[22nd General Election Research Series] Characteristics of Voters for the Two Major Parties in the 22nd General Election and Motivations for Voting and Abstention

I. Introduction

In the 22nd general election, the ruling People Power Party suffered a crushing defeat. The direct cause was the 'abstention' of voters who had supported Yoon Suk-yeol in the last presidential election. Assuming that voters from the last presidential election did not change their party affiliation in this general election, it is estimated that 80.9% of Yoon Suk-yeol voters voted for the People Power Party's district candidates, while 19.1% abstained. In contrast, 91.4% of Lee Jae-myung voters voted for the Democratic Party's district candidates, and 8.6% abstained.

Notably, in the Seoul district races, considered the most fiercely contested, 2,946,000 voters cast ballots for the Democratic Party candidates, a figure 0.67 percentage points higher than the 2,944,000 who voted for Lee Jae-myung. Conversely, 19.29% of Yoon Suk-yeol voters in Seoul abstained. The 2,628,000 People Power Party voters in Seoul represent 80.71% of the 3,256,000 Yoon Suk-yeol voters. In Seoul, Lee Jae-myung voters were mobilized and expanded, while Yoon Suk-yeol voters lost their fervor of support, with about one-fifth abstaining. The same phenomenon occurred in Busan, Daejeon, and South Chungcheong Province. In other regions as well, the behavior of Lee Jae-myung voters consolidating and expanding, and Yoon Suk-yeol voters fragmenting and abstaining, is prominent.

The general election turnout of 66.94% is 10.12 percentage points lower than the presidential election turnout of 77.06%. This result implies that (assuming all general election voters were also presidential election voters) 11.02% of voters who supported either Yoon Suk-yeol or Lee Jae-myung in the presidential election did not participate in the general election. Among these 11.02% of voters, Yoon Suk-yeol voters constitute a higher proportion than Lee Jae-myung voters. Ultimately, the reason for the People Power Party's defeat in this general election is the abstention of Yoon Suk-yeol voters and the high rate of voter participation among Lee Jae-myung voters. In all 16 regions except for Jeonnam, Yoon Suk-yeol voters abstained at a higher rate than Lee Jae-myung voters. Conversely, in all 16 regions except for Jeonnam, Lee Jae-myung voters not only participated at a higher rate than Yoon Suk-yeol voters but also, in Seoul, Busan, Daejeon, and South Chungcheong Province, more voters cast ballots for Democratic Party candidates than had voted for Lee Jae-myung in the presidential election.

It appears that Lee Jae-myung voters from the presidential election were mobilized by emotional factors such as disappointment and anger over the presidential election defeat and the ruling party's governance, leading to a high rate of participation in protest voting. In contrast, Yoon Suk-yeol voters failed to maintain or convert their support from the presidential election into votes of approval for the ruling party. Some appear to have chosen to abstain from the general election due to disappointment with the ruling party and a lack of 'hope and expectation'.

This study is broadly divided into two parts. The first part compares and analyzes the general election participation behavior of Yoon Suk-yeol voters and Lee Jae-myung voters in the 22nd general election by region, using statistical data from the National Election Commission. It explains the differences in general election participation between Yoon Suk-yeol and Lee Jae-myung voters through the phenomena of surge and decline in voter turnout and the hypothesis of emotional motivations for voting and abstention.

The second part analyzes voting and abstention. There are always voters who participate in elections (i.e., habitual voters), as well as voters who always abstain (i.e., habitual non-voters). There are also voters who selectively participate or abstain depending on the election (i.e., intermittent voters). If the election outcome is based on the default tendency of habitual voters to vote according to party affiliation, and the variables are the participation (or choice) of intermittent voters and their chosen party and candidate, then understanding the characteristics of these voter types—habitual voters, habitual non-voters, and intermittent voters—is key to explaining and predicting election results.

