TW
EN
News
About
Editorial Foreword
Open Access
Purpose
Editorial Foreword
Open Access
Purpose
Editorial Board
Edit Committee
Publication Ethics and Malpractice Statement
Publish
Edit Committee
Publication Ethics and Malpractice Statement
Publish
Contribution
Submission Guideline
Peer Review
Paper Format
Submission Guideline
Peer Review
Paper Format
Submit online
Issues
Online First
Current papers
Backissues
Online First
Current papers
Backissues
Community
Search results
home Home navigate_next Search results

Search results

Abstract

This study looks into issues evolving from dual-frame telephone surveys carried out in Taiwan. Some issues have been resolved, while others are still being explored. Directions of future studies are also suggested.
This study employs indicators of election prediction errors to evaluate the effectiveness and pros and cons of various estimation procedures for combining data sets collected from dual-frame telephone surveys.
There are basically three types of combining dual-frame survey data, namely, all landline survey data plus those of cell-phone only; all cell-phone survey data plus those of landline only, and combing both sets of data according to their coverage proportions in the population. This study shows that the landline phone survey samples differ systematically from the cell-phone survey samples in many demographical characteristics. Using a total predicting error indicator to compare the results of various estimation procedures in three election results, namely, the 2016 presidential election, the 2018 referendum on the same-sex marriage law, and the 2020 presidential election, this study finds the following: (1) The original data collected from the landline survey fares the best. (2) Adjusting data by applying weights derived from government household registration data enlarges the prediction errors for the landline phone survey but performs slightly better for the cell-phone survey. (3) Due to selection bias from cell-phone survey samples, the prediction errors tend to be further away from the true election results. The damage done does not redeem its good intention for correcting the coverage shortcomings caused by the traditional landline phone survey.
The study also suggests several issues to be explored in future studies: (1) Investigate further the characteristics of using both landline and cell-phone samples for landline and cell-phone surveys. (2) Investigate the mode effects on landline and cell-phone surveys. (3) Investigate the performance of dual-frame telephone surveys in areas other than electoral studies.

Keywords

Coverage rate
cell-phone only
dual-frame telephone survey
post-stratified estimating procedure
selection bias
election forecasting

Abstract

This paper examines the process of voters' voting decision and develops a way of predicting voters' voting decision. Most surveys on election prediction are based on respondents' answers on their vote choices. However, this paper indicates two serious problems in this way of election prediction. One problem comes from the indecision of most voters, and the other problems comes from measurement errors. To deal with this problem, this paper provides a way of estimating each voter's probabilities of voting for each candidiate. The author indicates that the probability of voting for a particular candidate is not 0, neither is 1. Instead, the probability of voting for a particular candidate ranges from 0 to 1. The author bases on four major variables--candidate evaluations, party identification, incumbent government satisfaction, and ethnicity-- to construct a vote choice model, and estimates the model by way of a multinomial logit model. Then, the author calculates individual voters' probabilities of voting for each of the candidates. The research finding shows that the chance of voting for a particular candidate is large if a voter has a high probability of voting for the candidate. Based on this principle, the paper predicts the outcomes of four elections and two of them are quite accurate.

Keywords

Voting decision
Voting behavior
Election forecast
Multinomial logit model

Abstract

Managers of campaign headquarters often find it difficult to balance resource allocations between mobilization and propaganda divisions. The issue is critical in the battlefield but usually over-looked by academic circle. The study looks into the phenomenon using both aggregate and survey data and tries to explore the possibility of utilizing results of opinion polls for evaluating candidate performance. Analysis of aggregate data confirms the strength of mobilization but finds its role less significant. Its share of vote getting is shrinking as time changes. On the other hand, analysis of survey data shows that opinion polls can actually reflect the effectiveness of candidates' strategy in image building. However, if a re-search work want to get the most out of opinion polls it must plan ahead, hands in hands with campaign strategy, and put in more ingredients of network and communication behaviors.

Keywords

Election forecasting
campaign strategy
aggregate data
survey data

Abstract

In election forecast, the ability to predict the intentions of undecided voters plays an important role in determining its accuracy. This article uses discriminant analysis with demographic variables, issue positions, and attitudinal variables, respectively, to classify voters who stated their voting preferences. We found that comparatively speaking, attitudinal variables can correctly classify voters with the highest percentage, followed by issue positions and demographic variables. However, none of the above models displayed satisfying results, not even the combined model including all three types of variables. Therefore, there are reasonable doubts employing these variables to construct models predicting the vote intention of undecided voters. The author then attempts to use feeling thermometer scores to classify voters who expressed their vote intention sand to predict the inclinations of undecided voters. After contrasting the results with the actual election outcome, we found that feeling thermometer scores have better discriminant power than previous models. The author also found that among the supporters of the three major candidates, ideologically speaking, Chen Shui-Bian's and Lian Chan's supporters are furthest away from each other on their average positions, while James Soong's supporters' average ideological position is closest to that of Lien Chan's supporters.

Keywords

presidential election
election forecast
feeling thermometer
discriminant analysis
demographic variables
issue position
attitudinal variables
political landscape

Abstract

Due to undervotes, misvotes, or switchvotes bias, many polling data users felt frustrated in using the past polling outcome to forecast the new election. It is commonplace for voters to note an early frontrunner in polls will be doomed to fall in the real election outcome. A beta-binominal distribution is suggested to model the accuracy of early poll outcome which strategically influences the polling data users such as political parties, candidates, and mass media in implementing the election campaign. We demonstrate the advantages of probabilistic distribution and Bayesian reasoning, and how to estimate the parameters from past data, in modifying the accuracy of prior poll outcomes. In comparison with the traditional frequency approach, beta-binominal mixture distribution imposes a statistical-adjusting framework with ability to proportionate a coherent mechanism that synthesizes the performances of prior votes. The empirical data sets include the 2004 US presidential election in Atlas Web and TVBS polls in 2006 Kaohsiung mayor election and 2008 presidential election in Taiwan. This paper describes the general fitting of beta-binomial distribution on both datasets and discusses fruitful avenues for future research.

Keywords

Public Polling
Voting Behavior
Election Forecasting
Mixture Distribution
Bayesian Statistics

Abstract

The advancement of information and communication technologies has greatly changed the lifestyle of people while using landline surveys in soliciting precise public opinion is becoming limited. As people use a variety of devices such as cellphones, internet phones and APPs in daily communication, problems of insufficient population coverage arise from relying only on landline phones to reach respondents. Therefore, a daunting task in the telephone polling industry is to ensure sample representation for obtaining precise population parameters. To achieve such an objective, a common practice by pollsters in Taiwan is to use household data as
weighting statistics. Many cases, however, have shown this practice to be inappropriate.
To solve the above-mentioned problem, this study proposes an estimation method based on a dual frame survey that combines landline phones and cellphones. We further use data from the 2016 presidential election to compare different estimations based on a dual frame survey. Our results demonstrate that a “landline survey supplemented by cellphoneonly” is the best combination, considering sample coverage and estimation error. The second-best alternatives are “cellphone survey supplemented by landline-only” and “use both landline and cellphone.” In other words, “the most economical and efficient” strategy of a dual frame survey is to conduct a traditional landline survey and incorporating cellphone-only respondents. The data collected in such combination not only reflect the characteristics of the population, but also cost much less than other strategies.

Keywords

coverage rate
cellphone-only
dual frame telephone survey
poststratified estimation procedures
the 2016 presidential election forecasting