Voter Authoritarianism in the 20th Presidential Election

1. Introduction

The 20th presidential election proceeded under the stigma of being the 'most unlikable election ever.' The two main candidates, Lee Jae-myung of the Democratic Party of Korea and Yoon Suk-yeol of the People Power Party, both outsiders or from non-mainstream factions within their parties, secured their nominations by rallying partisan supporters with their distinct, extreme, and authoritarian styles. The election campaign devolved into a mudslinging contest filled with scandals and negative attacks against opponents. A search of news articles from the year preceding the election (March 10, 2021 - March 9, 2022) using the news big data analysis service 'Big Kinds' revealed 33 articles mentioning Lee Jae-myung in conjunction with the term 'authoritarian,' and 118 articles mentioning Yoon Suk-yeol with the same term. Notably, Yoon Suk-yeol, a former Prosecutor General who ran on a platform of regime change, ultimately won the election with his charismatic persona, becoming the president-elect. Thus, the 20th presidential election can be characterized by the authoritarian aspects of the two main candidates' words, actions, tendencies, and policies. How, then, did voters respond to these authoritarian political styles?

The rise of authoritarianism is not unique to South Korea. Since the 2000s, the emergence of authoritarian leaders and parties, along with citizens who support them, has been observed globally in Europe, the United States, Latin America, and elsewhere (Jeong Dong-jun 2018, 2020; Cohen and Smith 2016; Conway and McFarland 2019; Donovan 2019; Steiner and Hillen 2021). In particular, analyses of the factors behind Donald Trump's election in the United States in 2016 suggested that voters with right-wing authoritarian tendencies coalesced around the far-right Trump (Choma and Hanoch 2017; Conway and McFarland 2019; Donovan 2019; Ludeke et al. 2018; MacWilliams 2016). In other words, behind the phenomenon of rising authoritarianism lie voters who lend their support to these authoritarian leaders. Recent research trends focus on individual-level 'authoritarian attitudes' to explain this bottom-up authoritarianism. Specifically, in relation to voting behavior, authoritarian attitudes are identified as a variable that independently influences voters' choices, separate from ideological leanings and policy stances (Conway et al. 2018; Conway and McFarland 2019; Cornelis and Van Hiel 2015; Choma and Hanoch 2017; Donovan 2019; Undzenas et al. 2021, etc.).

To what extent, then, do South Korean voters possess these authoritarian tendencies, and how did these tendencies influence their voting choices in the 20th presidential election? More specifically, to what degree do authoritarian attitudes manifest at the citizen level in South Korea, and in which groups are they particularly pronounced? What are the backgrounds that shape these authoritarian tendencies? And which candidate, Lee Jae-myung or Yoon Suk-yeol, received more votes from voters with high authoritarian tendencies in the 20th presidential election? To answer these questions, this study aims to measure citizens' authoritarian attitudes through a survey and statistically analyze their impact on voting choices. While attempts to identify individual-level authoritarian attitudes and explain their political characteristics and voting behavior have been undertaken in recent Korean academic research (Kwon Soon-hwan and Park Sang-hyun 2021; Lee Bo-mi and Ha Sang-eung 2018; Ha Sang-eung and Lee Bo-mi 2017), they remain insufficient. Research on authoritarian attitudes holds significant implications in relation to the 'crisis of democracy,' a topic of intense academic interest recently (Mounk 2018; Jeong Dong-jun 2020; Levitsky and Ziblatt 2018; Przeworski 2019). In a democratic system based on the will of the majority and democratic values, the sustainability of the system is precarious if the majority does not desire democracy or adheres to authoritarian values and norms. Therefore, understanding the extent and influence of current authoritarian attitudes among South Koreans is crucial for assessing the current state of South Korean democracy and charting a path forward.

2. Theoretical Discussion

① Concept of Authoritarian Attitudes

Research on individual-level authoritarian attitudes traces its origins to the work of Adorno et al. (1950). Based on Freudian theory, Adorno and his colleagues studied authoritarianism as a personality type, primarily to understand the attitude of prejudice. However, contemporary research on authoritarianism approaches it not as a personality type strongly influenced by genetic factors, but as a domain of 'attitudes' formed in conjunction with subsequent environmental factors (Ha Sang-eung and Lee Bo-mi 2017; Conway et al. 2021). Many scholars consider Altemeyer's (1996, 1998) work as the definitive starting point for modern research on authoritarian attitudes.

