做事测试评卷中的质量控制  

Quality Control for Ratings in a Performance Test

作　　者：刘建达[1] 杨满珍[2] 

机构地区：[1]广东外语外贸大学外国语言学及应用语言学研究中心,广东广州510420 [2]广东外语外贸大学英语教育学院,广东广州510420

出　　处：《外语电化教学》2010年第1期26-32,共7页Technology Enhanced Foreign Language Education

摘　　要：本文讨论了做事测试中人们普遍关心的一个核心问题——评卷质量。评卷人在评卷过程中常受自己的喜好、习惯、期望等等因素的影响，这些偏见往往会导致评卷误差，从而影响评卷质量。评卷是种复杂、易于犯错的认知活动，评卷误差主要来自于三个方面：评卷人、评卷过程、评分标准。本研究以写作测试为例，论述了如何通过多层面Rasch模型监控评卷质量，主要观察评卷人的严厉度是否前后一致、评卷人评卷是否有偏见、评卷人能否一致地、有效地使用评分标准、评卷人是否能很好区分考生的不同能力。This study describes a many-facet Rasch model for quality control for ratings in a performance test. Raters are engaged in a complex and error-prone cognitive process, and a rating involves an evaluative summary of past or present experiences in which the internal computer of the rater processes the input data in complex and unspecified ways to arrive at the final judgment. Rating errors come from three sources ： raters, rating procedures, and rating scales. This study illustrates the control of rating quality through many-facet Rasch analyses, focusing on raters＇ severity/leniency, centrality/restriction of range, accuracy/randomness, and halo effect.

关 键 词：评卷质量 多层面RASCH模型 做事测试 信度 

分 类 号：H319.3[语言文字—英语]