 2012 Taiwan Presidential Election, this study draws three preliminary conclusions regarding the perception of issue ownership in Taiwan:
First, from the voters’ perspectives, the two major parties do consistently “own” certain issues. Specifically, Taiwanese voters generally believe that the KMT is good at handling the “cross-Strait relations” issue, while the DPP is perceived as good at handling the “social welfare” issue.
Second, issue ownership advertising may reinforce voters’ beliefs on parties’ capabilities for handling certain issues. This finding echoes the previous studies by explaining why a party’s campaign messages tend to emphasize its advantageous issues rather than disadvantageous ones. Issue trespassing advertising may fix a party’s image on the disadvantageous issue dimension. Our data show that the KMT’s advertisements may have such an effect.
Third, “economy” or economic development was the most salient issue in Taiwan prior to the 2012 election. Although we did observe that the DPP supporters decreased their rating of the KMT’s capability of handling the “economy” issue after viewing advertisements, independent voters did not change their perceptions toward both parties’ capabilities of handling the “economy” issue at all. It seems more likely that when both parties address that most salient issue, the net advertising effect might be canceled out.
As a preliminary analysis, this study has three limitations: first, while we explored possible issue ownership perceptions in Taiwan and laid out different issue-based campaign strategies, we did not actually test the effect of issue-based campaign strategies on ultimate voting behavior. Future empirical research may specify how issue ownership perceptions affect voters’ attitudes, including their voting intentions. Second, we did not use a rigid experiment to see the real “treatment effect” (i.e., campaign strategy) on voters’ perceptions towards parties’ issue handling capabilities. In other words, our analyses did not control for other feasible factors that could affect voters’ perceptions before and after viewing the advertisements. For example, it is possible that an increase of party capacity rating simply reflects the better style or production of the advertisement, not the better content per se. Third, it is important to note that we only analyzed the data for one election. Thus, our findings are not necessarily robust if we incorporate more data from different elections. Based on this analysis, it will be interesting to test whether issue ownership perceptions as well as the effects of issue-based campaign strategies change over time.
This research was supported by the Ministry of Science and Technology of Republic of China (Taiwan) (MOST-102-2410-H-004-133-).
Eric Chen-hua Yu is an associate research fellow of the Election Study Center and jointly appointed as an associate professor of political science at National Chengchi University in Taipei, Taiwan. He is also affiliated with the University’s Taiwan Institute for Governance and Communication Research, a new multi-discipline research institute funded by Taiwan’s Ministry of Education. His research interests and recent publications have focused on electoral politics, public opinion, and public policy analysis. Yu received an MS in Public Policy Analysis from the University of Rochester and a Ph.D. in political science from Columbia University.
He can be reached at Election Study Center, National Chengchi University, No. 64, Sec. 2, Zhinan Rd., Wehnshan Dist., Taipei City 11505, Taiwan (ROC) or by e-mail at email@example.com.
Pei-chen Hsu received her MA and BA in political science from National Chengchi University in Taipei, Taiwan. Her master’s thesis analyzed the content of campaign advertisements of major candidates in Taiwan’s 2012 presidential election.
Ms. Hsu can be reached via <firstname.lastname@example.org.>
All correspondence concerning this article should be addressed to Eric Chen-hua Yu at Election Study Center, National Chengchi University, No. 64, Sec. 2, Zhinan Rd., Wehnshan Dist., Taipei City 11505, Taiwan (ROC) or by e-mail at email@example.com.
Date of Submission: 2018-12-11
Date of the Review Results: 2019-02-08
Date of the Decision: 2019-02-26
The survey was conducted by the following two procedures: first, we sent out approximately 8,000 invitations via email to the panelists of PollcracyLab. Then, 401 respondents answered our survey questionnaires. It is important to note that almost all types of Web surveys have the problem of sample selection bias. The data we collected here is no exception. Our sample does reflect a certain level of “digital divide.” That is, it tends to be biased toward young, highly educated, and urban demographics in comparison to the population. Yet, as the panelists of PollcracyLab are recruited via random digit dialing telephone surveys, the degree of bias here is less salient vis-à-vis the data collected via any opt-in Web survey platform.
Chang (2009) argued that these seven issue dimensions included the most important issues for Taiwanese voters in the 2008 presidential election. Regarding the most important issue in the campaign period of the 2012 presidential election, the 2012 post-election survey of Taiwan’s Election and Democratization Studies (TEDS2012) shows that economic development is the most salient issue for voters, followed by cross-Strait relations, unemployment, livelihood, and social inequality issues (Chu, 2012). Thus, in our survey experiment, by assuming that voters do care about the economy, cross-Strait relations, and social welfare issues, we selected both parties’ advertisements on those issues to see how voters reacted after viewing those advertisements.
Please see the Appendix for the subjects of the six advertisements selected for the survey as well as their possible issue-based campaign strategies. For detailed information about these advertisements, please contact the corresponding author.