Development of System for University Entrance Examinations by applying various methods of multivariate analysis

应用多种多元分析方法开发高考系统

• 批准号: 07308019
• 负责人: YANAI Haruo
• 金额: $ 4.74万
• 依托单位: National Center for University Entrance Examinations (NCUUE)
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (A)
• 财政年份: 1995
• 资助国家: 日本
• 起止时间: 1995 至 1997
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-07308019/
• 关键词: Swap rate; National Center Test; Covariance Rate; Multivariate Analysis; Methods of Equating; Item Response Theory; National Medical Practitioners Qualifying Examinations (NMPQE); 項目反応理論; 国試成績; 入試成績; ニューラル-ネットモデル; 適性検査; 入替り率; CBT; 欠損データ; 2次

During three years, we organized five meetings, in which 64 researches were presented. These researches were classified into the following two groups :The first is on development of new statistical methods which will be useful for applying to various kinds of entrance examination data. For example, application of convariance rates, and swap rate were proved to be useful for measuring the contribution of each of tests employed as the entrance examination tests. Further, a new method was introduced to adjust the low correlation computed between the records of entrance examination and school records after admission into University.The second is on development of computer software programs. They are termed as 1) Principal component analysis with missing values, 2) Partial credit model based on Item Response Theory, and 3) Program for Entrance Examination Data, Faculty of Medicine. In addition, we developed Computer System for University entrance examinations which covers the former three programs as its special cases.We hope that obtained results through our study contribute to improvement of entrance examination systems of various of universities in Japan.

在三年中，我们组织了五次会议，其中提出了64项研究。这些研究分为以下两类：第一类是开发新的统计方法，这将有助于应用于各种入学考试数据。例如，应用方差率和交换率被证明是有用的测量的贡献，每个测试作为入学考试测试。在此基础上，提出了一种新的方法来校正入学考试成绩与入学成绩之间的低相关性。二是计算机软件的开发。它们被称为1）缺失值的主成分分析，2）基于项目反应理论的部分学分模型，3）入学考试数据程序，医学系。此外，我们还开发了以上述三个项目为特例的大学入学考试计算机系统，希望通过我们的研究成果，对日本各大学的入学考试制度的完善有所贡献。

项目成果

前川 眞一: "SASによる多変量データ解析" 東大出版会, (1997)

Shinichi Maekawa：“使用 SAS 进行多元数据分析”，东京大学出版社，（1997 年）

Okada,A.: "University enrollment flow among the Japanese prefectures" Behaviormetrika. 23. 169-185 (1996)

Okada,A.：“日本各县之间的大学入学流动”Behaviormetrika。

Yanai,H.: "Generalized Canonical Correlation Analysis with Linear Constraints" Studies in Classification,Data Analysisj,and Knowledge Organization. (印刷中). (1997)

Yanai, H.：“具有线性约束的广义典型相关分析”分类、数据分析和知识组织研究（1997 年）。

Naoto Chino: Gendai-Suugakusya. Asymmetrical Multiple Dimensional Analysis, (1997)

千野直人：现代Suugakusya。

吉沢正: "日科技連" 多変量解析の事例集第2集, 210 (1997)

吉泽正：《日本科学技术学会》多元分析案例研究第2卷，210（1997）