What Were the Attitudes of Major Party Supporters in an "Unlikeable Election"?: Affective Polarization of Partisan Voters in the 20th Presidential Election

1. Introduction

The 20th presidential election, held on March 9, 2022, was conducted amidst significant criticism, with various allegations being raised against the candidates of the two major parties during the campaign. Consequently, domestic and international media outlets referred to this election as an "unlikeable election" or a choice of the "lesser evil" (Reuter 2022; The Korea Herald 2021; Yonhap News 2022; Chosun Ilbo 2021; Hankyung 2022).

However, despite these generally negative assessments of both major candidates, several noteworthy facts were observed. First, the high approval ratings for both candidates persisted even when negative news broke during the election (Hankook Ilbo 2021). The election outcome was similar. A high voter turnout of 77.1% was recorded despite the critical circumstances. Furthermore, the vote difference between candidate Yoon Suk-yeol (48.56%) and candidate Lee Jae-myung (47.83%) was less than 1%p, marking the smallest margin in the history of presidential elections. The high voter turnout, separate from the dislike for both candidates, and the fact that these two candidates garnered a total of 96.39% of the votes, raise questions about whether this election was indeed the most unlikeable election and a choice of the lesser evil, or if it exhibited unique characteristics distinguishing it from previous elections. Theoretically, under a plurality voting system, voters are known to tend to abstain from voting or choose a third-party candidate if they are dissatisfied with all candidates of the major parties forming the political competition or with the current political situation (Hethetington 1999; Himmelweit et al. 1985; Kang 2004; Southwell 1998; Webb 2005; Kang Won-taek 1998). Nevertheless, in this "unlikeable election," voters participated at a high rate and simultaneously chose one of the two candidates. In this choice between the two candidates, while it is possible that unaffiliated voters or supporters of third parties, who lack strong party preferences, exhibited behavior of choosing the lesser evil, or negative voting driven by negative partisanship (Abramowitz and Webster 2016), voters with a preference for the two major parties showed steadfast support for their chosen party and candidate even in an unlikeable election.

This paper, therefore, seeks to look beyond the surface of the "unlikeable election" by closely examining the political attitudes of voters supporting the two major parties during the 20th presidential election. The argument presented here is that the "unlikeable election" and the choice of the "lesser evil" were limited to supporters of minor parties and unaffiliated voters who did not support the two major parties. Partisan voters, on the other hand, made their "best" choice based on partisan "liking," as in any other election, while simultaneously displaying strong negative attitudes towards the opposing party and candidate. Thus, this paper aims to demonstrate that behind the assessment of an "unlikeable election" lay affective polarization (Iyengar et al. 2012), characterized by strong positive feelings towards one's own party and strong negative feelings towards the opposing party.

2. Affective Polarization of Voters

Existing research on voter polarization has attempted to theorize polarization based on ideology. For instance, Abramowitz and Saunders (1998) argued that polarization intensified as new social issues emerged, leading to the ideological divergence of the two parties. Consequently, voters became more aware of the differences between the parties and shifted their support to the party that better aligned with their ideological leanings, undergoing a gradual realignment. In subsequent research, Abramowitz (2010) pointed out the disappearance of moderate voters in the United States, leaving only voter groups divided into liberal and conservative ideological camps. However, criticism has been raised regarding the generalization that the public is divided into liberals and conservatives without moderates, as voter ideological distributions still show a high proportion of moderates in the middle.

Therefore, research by Fiorina et al. (2008; 2010) argued that despite the continued majority of moderate voters in the overall ideological distribution, the electoral system, which presents only two choices, leads to a superficial appearance of public polarization. They proposed the concept of "partisan sorting," suggesting that voters with ideological preferences increasingly supported parties aligning with those preferences, resulting in ideologically homogeneous party camps. However, the underlying ideological distribution of voters remained unimodal with a majority of moderates. Galston and Nivola (2006) pointed out that while polarization and partisan sorting are theoretically distinct concepts, it is becoming increasingly difficult to differentiate between the two in practice. Furthermore, it is difficult to explain the phenomenon of escalating inter-party conflict and confrontation, even when voters do not appear to be divided in their ideological distribution.

