[Public Opinion Brief No. 110] Key Results of the 1st Panel Survey for the 2012 General and Presidential Elections

[Public Opinion Brief No. 110] 2012 General and Presidential Elections Panel Survey Jointly Conducted by EAI, SBS, JoongAng Ilbo, and Hankook Research

1. Panel Survey and Research Team

2. Key Results of the 1st Survey

I. Panel Survey and Research Team

1. Panel Survey

■ EAI's election panel survey began with the 4th National Local Elections in 2006, when it first introduced the panel survey method in Korea, in collaboration with SBS, JoongAng Ilbo, and Hankook Research. The panel survey is virtually the only method that can overcome the limitations of one-off surveys, which only capture voting preferences at a specific point in time. This is because it allows for accurate identification of individual voter attitude changes and their patterns, as well as precise analysis of the causes of shifts in public sentiment over time.

■ Following the 2006 panel survey, EAI, in collaboration with SBS, JoongAng Ilbo, and Hankook Research, conducted panel surveys for the 2007 presidential election, the 2008 general election, and the 2010 local elections, receiving acclaim for opening new horizons in election research and reporting. The achievements of these panel surveys will continue in the 2012 general and presidential elections. EAI is conducting the "2012 General and Presidential Elections Panel Survey" in collaboration with SBS, JoongAng Ilbo, and Hankook Research. The survey will consist of a total of 7 waves, considering the general election on April 11 and the presidential election on December 19, for which a panel of 2,000 voters has been recruited.

(1) Conducting the Panel Survey

■ The joint panel survey by EAI, SBS, JoongAng Ilbo, and Hankook Research for the 2012 general and presidential elections will be conducted in a total of 7 waves. The first wave, starting immediately after March 29, when the campaign period for the April 11 general election begins, focuses on voting behavior such as support for general election candidates and parties while recruiting the surveyed voter panel. The second wave, conducted immediately after the April 11 general election, will inquire about voting behavior, evaluations, and support for presidential candidates. Waves 3 through 5, conducted between 150 and 20 days before the presidential election, will investigate various issues related to the December 19 presidential election, including hypothetical matchups between presidential candidates. Wave 6, conducted 7 days before the presidential election, will examine election forecasts and media influence. Immediately after the presidential election, the final seventh wave will assess the presidential election and explore expectations and prospects for the next government's administration.

[Figure 1] Conduct and Key Survey Contents of the Panel Survey

(2) Panel Composition and Characteristics

■ A total of 2,000 panelists were recruited nationwide, including from Jeju Island. The recruitment considered variables such as region, gender, age, education level, occupation, and housing tenure (owner-occupied, jeonse, monthly rent), reflecting the 2010 Population and Housing Census results from Statistics Korea. Regionally, Incheon and Gyeonggi had the largest number of panelists with 569, followed by Seoul with 416. Busan/Ulsan/Gyeongnam had 317, Daegu/Gyeongbuk had 207, Gwangju/Jeonbuk/Jeonnam had 205, and Daejeon/Chungbuk/Chungnam had 202. Gangwon had 61 panelists, and Jeju had 22.

■ Examining the panel composition by gender and age group, there are 1,010 female panelists and 990 male panelists. By age group, there are 369 individuals aged 19-29, 411 in their 30s, 440 in their 40s, 378 in their 50s, and 403 aged 60 and over.

[Figure 2] Panel Composition

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

2. Research Team

■ The research team primarily consists of Professor Lee Nae-young, Director of the EAI Public Opinion Analysis Center (Korea University), and Professor Kang Won-taek, Chairperson of the EAI Civic Panel Committee (Seoul National University), who have participated since the 2004 local election panel survey. Professors Kim Jun-seok (Dongguk University), Park Won-ho (Seoul National University), Yoon Kwang-il (Sookmyung Women's University), and Jang Seung-jin (Kookmin University) have newly joined. The total number of participating researchers is 23.

