Representative democracy is currently experiencing a backlash of populism, which is based on the mass public’s disenchantment with the established political system (e.g., Müller, 2016). Many citizens do not believe that professional politicians appropriately incorporate their voices in the decision-making process, and even cast doubts on the legitimacy of democratic governance itself. Scholars and pundits have extensively reported the causes and consequences of right-wing populism in western Europe (e.g., Mudde, 2013) and in the United States (e.g., Oliver & Rahn, 2016) and left-wing populism in southern Europe (e.g., Gerodimos, 2015), Latin America (e.g., Seligson, 2007), and the Philippines (e.g., Curato, 2017). This wave of populism is, to a lesser degree as of now, observed in other parts of the world, including the East Asia (e.g., Yoshida, 2019).
In an era of populism, it is essential to understand how to build and restore public support for the political system in a democratic country. This paper examines the effects of individuals’ personality traits on political support, three components of which are, following Easton (1975), confidence in political elites, confidence in government, and national pride. As an attempt to identify dispositional, psychological determinants of political support, this paper is in line with a well-established body of literature on the relationship between personality and political behavior (e.g., Gerber et al., 2011). Studying individuals’ personality in understanding political support can be justified because the ebbs and flows of political support depend on the communication between political elites and ordinary citizens. Identifying demographic characteristics (such as age, generation, income, or region) of people who are in favor of or against the current political system definitely helps in promoting trust in government. However, to find out individuals’ dispositional traits leading to higher levels of confidence and support will be even more helpful to deliver well-tailored political messages appealing to those high on those traits.
Theoretical Arguments and Hypotheses
Political Support: A Theoretical Framework
Citizens’ support for the political system is a backbone of democratic governance. In ideal conditions, citizens should be satisfied with the political process, where their own interests are well-represented by elected officials and should harbor high levels of a sense of belonging to their political community (Norris, 2011). However, in reality, citizens’ attitudes toward their governments often tend to be characterized by skepticism or cynicism, which is, needless to say, detrimental to the stability and legitimacy of democratic governance (Booth & Seligson, 2009; Mattes & Bratton, 2007). However, a careful observation leads us to detect considerable heterogeneity among citizens in expressing political support. Some citizens have developed strong attachments to political institutions, while remaining unconvinced by the performance or integrity of specific political leaders (e.g., the president). Conversely, others demonstrate favorable attitudes toward elected officials, but doubt whether the political process itself is appropriately designed and functioning to fully incorporate the will of the people.
A clear conceptualization of political support is thus necessary. Drawing on Easton (1975) theoretical framework, pre-existing research on political support has distinguished between two mutually related, yet distinct dimensions: (1) diffuse support, i.e., trust in regime and sense of belonging to a country; and (2) specific support, i.e., confidence in particular government officials and bodies.
Diffuse or generalized political support represents abstract feelings toward the country and its agencies. It helps citizens accept the legitimacy of the state and its officeholders, even when people are highly critical about incumbent party leaders or specific public policies. In this sense, diffuse political support is expected to be more stable and enduring, providing elected officials with the authority to act based on a long-term commitment and perspective.
When operationalizing diffuse political support in empirical research, it is necessary to distinguish regime support from the sense of belonging to a country. The former identifies attitudes toward the institutions and offices of government rather than the present officeholders—such as respect for the presidency rather than any particular president. The latter denotes the positive affect that the public feels toward their country, exemplified by national pride and identity (Haerpfer & Kizilova, 2014).
Specific support encompasses the public’s opinions toward the incumbents of political offices, or, in a broader sense, the pool of political elites from which government leaders are drawn. It is usually well-represented by retrospective or on-line evaluation of the performance of incumbent officeholders. In addition, as opposed to diffuse political support, specific support is malleable, depending on the changes of the political climate and other external factors such as terrorism and natural disaster.