II. Analysis of the 22nd General Election

1. Surge and Decline in Voter Turnout

<Table 1> presents statistical data on voter turnout from the National Election Commission. The number of eligible voters in the 22nd general election was 44,280,011, of whom 29,640,919 participated, and 14,639,092 did not. The voter turnout was 66.94%. Two years prior, in the presidential election, the number of eligible voters was 44,197,692, of whom 34,059,720 participated, resulting in a turnout of 77.06%. In the presidential election, Yoon Suk-yeol received 16,394,815 votes (48.56%), and Lee Jae-myung received 16,147,738 votes (47.83%). The difference in votes between the two candidates was 247,077 (0.73 percentage points).

The general election turnout was 66.94%, lower than the presidential election turnout of 77.06%. The number of general election voters was 4,418,801 fewer than the number of presidential election voters. According to the premise of this report, 4.418 million voters who voted in the last presidential election did not participate in this general election.

In the 20th general election, the total votes cast in the 254 electoral districts nationwide were 29,234,129. Of these, the Democratic Party secured 14,758,083 district votes, accounting for 50.5% of the vote share. The People Power Party received 13,179,769 votes, or 45.1%. In the 254 electoral districts nationwide, the difference in votes between the two parties was 1,578,314, indicating that the Democratic Party's district candidates received more votes than the People Power Party's district candidates. In this general election, both the People Power Party and the Democratic Party received fewer votes than their respective candidates did in the last presidential election. This is a result of the departure and abstention of presidential election voters, and Yoon Suk-yeol voters constitute a higher proportion of this departure and abstention.

<Table 1> Presidential and General Election Turnout by Region (Unit: Thousand persons, %)

Assuming 100% of Lee Jae-myung voters participated in this general election, it can be estimated that only 82.3% of Yoon Suk-yeol voters participated in the general election. The actual results suggest that 91.39% of Lee Jae-myung voters participated in the general election, while only 80.89% of Yoon Suk-yeol voters participated (<See Table 2>).

Turnout increases (surge) when comparing general election turnout to presidential election turnout, while turnout decreases (decline) when comparing presidential election turnout to general election turnout. This surge and decline phenomenon in voter turnout was observed in the changes in turnout between the quadrennial general elections and the midterm elections during a presidential term in the United States. Historical presidential and general elections in Korea also exhibit this pattern of turnout fluctuation.

The high turnout in presidential elections is a result of increased participation or enhanced voting motivation among intermittent voters, driven by high media and public interest, significant national issues, clear competition between candidates, and mobilization campaigns by parties and candidates. Conversely, midterm elections show a decrease in turnout due to the departure or abstention of intermittent voters who participated in the presidential election. In this general election, turnout decreased in all regions compared to the last presidential election.

2. Turnout Hypothesis

The issue is the asymmetric decrease in voter turnout among presidential election voters. That is, Yoon Suk-yeol voters and Lee Jae-myung voters did not abstain at the same rate; the abstention rate was higher among Yoon Suk-yeol voters.

It is a difficult task for a president elected in a presidential election to maintain support throughout their term. Many supporters withdraw or withhold their support from the initial fervor or enthusiastic backing due to anxiety, disappointment, dissatisfaction, or anger regarding the president's governance style or policy effects. While withdrawing or withholding support from the presidential election may not immediately translate into support for the opposition party, the likelihood of abstaining from continuous support voting in an election that serves as an interim evaluation of the president's governance is relatively high. Therefore, low voter turnout can be interpreted as being caused by a high abstention rate among ruling party supporters.

Conversely, voters who supported the opposition party in the presidential election and experienced 'defeat' are likely to maintain a critical perspective on the president's governance and policies, and express this through active protest votes.

The turnout hypothesis is an empirical inference based on past general election results in Korea, suggesting that generally, if turnout exceeds 60%, the 'judgment' frame against the ruling party operates, leading to an opposition victory. If turnout falls below 50%, the 'judgment' frame against the ruling party does not operate, resulting in a ruling party victory.