■ Chairperson: Lee Nae-young, Director of the EAI Public Opinion Analysis Center (Korea University)

■ Researchers: Kang Won-taek (Seoul National University), Kwon Hyuk-yong (Korea University), Kim Sung-tae (Korea University), Kim Min-jeon (Kyung Hee University), Kim Jun-seok (Dongguk University), Park Won-ho (Seoul National University), Park Chan-wook (Seoul National University), Seo Hyun-shin (Sungshin Women's University), Yoon Kwang-il (Sookmyung Women's University), Lee Gon-soo (EAI), Lee Woo-jin (Korea University), Lee Hyun-woo (Sogang University), Lim Sung-hak (University of Seoul), Jang Seung-jin (Kookmin University), Jeong Won-chul (EAI), Jeong Han-wool (EAI), Ji Byung-geun (Chosun University)

■ Media: Shin Chang-woon (JoongAng Ilbo), Hyun Kyung-bo (SBS)

■ Survey Team: Kim Choon-seok, Oh Seung-ho, Yoo Seok-sang (all from Hankook Research)

3. Overview of the 1st Survey

■ Panel Size: 2,000 voters aged 19 and over residing nationwide.

■ Sampling Method: Quota sampling, considering variables such as region, gender, age, education level, occupation, and housing tenure.

■ Sampling Error: Assuming random sampling, the maximum permissible sampling error at a 95% confidence level is ±2.2%.

■ Response Rate: 8.3%

■ Survey Period: March 30 to April 1, 2012 (3 days)

■ Survey Method: Computer-assisted telephone interview (CATI)

■ Survey Organization: Hankook Research Co., Ltd.

II. Key Results

1. Voting Intention and Behavior

Key Factor in Voting Decision: Candidate Factors Over Party Factors

■ For the April 11 general election to elect the 19th National Assembly members, 75.5% of the panelists responded that they would "definitely vote." This is 17.4 percentage points (p) higher than the 58.1% response rate for "definitely vote" in a survey conducted by the National Election Commission from April 1 to 3 among 1,500 voters nationwide. The response rate for "probably vote" was 12.4%, which, compared to "undecided" at 8.4%, "probably not vote" at 1.8%, and "will not vote" at 1.6%, also showed a high response rate. This indicates that the panelists participating in the "EAI SBS JoongAng Ilbo Hankook Research Joint 2012 General and Presidential Elections Panel Survey" are actively interested in politics.

■ Examining the voting decision criteria of these panelists, it is evident that candidate-specific factors overwhelmingly outweigh party affiliation factors. The response rate for considering the candidate's affiliated party was 18.5%, while the response rate for considering the candidate's ability and experience was 36.0%, their ideology and pledges 20.4%, their integrity 19.7%, and their electability and region of origin were 2.0% and 0.6%, respectively. Consequently, excluding other factors (1.9%) and don't know/no answer (0.9%) for candidate-specific factors, the total response rate for candidate-specific factors was 78.7%, more than double the party affiliation factors.

Saenuri Party and Democratic United Party in a Tight Race for Both District and Proportional Representation

Saenuri Party for those 50 and over, Democratic United Party for those 49 and under

■ A survey was conducted on which party's candidate respondents would vote for in the district constituency election if the election were held tomorrow. The results showed a very tight race between the Saenuri Party and the Democratic United Party candidates. The response rate for voting for a Saenuri Party candidate was 30.0%, and for a Democratic United Party candidate, it was 29.0%. With 29.3% of respondents remaining undecided, which is similar to the response rates for the two parties, it foreshadows that predicting the outcome of this general election will not be simple.

■ The situation was similar for the party vote for proportional representation seats. The response rate for voting for the Saenuri Party was 31.9%, and for the Democratic United Party, it was 31.2%. Those who had not yet decided accounted for 23.0%.

