home
Home
navigate_next
Search results
Search results
Abstract
Prior to the 1995 Legislative elections, the only party which could equalize the vote was the KMT. Because KMT could reduce the cost of winning a seat, it was always overrepresented. Although the opposition parties did not have strong grass-root organizations, they still looked for other ways to equalize the vote to prevent underrepresentation. In the 1995 elections, the New Party (NP) used a new method (forced vote distribution) to equalize their vote in Taipei City, Taipei County and Taoyuan County. This method enabled 12 of their 14 candidates to win. Why was the NP able to use forced vote distribution in the election? Did all nominees in a particular district winning seats mean the forced vote distribution had been effective? In this paper, I discuss the policy making procedure and the factors that influence the effect of the forced vote distribution. Finally, I give some advice to the parties that will use this campaign strategy in the future.
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.
Abstract
There are three types of error in Taiwan's election prediction:errorsfrom survey process,transplanting of measure scale,and no response voter.This paper try to reduce the error of election prediction via”Joined IdiosyncrasiesAdjusted Model”(JIA Model).JIA Model was operated bytwo stages.First,I compute a basic model,which reflects some generalfactors in every county.Secondly,I design extended model to adjust theoutput of basic model.In JIA Model,I try to use logistic regression tocompute the candidate's ballots,and present the final prediction in probability.In this paper,I try to explain in example of 1997 Taipei MayoralElection.I consider the factors of issue orientation,strategic voting,socialcontext and political map.JIA Model made the error less than samplingerrors.
