The Influence of the Prediction Nwes on the Election Return

Research Project

Project/Area Number	07610196
Research Category	Grant-in-Aid for Scientific Research (C)
Allocation Type	Single-year Grants
Section	一般
Research Field	社会学(含社会福祉関係)
Research Institution	Seikei University
Principal Investigator	YANAI Michio Seikei Univ.Faculty of Humanities professor, 文学部, 教授 (10054321)
Project Period (FY)	1995 – 1997
Project Status	Completed (Fiscal Year 1997)
Budget Amount *help	¥1,400,000 (Direct Cost: ¥1,400,000) Fiscal Year 1997: ¥400,000 (Direct Cost: ¥400,000) Fiscal Year 1996: ¥400,000 (Direct Cost: ¥400,000) Fiscal Year 1995: ¥600,000 (Direct Cost: ¥600,000)
Keywords	prediction news / election news / maiar constituency system / announcemert effect / election return / proportional representative / turnout
Research Abstract	I tried to contrast the prediction nwes on the election of the members of the House of Representatives and the election returns. There are 3 kinds of prediction news on the election. 1) prediction on the turnout of voters, 2) number of the parliamentary seats of each polotical party, 3) success or defeat of each candidate. There has been no problem to announce the first one. But there has been complains or pressure from the political parties to the announce of the second one. And also complains from the candidate camps. Concerning the second one, predictions proved considerably right from the standpoint of probability theory. But when there is a very big problematic issue, or when the specific party is going to take so many seats, the returns sometime differ from the predictions. Analysing these differences between predictions and returns under the Japanese major constituency system, the narrow margin of the votes divide the last-place successer and the runner-up candidate, and this narrow margins deside winners and losers in many constituencies allover the country, and then the predictions seriously fail in. The predictions about success or defest of each candidate proved considerably right. Sometime predictions fail badly and get much attention. In these cases the prediction announces that the certain candidate is top rated for the success, and then the supporters may abstain from voting, or vote for second most favorite candidate to get one more seat. For these predictions newspaper companies conduct scientific polls and statistical processing. So the predictions are very accurate. But in some cases predictions might be maipurated for some purpose.