1997 年 12 巻 p. 196-215,284
The objective of this study is to systematically analyze the results of the 15th Korean National Assembly election, using aggregation data and adopting the electoral district as a main unit of analysis. First, the effects of the electoral system are explored. Then, the variations in voter turnout and party votes among the districts are investigated in the multivariate context of analysis.
The electoral system for the national legislature has been revised several times apparently in a fair direction. In actuality, however, the system is not substan-tially fair in that the system favors larger parties. The system has consistently raised an entry barrier against new political forces, and become an object of controversy concerning the fairness of representation.
The analysis has shown that the rate of voter turnout in a district hinges largely on the urbanization level of the district: cetris paribus, the more urbanized a district is, the higher the voter turnout is. The relative composition of age groups is the second most important factor for explaining a district's voter turnout. A greater proportion of voters in their fifties means a higher rate of voter turn in an electoral district.
The district-level analysis in this study has also vindicated that a party's electoral showing is greatly determined by the district's region: regional voting was more pervasive in the 15th National Assembly election than in the previous one. But the urban-rural division of partisan support, usually salient in previous elections, has almost disappeared. There is no evidence that policy issues or social cleavage factors other than regionalism have significantly influenced voters' party support.
In brief, the recent election has suggested no clear dealigning or realigning tendencies in Korean party or electoral politics. The election can be characterized as a normal or maintaining election.