This study is provided by ICPSR. ICPSR provides leadership and training in data access, curation, and methods of analysis for a diverse and expanding social science research community.
Principal Investigator(s): East Asia Institute (South Korea); JoongAng llbo (South Korea); Seoul Broadcasting System (South Korea); Hankook Research Company (South Korea)
The South Korean Presidential Election Panel Study examined vote determinants of Korean voters and the causes and dynamics of changes in voter preferences. The survey for the 2007 Presidential Election Panel Studies in South Korea was conducted from April to December 2007 in six waves with a large-scale panel of 3,500 representing the nation’s gender, age, region, and education proportions.The study analyzed factors that influence the formation and change of voter preferences through three broad theoretical frameworks: (1) The sociological model that explains voter preference as a reflection of major social fragmentation (education, gender, income, religion, region, etc.); (2) The psychological model of the Michigan School that explains voter preference formation and change as activation of party identification in United States or Western elections, and regional identification in Korea as a proxy; (3) The rational voter model that posits that individuals, after calculating their own interests, support candidates or parties that possess the policies and ideology to maximize those interests. Tracking core questions are questions that repeatedly track the change in vote determinants, the core project of the current study. The tracking core questions that will track public opinion change over the entire period of the panel surveys consist of questions about attitudes on candidate factors, political party factors, election campaigns, issues, and policies. Whenever possible, the questions use the wording of international surveys such as the ANES questionnaire to allow international comparisons, but when not appropriate to the reality of Korean elections, new question items are developed. The current study analyzed which of the three models is the most appropriate in explaining Korean elections and voter preference.
These data are available only to users at ICPSR member institutions. Because you are not logged in, we cannot verify that you will be able to download the data.
East Asia Institute (South Korea), JoongAng llbo (South Korea), Seoul Broadcasting System (South Korea), and Hankook Research Company (South Korea). South Korean Presidential Election Panel Study: Six Waves, 2007. ICPSR26661-v1. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2010-03-02. http://doi.org/10.3886/ICPSR26661.v1
Persistent URL: http://doi.org/10.3886/ICPSR26661.v1
This study was funded by:
- JoongAng llbo (South Korea)
- Seoul Broadcasting System (South Korea)
Scope of Study
Smallest Geographic Unit: county
Date of Collection:
Universe: All individuals over the age of 19 who were eligible to vote in South Korea.
Data Types: survey data
Sample: About half (49 percent) of the sample was newly selected while extracting the other half (51 percent) from the Hankook Research master sample which possesses a relatively high panel retention rate allowed for minimal panel attrition. The Hankook Research master sample consists of a pool of around 130,000 respondents, which represent the national population based on a combination of gender, age, and region. The panel study sample was selected using stratified random sampling. The data for setting the quotas were determined from the Resident Registration data as of December 31, 2006, published by the Ministry of Government Administration and Home Affairs, with the quotas set according to demographic proportions by age, by gender, and by region. The sample was allocated by crossing the proportions of each region (the 16 metropolitan self-governing bodies), gender (male/female), and age group (29 years of age and under, 30-39 years of age, 40-49 years of age, 50 years of age and over). The quotas were applied separately to each sample group from the master sample and from the newly selected sample.
Weight: Four variables (gender, age, region, and education) were weighted according to their national proportions and applied to the responses of the remaining respondents. The final analysis data was calculated by adding a weighting factor based on the education level distribution of all respondents to the weighting factors by each region, gender, and age combination that formed the basis of the sampling quotas.
Mode of Data Collection: computer-assisted telephone interview (CATI)
Response Rates: The final panel retention rate for the South Korean Presidential Election Panel Study, 2007, was 60.3 percent, thereby meeting the 60 percent goal established in the planning stages. When panel attrition rates for each wave are compared, the 17 percent drop between Waves I and 2 was the highest attrition rate, and then in Wave III 11 percent (of the original sample size) dropped. On the other hand, from Wave IV on, the panel attrition rate decreased drastically to within 5 percent. It may be considered that in order to raise the panel retention rate it is important to reduce the attrition rate early on in the panel study such as in Wave II or Wave III. Note: More detailed information is contained in the codebook.
Extent of Processing: ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major statistical software formats as well as standard codebooks to accompany the data. In addition to these procedures, ICPSR performed the following processing steps for this data collection:
- Checked for undocumented or out-of-range codes.
Original ICPSR Release: 2010-03-02
Related Publications (see Notes)
- List all ~13 citations associated with this study
Most Recent Publications
- Citations exports are provided above.
Export Study-level metadata (does not include variable-level metadata)
If you're looking for collection-level metadata rather than an individual metadata record, please visit our Metadata Records page.