Political System Change and Electoral Behavior: an Analysis Of Voting Behavior in the 1995 Legislative Yuan Election

Nai-teh Wu, research fellows from Academia Sinica, Prof. Yung-tai Hung, Director of the Election Study Center at National Chengchi University, Prof. Ming-tong Chen from the NTU Graduate Institute of San Min Chu I, and Prof.

Ying-lung You from the Department of Political Science at Soochow University also assisted on the project. The purpose of this project was to develop a better understanding of the relationship between the voting behavior of the electorate and specific characteristics of the political system. Towards this end, we conducted an empirical study of Taiwan\'s 1995 Legislative Yuan Election campaigns.

Our study investigated several different research questions. First, we analyzed the relationship between the socio-economic background of voters (including their class characteristics and group identity) and their party affiliations. We also investigated the origins of party preference, including the role of party images and strong party identity.

In addition, we researched the relationship between ethnic and national identities and voting behavior. This was accomplished by measuring the extent to which voters identified with either \'Taiwanese\' or \'Chinese\' ethnic identity and the extent to which voters supported either Taiwanese independence or Chinese reunification. The phenomenon of the \'Lee Teng-hui Complex\' was another major topic considered in this study.

We measured the effects of the \'Lee Teng-hui Complex\' using political-psychological models of voting behavior, and used our findings to analyze the relationship between mass psychology and voting behavior. In addition, we observed the role of political culture in the attitudes of voters. We also obtained measures of the political participation, political mobilization, and the personal characteristics of candidates in our study.

Finally, we measured the positions of voters on some special issues such as the politics of money, the problem that none of the three parties can win half of the seats, the grand coalition cabinet, and the split of the KMT. Our research methodology utilized the multiple stage sampling method, which included the principles of probabilities proportional to size (PPS), stratified sampling, and systematic sampling. We designed our questionnaire to include both structured and open-ended questions.

The questionnaire asked voters what they most liked or disliked about the parties (particularly focusing on the main parties), their impressions of the candidates, and the origins of these impressions. The questionnaire also asked respondents to rank the most pressing social and political issues facing Taiwan at the time of the 1995 election.