■ While the overall response rates show a close race, distinct differences are observed across age groups. For the district constituency election, among respondents aged 20s, the response rate for the Saenuri Party was 10.6%, while for the Democratic United Party, it was 32.9%. In their 30s, a gap emerged between the two parties, with the Saenuri Party at 19.6% and the Democratic United Party at 37.7%. Although the gap between the two parties narrowed for those in their 40s compared to the 20s and 30s, it remained over 10 percentage points, with the Saenuri Party at 22.0% and the Democratic United Party at 35.0%. The gap between the Saenuri Party and the Democratic United Party reverses for those in their 50s and 60s and over. Specifically, in their 50s, the Saenuri Party garnered 41.8%, while the Democratic United Party received 23.8%. For those aged 60 and over, the Saenuri Party accounted for 51.9% and the Democratic United Party 16.8%, effectively offsetting the votes lost by the Saenuri Party among those in their 20s to 40s.

■ The results for the party vote for proportional representation seats are very similar to those for district constituency candidates. In their 20s, the Saenuri Party received 10.6% and the Democratic United Party 37.8%. In their 30s, the Saenuri Party was at 18.9% and the Democratic United Party at 36.8%. In their 40s, the figures were 24.9% and 36.2%, respectively, with the Democratic United Party maintaining its lead. However, the situation reverses for those in their 50s and 60s and over, where the Saenuri Party leads the Democratic United Party by a significant margin. For those in their 50s, the response rate for the Saenuri Party was 44.0%, while for the Democratic United Party, it dropped to 26.4%. For those aged 60 and over, the gap widens further, with the Saenuri Party's response rate at 56.6% compared to the Democratic United Party's 20.3%. Consequently, it can be confirmed that the Saenuri Party gains relatively strong support from those aged 50 and over, while the Democratic United Party does so from those aged 40 and under.

[Figure 3] Regional and Proportional Voting Preferences by Age Group (%)

f2e34dd1a320d063

f2e34dd1a320d063

f2e34dd1a320d063

The Hidden Vote Effect of the 20s and 30s

■ Another important variable that indicates which age group will have a greater turnout at the polling stations, thus becoming a determining factor in the seat distribution between the two parties, is the "undecided" response rate of 29.3%. When this 29.3% of undecided voters for district constituency candidates is analyzed by age group, the 20s accounted for 42.8% and the 30s for 30.3%. This differs from the 22.7% for those aged 60 and over and 25.3% for those in their 50s.

■ The situation is similar for the party vote for proportional representation seats. While 34.4% of those in their 20s and 26.2% of those in their 30s responded as undecided, the response rate for those aged 60 and over was 16.6%, and for those in their 50s, it was 18.6%, which are relatively lower compared to the 20s and 30s.

[Figure 4] Undecided Voters for District and Proportional Representation Elections by Age Group (%)

■ Among these individuals in their 20s and 30s, the response rate for "definitely vote" is relatively low at 62.1% and 63.3%, respectively, compared to 91.3% for those aged 60 and over and 81.4% for those in their 50s. Depending on the situation during the remaining campaign period, the possibility of these individuals in their 20s and 30s heading to the polls cannot be ruled out. In fact, the response rate for "probably vote" is relatively high at 19.6% for those in their 20s and 19.0% for those in their 30s, compared to 9.9% for those in their 50s and 3.4% for those aged 60 and over. However, even if these hidden votes are more likely to favor the Democratic United Party, there is a prerequisite that the act of converting these potential votes into actual votes, i.e., the turnout of the 20s and 30s, must materialize.

[Figure 5] Key Characteristics of Voting Intention by Age Group (%)

2. Evaluation of Nomination Results and Election Outlook

Disappointing Nomination Results

■ Ahead of the general election, both the Saenuri Party and the Democratic United Party suffered from nomination disputes. Both parties proclaimed a nomination revolution, but the mainstream evaluation was that it was a 'dragon's head and a snake's tail.' The panelists' evaluations were also predominantly negative rather than positive. Relatively, the evaluation of the Democratic United Party's nomination results was more negative, which is a point of difference. For the Democratic United Party's nominations, the 'did not do well' (very poorly + poorly) evaluation was 40.7%, which was 13.6 percentage points (p) higher than the 'did well' (very well + generally well) evaluation of 27.1%. For the Saenuri Party, the 'did not do well' (very poorly + poorly) evaluation was 32.7%, and the 'did well' (very well + generally well) evaluation was 32.4%.

