ARRG法在消除随机猜答影响中的应用研究综述

Rasch模型框架下,考生能力和题目难度的不匹配会导致考生猜测作答现象的出现,ARRG法的提出即是为了消除考生猜测作答对参数估计的影响。ARRG法通过原始分析、裁剪分析、锚分析和全锚分析,能够实现题目难度和能力参数的调整,该方法中概率阈限的设置、样本量的选择及结果的比较也是需要关注的重点。

Review of ARRG Method to Remove the Influence of Random Guessing Responses

Under the framework of Rasch model,the mismatch between candidates’ability and question difficulty will lead to the phenomenon of candidates’guessing responses.The ability removing random guessing(ARRG)method is proposed to eliminate the influence of candidates’guessing responses on parameter estimation.ARRG method can adjust the problem difficulty and ability parameters through original analysis,tailored analysis,anchored analysis and all-anchored analysis.At the same time,the setting of probability threshold,the selection of sample size and the comparison of results in this method are also the key points that need to be paid attention to.