The results of the current general election support this turnout hypothesis. The general election turnout indicates that Lee Jae-myung voters agreed with the 'regime judgment' and actively participated, while a considerable number of Yoon Suk-yeol voters chose to abstain rather than agree with 'opposition party judgment.' Ultimately, the outcome of the 22nd general election appears to have been determined by the proportion of Yoon Suk-yeol voters and Lee Jae-myung voters who participated or abstained in the last presidential election.

<Table 2> shows the proportion of People Power Party/Yoon Suk-yeol voters and Democratic Party/Lee Jae-myung voters in the presidential and general elections by region. The regions where the Democratic Party's district vote share in the general election exceeded Lee Jae-myung's vote share were Seoul (100.67% of presidential election vote share), Busan (103.97%), Daejeon (100.69%), and South Chungcheong Province (102.31%). In other regions, the participation rate of Lee Jae-myung voters in the general election was high, in descending order: Jeonnam (70.78%), Gwangju (74.34%), North Jeolla Province (80.51%), Gangwon (93.1%), Daegu (93.33%), Gyeonggi (94.33%), Ulsan (96.3%), North Chungcheong Province (97.15%), and Incheon (97.59%). While the turnout rates in these regions show considerable variation, excluding Jeonnam, Gwangju, and North Jeolla Province, other regions showed a turnout rate exceeding 90% of the presidential election turnout.

Conversely, the turnout for People Power Party candidates in the general election, relative to Yoon Suk-yeol voters, was as follows: Daegu (45.17% - lowest), Gwangju (50.81%), Jeonnam (71.23%), North Jeolla Province (73.3%), Daejeon (74.57%), North Gyeongsang Province (74.7%), Jeju (75.72%), Ulsan (79.80%), Seoul (80.71%), Busan (81.50%), South Chungcheong Province (82.34%), North Chungcheong Province (82.62%), South Gyeongsang Province (83.02%), Gyeonggi (83.13%), Incheon (84.98%), and Gangwon (86.79%). The People Power Party's vote share did not exceed 90% of the presidential election vote share in any region. Notably, excluding Jeonnam Province, Lee Jae-myung voters participated in the general election at a higher rate than Yoon Suk-yeol voters in all other regions.

<Table 2> Ratio of General Election Voters to Presidential Election Voters by Region (Unit: Thousand persons, %)

* Dark shading indicates regions where turnout was higher than the opposing candidate/party.

<Table 3> shows the change in voter turnout from the presidential election to the general election by region. While the general election turnout decreased in all regions compared to the presidential election turnout, the ratio of increase/decrease in Democratic Party votes compared to Lee Jae-myung's votes did not 'decrease' but rather 'increased' (excluding Gwangju, Jeonnam, North Jeolla Province, and North Gyeongsang Province). For example, in Seoul, the general election turnout was 88.54% of the presidential election turnout, a 'decrease,' but the Democratic Party's vote share actually 'increased' by 12.13 percentage points compared to Lee Jae-myung's vote share. However, the People Power Party's vote share 'decreased' by 7.83 percentage points compared to Yoon Suk-yeol's vote share. This pattern was observed in all regions nationwide (excluding Gwangju, Jeonnam, North Jeolla Province, and North Gyeongsang Province) (Daejeon: 14.02%P, South Chungcheong Province: 12.80%P, Busan: 15.47%P).

<Table 3> Change in Voter Turnout from Presidential Election to General Election by Region (Unit: %, %P)

3. Motivations for Voting and Abstention

There was a strategic judgment by the opposition party and some assertions that this general election, serving as an interim evaluation of the president, would manifest as protest votes or punitive votes, driven by disappointment and anger towards the ruling party, along with high voter turnout. Below, we analyze the motivations for voters' participation and abstention using survey data.