■ Of course, these evaluations show some differences depending on the party supported. First, regarding the party supported, Saenuri Party supporters responded that the Saenuri Party's nominations were 'done well' (very well + generally well) at 65.2%, which was 47.4 percentage points (p) higher than the 'did not do well' (very poorly + poorly) response of 17.8%. For Democratic United Party supporters, 'did well' (very well + generally well) was 19.3%, while 'did not do well' (very poorly + poorly) was 44.5%.

■ Among Democratic United Party supporters, the proportion of respondents who answered that the Democratic United Party's nominations were 'done well' was 47.3%, but the proportion who answered 'did not do well' was also 38.0%, showing a difference of only 9.3 percentage points (p) and a not-low response rate. In contrast, for the Saenuri Party's nominations, the 'did well' evaluation was 20.6%, and the 'did not do well' evaluation was 42.8%. These results show that the Democratic United Party's nominations did not receive a warm welcome even from its supporters compared to the Saenuri Party.

[Figure 6] Evaluation of Saenuri Party Nomination Results (%)

Note) 'Did well' is the sum of 'very well' and 'generally well,' and 'did not do well' is the sum of 'very poorly' and 'poorly.'

[Figure 7] Evaluation of Democratic United Party Nomination Results (%)

Note) 'Did well' is the sum of 'very well' and 'generally well,' and 'did not do well' is the sum of 'very poorly' and 'poorly.'

Democratic United Party + Unified Progressive Party are positive, but Saenuri Party has the highest possibility of becoming the leading party

■ Regarding the candidate unification between the Democratic United Party and the Unified Progressive Party in this general election, 50.7% held the position that it was desirable (very desirable + generally desirable). The number of respondents who answered it was not desirable (not at all desirable + not very desirable) was 27.2%. This evaluation is clearly evident among supporters of both parties. Among Democratic United Party supporters, it received a 'desirable' (very desirable + generally desirable) evaluation with a response rate of 75.9%, and among Unified Progressive Party supporters, it was 89.8%.

■ However, despite this candidate unification between the Democratic United Party and the Unified Progressive Party, the Saenuri Party had the highest possibility of becoming the leading party. This is because 46.3% expected the Saenuri Party to become the leading party. The proportion that chose the Democratic United Party was 39.1%, which is 7.2 percentage points (p) lower than the Saenuri Party.

■ These projections were even stronger among Saenuri Party supporters. Among Saenuri Party supporters, 74.5% expected the Saenuri Party to become the leading party. Among Democratic United Party supporters, the response rate expecting the Saenuri Party to become the leading party was 32.4%. Conversely, among those who expected the Democratic United Party to become the leading party, the proportion of Democratic United Party supporters was 59.9%. The proportion of Saenuri Party supporters was only 13.3%. This is evidence of the high cohesion and confidence of Saenuri Party supporters.

[Figure 8] Possibility of Becoming the Leading Party (%)

Saenuri Party candidates also have the highest possibility of winning in terms of constituency

■ This strength of the Saenuri Party was also evident in the possibility of winning for the incumbent's constituency in the panelists' residential areas. The response rate that considered the possibility of Saenuri Party candidates winning to be high was 43.7%, which was higher than the 34.9% for the Democratic United Party.

