Bayesian item selection criteria for adaptive testing

被引:101
作者
van der Linden, WJ [1 ]
机构
[1] Univ Twente, Dept Educ Measurement & Data Anal, NL-7500 AE Enschede, Netherlands
关键词
adaptive testing; item response theory; Bayesian statistics; item selection criteria;
D O I
10.1007/BF02294775
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Owen (1975) proposed an approximate empirical Bayes procedure for item selection in computerized adaptive testing (CAT). The procedure replaces the true posterior by a normal approximation with closed-form expressions for its first two moments. This approximation was necessary to minimize the computational complexity involved in a fully Bayesian approach but is no longer necessary given the computational power currently available for adaptive testing. This paper suggests several item selection criteria for adaptive testing which are all based on the use of the true posterior. Some of the statistical properties of the ability estimator produced by these criteria are discussed and empirically characterized.
引用
收藏
页码:201 / 216
页数:16
相关论文
共 21 条