Altemeyer defines authoritarian attitudes through three sub-concepts (Kwon Soon-hwan and Park Sang-hyun 2021; Ludeke et al. 2018). First is 'authoritarian aggression,' which refers to the tendency to desire firm responses to phenomena that disrupt social order, such as crime, political, and social unrest. This aggression leads to attitudes such as advocating for harsh punishments for criminals or opposing immigration. Second is 'conventionalism,' the tendency to dislike change and respect tradition and conservative values. This tendency is linked to opposing progressive social phenomena like abortion and homosexuality by upholding existing social, religious traditions, and family values. Third is 'authoritarian submission,' the tendency to obey and conform to the authority and hierarchical order imposed by society. This submissive tendency manifests as an attitude of respecting existing political institutions and social order and opposing resistance to authority.

Individual-level authoritarian attitudes are almost exclusively discussed in relation to right-wing conservatism (Kwon Soon-hwan and Park Sang-hyun 2021; Lee Bo-mi and Ha Sang-eung 2018; Ha Sang-eung and Lee Bo-mi 2017; Conway and McFarland 2019; Conway et al. 2021; Cornelis and Van Hiel 2015; Duckitt et al. 2010; Duckitt 2013; Ludeke et al. 2018). In Western capitalist societies, in particular, the term 'Right-Wing Authoritarianism' has often been used as a synonym for authoritarian attitudes. According to Conway et al.'s (2018) meta-analysis of authoritarian attitude scales, approximately 69% of recently published studies use the Right-Wing Authoritarianism scale for measurement (1051-53).[1]

The close association between authoritarian attitudes and conservative ideology can be understood in two ways. First, the inherent characteristics of authoritarian attitudes align with the values of conservatism. While the values of conservatism may vary by time and place, they generally share common traits such as opposing social change, respecting traditional values, and tolerating hierarchical order and unequal structures (Conway et al. 2018; Jost et al. 2003). Considering that authoritarian attitudes, as previously mentioned, possess characteristics of 'conventionalism'—disliking change and upholding traditional values and order—and 'submission'—conforming to societal authority and hierarchy—there is a conceptual overlap between the two. Second, consideration must be given to the causes underlying the manifestation of authoritarian attitudes. Existing research on authoritarian attitudes posits that 'perception of threat' is a primary background factor that increases authoritarian levels at the individual level (Choma and Hanoch 2017; Conway and McFarland 2019; Duckitt et al. 2002; Duckitt et al. 2010; Feldman 2003). The perception that the world is a dangerous place, and particularly the belief that one's in-group's values and socioeconomic status are threatened by out-groups, strengthens the motivational goal of pursuing in-group cohesion and collective security, thereby shaping and reinforcing authoritarian attitudes. In other words, as the perception of threat increases, individuals tend to uphold existing values and order to ensure the survival of their group, leading to authoritarian attitudes that seek strong leaders and authority to resolve such threatening situations.

② Previous Research on Authoritarian Attitudes

At the individual level, authoritarian attitudes are understood as dispositions situated in the middle of the causal chain, formed by the interaction of genetic/temperamental factors, such as the 'Big-Five personality traits,' with various environmental factors, influencing our political behavior and attitudes (Ha Sang-eung and Lee Bo-mi 2017). In this regard, authoritarian attitudes are sometimes referred to as 'characteristic adaptations' (Ha Sang-eung and Lee Bo-mi 2017; McAdams and Pals 2006) or considered 'social attitudes' based on a society's values and norms (Duckitt et al. 2010; Conway et al. 2021). Thus, authoritarian attitudes function as a dependent variable shaped by various factors like an individual's genetic disposition or socioeconomic background, while simultaneously acting as an independent variable influencing diverse types of political behavior.

Factors influencing the formation of authoritarian attitudes as a dependent variable include the aforementioned perception of threat. The most representative political behavior influenced by authoritarian attitudes as an independent variable is voting choice. Specifically, right-wing authoritarianism, which possesses both authoritarian attitudes and socio-cultural conservative attributes, has been found to increase the probability of voting for far-right parties and politicians emerging in many democratic countries in Europe, the Americas, and Latin America today (Choma and Hanoch 2017; Cohen and Smith 2016; Conway and McFarland 2019; Cornelis and Van Hiel 2015; Donovan 2019). Furthermore, it is known to influence referendums, such as the Brexit vote (Undzenas et al. 2021), and also non-conventional political participation beyond voting (Kwon Soon-hwan and Park Sang-hyun 2021).

