• Survey Research
• /
• v.4 no.2
• /
• pp.87-102
• /
• 2003

Till now, much progress has been made in election forecasting. But fixed-line telephone survey has limitations, because it has become more and more difficult to get a representative sample, Exit poll has been considered to provide a new solution. From the beginning, exit poll provided more accurate predictions than those based on surveys using fixed-line phone. In 2002 presidential election, prediction based on exit poll was perfect. Its predictive error was zero. This paper describes how the exit poll was done in 2002 presidential elections. Specifically, we are to show the estimating procedures as well as sampling and polling process. Among many factors, sampling procedure has been fond to be the most important factor in exit poll accuracy.

• Asian Journal for Public Opinion Research
• /
• v.11 no.4
• /
• pp.274-303
• /
• 2023

The third-person perception phenomenon can consistently be found in opinion poll news, but it remains unknown what determines the degree of the third-person perception (TPP) about specific election poll news. We investigated how respondents' preferred candidate's status in the poll affects the perceived impact of polling news on both themselves (PMI1) and on others (PMI3) as well as TPP (PMI3 - PMI1). We also examined the effect of subjective political knowledge and the perceived level of political knowledge of others on TPP. An online experiment was conducted in the context of a gubernatorial election in South Korea, in which the leading candidate in the poll and the question order (self-question first vs. other-question first) were manipulated. The results indicated that PMI1 and PMI3 were greater when the respondent's preferred candidate was leading in the poll. TPP did not differ depending on subjective knowledge, but it was greater when the others were non-experts (vs. experts). Lastly, question order was found to be a method factor that affected both PMI1 and PMI3. The theoretical and practical implications of these findings are discussed.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2000.11a
• /
• pp.21-39
• /
• 2000

Researchers in sample surveys have long recognized that what people say they are going to and what they do are not always the same, to say the least. Naturally this inconsistency distorts the survey results, especially in election poll surveys in Korea. How can one cope with this measurement bias\ulcorner This paper suggests two ideas: election poll tree and using posterior probability. By using two ideas the inconsistency of the poll survey results could be improved much or less.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2000.06a
• /
• pp.35-57
• /
• 2000

Researchers in sample surveys have long recognized that what people say they are going to and what they do are not always the same, to say the least. Naturally this inconsistency distorts the survey results, especially in election poll surveys in Korea. How can one cope with this measurement bias\ulcorner This paper suggests two ideas: election poll tree and using posterior probability. By using two ideas the inconsistency of the poll survey results could be improved much or less.

• Survey Research
• /
• v.1 no.2
• /
• pp.35-57
• /
• 2000

Researchers in sample surveys have long recognized that what people say they are going to and what they do are not always the same, to say the least. Naturally this inconsistency distorts the survey results, especially in election poll surveys in Korea. How can one cope with this measurement bias? This paper suggests two ideas: election poll tree and using posterior probability. By using two ideas the inconsistency of the poll survey results could be improved much or less.

• Korean journal of communication and information
• /
• v.62
• /
• pp.159-178
• /
• 2013

The current study examined the effects of the voters' perception of television news coverage of election poll results on their political participation intention. 700 voters participated in a telephone interview three weeks before the 2012 Korean presidential election. A structural equation modeling with the nationally representative sample was performed. The findings indicate the respondents were more likely to evaluate television news coverage of election poll results negatively when the news coverage presented that the candidate they supported was behind in the race, and the negative evaluation was linked to a greater third-person perception. The third-person perception, in turn, had an indirect effect on political participation intention through negative emotional responses. The results imply that voters' political position influences their perception of the television news coverage of election poll results, and this perception can have indirect effects on political participation.

• Survey Research
• /
• v.4 no.1
• /
• pp.31-48
• /
• 2003

We consider a history of exit poll which is generally used in the election forecasting survey. And we discuss some problems and improving items based on the results of executed in Korea. In addition, we consider the errors due to the insufficient preparation for exit poll and methods to reduce these errors. From this study, we expect the election forecasting will be more accurate.

• Survey Research
• /
• v.11 no.3
• /
• pp.33-55
• /
• 2010

In this study, we newly define the Total Survey Error(TSE) in exit poll and investigate the TSEs of the exit poll survey for the 18th general election of 2008 to analyse the cause of the exit poll prediction error. To explore the main cause and effect of the total survey error, the total survey error was divided by the sampling error which comes from sampling process of poll stations and the non-sampling error which comes from selecting voter and collecting responses from sampled voters in each electoral district. We consider the relationship between non-response rates and total survey error as well as non-sampling error. Also, we study the representativeness of the exit poll sample by comparing the sex/age distribution of the exit poll data and the National Election Commission poll data.

• Survey Research
• /
• v.12 no.3
• /
• pp.25-48
• /
• 2011

In this paper, we overview the sample design, sampling error, non-response rate and prediction errors of the exit poll conducted for 2010 local elections and discusses how to detect a prediction bias in exit poll. To investigate the bias problem in exit poll in regional(Si-Do) level, we analyze exit poll data for 2007 presidential election and 2006 local elections as well as 2010 local elections in Korea. The measure of predictive accuracy A proposed by Martin et al.(2005) is used to assess the exit poll bias. The empirical studies based on three exit polls clearly show that there exits systematic bias in exit poll and the predictive bias of candidates affiliated to conservative party (such as Hannara-Dang) is serious in the specific regions. The result of this study on systematic bias will be very useful to improving the exit poll methodology in Korea.

• Survey Research
• /
• v.12 no.1
• /
• pp.1-30
• /
• 2011

The purposes of the present study were to analyze correlation of vote behavior and attitude and vote intention in the pre-survey, and to investigate the efficient method of predicting the voting result from the pre-surveys. The previous attitude is measured by the support for the candidate, political self-confidence, self-efficacy and opinion on present issues. The vote intention is surveyed by the past election participation and degree of election interest. Real voting behavior is surveyed by the post enumeration, and the pre-survey and both post-survey are conducted to the same person to analyze the correlation of voting behavior and pre-survey. The real election participation is highly correlated with vote intention, election interest and past election participation. Almost respondents did not change the supporting candidate from the poll survey to the election vote. It is shown that the voting behavior at election of the nonrespondent of pre-survey can be predicted with the demographic charater and attitude of present issues.