Personality traits are defined as “dimensions of individual differences in tendencies to show consistent patterns of thought, feelings, and actions” (McCrae & Costa, 1990, p. 23). In measuring this elusive concept, the five-factor model, i.e., the “Big Five” model, has widely been used since the late 1980s. The Big Five model is rooted in a questionnaire-based lexical analysis, which employs factor analysis in order to identify several distinct groups of descriptors of personality characteristics. By doing so, the “Big Five” model is known to constitute five “broad domains, collectively representing a hierarchy that organizes and summarizes the vast majority of subsidiary traits” (Mondak, 2010, p. 25).
The five dimensions of personality are Agreeableness, Conscientiousness, Openness to Experience, Extraversion, and Emotional Stability (or, as its opposite, Neuroticism). Agreeableness means being compassionate, modest, accommodating, trusting, and cooperative. Conscientiousness mainly refers to a tendency to act dutifully, to show self-discipline, and to facilitate task-and-goal-directed behavior. Openness deals with the degree to which individuals are open-minded to new experiences. Extraversion refers to a tendency to seek the company of others and external stimulation. Finally, Emotional Stability chiefly refers to controlling negative emotions like anxiety, depression, anger, discontent, and irritation (Funder, 2008).
In the field of political science, there are now a vast amount of studies that report a direct, significant effect of personality on political behavior, including, but not limited to, political ideology, party identification, political efficacy, political discussion, vote choice, voter turnout, and non-electoral political participation (e.g., Gerber et al., 2011) for a review. It is also known that personality traits exert a direct or moderating influence on citizens’ opinion on a specific political issue such as the usage of military force in foreign affairs (Schoen, 2007), immigration (Vecchione et al., 2012), and attitudes toward political system such as direct democracy in Switzerland (Freitag & Ackermann, 2016).
A brief review of the Big Five personality traits suggests that personality should be associated with individuals’ political support. People high on Agreeableness usually show a caring and kind orientation toward other people, and therefore are more likely to excel in collaborative work and less likely to have interpersonal problems than those low on Agreeableness. The key facets (i.e., sub-dimensions) of Agreeableness are trust and compliance, which are expected to lead to higher levels of confidence in political figures and institutions in general. Meanwhile, people high on Conscientiousness, who are inclined to hard work, dutifulness, and order, are known to perform better in schools and workplaces and tend to live longer—presumably due to self-discipline and good lifestyle—than their counterparts. Two facets of Conscientiousness are dutifulness and order, which are also expected to be positively associated with political support. People high on Emotional Stability are also expected to demonstrate higher levels of political support because they are relatively free from discontent and anger initiated by external stimuli. However, hypotheses regarding Openness and Extraversion may not be clear in their directions. People high on Openness are curious and eager to digest information available in the political arena, and therefore are more likely or less likely to express political support, depending on the situations (such as political ideology of the elites or macro-economic conditions). Likewise, people high on Extraversion are gregarious and willing to be surrounded by other people, and therefore are exposed to diverse information, which can lead to lower or higher levels of political support.
A similar logic can be applied to the hypothesis regarding national pride (Wang & Weng, 2018). It is reasonable to hypothesize that both Agreeableness and Conscientiousness are associated with national pride, given the peculiar political context of South Korea. South Korea is one of the most racially and ethnically homogeneous countries in the world. Since the division between two Koreas in 1948, South Koreans have been expected to demonstrate a dedicated allegiance to their country to consolidate the legitimacy of its capitalist economy and democratic government vis-à-vis the communist regime of North Korea. Given that a strong sense of belonging to the country has been a social norm in South Korea, both Agreeableness—characterized by conflict avoidance (e.g., van de Vliert & Euwema, 1994)—and Conscientiousness—characterized by norm-abiding attitudes (e.g., Roccas et al., 2002)—are expected to positively correlate with national pride. Further, lower levels of stress, dissatisfaction, and depression characterized by emotionally stable individuals is also expected to be positively associated with national pride.
