2018 Fiscal Year Final Research Report
Development of a Linkability Analysis Method on Reliability of Linked Scores
Project/Area Number |
15K04052
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Research Field |
Educational psychology
|
Research Institution | Kyushu University (2016-2018) Niigata University (2015) |
Principal Investigator |
Sato Yoshikazu 九州大学, アドミッションセンター, 教授 (00300517)
|
Co-Investigator(Kenkyū-buntansha) |
柴山 直 東北大学, 教育学研究科, 教授 (70240752)
|
Project Period (FY) |
2015-04-01 – 2019-03-31
|
Keywords | 教育測定 / 古典的テスト理論 / 対応づけ / 等化 / 対応づけ可能性(対応づけの実行可能性) / テストの信頼性 / 対応づけ得点の信頼性 |
Outline of Final Research Achievements |
When two different tests are linked, we can see the correspondence between the scores of Test X and Test Y, such as 80 points of Test X correspond to 85 points of Test Y. For meaningful linking, it is necessary to examine the linkability, or the feasibility of linking, between the two tests in a comprehensive way. The purpose of this study was to incorporate a new perspective on the reliability of linked scores into linkability analysis. We defined a measurement model of linked scores and then formulated basic indices for the reliability of linked scores. In addition, we showed that the indices enable us to perform linkability analysis in terms of the reliability of linked scores.
|
Free Research Field |
教育測定学
|
Academic Significance and Societal Importance of the Research Achievements |
大学英語とTOEICやTOEFLとの対応づけや,全国学力調査と地方自治体の学力調査との対応づけなど,近年,ますます対応づけへの関心が高まりつつある.対応づけを伴うテストの場合,私たちの関心の多くは共通尺度における受験者の得点,すなわち対応づけ得点にある.本研究では,これまで十分な議論のなかった「対応づけ得点の信頼性」に着目し,定式化した指標を対応づけ可能性分析に応用した.これらの研究成果により,意味ある対応づけが可能なテストの組み合わせかどうかを判断できるため,対応づけの濫用を防ぐなどの効果が期待できる.
|