摘要
大学各门课程的考试一般应为目标参照性考试,这种考试旨在检测学生有没有达到教学目标的要求或达标的程度,因而分数的分布不应一味地追求正态分布,其难度和区分度的要求也与常模参照性考试有所不同,信度的估计应采用全等性信度而不应采用相似性信度。
The tests of all subjects in universities are objective-referenced in general, which.aim at examining whether students achieve the teaching target or the extent of attaining the goal, so normal distribution should not be only criterion in the analysis of the distribution of students＇ marks. Moreover, the standards of difficulty and distinction is different from norm-referenced testing, Congruent reliability, instead of similar reliability, should be employed in the assessment of reliability
出处
《合肥工业大学学报（社会科学版）》
2006年第2期19-21,共3页
Journal of Hefei University of Technology(Social Sciences)
关键词
常模参照性考试
目标参照性考试
正态分布
信度
norm referenced testing
objective-referenced testing
normal distribution
reliability