Method and Procedure
Data come from the 2009, 2011, and 2012 Korean General Social Survey (KGSS). The KGSS—a face-to-face, nationally representative survey—has been conducted every year since 2003, and its sampling procedure and interviewing methods are virtually identical to those of the General Social Survey (GSS) in the United States. The respondents include 1,599 participants in 2009 (response rate: 64%), 1,535 participants in 2011 (response rate: 61%), and 1,396 participants in 2012 (response rate: 56%).
Personality. The ten-item personality inventory (TIPI) (Gosling et al., 2003) was translated into Korean, and included in all three surveys to assess personality traits. Each item contained a self-reported statement that reflects the Big Five personality dimensions and was rated on a seven-point scale. Responses were rescaled to range from 0 to 1. The bivariate correlations between the couple of items that form each of the Big Five were modest, ranging from 0.24 to 0.45. Though not ideal, inter-item correlations between the two items used to measure each trait may be less informative of the items’ reliability, because the TIPI originally intended to cover a variety of “facets” of each dimension of personality with two pairs of adjectives (Gosling, 2009). Table 1 presents the descriptive statistics of the Big Five and correlations among them. It is notable that some of the five personality dimensions were negatively correlated in the Korean TIPI (see Appendix A for the questionnaire). This is presumably because of the Confucian tradition, where both Extraversion and Openness are likely to be viewed negatively (Ha et al., 2013). The descriptive statistics suggest that personality is stable—at least at the aggregate level—across the three survey years.
Confidence in Blue House. In order to measure specific political support, one survey item tapping self-reported confidence in the Blue House (the executive office and official residence of the South Korean president) was selected. This survey question was based on a three-point scale (1 = no confidence at all; 3 = strong confidence). Overall, the levels of confidence of participants remained low across three years (M = 1.61, SD = 0.64 in 2009; M = 1.71, SD = 0.63 in 2011; and M = 1.61, SD = 0.63 in 2012). In all three years, the president was a conservative politician, Lee Myung-bak.
Confidence in government. Another survey question asking public confidence in the government (presumably encompassing all three branches of the government) served as a proxy for regime support, a narrower type of diffuse political support. Like the case of confidence in the Blue House, participants consistently reported low levels of confidence in the government (M = 1.59, SD = 0.61 in 2009; M = 1.62, SD = 0.59 in 2011; and M = 1.63, SD = 0.61 in 2012).
National pride. One survey item was used to assess national pride, a broader type of diffuse political support. This item contained a self-reported statement that indicates how proud a participant is of being a Korean citizen on a four-point scale (1 = not proud at all; 4 = very proud). A large proportion of participants expressed high levels of national pride: 81.8% of the participants in the 2009 sample reported they were either “very proud” or “proud” of being Korean (M = 3.09, SD = 0.72), compared to 85.2% of the participants in the 2011 survey (M = 3.20, SD = 0.71) and 85.0% of those in 2012 (M = 3.25, SD = 0.74) did so.
Demographic covariates. A set of socio-demographic variables were considered: (1) age (M = 44.0, SD = 15.3 in 2009; M = 46.0, SD = 16.4 in 2011; M = 50.6, SD = 18.2); (2) income—coded in twenty-two categories, ranging from 0 to 21 (M = 8.21, SD = 5.01 in 2009; M = 8.37, SD = 5.23 in 2011; M = 8.09, SD = 5.22); and (3) gender (male = 48.0%, female = 52.0% in 2009; male = 44.9%, female = 55.1% in 2011; male = 44.2% female = 55.8% in 2012) and education—coded in eight categories, ranging from 0 to 7 (M = 3.61, SD = 1.49 in 2009; M = 3.54, SD = 1.59 in 2011; M = 3.01, SD = 1.68 in 2012)—were included as standard demographic covariates. In addition, three other variables were controlled: political ideology, ranging from 1 “very progressive” to 5 “very conservative” (M = 2.99, SD = 0.98 in 2009; M = 2.99, SD = 1.00 in 2011; M = 3.03, SD = 1.04 in 2012); satisfaction with personal finance, ranging from 1 “not satisfied at all” to 5 “very satisfied” (M = 2.93, SD = 1.06 in 2009; M = 2.91, SD = 1.13 in 2011; M = 2.85, SD = 1.10 in 2012); and partisanship that contained four values, i.e., identification with the ruling party, identification with the major opposing party, identification with other minor parties, and independents.