In addition to voting behavior, authoritarian attitudes are closely related to various types of political attitudes. 'Rigid attitudes' such as prejudice and dogmatism (Conway et al. 2018), intolerant attitudes towards those who undermine in-group order and cohesion (Crawford and Pilanski 2014), support for populist parties and populist attitudes (Akkerman et al. 2017; Vasilopoulos and Jost 2020), support for institutions that suppress civil liberties (Lee Bo-mi and Ha Sang-eung 2018), and opposition to social redistribution policies (Jedinger and Burger 2018) have all been found to increase with authoritarian attitudes.

While some studies on citizen-level authoritarian attitudes have been conducted in Korean academia, they remain limited. Recent studies on South Koreans' authoritarian attitudes include those by Ha Sang-eung and Lee Bo-mi (2017), Lee Bo-mi and Ha Sang-eung (2018), and Kwon Soon-hwan and Park Sang-hyun (2021). Ha Sang-eung and Lee Bo-mi (2017) found that stronger right-wing authoritarianism and 'Social Dominance Orientation' led to higher favorability towards conservative parties and politicians, while weaker levels led to higher favorability towards progressive parties and politicians. Lee Bo-mi and Ha Sang-eung (2018) argued that individuals with strong right-wing authoritarian tendencies are more likely to support institutions that restrict civil liberties. Kwon Soon-hwan and Park Sang-hyun (2021), who studied the relationship between right-wing authoritarianism and political participation, found that right-wing authoritarianism negatively impacts all types of political participation, regardless of conventional or unconventional participation. Although these studies present meaningful findings regarding the political characteristics of right-wing authoritarianism among South Koreans, their number is absolutely insufficient. Furthermore, they have not empirically analyzed voting choice, a political behavior known to be significantly influenced by authoritarian attitudes. This study aims to complement existing research by identifying the authoritarian attitudes present among South Korean citizens and analyzing how these attitudes translated into choices in the 20th presidential election.

3. Data and Variables

This analysis utilizes data from the '2022 EAI Presidential Panel Survey,' conducted by Korea Research and commissioned by the East Asia Institute (EAI). The survey was administered via telephone interviews in two waves before and after the presidential election (Wave 1: January 12-15, 2022; Wave 2: March 10-15, 2022). A sample of adults aged 18 and above nationwide was randomly selected, yielding 1,515 responses in Wave 1 and 1,104 responses in Wave 2. However, this analysis does not utilize the longitudinal aspect of the panel survey, as it does not analyze changes between the two waves. When questions from both Wave 1 and Wave 2 were used for independent variables, the results are presented distinguishing between them.

To examine the impact of authoritarian attitudes, the core analysis of this study, on voting choices, 'voting choice' was first used as the dependent variable. For intuitive interpretation, voting choice was operationalized as a binary variable: '1' for voting for Yoon Suk-yeol, the winner of the recent election, and '0' for voting for Lee Jae-myung. While a more precise analysis would ideally consider votes for other candidates or non-voting behavior, only 29 respondents (2.69% of the total) voted for 'other candidates,' and 43 respondents (3.95%) indicated they 'did not vote' among the 1,093 respondents who answered the relevant question in this survey.[2]For analytical convenience, these respondents were treated as missing values.[3]A binary variable of 0 (Lee Jae-myung vote) and 1 (Yoon Suk-yeol vote) was constructed from the remaining respondents.

Next, 'authoritarian attitudes,' the core variable of this analysis, was used as the independent variable. Altemeyer's 'Right-Wing Authoritarianism scale' (Altemeyer 1996, 1998), mentioned earlier, is actively used in many studies to measure authoritarian attitudes (Conway et al. 2018). Based on the three sub-concepts of authoritarianism he defined, Altemeyer constructed a total of 30 items covering politics, economics, society, and culture. These 30 items continue to serve as a benchmark for measuring authoritarian attitudes, with subsequent studies using them in whole or in part[4]or with some modifications to the wording[5]The 'child-rearing values' scale developed by Feldman and Stenner (1997; Stenner 2005) is another frequently used scale for authoritarian attitudes. This scale consists of questions asking how important it is to raise children to obey parental and societal authority compared to raising them to be independent and autonomous.[6]