In the evaluation of the president's performance, there was no statistically significant difference between voters who participated in the general election and those who abstained among Yoon Suk-yeol voters. This means that Yoon Suk-yeol voters who abstained did not do so because of negative evaluations of the president. In contrast, there was a statistically significant difference between voters who participated in the general election and those who abstained among Lee Jae-myung voters in their evaluation of the president's performance. That is, the more negatively they evaluated the president's performance, the more they participated in voting; if the degree of negative evaluation was weaker, they abstained.

<Table 4> Evaluation of Presidential Performance by Voter Type

0=Very Poor, ~ 10=Very Good.

There was a statistically significant difference between voters who participated in the general election and those who abstained among Yoon Suk-yeol voters regarding the degree of 'anger' felt towards the Democratic Party. Yoon Suk-yeol voters who felt less anger towards the Democratic Party abstained, while those who felt strong anger towards the Democratic Party voted for the People Power Party. On the other hand, there was a statistically significant difference between voters who participated in the general election and those who abstained among Yoon Suk-yeol voters regarding the degree of 'hope and expectation' felt towards the People Power Party. Yoon Suk-yeol voters who felt greater hope and expectation towards the People Power Party participated in the election, while those who felt less hope and expectation abstained.

<Table 5> Anger and Hope/Expectation of Yoon Suk-yeol Voters

0=Do not feel at all, ~ 10=Feel very strongly.

There was a statistically significant difference between voters who participated in the general election and those who abstained among Lee Jae-myung voters regarding the degree of 'anger' and 'worry and anxiety' felt when thinking of the People Power Party. That is, Lee Jae-myung voters who felt strong anger and worry/anxiety towards the People Power Party participated in the general election, while those who did not feel these emotions abstained.

<Table 6> Anger and Hope/Expectation of Lee Jae-myung Voters

0=Do not feel at all, ~ 10=Feel very strongly.

4. The 22nd General Election as an Extension of the 20th Presidential Election

The 22nd general election was held 25 months after the 20th presidential election on March 9, 2022. It served as an interim evaluation during the mid-term of the presidential term. Furthermore, Lee Jae-myung, who narrowly lost the last presidential election by 0.73 percentage points and is now the leader of the opposition Democratic Party, campaigned under the banner of 'regime judgment.' As mentioned earlier, Lee Jae-myung voters were mobilized and expanded, while Yoon Suk-yeol voters had a relatively higher abstention rate. <Table 7> clearly illustrates these characteristics of the current general election.

<Table 7> Characteristics of Yoon Suk-yeol-People Power Party Voters and Lee Jae-myung-Democratic Party Voters

Yoon Suk-yeol-People Power Party voters and Lee Jae-myung-Democratic Party voters are groups of voters with very different characteristics. They are highly heterogeneous voter groups, clearly distinguished in terms of socioeconomic status, political attitudes, and policy preferences.

Compared to Lee Jae-myung-Democratic Party voters, Yoon Suk-yeol-People Power Party voters are statistically significantly older, have a higher rate of homeownership, and possess higher total assets. Yoon Suk-yeol-People Power Party voters have a statistically significantly higher level of political knowledge and a clearly conservative political ideology compared to Lee Jae-myung-Democratic Party voters. Additionally, they exhibit contrasting emotional attitudes regarding positive presidential performance evaluations, low party favoritism difference, and levels of anger and expectation/hope towards the Democratic Party and the People Power Party. On the 10 policy issues listed above, both groups consistently demonstrate statistically significant conservative and progressive stances, respectively.

To verify the relative influence of these variables characterizing the two groups, a statistical model was designed and executed. Since the dependent variable is binary, a logit regression model was constructed as follows to analyze the relative influence of the variables analyzed above. The independent variables are those that showed statistically significant differences, with other variables added as control variables. The statistical model is as follows:

Where,

Yi : Yoon Suk-yeol-People Power Party voter (=1), Lee Jae-myung-Democratic Party voter (=0)

Xi : Age, average monthly household income, homeownership (=1), total household assets, education level, political interest, political knowledge, political ideology, unaffiliated (=1), party favoritism difference, presidential performance evaluation, anger-Democratic Party, anger-People Power Party, hope and expectation-Democratic Party, hope and expectation-People Power Party, preference on 10 policy issues.

