基于CTT和IRT的意大利语试题质量研究  

Analysis and Research on the Quality of Italian Language Test based on CTT and IRT

作　　者：刘春红 LIU Chun-hong(School of Continuing Education,Beijing Language and Culture University,Beijing,China 100083)

机构地区：[1]北京语言大学培训学院,北京100083

出　　处：《现代教育技术》2018年第8期72-77,91,共7页Modern Educational Technology

基　　金：北京语言大学校级项目"中央高校基本科研业务费专项资金"(项目编号:16YJ090002)的资助

摘　　要：文章基于经典测量理论和项目反应理论,运用项目分析技术和测验等值技术,对2010~2016年意大利语初级考试试题进行难度及区分度分析,并将IRT与CTT的难度及区分度进行比较,然后采用垂直等值法分析不同年份间考题的稳定性。最后,文章结合二语习得理论与意大利语学科特点,提出完善试题难度及区分度的方法。作为国内首例意大利语试题质量研究,文章可为意大利语考题的编制、建立和健全意大利语等级考试题库提供有益的实践尝试。Drawing upon the Classical Testing Theory（CTT） and the Item Response Theory（IRT）, and using item analysis techniques and test equivalence techniques, the research analyzed the difficulty and discrimination of the items extracted from preliminary Italian tests from 2010 to 2016. Then, a comparison on the difficulty and discrimination between IRT and CTT was carried out, and the stability of items was analyzed by using vertical equivalence techniques. Finally, combined the second language acquisition theory and the characteristics of the Italian language, strategies for improving the difficulty and discrimination of the test items were proposed. As the first domestic research on quality analysis of Italian test, this article could serve as a useful attempt to provide empirical basis for the preparation, establishment and improvement of the item bank construction of Italian language test.

关 键 词：经典测量理论(CTT) 项目反应理论(IRT) 国家公派留学意大利语考试 意大利语初级考试 

分 类 号：G40-057[文化科学—教育学原理]