A series of ordered logit models were utilized to reflect the fact that all three political support variables are ordered variables. The region-level fixed effects (7 regions in total) were considered to make sure that the results were not the products of some correlation between personality and some unobserved contextual factors (e.g., region-level geographical and cultural differences) that might affect political support. The analysis reported robust standard errors clustered at the region level to allow for the interdependence of participants in a given region. The data were pooled, considering survey year fixed effects by including survey year dummies in the models.
Table 2 presents the results from data analysis. The findings reveal that Agreeableness is the most consistent predictor of political support in South Korea. Controlling for other factors (even including political ideology and partisanship variables), individuals high on Agreeableness are more likely to demonstrate confidence in the Blue House, confidence in the government, and national pride. This result is consistent with the image of an agreeable person, who is trustworthy, gets along well with others, and tends to avoid conflict in everyday life.
Other personality variables seem to be associated with political support, but the relationship is less consistent than the case of Agreeableness. Given that people high on Conscientiousness tend to be politically conservative, it is understandable that they reveal higher levels of confidence in the government and national pride, because all three surveys were conducted under the conservative regime. However, the lack of a statistically significant finding regarding Conscientiousness and confidence in the Blue House suggests that the effects of Conscientiousness on political support are rather weak, compared with those of other variables included in the models. Similarly, a lack of statistical significance regarding Emotional Stability and national pride prohibits a clearer interpretation. In the cases of Openness and Extraversion, even the directions of regression coefficients are not consistent. In sum, among the Big Five personality dimensions, only Agreeableness exerts a positive influence on political support, independent of other potential determinants.
Figure 1 shows the changes in predicted probabilities as Agreeableness increases. The effects of Agreeableness on political support are quite substantial. For example, for a hypothetical individual whose score of Agreeableness is as low as possible, her probability of choosing “not confident at all” on the confidence in the Blue House question is almost 60%, but it is 40% in the case of another hypothetical individual whose score of Agreeableness is as high as possible. Likewise, there is approximately a 10%-point difference between the confidence in the government of those at the minimum level of Agreeableness (50% chance of choosing “not confident at all”) and those with the maximum level of Agreeableness (40% chance of choosing “not confident at all”). When the value of Agreeableness is at the minimum, the probability of thinking “I am very proud of our nation” is only 15%, but it increases up to 50% when the value of Agreeableness is at its maximum.
The present study shows that after controlling for several socio-demographic factors, the Big Five personality traits affect political support, i.e., confidence in the president, confidence in the government, and national pride. The effect of Agreeableness is especially notable, since it turns out to be consistently associated with higher levels of political support. Such a strong relationship may be due to two core facets of Agreeableness—trust and compliance—activated in expressing public support for political system. Additionally, other personality traits sometimes appear to be associated with political support. Post-hoc interpretation of these findings is possible (e.g., the impact of the change of political climate on the relationship between Extraversion and national pride), but it is admittedly hard to offer theoretically coherent arguments here.