■ However, these possibilities vary by residential region. In Seoul, Incheon/Gyeonggi, and Daejeon/Chungbuk/Chungnam, the superiority between the Saenuri Party and the Democratic United Party was not clear. In Seoul, the Saenuri Party was 37.9% and the Democratic United Party was 40.9%. In Incheon/Gyeonggi, the Saenuri Party was 41.0% and the Democratic United Party was 38.8%. In Daejeon/Chungbuk/Chungnam, the Saenuri Party was 33.4% and the Democratic United Party was 35.8%. The Saenuri Party's dominant regions were Daegu/Gyeongbuk and Busan/Ulsan/Gyeongnam, with 78.4% and 68.8%, respectively. The Democratic United Party's dominant regions were Gwangju/Jeonbuk/Jeonnam and Gangwon/Jeju, with response rates of 75.7% and 41.8%, respectively.

[Figure 9] Possibility of Winning by Residential Region (%)

3. Election Issues and Agendas

The 'Judgment of the Lee Myung-bak Administration' and 'Irresponsible Opposition' gained consensus

Judgment of the Lee Myung-bak Administration showed generational differences, while 'Irresponsible Opposition' showed consistent results across generations

■ An investigation into the 'Judgment of the Lee Myung-bak Administration' advocated by the opposition and the 'Irresponsible Opposition' advocated by the ruling party for this general election revealed that it was virtually a draw. Regarding the 'Judgment of the Lee Myung-bak Administration,' the response rate for agreement (very agree + generally agree) was 63.0%, while disagreement (not at all agree + not very agree) was 33.4%.

■ For the 'Irresponsible Opposition,' agreement (very agree + generally agree) was 58.5%, and disagreement (not at all agree + not very agree) was 37.1%. These results can be interpreted as both claims gaining consensus from voters.

■ Of course, the differences according to the supported party are relatively distinct. Among Saenuri Party supporters, the proportion who agreed (very agree + generally agree) with the 'Judgment of the Lee Myung-bak Administration' was 40.5%. Among Democratic United Party supporters, 81.4% agreed (very agree + generally agree). For the 'Irresponsible Opposition,' 44.4% of Democratic United Party supporters expressed agreement, and the response rate among Saenuri Party supporters was 73.7%.

■ It can be seen that a consensus in the 40% range is formed for both claims among both Saenuri Party and Democratic United Party supporters, which can be seen as evidence that many voters view the current political situation negatively.

[Figure 10] Evaluation of the Judgment of the Lee Myung-bak Administration by Supported Party (%)

[Figure 11] Evaluation of Irresponsible Opposition by Supported Party (%)

■ Generational differences are also distinct. Regarding the 'Judgment of the Lee Myung-bak Administration,' 73.6% of those in their 20s, 75.9% of those in their 30s, and 69.1% of those in their 40s agreed (very agree + generally agree). Even among those in their 50s, the response rate was 56.7%, surpassing the 38.8% who disagreed. The situation only improved among those aged 60 and over, where the agreement rate (very agree + generally agree) dropped sharply to 36.9%.

■ For the 'Irresponsible Opposition,' similar to the 'Judgment of the Lee Myung-bak Administration,' there is a consensus, but generational analysis also reveals some differences. The response rate for agreement was relatively consistent across all age groups. For those in their 20s, the agreement rate (very agree + generally agree) was 64.8%. For those in their 30s, it was 58.0%, for their 40s, 54.8%, and for their 50s, 58.9%. Among those aged 60 and over, the response rate was 57.0%.

■ Consequently, individuals in their 20s and 30s showed the highest consensus on both the 'Judgment of the Lee Myung-bak Administration' and the 'Irresponsible Opposition,' indicating that they hold a relatively more critical stance compared to other generations regarding the current political situation.

Economic Development Issue Gained Traction

■ Regarding important issues, economic growth was cited as the most important, with a response rate of 22.4%. Regional development/pledges ranked second with 17.8%. Other issues that emerged as hot topics included the Blue House civilian surveillance and power corruption, cited by 13.4%, and universal welfare and the KORUS FTA, cited by 13.3% and 12.1%, respectively. North Korea's security threat (8.7%) and chaebol reform (5.4%) lagged behind other issues. These results suggest the high level of expectations voters have for economic development.

