摘要
本研究从概推性和一致性角度考察雅思考试写作卷与中国英语能力等级量表对接的效度。根据3名量表专家对考试任务与617条量表描述语的关联程度的判断,考察对接结果的概推性;根据16名对标专家对100篇样文所对应的量表等级的判断,从专家判断的一致性、专家判断与样文实际得分的一致性和临界分数的标准误差等角度考察对接结果的一致性。结果发现,目标考试的两项写作任务与量表描述语的内容关联符合对接研究的要求,对接结果一致性高,但结果不宜概推到不同文本类型的写作任务。研究表明,概推性和一致性是考察产出型语言考试与语言标准对接效度的重要证据。
This study explores the generalizability and consistency of the results from linking the IELTS writing test to Chinas Standards of English Language Ability(CSE). Three CSE specialists judged the degree of congruence between sample IELTS writing tasks and 617 relevant CSE descriptors to generate evidence of generalizability;16 panelists benchmarked 100 sample IELTS writings in a standard setting session to provide data for assessing consistency from three perspectives: consistency of panelists judgments, consistency between panelists judgments and empirical scores, and the standard error of cut scores. The evidence showed that the target test tasks were congruent with the relevant CSE descriptors in content coverage, the linking results were consistent, but they should not be generalized to writing tasks involving different text types. The study demonstrates that generalizability and consistency are important validity issues in linking tests of productive language skills to language standards.
作者
蔡宏文
CAI Hongwen(National Key Research Centre for Linguistics & Applied Linguistics,Guangdong University of Foreign Studies,Guangzhou 510420,P.R.China)
出处
《现代外语》
CSSCI
北大核心
2019年第5期709-721,共13页
Modern Foreign Languages
基金
广东外语外贸大学特色创新项目“大规模英语考试中的评分员认知及其对评分结果的影响”(15T29)
教育部人文社会科学重点研究基地重大项目“外语能力测评研究“(19JJD750007)的阶段性成果
关键词
产出型技能
概推性
一致性
中国英语能力等级量表
对接使用论证
productive language skills
generalizability
consistency
Chinas Standards of English Language Ability
Linking Use Argument
