摘要
阅读能力能在很大程度上体现出外语学习者的水平。因此,对阅读能力的测试是英语教学中的一项艰巨的任务。测试阅读能力的题型很多,但哪种题型效度较高历来争论也较多。本文通过实证研究的方法对多项选择题、填空题和问答题等三种常见的阅读题型的效标关联效度进行了全面的检验。结果表明,在内容效度和建构效度相同的情况下,填空题和问答题的效标关联效度均令人满意,其中,填空题的效度更高,而多项选择题的效标关联效度不尽如人意。
Reading ability can show the learners＇ English level to a large extent. Therefore, the testing of reading ability is a demanding task. Among the various reading item types, which one is more valid has evoked a lot of controversies. This paper has investigated such criterion-related validity as concurrent validity and predictive validity of multiple-choice questions, close tests and short-answer questions. It is shown in this research that in spite of the same content validity and construct validity, the criterion-related validity of the close tests and short-answer questions is satisfying, with that of the close tests the highest, while that of MC item types is the lowest, which does not show any significant correlation.
出处
《西安外国语大学学报》
2012年第4期80-83,共4页
Journal of Xi’an International Studies University
关键词
阅读题型
同期效度
预测效度
效标关联效度
reading item types
concurrent validity
predicative validity
criterion-related validity