Unfortunately, this survey does not include items based on previous research. However, similar to the logic underlying those studies, items related to the sub-concepts of authoritarianism—strong adherence to law and order ('aggression'), respect for tradition and conservative values ('conventionalism'), and conformity to authority and social systems ('submission')—were identified to construct an authoritarian attitude index. The 8 items initially considered for this analysis are presented in Table 1 below. However, based on Cronbach's alpha analysis, which assesses the internal validity of the index through the correlation between items and the latent variable, three items (numbers 3, 4, and 8) showed relatively low item-rest correlation coefficients and were excluded from the index construction (Nunnally and Bernstein 1994).[7]Therefore, using the remaining 5 items, an index was constructed by assigning 1 point when a respondent selected an option highly associated with authoritarianism (refer to the fifth column of the same table) (a 6-point scale from 0 to 5, with higher scores indicating greater authoritarianism).

[Table 1] Survey Items Considered for Constructing the Authoritarian Attitude Index

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

Note) * indicates items used in the first survey.

[Table 2] Summary Statistics of Variables Used in the Analysis

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

Note: * indicates items used in the first survey.

The analysis proceeds as follows. First, a series of descriptive analyses will be conducted to ascertain the extent to which authoritarian tendencies are present among our citizens and in which groups they are more pronounced. Subsequently, the analysis will examine the background factors contributing to these authoritarian tendencies, with a particular focus on the threat perception theory explained earlier. Finally, to examine the core analysis of this study—the impact of authoritarian tendencies on voting choices—a multivariate regression analysis will be performed, controlling for all the variables presented above.

4. Analysis Results

① Distribution of Authoritarian Tendencies

First, we examined the extent to which respondents' authoritarian tendencies were manifested in this survey. The distribution (%) of respondents for each value of the authoritarian tendency index, ranging from 0 to 5, is shown in [Figure 1]. The majority of respondents were concentrated in the middle scores of '2' and '3', while scores of '4' and '5' combined accounted for 16.7% of the respondents. It is difficult to directly compare the magnitude of this proportion as there are no other studies that have constructed an index using the same items. However, some studies measuring authoritarian tendencies using Altemeyer's Right-Wing Authoritarianism scale show the following average scores: Ludeke et al. (2018) reported an average of 0.45 on a 0-1 point scale for US citizens, Ha Sang-eung and Lee Bo-mi (2017) reported an average of 0.42 on a 0-1 point scale for Koreans, and another study by Lee Bo-mi and Ha Sang-eung (2018) reported an average of 3.84 on a 1-7 point scale (which simplifies to 0.47 on a 0-1 point scale). Considering that the average value of this index, when simplified to a 0-1 point scale, is 0.41, the observed level of authoritarian tendencies can be considered relatively high.

[Figure 1] Distribution of Respondents by Authoritarian Tendency Index (%)

An examination of which social groups exhibit higher authoritarian tendencies revealed, as shown in [Table 3], that authoritarian tendencies differed significantly by gender, age, ideology, and party affiliation, and these differences were statistically significant. First, regarding gender, men (2.23) had a higher average authoritarian index than women (1.85). By age group, individuals in their 20s and 30s showed higher authoritarian tendencies compared to other age groups. This result is somewhat surprising considering that authoritarian attitudes typically strengthen with age (Ludeke et al. 2018). When observing that those aged 60 and above recorded an average of 2.04, similar to the overall average, the authoritarian tendencies of those in their 20s and 30s, particularly the 20s, are notably high. This appears to be related to the recent trend of conservatism among younger generations, which has become an issue in Korean society. Continuous observation is necessary, as these authoritarian tendencies, if sustained, could increase the overall level of authoritarianism in Korean society.

Regarding the distribution by ideology, authoritarian tendencies were significantly higher among conservatives (scores 6-10 on an 11-point scale) than among progressives (scores 0-4 on an 11-point scale). For conservatives, the average authoritarian index was 2.80, nearly three times higher than that of progressives (1.04). The statistical significance through the ANOVA model also showed the highest F-value among the surveyed variables, at 162.56. This is likely because authoritarian tendencies themselves are deeply intertwined with right-wing conservatism, as described above. This is because the conservative values of respecting tradition and upholding hierarchical social order share many connections with the authoritarian items used in this survey. Furthermore, the term 'authoritarian government' in the item 'preferred political system,' which is one of the items used to construct the index, almost exclusively evokes 'right-wing authoritarian government' within the historical context of South Korea, which could partially explain these results. In a similar vein, supporters of the conservative People Power Party (2.94) showed higher authoritarian tendencies than supporters of the Democratic Party (1.16). Given that ideological factors strongly influence party affiliation, these results can also be explained by the relationship between authoritarianism and conservatism described above.

