摘要
针对目前结构效度定量研究方法的不足,本文尝试从受试答题过程的角度对TEM-4完型填空测试进行了结构效度研究。实验结果表明,TEM-4完型填空测试很好地考查了受试运用冗余信息的能力,但是在考查受试辨析词汇能力方面尚有改进的余地。在分析受试答题过程的基础上,结合Bachman和Cohen对结构效度的定义,我们尝试性地提出了一个计算结构效度系数的公式。
This paper reports on a process-oriented study of the construct validity of the cloze test in TEM-4 (2003). Employing an immediate introspective verbal report procedure, the study was conducted among 20 second-year English majors from the PLA University of Foreign Languages. Results show that the construct of using redundancy information in the cloze test was well measured and this was not the case with that of distinguishing between options, particularly between vocabulary options, suggesting that there is still room for improvement in the design and assessment of TEM-4 cloze test. And on the basis of the analysis of the testees' actual test-taking processes and in line with Bachman and Cohen's definition of construct validity, a tentative formula is put forward for the calculation of construct validity index (CVI).
出处
《现代外语》
CSSCI
北大核心
2006年第1期71-77,共7页
Modern Foreign Languages