In response, Iyengar et al. (2012) proposed a new perspective: voter polarization manifests not in the ideological dimension but in the affective dimension. They argued that even though moderates still constitute the majority in the ideological distribution of voters, the substantial form of voter polarization is affective polarization, characterized by intensified animosity and negative attitudes among partisan voters, ultimately widening the gap between positive attitudes towards one's own party and negative attitudes towards the opposing party. Several studies using South Korea as a case study have also identified the patterns of affective polarization among voters and examined its political effects. First, there are studies that identify the components or psychological underpinnings of partisan polarization among South Korean voters (Gil Jeong-ah and Ha Sang-eung 2019; Jang Seung-jin and Seo Jeong-gyu 2019). Additionally, research has analyzed how the effects of this partisan polarization extend to non-political domains, reinforcing negative perceptions of those with different political leanings (Jang Seung-jin and Jang Han-il 2020). In summary, these studies highlight the intensification of inter-party hostility and hatred due to partisan polarization.

3. Liking for the Two Major Parties and Candidates

① All Voters

[Table 1] presents the average liking for the two major parties and their candidates. Respondents chose a score between 0 (most disliked) and 10 (most liked). The average scores for all subjects were below 5. The average scores for the two candidates were slightly higher than for the two parties. However, there were no significant differences among the two parties and the two candidates.

[Table 1] Average Liking for the Two Major Parties and Candidates

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

The following [Figure 1] shows the distribution of liking for the two major parties and candidates as a histogram. Reflecting the prevailing sentiment of dislike or choice of the lesser evil in this election, the proportion of respondents with the most negative attitude was highest for both parties and candidates. The proportion of respondents expressing the most negative sentiment towards the Democratic Party and the People Power Party was 21.04% and 24.25%, respectively. For candidates Lee Jae-myung and Yoon Suk-yeol, these proportions were 25.72% and 25.07%. The distribution of liking for all subjects showed a very similar pattern.

[Figure 1] Distribution of Liking for the Two Major Parties and Candidates

② By Respondent's Party Support

Meanwhile, the following [Table 2] analyzes the affective attitudes towards the four subjects, categorized by the respondent's party support. Although the overall average values for the two parties and the two candidates were similar, significant differences were observed based on party preference. Respondents supporting either of the two major parties showed high liking for their own party and candidate, while displaying considerably low average scores for the opposing party and candidate. Therefore, it can be confirmed that the difference between positive sentiment towards one's own party and negative sentiment towards the opposing party is large among supporters of the two parties. Thus, it is necessary to pay more attention to the partisan differences within attitudes towards each subject, rather than the overall negative sentiment towards the two parties and candidates.

[Table 2] Average Liking for the Two Major Parties and Candidates: By Respondent's Party Support

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

[Figure 2] and [Figure 3] present histograms illustrating the distribution of favorability toward the two political parties and two candidates, categorized by respondents' party preference. As observed in [Figure 1], the proportion of the most negative attitude was highest for both parties and candidates. However, in [Figure 2] and [Figure 3], which are depicted by party support, a significant difference was found between supporters of the Democratic Party and the People Power Party. Looking at [Figure 2-1] and [Figure 3-1], which target favorability toward the Democratic Party and Lee Jae-myung, the proportion of negative attitudes toward the Democratic Party and Lee Jae-myung markedly decreased when only Democratic Party supporters were isolated. Conversely, among People Power Party supporters, the proportion of such negative attitudes significantly increased. In contrast, in the histograms for favorability toward the People Power Party and Yoon Suk-yeol, [Figure 2-2] and [Figure 3-2], negative attitudes were rarely observed among People Power Party supporters, while negative attitudes toward them significantly increased among Democratic Party supporters. Therefore, it can be inferred that the negative sentiment towards the two parties and two candidates, observed on average in [Figure 1], originates from respondents supporting the opposing party. In summary, behind the dislike for the two political factions lies affective polarization (Iyengar et al. 2012), characterized by a large difference between positive sentiment toward one's own party and negative sentiment toward the opposing party among partisan voters.