<Table 8> Logit Regression Analysis Results

(1=Strongly Agree ~ 4=Strongly Disagree)

Policy preferences did not play a major role in why Lee Jae-myung voters participated in this general election and chose the Democratic Party candidate (in district races). They can be seen as having voted for the Democratic Party candidate based on strong party-based emotional attitudes, including progressive ideology, very strong party favoritism difference, strong negative evaluation of the president's performance, strong anger towards the People Power Party, and high hope and expectation towards the Democratic Party (conversely, low hope and expectation towards the People Power Party).

5. Conclusion: Summary of the 22nd General Election Results

While approximately 80% of Yoon Suk-yeol voters voted for the People Power Party (the number of general election People Power Party voters was 3,215,000 fewer than the number of presidential election Yoon Suk-yeol voters), over 90% of Lee Jae-myung voters voted for the Democratic Party (the number of general election Democratic Party voters was 1,390,000 fewer than the number of presidential election Lee Jae-myung voters). The asymmetric participation and abstention between Yoon Suk-yeol voters and Lee Jae-myung voters were the determining variables for the general election results. As a result, the Democratic Party's district candidates received 1,578,000 more votes than the People Power Party's district candidates.

So, why did approximately 20% of Yoon Suk-yeol voters abstain? Conversely, why was the abstention rate for Lee Jae-myung voters 10% or less?

Among Yoon Suk-yeol voters who abstained from the general election, there was no statistically significant difference in their evaluation of the president's performance compared to those who participated (the positive evaluation of the president's performance did not differ significantly between participants and abstainers). The reason for their abstention was also not 'anger' towards the Democratic Party. This explains why frames such as 'opposition party judgment' or 'judgment of the corrupt' or 'judgment of criminals' had little effect. However, Yoon Suk-yeol voters who abstained had statistically significantly lower 'hope and expectation' towards the People Power Party or the Yoon Suk-yeol administration compared to those who participated. In other words, they lacked the motivation to vote in the general election 'for' the People Power Party or 'for' the future of the People Power Party. The primary reason for their choice of 'abstention,' a form of 'non-participation' or 'non-decision-making,' appears to be the difficulty in maintaining 'hope and expectation' towards the ruling party, rather than a negative evaluation of the incumbent president or anger towards the opposition party. The People Power Party's election campaign seems to have been far too 'blunt' to comprehend these 'complex and sensitive emotions towards the future'.

Over 90% of Lee Jae-myung voters who participated in the general election are typical Democratic Party supporters. These are voters who evaluate the president's performance negatively, feel 'anger' towards the People Power Party, and experience 'worry and anxiety.' Their participation in voting laid the foundation for the Democratic Party's victory. However, the approximately 10% of 'minority' voters among Lee Jae-myung supporters who abstained are those who do not have a statistically significant negative view of the president and do not feel statistically significant 'anger' or 'worry and anxiety' towards the People Power Party.

Ultimately, this general election resulted in the 'majority' of Lee Jae-myung voters consolidating and 'also' winning the 'numbers game' against the 'majority' of Yoon Suk-yeol voters through general election participation. This was then converted into an overwhelming majority of seats within the 'first-past-the-post' electoral system's 'minimum winning margin' rule. As about one-fifth of Yoon Suk-yeol voters abstained from the general election, the People Power Party had fewer voters in absolute numbers than the Democratic Party and lost by small margins in many competitive district races.

Furthermore, the 'minority' of both Lee Jae-myung voters and Yoon Suk-yeol voters, namely those who abstained in this general election, appear to be the 'central axis' or 'balancer' for the stability of Korean politics. Could the implication of this general election be the paradox that abstentionists act as a balancer, suppressing political instability and polarization?

