The Deactivated Voter and Attitudinal Ambivalence

다국어 초록 (Multilingual Abstract)

At issue in this dissertation are understandings about how people change their voting decisions and what mechanisms underlie this dynamic process. The major distinction penetrating this project is between activation and deactivation. A voter is defined as "activated" if he/she changes a voting decision in the direction of his/her own party candidate, while the voter who moves in the opposite direction, toward the opposite party candidate, is a "deactivated voter."
The main objective of this paper is to identify the systematic factor that distinguishes the deactivated voters from the activated ones. As for the underlying mental mechanism of activation, attitude consistency theory and the prior-driven information updating process in general, and in the electoral context the role of party identification as a political prior driving subsequent perceptions are first reviewed. Like most attitude theories, these biased information perspectives are commonly based on the reciprocal-bipolar perspective of the human evaluative process. However, recent developments in cognitive science have challenged these bipolar perspectives of attitude.
Conventionally, attitudinal ambivalence is defined as the simultaneous existence of positive and negative attitude toward a single attitude object. Extending the definition of ambivalence beyond a single object, "ambivalent party attitude" is defined as conflicting or inconsistent attitudes toward the two major parties. In this paper it is proposed that ambivalent party attitude, by moderating the role of party identification as a prior, functions as the underlying factor of deactivation.
Utilizing multinomial and ordered logit estimation and the 1980 NES Major Panel Studies, I attempt to provide a more comprehensive picture of the dynamics of voting decisions over an election year than past studies have done. Consistent with past studies, the results show that activation was the major dynamics in the 1980 presidential election. Contrary to previous research, however, between-party difference in terms of activating their party identifiers, an asymmetric activation, was the main factor in explaining Reagan's unexpected landslide victory. The finding of an asymmetric activation is also consistent with the retrospective voting perspective in that if the party-line-based symmetric activation perspective is true, then there is little room left for the objective national conditions to have a substantial impact on election outcomes.
Ambivalent party attitude as well as strength of party identification emerges as the main determinant in explaining within-party variation in terms of party-line voting and perception. The results from five recent presidential elections confirm the association between ambivalent attitude and defection. In addition, the paper also shows that ambivalent attitude is associated with less biased perceptions and attitudes. The findings reported in this paper suggests that those with ambivalent party attitude may be at the core of responsible voters in V.O.Key's term.
The finding that, contrary to the expectation derived from the conventional bipolar perspective of attitude toward parties, attitudinal ambivalence has an independent consequence on individual voting, supports the bivariate perspective of attitude, and thus, merits future research in this line.

목차 (Table of Contents)

• Table of Contents = vi
• List of Figures = vii
• List of Tables = viii
• Acknowledgements = x
• CHAPTER 1 Introduction = 1
• CHAPTER 2 Deactivation in the 1980 Presidential Election = 6
• Why a Unified Dependent Variables in Studying the Change in Voting Decisions? = 6
• The Change in Voting Decision in the 1980 Presidential Election = 11
• CHAPTER 3 Deactivation and Attitudinal Ambivalence = 24
• Cross Pressures, Attitude Conflict, and Attitudinal Ambivalence = 25
• Structure of Ambivalence and Its Behavioral Consequences = 28
• Operationalization of Meta Ambivalence = 33
• Model Specification and Results = 36
• CHAPTER 4 Ambivalence, as a Moderator on Perceptional Polarization = 48
• Attitude Consistency, Perceptional Polarization, and Activation = 49
• Attitude Inconsistency or Ambivalence, and Moderation = 57
• Model Specification and Results = 63
• CHAPTER 5 Electoral Consequences of Ambivalence in Recent Elections: 1984-2000 = 86
• CHAPTER 6 Conclusion and Implications = 107
• References = 112