[Figure 2] Distribution of Favorability Toward the Two Parties: By Respondent's Party Support

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

[Figure 3] Distribution of Favorability Toward the Two Candidates: By Respondent's Party Support

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

4. Satisfaction with Democracy and Favorability Toward Parties and Candidates

① Satisfaction with Democracy

This section examines respondents' satisfaction with South Korea's current democracy and its relationship with partisan attitudes. Satisfaction with democracy (Anderson and Guillory 1997; Anderson et al. 2005) is understood as a component of democratic support. Critical evaluations of the two major parties and candidates can be predicted to negatively influence voters' attitudes toward democracy.

[Table 3] Mean Satisfaction with Democracy

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

Meanwhile, in representative democracy, where the party that wins the election forms the government and manages state affairs, partisan differences inherently exist among voters regarding compliance with government policies, trust in government, and ultimately, democratic support. Anderson et al. (2005) and Anderson and LoTempio (2002) named this the Winner-Loser Gap. Satisfaction with democracy, which this paper aims to examine, was measured on an 11-point scale, ranging from 0 for not at all satisfied, through 5 for neutral, to 10 for completely satisfied. First, looking at the average values, as shown in [Table 3], supporters of the ruling Democratic Party of Korea expressed the most positive attitude toward democracy, while supporters of the People Power Party, the main opposition party forming the political competition, expressed the most negative attitude. However, since these results were derived from a bivariate relationship without considering the influence of other variables, they are unlikely to be considered significant differences given the observed differences and the magnitude of the F statistic.

[Figure 4] shows the distribution of responses regarding satisfaction with democracy. Among all respondents, the proportion who answered 'neutral' was the highest at 30.21%, and the proportion generally decreased as it moved toward the extreme values. However, more respondents were distributed in the positive evaluation scores than in the negative evaluation scores of democracy.

[Figure 4] Distribution of Satisfaction with Democracy

[Figure 5], which illustrates satisfaction with democracy categorized by respondents' party support, shows no significant differences among supporter groups. In each group, the proportion of 'neutral' responses was the highest, and the pattern of gradual decrease in proportion as it moved toward scores of 0 and 10, with more respondents distributed in positive evaluations than negative ones, was observed consistently across all groups. This pattern was also similar to [Figure 4] for all respondents. Therefore, based on [Table 3], [Figure 4], and [Figure 5], it is concluded that there are no significant differences in satisfaction with the current operation of democracy in South Korea based on party preference.

The [Figure 6] below presents a scatter plot illustrating the relationship between favorability toward the two major parties and two candidates, and satisfaction with democracy. There appears to be no particular relationship between favorability toward any specific party and satisfaction with democracy.

[Figure 5] Distribution of Satisfaction with Democracy: By Respondent's Party Support

[Figure 6] Scatter Plot of Party and Candidate Favorability and Satisfaction with Democracy

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

② Satisfaction with Democracy and Favorability Toward Major Parties and Candidates

Meanwhile, [Figure 7] and [Figure 8] present interesting results. Despite the lack of significant differences in satisfaction with democracy based on respondents' party preference, supporters of the two major parties associate satisfaction with democracy with positive attitudes toward their preferred party and candidate, and negative attitudes toward the opposing party and candidate. First, as observed in [Figure 4] and [Figure 5], since more respondents are distributed in the positive satisfaction range for democracy, the scatter plots in [Figure 7] and [Figure 8] are primarily distributed above the 5-point mark. However, the top-left graph in [Figure 7] shows that supporters of the Democratic Party of Korea exhibit high satisfaction with democracy and high favorability toward the Democratic Party of Korea, while displaying a negative attitude toward the People Power Party. Conversely, the top-right graph shows that supporters of the People Power Party exhibit high satisfaction with democracy along with high favorability toward the People Power Party, and antipathy toward the Democratic Party of Korea. Supporters of other parties and independents do not exhibit a biased attitude toward either party, as shown in [Figure 6-1] above.