III. Characteristics of Abstentionists

Representative democracy presupposes voter equality, but the personal resources and institutional incentives related to socioeconomic status required for voting are not equally distributed among all voters. It is a well-known fact that socioeconomic inequality acts as a barrier to exercising the fundamental right to vote (Wolfinger and Rosenstone 1980, Blais 2000, Bartels 2012). Furthermore, voting and abstention are related to voters' political attitudes and predispositions, such as political interest, political knowledge, political efficacy, and political trust (Niemi et al. 1991, Prior and Bougher 2018, Blais and Achen 2019). On the other hand, voting and abstention are also the results of strategic choices by voters, parties, and candidates, and are also effects of persuasion and mobilization depending on what campaign messages are produced, through which channels (or media), and targeting whom, and these effects are also asymmetrical (Rosenstone et al. 1986, Green and Gerber 2003, Kreiss 2016).

Below, we analyze the characteristics of abstentionists in the 22nd general election and abstentionists in general. We attempt a comparative analysis across various dimensions to analyze the characteristics of abstentionists and the reasons for their abstention. The comparative pairs for this purpose are [General Election Voters vs. General Election Abstentionists], [People Power Party Voters vs. Abstentionists], [Democratic Party Voters vs. Abstentionists], [Habitual Voters vs. Habitual Abstentionists], and [Intermittent Voters vs. Habitual Abstentionists].

1. Characteristics of Abstentionists in the 22nd General Election

<Table 9> shows the differences between voters and abstentionists in the 22nd general election. Participants in the general election are voters with relatively higher socioeconomic status (SES) compared to abstentionists. These are voters with higher age, higher (household) income, higher homeownership rates, and higher asset ownership. The statistically significant differences in socioeconomic status among them show that Korean elections are a system where the voices of citizens with relatively lower socioeconomic status are not reflected in the election results.

On the other hand, abstentionists had statistically significantly lower levels of political interest and political knowledge compared to voters. Furthermore, abstentionists were statistically significantly more likely to be unaffiliated with any party, and they showed small differences in favorability towards the two major parties. In other words, abstentionists in this general election were 'cognitively unaffiliated' voters who generally held non-attitudes and non-emotions towards politics and political parties.

Additionally, abstentionists showed no statistically significant differences from voters in their evaluation of the president's job performance, anger towards the two major parties, and expectations and hopes for the two major parties. This suggests that emotional factors regarding the president and the two major parties were not the criteria for deciding whether to vote or abstain.

<Table 9> Characteristics of Voters and Abstentionists in the 22nd General Election

*Shaded areas indicate statistically significant differences.

The first column of <Table 10> presents the results of a logit regression analysis model with voting and abstention in the 22nd general election as the dependent variable. The logit regression model is as follows:

Here,

Yi : General election voter (=1), General election abstentionist (=0)

Xi : Age, average monthly household income, homeownership (=1), total household assets, education level, political interest, political knowledge, political ideology, unaffiliated with party (=1), difference in party favorability, presidential job performance evaluation, anger-Democratic Party, anger-People Power Party, expectations and hopes-Democratic Party, expectations and hopes-People Power Party, preference for 10 issue policies.

Voters with lower age, lower political interest, and lower political knowledge, and smaller differences in favorability towards the (two major) parties, were more likely to abstain. Notably, excluding age, all socioeconomic status variables such as average monthly household income, homeownership, and total assets were not statistically significant in the regression model evaluating the relative influence of independent variables. Ultimately, the core of abstention lies in variables related to attitudes and predispositions, such as political interest, knowledge, and differences in party favorability, rather than socioeconomic status. In summary, abstentionists in the 22nd general election are voters characterized by apathy towards politics, ignorance, and emotional indifference (towards party favorability).