The finding that people high on Agreeableness are more likely to hold high levels of political support suggests that mobilizing them in the political arena is essential to maintain political stability. As a matter of fact, a recent study shows that Agreeableness is negatively associated with support for the populist ideas, suggesting that citizens high on Agreeableness can function as defenders of representative democracy (Bakker et al., 2016). However, there are three issues requiring further consideration. First, Agreeableness is also characterized by conflict avoidance, and therefore it is likely for people high on Agreeableness may not to participate in the political process as it becomes more and more polarized and contentious (Ha et al., 2013). That said, once the political arena has been ideologically polarized, appealing to voters high on Agreeableness becomes a daunting task. Second, even though the results lead us to believe that national pride is a type of authentic pride, e.g., a benign, genuine esteem for one’s own country, one cannot rule out the possibility that national pride connotes prejudice against other nations. If this is the case, successful mobilization of voters high on Agreeableness may yield nationalistic policies, which have the potential to harm democratic governance. Third, previous research in the field of political psychology has reported that Agreeableness is negatively associated with social dominance orientation (SDO), rejecting endorsements of economic inequality and laissez-faire conservatism (e.g., Duckitt & Sibley, 2010). So, the efforts of mobilizing agreeable citizens may lead to ideological conflict regarding economic issues, characterized by the tension between governmental intervention and the free market. In sum, the findings of this study ask us to design political communication strategies mobilizing voters high on Agreeableness in order to maintain political stability and legitimacy, but some potential negative side effects should not be underestimated.
Despite some potentially important contributions of the present study, its weaknesses should not be understated. First, it is clearly not satisfactory to rely on just one survey question to measure political support, particularly a multi-faceted concept such as national pride. A set of questions such as the domain-specific national pride battery in the International Social Survey Programme (ISSP) will serve better to confirm whether national pride actually functions as a type of authentic pride. More generally, in order to discern the differences between authentic aspects of national pride and hubristic ones, a carefully designed measure—for example, uncritical patriotism and constructive patriotism scales (Schatz et al., 1999)—needs to be utilized.
Moreover, it is worth noting that the effects reported above were obtained despite the fact that there must have been some noise in the personality trait measures employed in this study. Substantial methodological and technical difficulties certainly exist to measure personality traits across cultures. One possibility for lack of evidence in support of the hypothesis regarding Conscientiousness may be because both Conscientiousness and Agreeableness share some meanings in a short battery. For example, the term “dependable” intends to measure Conscientiousness, but its meaning (e.g., trustworthy or reliable) can correspond to some facets of Agreeableness. In this vein, a longer battery of personality is more desirable than the TIPI or other short measures, and even essential to conduct facet-level analysis. But, in practice, it is not easy to incorporate a longer version in a nationally representative, face-to-face survey.
Though the present study is based on three cross-sectional surveys and a correlational analysis, it is safe to assume that the effect of personality traits on political support is causal, because it is widely known that personality variation emerges prior to variations in socio-political orientations (e.g., Gerber et al., 2010). Further study using longitudinal data is necessary to identify the casual relationship, because experimental research may not be a viable option, as personality traits are virtually impossible to experimentally manipulate in the lab. Also, due to the fact that personality remains stable over time, it is exceedingly difficult to take advantage of any exogenous shock to conduct a natural experiment. However, it does not preclude us from examining the interactive effects between personality and environment. Further research needs to be conducted to clarify the conditional effects of contextual factors regarding the relationship between personality traits and political behavior.
Shang E. Ha (Ph.D. University of Chicago) is an associate professor of political science at Sogang University, South Korea. His interests include political psychology, public opinion, and voting behavior. He has authored numerous articles in journals such as American Political Science Review, Political Research Quarterly, American Politics Research, and Political Psychology on topics ranging from personality traits to voting behavior in local elections. He is currently working on papers about the association between personality traits and political behavior, the antecedents and consequences of populist attitudes, and the effects of fact-checking on political behavior of misinformed citizens, all using large-scale, face-to-face or online surveys conducted in South Korea.
He can be reached at Department of Political Science, Sogang University, 35 Baekbeom-ro, Mapo-gu, Seoul, Korea, 04107 or by e-mail at email@example.com.
All correspondence concerning this article should be addressed to Shang E. Ha at Department of Political Science, Sogang University, 35 Baekbeom-ro, Mapo-gu, Seoul, Korea, 04107 or by e-mail at firstname.lastname@example.org.
Date of submission: 2019-09-02
Date of the review result: 2019-10-17
Date of the decision: 2019-11-14