■ However, this does not mean that issues other than economic development are in a stable phase. They are merely submerged for the time being and have the potential to re-emerge and reignite controversy. For the KORUS FTA, the response rates for implementing the original agreement, considered the ruling party's proposal, were 8.6% and 46.6% for implementing it first and then discussing issues. The opposition's proposals, 'the FTA must be immediately abolished' and 'it must be renegotiated and ratified again,' were 6.9% and 34.9%, respectively. These results indicate a standoff between the ruling and opposition parties' proposals. ■ Regarding chaebol regulation, the opinion that regulations on chaebols should be strengthened compared to the present was 42.6%. The remaining response rates were concentrated on maintaining the current level or easing regulations, indicating a sharp division in stances.

■ However, regarding North Korea policy and welfare policy, opinions tend to converge. For North Korea policy, 61.7% supported strengthening exchanges and cooperation between the South and North, which was 31.9 percentage points (p) higher than the 29.8% who supported maintaining or strengthening a hardline policy toward North Korea.

■ Regarding welfare policy, the stance of providing welfare services such as medical care, childcare, and education equally to all citizens (29.8%) was less prevalent than the stance of providing them selectively to urgent sectors and groups (69.4%).

[Figure 12] Attitude by Issue (%)

4. Political Efficacy

'High Expectations' but 'Low Satisfaction'

■ Voting in elections and the act of voting can be seen as related to efficacy. The political efficacy of the panelists in this survey can be summarized as 'high expectations' but 'low satisfaction.' First, regarding the statement 'People like us have no right or ability to speak about what the government does,' the response rate for disagreement (not at all disagree + not very disagree) was 51.8%, which was higher than the 45.8% who agreed (very agree + generally agree). In particular, 70.9% of those in their 20s and 59.6% of those in their 30s disagreed, indicating an active stance in expressing opinions about government authority, i.e., a form of political participation. In contrast, among those aged 60 and over, 41.9% disagreed, and among those in their 50s, 37.3% disagreed, indicating a relatively passive stance compared to those in their 20s and 30s.

■ In terms of voting perception, the majority of citizens take an active stance. This is because 84.1% disagreed (not at all disagree + not very disagree) with the statement 'Since a lot of people vote, it doesn't matter whether I vote or not.' The response rate for agreement (very agree + generally agree) was only 15.0%. Similar results were observed across all age groups. Among those in their 20s, 86.8% disagreed; in their 30s, 97.3%; in their 40s, 85.9%; in their 50s, 77.8%; and among those aged 60 and over, 82.5% disagreed.

■ Unlike the expectations regarding political efficacy mentioned above, disappointment with the political reality has increased. Regarding the statement 'In Korea, a few people control politics regardless of the will of the majority of citizens,' 82.1% agreed (very agree + generally agree). The response rate for disagreement (not at all disagree + not very disagree) was only 15.0%. This reflects the disappointment that politics is not sensitive to the voices of the majority of citizens.

■ Along with politics, disappointment with politicians is also widespread among the general public. Regarding the statement 'Generally, politicians are significantly different after being elected compared to their actions during the election,' 93.4% agreed (very agree + generally agree). Those who disagreed (not at all disagree + not very disagree) accounted for only 5.5%.

[Figure 13] Political Efficacy Survey Results (%)

People like us have no right or ability to speak about what the government does

Since a lot of people vote, it doesn't matter whether I vote or not

In Korea, a few people control politics regardless of the will of the majority of citizens

Generally, politicians are significantly different after being elected compared to their actions during the election

Note) 'Agree' is the sum of 'very agree' and 'generally agree,' and 'disagree' is the sum of 'not at all disagree' and 'not very disagree.'}]} is not valid JSON. Please provide a valid JSON object. The error is: Expecting property name enclosed in double quotes: line 1 column 2 (char 1)