学业水平测试中作文评分误差的多面Rasch分析  

The Measurement of Writing Scoring Quality with the Many-Facet Rasch Model

作　　者：刘红云[1] 陈阅[2] 骆方[1] 王云峰[3] 

机构地区：[1]北京师范大学心理学院,应用实验心理北京市重点实验室,北京100875 [2]中科院心理研究所,北京100192 [3]北京教育科学研究院,北京100031

出　　处：《心理科学》2010年第4期925-927,共3页Journal of Psychological Science

基　　金："教育部课程与教材研究中心"主持的"建立中小学生学业质量分析;反馈与指导系统"项目成果之一;国家自然科学基金项目(30870784)资助

摘　　要：使用多面Rasch模型,从评分量表、评分员等层面对参与2007年八年级语文学业水平测试作文评分的17名评分员的评分情况进行了研究。结果发现:(1)评分员的评分等级所对应的能力值呈正常的变化趋势,大部分评分员有较好的内部一致性;(2)不同评分员的宽严程度有显著差异,评分员之间的一致性整体较好;(3)此外,本文还就评分内部一致性较差的几个评分员的评分做了进一步研究。Based on the scoring data of writing tests of the Chinese Students＇ Academic Achievement Assessment, the writing scoring quality of 17 raters were analyzed with the Many-Facet Rasch Model. The results revealed that （1） Corresponding to the students＇ ability, the trend of the raters＇ grading was normal, and most raters exhibited an intra-rater stability. （2） The raters were significantly different in their strictness, but the consistence of the inter-rater scoring was generally good. （3）Besides, the paper did further studies on the raters with low rating reliability.

关 键 词：作文 评分误差 项目反应理论 多面RASCH模型 

分 类 号：B842[哲学宗教—基础心理学]