[Table 3] Authoritarian Tendency Index by Socio-Demographic Variables

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

Note: *** p-value<0.01, ** p-value<0.05, * p-value<0.1

② Background Factors of Authoritarian Tendencies

If so, where do these authoritarian tendencies originate? In other words, what are the causes that form these authoritarian tendencies? As discussed in the theoretical section, among the many factors explaining authoritarian tendencies, scholars particularly focus on the perception of threat. The correlation between threat perception and authoritarian tendencies has been demonstrated through various studies (Choma and Hanoch 2017; Conway and McFarland 2019; Duckitt et al. 2002; Duckitt et al. 2010; Feldman 2003). In the current Korean context, threats that can be perceived by the in-group can be broadly categorized into threats due to economic hardship and health threats due to COVID-19. That is, the more one feels that their personal or national economic situation has worsened, and the more they perceive the government's response to COVID-19 as inadequate, the greater the threat they feel, which can lead to an increase in authoritarian tendencies. Based on these considerations, we examined the average authoritarianism index according to retrospective evaluations of household and national economies over the past five years and the government's COVID-19 response. The analysis results confirmed this prediction. As presented in [Table 4], the authoritarianism index was higher as the evaluations of household and national economies and the government's COVID-19 response became more negative. These differences were statistically significant in all cases.

[Table 4] Authoritarianism Index by Economic Evaluation and Government COVID-19 Response Evaluation

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

Note: *** p-value<0.01, ** p-value<0.05, * p-value<0.1

3. Authoritarian Tendencies and Voting Choice

To examine the influence of these authoritarian tendencies on voters' choices in the 20th presidential election, a multivariate regression analysis was conducted, including control variables. Given that the dependent variable is dichotomous, a nonlinear logistic regression model was employed. An initial analysis of the basic model, which included only control variables believed to influence voting behavior, yielded the results presented in Model 1 of [Table 5]. The analysis indicated that older age, conservative ideological orientation, support for the People Power Party, and a negative evaluation of President Moon Jae-in's and the current government's COVID-19 response were statistically significant predictors of a higher probability of voting for Yoon Suk-yeol. Conversely, individuals from the Honam region and supporters of the Democratic Party of Korea showed a significantly higher probability of voting for Lee Jae-myung. Contrary to expectations, however, socioeconomic background variables such as gender, education, and income, [8]as well as the Yeongnam region of origin and evaluations of household and national economic conditions, did not show statistically significant influence. The regionalism voting trend in the Yeongnam region, particularly Busan, Ulsan, and Gyeongnam, has been notably observed since the 2018 local elections (Kang, 2019; Jeong, 2018; Jeong & Lee, 2018), and this trend was observed to some extent in this presidential election as well. Furthermore, the influence of economic evaluations was relatively weaker compared to presidential and COVID-19 evaluations, suggesting that the judgment of the Moon Jae-in administration[9]and the COVID-19 crisis served as major issues in this election.

As the logistic model is nonlinear, the magnitude of the coefficients alone does not accurately reflect the degree of influence. Therefore, by calculating the odds ratio (in parentheses in the table), which represents the increase in the probability of the dependent variable changing from 0 to 1 for a one-unit increase in the independent variable, party affiliation emerged as the variable with the strongest influence among those with statistically significant results. The party affiliation variable, surveyed as 'Party Supported,' showed odds ratios of '9.32' for the People Power Party and '0.28' for the Democratic Party of Korea, respectively, indicating an overwhelming influence compared to other variables, even when considering the scale differences.[10]Following this, ideological orientation showed a significant odds ratio (1.46), revealing that this presidential election was heavily influenced by party affiliation and ideology more than other socioeconomic background variables. This aligns with the phenomenon of partisan polarization observed in democratic countries worldwide, including South Korea, in recent years (Gil & Ha, 2019; Jeong, 2018; Jang & Seo, 2019; Dinkelberg et al. 2021; Fiorina 2017). Consequently, how to reconcile these strong partisan and ideological conflicts and manage state affairs will be a crucial task for the next administration.