[Figure 7] Party Favorability and Satisfaction with Democracy: By Party Support

This pattern is identically observed in [Figure 8]. While points are predominantly distributed above the 5-point mark for satisfaction with democracy, the top-left graph shows supporters of the Democratic Party of Korea exhibiting high satisfaction with democracy and high favorability toward Lee Jae-myung, while displaying a negative attitude toward Yoon Suk-yeol. Conversely, the top-right graph shows supporters of the People Power Party exhibiting high satisfaction with democracy along with high favorability toward Yoon Suk-yeol, and antipathy toward Lee Jae-myung. Supporters of other parties and independents do not exhibit a biased attitude toward either candidate, as shown in [Figure 6-2] above.

[Figure 8] Candidate Favorability and Satisfaction with Democracy: By Party Support

In summary, the results from [Figure 7] and [Figure 8], which contrast the two parties and two candidates, particularly among supporters of the major parties, indicate that while there are no differences in the evaluation or satisfaction with the current operation of democracy in South Korea based on partisan preference, partisan voters can be inferred to make different attributions regarding the operation of democracy. Specifically, by confirming that partisan voters associate positive emotions toward their preferred party and candidate, and negative emotions toward the opposing party and candidate, with satisfaction with democracy, it can be predicted that for them, negative attitudes toward democracy will be formed not by a simultaneous dislike of both parties strengthening negative attitudes toward democracy, but when the difference in dislike for the opposing party compared to the favorability of the preferred party widens.

[Table 4] OLS Regression Analysis of Favorability Toward the Two Major Parties and Two Candidates and Satisfaction with Democracy

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

*** p<0.001, ** p<0.01, * p<0.05, † p<0.1

Reference Category: Democratic Party Partisanship

Accordingly, an OLS regression analysis was conducted with satisfaction with democracy as the dependent variable, as shown in [Table 4]. First, looking at [Table 4], which shows the results of analyses where favorability toward the two parties and the two candidates were entered separately, statistically significant positive coefficients were estimated for all. This means that favorability toward the two parties and the two candidates has a positive impact on satisfaction with democracy, or conversely, that as respondents' antipathy toward the two parties and the two candidates increases, their attitude toward democracy becomes more negative. It also appears that respondents do not associate satisfaction with democracy with only one political faction.

However, when respondents supporting the two major parties were analyzed separately, as shown in [Table 5], no statistically significant relationship was found between favorability toward the Democratic Party and favorability toward Lee Jae-myung and satisfaction with democracy among Democratic Party supporters. Meanwhile, among these individuals, only favorability toward the People Power Party and favorability toward Yoon Suk-yeol influenced satisfaction with democracy.

[Table 5] OLS Regression Analysis of Favorability Toward Two Parties and Two Candidates and Satisfaction with Democracy:

Supporters of the Democratic Party

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

*** p<0.001, ** p<0.01, * p<0.05, † p<0.1

Note: The same control variables as in the analysis for [Table 4] were included but not reported in the table.

However, the analysis of supporters of the People Power Party, as shown in [Table 6], yielded the opposite results. Among these individuals, only favorability toward the Democratic Party and favorability toward Lee Jae-myung were associated with satisfaction with democracy, while favorability toward the People Power Party and favorability toward Yoon Suk-yeol did not affect satisfaction with democracy.

[Table 6] OLS Regression Analysis of Favorability Toward Two Parties and Two Candidates and Satisfaction with Democracy:

Supporters of the People Power Party

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

*** p<0.001, ** p<0.01, * p<0.05, † p<0.1

Note: The same control variables as in the analysis for [Table 4] were included but not reported in the table.

Judging from the analysis results in [Figures 6], [Figures 7], [Figures 8], and [Tables 4], [Tables 5], and [Tables 6], it can be inferred that for supporters of the Democratic Party and supporters of the People Power Party, the difference between favorability toward their preferred group and antipathy toward the opposing group, rather than mere antipathy toward the two parties, was a more significant factor.

Therefore, in the next step, an analysis was conducted using affective polarization toward parties and candidates, i.e., relative favorability measured by the difference between favorability toward one's own party and candidate and favorability toward the opposing party and candidate, as the independent variable. It was found that as respondents' affective favorability increased, i.e., as the variance in relative favorability increased, their satisfaction with democracy decreased.