2. People Power Party Voters/Democratic Party Voters vs. Abstentionists

The second and third columns of <Table 10> present the results of logit regression models with People Power Party voters/Democratic Party voters vs. abstentionists as the dependent variable. People Power Party voters were statistically older than abstentionists, whereas age did not affect voting or abstention for Democratic Party voters and abstentionists. These results imply that the effect of higher age was significant when comparing People Power Party voters with Democratic Party voters and abstentionists, respectively. Similarly, People Power Party voters are asset-rich voters and possess the highest level of political knowledge compared to Lee Jae-myung voters and abstentionists, respectively. On the other hand, Democratic Party voters have the highest level of political interest compared to People Power Party voters and abstentionists, respectively. In summary, People Power Party voters are 'older (average age 55.6, Democratic Party voters 48.9, abstentionists 42.7), wealthier, and more knowledgeable' citizens compared to Democratic Party voters and abstentionists. They are 'special' groups deviating from the citizen average based on socioeconomic status.

The effect of political ideology shows a positive value in the comparison of People Power Party voters versus abstentionists, and a negative value in the comparison of Democratic Party voters versus abstentionists, suggesting that the more conservative one is, the more likely they are to be a People Power Party voter, and the more progressive one is, the more likely they are to be a Democratic Party voter, indicating ideological polarization between the two groups. Similarly, People Power Party voters and Democratic Party voters are groups that show polarized evaluations regarding presidential job performance and expectations for the two parties. Abstentionists are positioned between People Power Party voters and Democratic Party voters in terms of political ideology, presidential job performance evaluation, and emotions (expectations and hopes) towards the two major parties. With abstentionists in the middle, People Power Party voters and Democratic Party voters are located at opposite extremes.

In summary, when comparing the characteristics of abstentionists with those of People Power Party voters and Democratic Party voters through a double comparison, abstentionists are not distinguished from Democratic Party voters in terms of age, political knowledge, and assets. That is, abstentionists are not younger, less knowledgeable, or poorer than Democratic Party voters. Democratic Party voters are a group similar to abstentionists in terms of socioeconomic status, unlike voters of both major parties. However, abstentionists are characterized by ideological non-attitudes and emotional indifference towards both parties, unlike voters of the two major parties.

3. 'Habitual' Abstentionists and 'Intermittent' Abstentionists

The fourth and fifth columns of <Table 10> present the results of logit regression models with habitual voters/intermittent voters vs. habitual abstentionists as the dependent variable, respectively. Habitual abstentionists are not statistically distinguishable from intermittent voters in terms of age, political interest, and favorability towards the two major parties. In other words, habitual voters are voters with high age, high political interest, and significant differences in favorability towards the two major parties. Furthermore, habitual abstentionists are unaffiliated with parties and have statistically significantly lower political knowledge compared to habitual voters and intermittent voters. In summary, the criteria distinguishing intermittent voters from habitual abstentionists are party affiliation and political knowledge level; in this sense, habitual abstentionists can be called 'cognitively unaffiliated.' Socioeconomic status was not statistically significant here.

<Table 10> Logit Regression Analysis Results

+p<0.06 *p<0.05 **p<0.01 ***p<0.001

IV. Summary and Conclusion

Abstentionists in the 22nd general election are generally young citizens with low levels of political knowledge, who are unaffiliated with parties and emotionally indifferent towards parties. On the other hand, Democratic Party voters in the 22nd general election are voters with low political knowledge and high political interest, progressive political ideology, and emotional investment in parties and consequent biases. People Power Party voters are voters with high political knowledge, high assets, high age, conservative political ideology, and emotional investment in parties and consequent biases. The above is summarized in <Table 11>.

