大学英语四、六级计算机口语测试效度、信度和可操作性研究

Study of the validity, reliability and practicality of the computer-based oral proficiency test in CET

不久前，四、六级考试委员会正式出台了《全国大学英语四、六级考试改革方案（试行）》，明确提出“考委会将积极研究开发计算机口语测试，以进一步扩大口语考试规模”。本文通过计算机口语考试和目前的面试型口语考试的对比研究，发现考生在这两个考试中的成绩相关达到0．7058，97．80％的考生得到的成绩等级完全一致或只差半级。更重要的是，通过分析和对比研究，发现计算机考试无论在效度和信度方面都将超过目前四、六级面试型口试；在考试成本上也有较大的降低。文章对如何实行大规模口语考试提出了建议。

The reform of the present face-to-face CET Spoken English Test has now been put on the agenda of the CET committee, because, administered on a rather smaller scale due to the limited number of examiners and insufficient fund, it fails to meet the demand of the increasing number of candidates. The way out is to develop the computer-based oral proficiency test. The comparative study of the two tests is carried out in terms of validity, reliability and practicality. It shows that COPT is a successful surrogate for the CET-SET. Statistical evidence is quite encouraging： the correlation of the two tests are 0. 7058 and their rates in 97. 80% of the occasions either completely agree with each other or deviate by only haft a grade. Most important, the analysis proves the higher validity and reliability of COPT and its lower cost.