AUTHOR=Xiong Jianhua , Ding Shuliang , Luo Fen , Luo Zhaosheng 

TITLE=Online Calibration of Polytomous Items Under the Graded Response Model

JOURNAL=Frontiers in Psychology

VOLUME=Volume 10 - 2019

YEAR=2020

URL=https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2019.03085

DOI=10.3389/fpsyg.2019.03085

ISSN=1664-1078

ABSTRACT=Computerized adaptive testing (CAT) is an efficient testing mode，which allows each examinee to answer appropriate items on the basis of his or her latent trait level. The implementation of CAT requires a large-scale item pool, and item pool needs to be frequently replenished with new items to ensure test validity and security. Online calibration is a technique to calibrate the parameters of new items in CAT, which seeds new items in the process of answering operation items, and estimates the parameters of new items through the response data of examinees on new items. The most popular estimation methods include one EM cycle method (OEM) and multiple EM cycle method (MEM) under dichotomous item response theory models. This study extends OEM and MEM to the graded response model (GRM), a popular model for polytomous data with ordered categories. OEM and MEM are all iterative algorithms, the accuracy of estimation depends on the accurate initial value of item parameters, then, a new method for calculating the initial value of item parameters is presented. The simulation research is carried out under different categories numbers and sample sizes. The results show that the accuracy of new items parameters calibrated by MEM algorithm under GRM can be acceptable, and the accuracy of the parameters of the new item increases with the small increase of the calibration sample size.