As in the previous case, an analysis was conducted targeting only voters supporting the two major parties. As can be confirmed through [Table 8], among these individuals, affective polarization, i.e., a larger variance in relative favorability, was associated with a more negative attitude toward democracy. Furthermore, the magnitude of the coefficient values for the two affective polarization variables increased compared to [Table 7] above, which ultimately signifies that the behavior linking affective polarization and satisfaction with democracy is occurring, particularly among supporters of the major parties. In summary, among supporters of the two major parties, a different behavior pattern emerges where a negative attitude toward democracy is formed due to a negative attitude toward the opposing party, not the party they support.

[Table 7] Relative Favorability (Affective Polarization) Toward Parties and Candidates and Satisfaction with Democracy

OLS Regression Analysis

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

*** p<0.001, ** p<0.01, * p<0.05, † p<0.1

Reference Category: Democratic Party's Sense of Party Identity

[Table 8] Affective Polarization (Party) and Satisfaction with Democracy

OLS Regression: Supporters of the Democratic Party and the People Power Party

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

*** p<0.001, ** p<0.01, * p<0.05, † p<0.1

Note: Control variables identical to those in the analysis for [Table 7] are included but not reported in this table.

5. Conclusion

The analytical results presented in this paper illuminate the intricacies of the 20th presidential election, often characterized as an "election of dislike" and a choice of the lesser evil. The candidates from the two major parties were both embroiled in scandals during the campaign, creating a surface impression of widespread negative sentiment towards both parties and candidates. Consequently, predictions persisted that voters would choose the candidate they disliked less, the "lesser evil," leading to the prevailing assessment that this election unfolded in a unique manner compared to previous ones. However, the "election of dislike" and the "choice of the lesser evil" likely occurred only to a limited extent among supporters of minor parties and unaffiliated voters who did not support the two major parties. Partisan voters, as in any election, made their "best" choice based on partisan "liking," with strong dislike directed towards the opposing party being a cross-cutting sentiment. This implies that, on average, the observed high levels of dislike were solely generated by supporters of the opposing party, while strong preferences for their own party and candidate remained robust. Nevertheless, when examined without regard to party preference, high levels of liking and dislike for each target object offset each other, resulting in similar degrees of negative sentiment between parties and candidates. Ultimately, in the 20th presidential election, where negative sentiment towards both candidates was prevalent, the assessment of an "election of dislike" masks a more significant phenomenon: affective polarization (Iyengar et al. 2012), characterized by strong liking for one's own party and strong dislike for the opposing party, particularly among partisan voters.■

References

Kang, Won-taek. 1998. “Political Expectation Level and Protest Voting: The Logic of Support for a Third Party under a Simple Plurality System.” *The Journal of Korean Political Studies* 32(2), 191-210.

Gil, Jeong-ah, and Sang-eung Ha. 2019. “Attribution of Responsibility Based on Partisan Bias: Focusing on Perceptions of Inter-Party Conflict and Party Liking.” *Korean Journal of Legislative Studies* 25(1), 46-78.

Jang, Seung-jin, and Jeong-gyu Seo. 2019. “The Dual Structure of Partisan Polarization: Political Identity, Policy Preferences, and Political Sophistication.” *Korean Association of Party Studies* 18(3), 5-29.

Jang, Seung-jin, and Han-il Jang. 2020. “Non-Political Effects of Partisan Polarization.” *The Journal of Korean Political Studies* 54(5), 153-175.

Abramowitz, Alan I. 2010. The Disappearing Center: Engaged Citizens, Polarization, and American Democracy. New Haven: Yale University Press.

Abramowitz, Alan I., and Kyle L. Saunders. 1998. “Ideological Realignment in the US Electorate.” Journal of Politics 60(3), 634-652.

Abramowitz, Alan, and Steven Webster. 2016. “The Rise of Negative Partisanship and the Nationalization of U.S. Elections in the 21st century.” Electoral Studies 41, 12-22.

Anderson, Christopher J., Andre Blais, Shaun Bowler, Todd Donovan, and Ola Listhaug. 2005. Losers’ Consent: Elections and Democratic Legitimacy. Oxford University Press.

Anderson, Christopher J. & Andrew J. LoTempio. 2002. “Winning, Losing and Political Trust in America.” British Journal of Political Science 32, 335-351.

Anderson, Christopher J. and Christine A. Guillory. 1997. “Political Institutions and Satisfaction with Democracy: A Cross-National Analysis of Consensus and Majoritarian Systems.” American Political Science Review 91(1), 66-81.

Fiorina, Morris P., Samuel J. Abrams, and Jeremy C. Pope. 2008. “Polarization in the American Public: Misconceptions and Misreadings.” Journal of Politics 70(2), 556-560.

Fiorina, Morris P., Samuel J. Abrams, and Jeremy C. Pope. 2010. Culture War? The Myth of a Polarized America. New York: Pearson Longman.

Galston, William A. and Pietro S. Nivola. 2006. “Delineating the Problem.” in Nivola, Pietro S. and David W. Brady. eds. Red and Blue Nation? Characteristics and Causes of America’s Polarized Politics. Washington, D.C.: Brookings Institution Press, 1-47.

Hetherington, Marc. 1999. “The Effect of Political Trust on the Presidential Vote, 1968–96.” American Political Science Review 93(2): 311–326.

Himmelweit, Hilde, Patrick Humphreys and Marianne Jaegar. 1985. How Voters Decide. Milton Keynes: Open University Press.

Iyengar, Shanto, Gaurav Sood, and Yphtach Lelkes. 2012. “Affect, Not Ideology: Social Identity Perspective on Polarization.” Public Opinion Quarterly 76. No. 3, 405-431.

Kang, Won-Taek. 2004. “Protest Voting and Abstention under Plurality Rule Election: An Alternative Public Choice Approach.” Journal of Theoretical Politics 16(1): 79-102.

Southwell, Priscilla L. 1998. “The Electoral Consequences of Alienation: Nonvoting and Protest Voting in the 1992 Presidential Race.” Social Science Journal 35: 44-51.

Webb, Paul. 2005. “The Continuing Advance of the Minor Parties.” Parliamentary Affairs 58(5): 757–775.

Yonhap News. 2022. “Record-High Disliked Election, Record-High Close Race... 20th Presidential Election Shatters Records.” March 10. https://www.yna.co.kr/view/AKR20220310078300001?input=1195m

Chosun Ilbo. 2021. “Presidential Election Descends into a Contest of the 'Lesser Evil'... Will it be a Record-High 'Disliked Election'?” October 10. https://biz.chosun.com/distribution/channel/2021/10/10/QDKLGA2HZRBPHD5USEYRFBRTLQ/

Korea Economic Daily. 2022. “‘Close Contest, Divided Presidential Election’... ‘Election of the Lesser Evil’.” March 10. https://n.news.naver.com/article/015/0004672318

Hankook Ilbo. 2021. “The Presidential Election is Strange... Lee Jae-myung and Yoon Suk-yeol's Approval Ratings Remain 'Untouchable' Despite Scandals.” October 5. https://www.hankookilbo.com/News/Read/A2021100417430002584?did=NA

Reuters. 2022. “S.Korean Voters Hold Noses as Rivals Land Low Blows in 'Unlikeable Election.'https://www.reuters.com/world/asia-pacific/skorean-voters-hold-noses-rivals-land-low-blows-unlikeable-election-2022-02-03/

The Korea Herald. 2021. “‘Lesser Evil’? Disapproval of Presidential Candidates Rises.”http://www.koreaherald.com/view.php?ud=20211024000193

■ Author: Gil Jeong-ah_Research Professor, Institute of Government Studies, Korea University. Teaches voter political attitudes, Korean politics, and quantitative analysis. Holds a Ph.D. in Political Science from Seoul National University. Primary research interests include voter political attitudes, partisan polarization, and the political accountability of representative democracy. Has published articles in numerous journals, including Social Science Research, Korean Political Science Review, Korean Party Studies Review, and Korean Journal of Legislative Studies.

■ Editor: Jeon Ju-hyeon _EAI Research Fellow

    Contact: 02 2277 1683 (ext. 204) | jhjun@eai.or.kr