综合型语言考试与语言能力标准对接研究——以英语听说考试为例  

Exploring the alignment between integrated language tests and language standards:Using an integrated listening and speaking test as a case study

作　　者：陈大建 胡杰辉[1] CHEN Dajian;HU Jiehui

机构地区：[1]电子科技大学外国语学院,四川成都611731

出　　处：《外语界》2023年第3期87-96,共10页Foreign Language World

基　　金：四川省社会科学研究“十四五”规划2021年度项目“英语听说融合智能测评体系构建研究”(编号SC21WY011);教育部产学合作协同育人项目“基于CSE量表的英语听说智能测评研究”(编号202102132011)的资助。

摘　　要：本研究以英语听说综合型考试与《中国英语能力等级量表》(CSE)的对接为例,探索综合型语言考试与语言能力标准的对接方法,并检验对接效度。听力和口语两组共24名专家根据55名考生的330条作答数据,运用整体表现法作出CSE等级判断。结果发现,听力组和口语组专家判断之间信度系数分别为0.66和0.92,专家严厉度、模型拟合度、临界分数标准误差均在合理范围之内,表明对接结果内部效度较高。本研究考试对接结果与托福口语考试、大学英语四级听力考试对接结果在一个等级差内的一致率分别为85.5%和85.0%,说明对接结果外部效度合理。专家问卷结果显示整体对接程序效度高。研究进一步讨论了整体表现法的应用效果和综合型语言考试的对接难点。Through a case study of linking an integrated English listening and speaking test to the China's Standards of English Language Ability,this study explores the standard setting method for aligning integrated language tests and language standards,and examines the alignment validity.Using the Body of Work Method,a total of 24 panelists judge the listening and speaking ability levels of 55 test takers individually based on their testing performance on 330 items.It finds that the panelists'judgement of the listening group and speaking group reach consistency of 0.66 and O.92 respectively,and the panelists'leniency,model fit,and cut score standard error measures are within acceptable ranges,indicating relatively high internal validity.The alignment results show a relatively high adjacent level consistency with the TOEFL speaking test(85.5%)and CET 4 listening test(85.0%),suggesting reasonable external validity.The questionnaire data reveals high procedural validity.The study further discusses the validity of the Body of Work Method and the alignment difficulties of integrated language tests.

关 键 词：听说综合型考试 中国英语能力等级量表 对接研究 整体表现法 

分 类 号：H319[语言文字—英语]