摘要
计算机自适应性测试中的连续概率比测试模式,是一种适用于普通教师利用网络技术在班级教学这样的小规模测试中评估学生语言学习效果的测试模式。其基本原理就是估计被试连续测试时答对与答错的概率,然后与"掌握"和"未掌握"两个相互对立的假设作比较而产生相应的决策。它一方面可以弥补基于IRT测试模式应用范围的不足,另一方面可以更好地帮助教师完成对于学生语言能力的评估。
The SPRT model in the CAT is a test model for teachers to assess students＇ language learning results in small-scale tests such as classroom tests through network technology. Its primary scenario is to estimate testee＇ s right/wrong possibility of response to items in a sequential test, and then to make a relevant decision after comparing with two contrary hypotheses of ＂mastery＂ and ＂nonmastery＂. SPRT not only can expand the application beyond the traditional test model based on IRT, but can better help teachers assess students＇ language ability.
出处
《考试研究》
2012年第4期61-67,共7页
Examinations Research
