Dually Adaptive Online IRT Testing System to the Large Scale Data


Yoshiko Tokusada, Hideo Hirose


2nd International Symposium on Applied Engineering and Sciences (SAES2014), Big Data Session 2, December 20-21, 2014, Fukuoka, Japan

The item response theory (IRT) provides us not only the abilities of examinees but also the difficulties of items (problems), and it is believed that the results are fairer and more accurate than those by the classical test methods. Adaptive testing using the IRT selects the most appropriate items to examinees automatically, resulting more accurate ability estimation and more efficient test procedures, where the term gadaptiveh means the adequate item selection. The difficulties of the items are determined somehow, e.g., by using the monitor test, in advance. Such adaptive testing fits well online test systems. However, as the number of examinees is growing in online testing, the difficulty values measured by the monitor test will possibly be different from those assessed by the new examinees. Then, calibration of the difficulty values may be required. For such conditions, we use the dually adaptive online IRT testing system, where gdually adaptiveh means that one is targeted to the adequate item selection and the other is targeted to the adjustment of the difficulty values for items. The monitor tests are not necessarily required for the newly added items in the proposed system. We have investigated how this system works using the simulation study, and we applied this method to the high-school mathematics testing case and university mathematics case. This system can be applicable to many fields such as medical area, human resource division, a variety of tests. In this paper, we focus on the large scale data.

Key Words



Times Cited in Web of Science:

Cited in Books: