On-the-Fly Parameter Estimation Based on Item Response Theory in Item-Based Adaptive Learning Systems

Autor/inn/en	Jiang, Shengyu; Xiao, Jiaying; Wang, Chun
Titel	On-the-Fly Parameter Estimation Based on Item Response Theory in Item-Based Adaptive Learning Systems
Quelle	(2022), (21 Seiten) PDF als Volltext Verfügbarkeit
Zusatzinformation	Weitere Informationen
Sprache	englisch
Dokumenttyp	gedruckt; online; Monographie
Schlagwörter	Item Response Theory; Item Banks; Bayesian Statistics; Learning Management Systems; Electronic Learning; Individualized Instruction; Algorithms; Evaluation Methods + Suchen Sie Ihr Suchwort? Item-Response-Theorie; Individualisierender Unterricht; Algorithm; Algorithmus
Abstract	An online learning system has the capacity to offer customized content that caters to individual learner's need and has seen growing interest from industry and academia alike in recent years. Different from traditional computerized adaptive testing setting which has a well-calibrated item bank with new items periodically added, online learning system has two unique features: (1) the number of items is large, and they are likely not gone through costly field testing for item calibration; and (2) individual's ability may change due to learning. Elo rating system has been recognized as an effective method for fast update of item and person parameters in online learning system to enable personalized learning. However, the updating parameter in Elo has to be tuned post-hoc, and Elo is only suitable for Rasch model. In this paper, we propose to use a moment-matching Bayesian update algorithm to estimate item and person parameters on the fly. With sequentially updated item and person parameters, a modified maximum posterior weighted information criterion (MPWI) is proposed to adaptively assign items to individuals. The Bayesian updated algorithm along with MPWI is validated in a simulated multiple-session online learning setting and the results show that the new combo can achieve fast and reasonably accurate parameter estimations that are comparable to random selection, match-difficulty selection as well as traditional online calibration. Moreover, the combo could still function reasonably well with as low as 20% of items being pre-calibrated in the item bank. [This is the online first version of an article published in "Behavior Research Methods."] (As Provided).
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2024/1/01