A Large Scale Testing System for Learning Assistance and Its Learning Analytics

廣瀬 英雄
Hideo Hirose


統計数理, Vol.66, No.1, pp.79-96, (2018.6)



One of the most crucial issues in universities where a variety of enrolled students shall be educated to a level of universities' diploma policy is to identify students at risk for failing courses and/or dropping out early, to take care of them, and to reduce their risks. For this purpose, Hiroshima Institute of Technology implemented the newly developed online testing system using the item response theory for students' ability evaluation into the follow-up program for fundamental undergraduate education in April 2016, and the system has been well operating since then. The subjects are analysis basic (calculus) and linear algebra. The learning data have been accumulated sufficient for assessing primary learning analytics. In this paper, we first demonstrate our case as a large scale testing system for accumulating learning data steadily, and then describe whether we can identify students at risk by analyzing such data in the early stages. It is worth mentioning that the risk factors we have been ambiguously felt have revealed by using the data statistically. Although the subjects are mathematics, this kind of system will easily be applied to other subjects such as Statistics, Statistics Education, and STEM (Science, Technology, Engineering and Mathematics).

Key Words

ラーニングアナリティックス, ドロップアウト, フォロー アッププログラム, 項目反応理論, 習熟度確認テスト, アダプティブテ スティング

learning analytics, drop-out, follow-up program, item response theory, learning check testing, adaptive testing.



Times Cited in Web of Science:

Times Cited in Google Scholar:

Cited in Books: