【摘要】 选题策略是计算机化自适应测验(CAT)的核心.该文提出了一种新的选题策略,是一种相对严格的"升a"方法,它选择区分度参数的百分等级尽可能接近测验进程的项目,而且还可以通过调整控制参数的取值来满足不同测验场景的需求.Monte Carlo实验结果表明:该方法在测验精度、项目曝光率控制和题库利用率等方面均表现良好.更多还原

【Abstract】 The item selection strategy is an important content in computerized adaptive testing(CAT).A new item selection strategy about composing the percentile rank of the discrimination parameter for an item in the bank with the particular examinee′s test process is introduced in this paper.The control parameter can be changed to meet the different test situations.New method has better performance through the Monte Carlo simulation method on improving the test precision, controlling item exposure and the utilization of the item bank.更多还原