• 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.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.

• 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.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.12 no.1
• /
• pp.141-152
• /
• 2011

During major elections, three terrestrial broadcasting stations in Korea have covered the progresses of election results by announcing the simple sum of ballot counts of all ballot counting stations. The current approach, however, does not properly reflect the actual ballot count differences by ballot counting location, leading to cause unnecessary but possible confusions. In addition, the current coverage approach restricts the broadcasters from using regional poll data gained through exit polls by letting them to use the significant information on a one-off purpose to announce the initial prediction of the poll results and to fully disregard the exit poll results during the ballot counting process. Based on the understanding, this paper is designed to suggest a Bayesian approach to consolidate the exit poll results with the progressive ballot counting results and announce them as such. The suggested consolidation approach is expected to mitigate or avoid the possible confusions that may arise in connection with the different ballot counting paces by ballot counting station.

• Survey Research
• /
• v.6 no.2
• /
• pp.1-32
• /
• 2005

The accuracy of exit poll mainly depends on the sampling method of voting places. For exit poll, we propose a probability sampling method of selecting voting places as an alternative to the bellwether polling place sampling. Through an empirical study based on the 2004 general election data, the efficiency of the suggested systematic sampling from ordered voting places was evaluated in terms of mean prediction error and it turns out that the proposed sampling method outperformed the bellwether polling places sampling. We also calculated the variance of estimator from the proposed sampling, and considered the sample size problem to guarantee the target precision using the design effect of the proposed sample design.

• Survey Research
• /
• v.5 no.2
• /
• pp.3-30
• /
• 2004

In Korea, bellwether voting places were selected for exit poll based on the past voting results. Sometimes, voting place stratification were used to improve the exit poll performance. The sampled voting places are intended to mirror the general voters of the entire electoral district. But few studies have been done as to which sampling method works better. This study compared the four sampling methods-bellwether voting place sampling method, random sampling method, stratified bellwether sampling method and systematic sampling from ordered voting places method. When we applied the four methods to the 2004 general election data, the systematic sampling from ordered voting places method outperformed the other three sampling method. Also, we found that the additional sampling of voting places over nine contribute little to the accuracy of the estimation.

• Survey Research
• /
• v.8 no.1
• /
• pp.55-79
• /
• 2007

This article attempts to provide an overview of the exit polls for the 2006 local elections in Korea. The sampling method, sampling error, non-response rate, and prediction error of the exit polls are reviewed. Also, we explore the fact that the propensity to vote varies according to age and gender of voters. In terms of age and gender, the representativeness of the sample is investigated by comparing to the data released by the National Election Commission. Through this empirical research, we show that the exit poll samples are unbalanced in terms of age and this unbalance may be one of the causes of bias occurred in the prediction of the 2006 local election results. The design effects of the sample design implemented for the exit polls are also examined.

• Communications for Statistical Applications and Methods
• /
• v.26 no.3
• /
• pp.261-272
• /
• 2019

In this paper, we propose a procedure to build a prediction interval of the sum of dependent binary random variables over a graph to account for the dependence among binary variables. Our main interest is to find a prediction interval of the weighted sum of dependent binary random variables indexed by a graph. This problem is motivated by the prediction problem of various elections including Korean National Assembly and US presidential election. Traditional and popular approaches to construct the prediction interval of the seats won by major parties are normal approximation by the CLT and Monte Carlo method by generating many independent Bernoulli random variables assuming that those binary random variables are independent and the success probabilities are known constants. However, in practice, the survey results (also the exit polls) on the election are random and hardly independent to each other. They are more often spatially correlated random variables. To take this into account, we suggest a spatial auto-regressive (AR) model for the surveyed success probabilities, and propose a residual based bootstrap procedure to construct the prediction interval of the sum of the binary outcomes. Finally, we apply the procedure to building the prediction intervals of the number of legislative seats won by each party from the exit poll data in the $19^{th}$ and $20^{th}$ Korea National Assembly elections.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.1
• /
• pp.53-64
• /
• 2014

One of the major problems to forecast election, especially based on survey, is nonresponse. We may have different forecasting results depend on method of imputation. Handling nonresponse is more important in a survey about sensitive subject, such as presidential election. In this research, we consider a model based method of nonresponse imputation. A model based imputation method should be constructed based on assumption of nonresponse mechanism and may produce different results according to the nonresponse mechanism. An assumption of the nonresponse mechanism is very important precondition to forecast the accurate results. However, there is no exact way to verify assumption of the nonresponse mechanism. In this paper, we compared the accuracy of prediction and assumption of nonresponse mechanism based on the result of presidential election exit poll. We consider maximum likelihood estimation method based on EM algorithm to handle assumption of the model of nonresponse. We also consider modified within precinct error which Bautista (2007) proposed to compare the predict result.