We present a new item response model and apply it to develop the Automated Test Maker (ATM). Testing itself in language education has a long history, and a variety of models based on the Item Response Theory have been developed. These sophisticated mathematical models contributed to the emergence of the global standardized tests in language education. As a result, the standardized tests have become powerful tools that turn classroom-based practices to subject-specific test performance and overwhelm curriculum-based and teacher-made tests regardless of how different they are. Testing is an area of study or a tool that makes it possible to monitor and improve teaching practices in the classroom. However, it well left behind practitioners and students in the process of its development with complexity. To ameliorate the situation, we applied the Elo Rating System (ERS) and developed the ATM. The ERS was originally developed by Elo in 1978 to rate chess players in a fair manner. The ERS uses a simple formula and has been widely employed for ranking players in online games or sports and language education. The ATM automatically generates a set of multiple-choice questions, reading comprehension tests, and computer adaptive tests. The test item difficulty parameters are automatically estimated in response to students’ responses and we have come to monitor students’ performance real-time along with lessons. These new features made it possible to integrate practice, assessment, and technology towards better practices and contributed to the improvement of students’ English vocabulary learning.

Authors:
Sei Sumi, Kwansei Gakuin University, Japan
Tomomi Deguchi, Kindai University, Japan

About the Presenter(s)
Sei Sumi, Ph.D., is a Full-time Lecturer in the School of Biological and Environmental Sciences, Kwansei Gakuin University, Japan. He received MA from Tokyo Metropolitan University and Ph.D. from Kansai University.

Connect on Linkedin
https://www.linkedin.com/in/sei-sumi-a6b03716/

Connect on ResearchGate
https://www.researchgate.net/profile/Sei-Sumi

Additional website of interest
http://lang-tech.net/

See this presentation on the full schedule – Schedule

Bookmark