[Table 5] Logistic Regression Analysis of Voting for Yoon Suk-yeol

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

Note) *** p-value<0.01, ** p-value<0.05, * p-value<0.1

Next, we analyzed the data by adding the authoritarianism tendency index, the core variable of this analysis. As shown in Model 2, the authoritarianism tendency index (bolded in the table) showed a statistically significant positive coefficient, indicating that respondents with stronger authoritarian tendencies were more likely to vote for candidate Yoon Suk-yeol than candidate Lee Jae-myung. The odds ratio was 1.36, which, even considering the difference in scale between variables, was analyzed to have a high level of influence, following partisanship and ideological orientation. Figure 2, which graphically represents the results of Model 2, shows that as authoritarian tendencies increase, the probability of voting for candidate Yoon Suk-yeol increases in a nearly linear upward trend. This result appears to be because authoritarian tendencies themselves show a high correlation with conservatism, as mentioned earlier. In other words, the conservative values of valuing tradition and upholding the hierarchical order of society are also connected to the concept of authoritarianism; thus, it can be interpreted that voters with high authoritarian tendencies cast their votes for the conservative candidate, Yoon Suk-yeol. However, as argued in some studies (Dusso 2016; Ludeke et al. 2018), authoritarian tendencies were not synonymous with or encompassed by conservative ideology. The influence of authoritarian tendencies remained significant and high even when controlling for ideology and partisanship variables, as in Model 2. In other words, it can be said that the influence of authoritarian tendencies, which operate independently of ideology or partisanship in voting choices, manifested in this presidential election.

[Figure 2] Probability of Voting for Candidate Yoon Suk-yeol by Authoritarianism Tendency (95% Confidence Interval)

5. Conclusion and Discussion

This study examined the degree of authoritarian tendencies among Korean citizens through a survey for the 20th presidential election, exploring the background of these tendencies and their influence on voting choices. The analysis revealed a considerable level of authoritarian tendencies, even when compared to similar previous research. These tendencies were particularly high among men, the younger generation (20s-30s), ideological conservatives, and supporters of the People Power Party. Authoritarian tendencies are known to strengthen when individuals perceive threats more acutely; this survey also found that authoritarian tendencies increased when respondents evaluated the economic situation and the government's COVID-19 response negatively. Finally, the analysis of voting choices indicated that authoritarian tendencies had an independent and significant impact even after controlling for other variables. Through logistic regression analysis, voters' authoritarian tendencies significantly increased the probability of voting for candidate Yoon Suk-yeol, and the magnitude of this influence was the second largest after partisanship and ideology variables. These results demonstrate that authoritarian tendencies are closely related to conservative ideology.

Despite these empirical findings on the authoritarian tendencies of Koreans, this study has the following limitations. First, it failed to consider the variation in authoritarianism levels between candidates in the dependent variable. Voters with authoritarian tendencies are more likely to vote for candidates who are more authoritarian; therefore, the hypothesis regarding this relationship can only be established if it is clear which candidate is relatively more authoritarian. However, this analysis, which constructed the dependent variable solely based on choices between the two candidates, did not accurately verify this hypothesis. To address this, a separate analysis of which of the two candidates is more authoritarian, or a survey asking respondents' subjective perceptions of this, would be necessary. Unfortunately, the former is beyond the scope of this analysis, and the latter could not be performed because the relevant question was not included in the survey. Second, there is the issue of measurement. As mentioned in the main text, the items used in this analysis to construct the authoritarianism tendency index are not established items from previous research. Although the index was constructed based on the sub-concepts of authoritarianism covered in previous studies as much as possible, and the results were not significantly different from those of other studies, this measurement issue needs to be improved for better comparative research. Lastly, as mentioned earlier, the consideration of left-wing authoritarianism was insufficient. While authoritarian tendencies are closely related to the right-wing, they are not exclusive to it. Depending on the situation, authoritarian tendencies can also be observed in progressive groups. Therefore, future research needs to consider left-wing authoritarianism and reflect this in measurement and analysis for studies targeting Koreans.

Despite these limitations, the findings of this study offer certain implications for contemporary Korean politics. Many scholars argue that the rise of authoritarianism today is not solely the responsibility of political elites but also of the authoritarian-minded citizens who support them. This study's findings, which show a high level of authoritarian tendencies and a significant impact on voting behavior, demonstrate that South Korea is no exception. Amidst deepening political polarization through several recent changes in government, political parties and politicians exhibiting more extreme and authoritarian leadership are gaining significant support. Particularly in the primary election process, where the influence of partisan supporters is strong, candidates with more extreme stances repeatedly achieve victory. The finding that these tendencies are high among the younger generation (20s-30s) is also a cause for concern, foreshadowing a potential worsening of this trend in the future. In a democracy where the majority opinion and democratic norms are crucial, if the majority of the public cheers for leaders who make policy decisions and solve problems in an authoritarian rather than a democratic manner, the democratic system itself cannot be guaranteed. We have already experienced the collapse of numerous nascent democracies in the former Soviet Union, Eastern Europe, Latin America, and Southeast Asia. Amidst the rise of authoritarianism from both the top and bottom, we must be aware that we are not an exception and remain vigilant. ■

[1] Recently, studies suggesting that authoritarian tendencies are not exclusive to the right-wing have emerged, drawing attention to the concept of 'Left-Wing Authoritarianism' (Conway et al. 2018; Conway and McFarland 2019; Conway et al. 2021). However, this falls outside the scope of the current study's topic, and a detailed discussion is omitted.

[2] Considering that only 3.95% of respondents answered 'did not vote,' this is a very small figure given that the final voter turnout for this presidential election was 77.1%. This suggests that the common phenomenon of over-reporting of voter turnout in survey research (Lee Hyun-woo and Jeon Si-hong 2010; Duff et al. 2007) also occurred in this survey, which warrants caution in interpreting the analysis results.

[3] The analysis results did not change significantly even when missing values were treated as '0' instead of being excluded.

[4] Sibley et al. (2007) used 10 out of 30 items, and these ten items were also utilized in domestic research by Ha Sang-eung and Lee Bo-mi (2017). Manganelli Rattazzi et al. (2007) also reduced Altemeyer's questionnaire items to a total of 14 items, consisting of 7 items on aggression/submission and 7 items on conservatism. The former 7 items were included in the 2016 Korean General Social Survey and have been used in studies by Kwon Soon-hwan and Park Sang-hyun (2021), Lee Bo-mi and Ha Sang-eung (2018), among others.

[5] For example, Conway and colleagues (Conway et al. 2018, etc.), who are conducting pioneering empirical research on left-wing authoritarianism, modified the right-wing expressions in 20 items from the existing Altemeyer scale to left-wing expressions to simultaneously measure both right-wing and left-wing authoritarian tendencies.

[6] The child-rearing values scale is criticized for focusing only on 'obedience,' one of the sub-concepts of authoritarian tendencies (Ludeke et al. 2018).

[7] When using the selected 5 items, the Cronbach's alpha value was 0.56, which was higher than 0.44 when using all 8 items. Of course, 0.56 cannot be considered high by conventional standards, but the related items included in this survey were limited, and since these items were considered to have sufficient theoretical relevance to the concept of authoritarianism, the index was constructed based on these, albeit insufficient, items.

[8] Although this presidential election was noted for conflicts between generations and genders, various tests were conducted on different generations (20s, 20s-30s, 60s and older), gender variables, and interaction variables between each item, but most yielded no significant results. In the case of gender variables, when variables such as ideology, partisanship towards the People Power Party, and presidential evaluation were added, their significant influence disappeared. This suggests that political orientation and evaluation, rather than gender itself, played a more significant role. Of course, this may be a result of over-reporting of voter turnout, and further confirmation with other research findings may be necessary.

[9] In the survey item 'This presidential election is a referendum on the Moon Jae-in administration,' 52.6% of respondents agreed (scored 6 or higher on a scale of 0: 'Strongly Disagree' to 10: 'Strongly Agree').

[10] The specific interpretation is as follows: supporting the People Power Party increases the probability of voting for candidate Yoon Suk-yeol by 9.32 times (832% increase), while supporting the Democratic Party of Korea decreases the probability of voting for the same candidate by 72%.

■ Author: Jeong Dong-junHe is currently a professor in the Department of Social Education at Inha University, teaching comparative politics, political processes, and political systems. He received his Ph.D. in Political Science (Comparative Politics) from the University of Florida and worked as a senior researcher at the Institute for Peace and Unification Studies at Seoul National University before joining Inha University. His main research interests include post-communist democratization, elections and parties, and civil society and political attitudes. He has published numerous articles in international and domestic journals, including Comparative Politics, Perspectives on Politics, and Electoral Studies.

■ Editor: Jeon Ju-hyeon_EAI 연구원

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