【摘要】 若研究或使用CAT时将项目参数的估计值认为是真实值,则会产生所谓的机会红利(capitalization on chance),国内尚未报道该方面的研究.在定长和不定长CAT测验中考察引入曝光因子的多种分层化选题策略和随机选题策略,综合比较它们在测验精度、题库利用率和机会红利等评价指标中的表现,发现引入曝光因子的2种动态分层方法有更好的表现.更多还原

【Abstract】 Previous studies of CAT have treated item parameter estimates as if they are the true population parameter values,but capitalization on chance may occur. In this article,it is examined that the performance of several stratified methods under more realistic conditions where item parameter estimates instead of true parameter values are used in the CAT. To improve the bank utilization and reduce the captalization on chance,Monte Carlo simulations manipulated some key factors based on 3-parameter Logistic model: fixed or variable test,stratified item pool methods,use or not use of the exposure control factor. The results showed that the two dynamical stratified methods in conjunction with exposure control factor are better.更多还原