<Table 11> Characteristics of Abstentionists, People Power Party Voters, and Democratic Party Voters

The voter characteristics and resulting voting choices appear to have several implications. First, the central role of political knowledge in voting and abstention. Generally, well-structured political knowledge is a key variable enabling consistent political choices and a major driver of increased political participation. This general election confirms previous research. However, attention should be paid to the combination of low political knowledge and high political interest among Democratic Party voters, and the combination of high political knowledge, high assets, and high age among People Power Party voters. If the former represents an 'excess of political self-awareness,' the latter represents a 'bias in political support.' The former tends to amplify passionate support and opposition based on misinformation and disinformation, and the potential for distorted policy and political preferences and self-deceptive choices. The latter tends to amplify the potential for privileged/entrenched self-confidence or arrogance, misjudgment that substitutes differences in preference for knowledge, and self-indulgent/satisfied choices. Both are far from the image of 'rational and reasonable citizens' that a democratic society presupposes.

Second, both People Power Party voters and Democratic Party voters exhibit emotional investment in the competition and confrontation between parties, and consequent confirmation bias. It appears that choices made through 'emotional heuristics (or schemas)' anchored to 'parties' for new issues and phenomena not covered by existing political ideology are being amplified and produced within the 'echo chambers' of the new media environment. This party-centric or strong 'emotional investment' in parties appears to be replacing 'cognitive-ideological understanding' while simultaneously reinforcing 'ideological-group bias.' Tracking and analysis of this new phenomenon seem necessary.

Third, there is a clear difference between the Democratic Party voter group and the People Power Party voter group. Along with ideological polarization, emotional polarization between them appears to be the strongest variable determining their choices. The spread and reinforcement of negative effects stemming from the exclusion and hatred of other groups caused by extreme emotional polarization not only hinder social integration but also seem to structure a political environment that taboos any political compromise between parties. Extraordinary decisions are needed for Korean society and politics. ■

References

Aldrich John H, James M. Montgomery Wood. 2011. “Turnout as a habit.” Political behavior. 33: 535-63.

Bartels Larry M. 2012. The New Gilded Age: From Unequal Democracy. Princeton University Press.

Blais A. 2000. To Vote Or Not To Vote? The Merits and Limits of Rational Choice. Pittsburgh: Univ. Pittsburgh Press.

Blais A, Achen CH. 2019. “Civic duty and voter turnout.” Political Behavior. 41: 473-97.

Born R. 1990. “Surge and decline, negative voting, and the midterm loss phenomenon: A simultaneous choice analysis.” American Journal of Political Science. Aug 1: 615-45.

Campbell James E. 1987. The revised theory of surge and decline. American Journal of Political Science. Nov 1: 965-79.

Dalton Russell J. 2007. “Partisan mobilization, cognitive mobilization and the changing American electorate.” Electoral Studies. 26(2) :274-86.

Fowler James H. 2006. “Habitual voting and behavioral turnout.” The Journal of Politics. 68(2): 335-44.

Green Donald P, Gerber Alan S, Nickerson David W. 2003. “Getting out the vote in local elections: Results from six door-to-door canvassing experiments.” The Journal of Politics. 65(4): 1083-96.

Green Donald P, Shachar R. 2000. “Habit formation and political behaviour: Evidence of consuetude in voter turnout.” British Journal of Political Science. 30(4): 561-73.

Kreiss Daniel. 2016. Prototype politics: Technology-intensive campaigning and the data of democracy. Oxford University Press.

Niemi Richard G, Reed David R, Weisberg Herbert F. 1991. “Partisan commitment: A research note.” Political Behavior. 13: 213-21.

Prior M, Bougher Lori D. 2018. “Like they’ve never, ever seen in this country? Political interest and voter engagement in 2016.“ Public Opinion Quarterly. 82(S1):822-42.

Rosenstone Steven J, Hansen John M, Kinder Donal R. 1986. “Measuring change in personal economic well-being.” Public Opinion Quarterly. 50(2): 176-92.

Wolfinger, R.E. and Rosenstone, S.J., 1980. Who votes?. Yale University Press.

■ Author: Yoo Jaesung_Professor, Department of International Studies, Keimyung University

■ Responsible Editor: Kim Sunhee_EAI 연구원

    Contact: 02-2277-1683 (ext. 209) shkim@eai.or.kr