대학수학능력시험(영어영역)의 듣기 문항 개선에 대한 연구

Abstract

This study investigated the problems that the current College Scholastic Ability Test (CSAT) in Korea has, focusing on the English listening test, and suggests new means for the English listening test with the evidence for the validity of the new types of questions. In Korea, the CSAT has an enormous influence on the goal and contents of courses, as well as the methods of teaching and learning. However, though the objectives of English teaching and learning in high schools have changed a lot in the last two decades, the CSAT in Korea has not kept up with the advancements. As a result, students are trained not for communication ability, but for the several fixed types of questions, which have a low degree of difficulty. This is mainly due to the fact that questions and the items in multiple choice questions are presented in written Korean. In this study, two new types of questions are proposed for the English listening test in the CSAT in order to raise the validity of the test and ultimately, to present a better model of English education. In the experiment, after the main passages were read, the questions were also given in spoken English, rather than in written Korean. Another change was that a passage was followed by two questions instead of only one question. The experiment was conducted with 266 senior high school student participants, who were divided into the experimental and the control groups. The two groups were not different from each other in the previous trial test. For the experimental group, two questions were given in spoken English after one passage and for the control group, one question was given in written English after one passage. The comparison of the results shows that the two groups were significantly different from each other in terms of the average score, the items discrimination index and items difficulty level. The new test done by the experimental group was more difficult than that of the control group. In addition, the items discrimination index was raged over .4, which indicates that the questions are good enough to evaluate the ability of students. On the basis of the results, the findings suggest that it is appropriate to give questions in spoken English. In addition, considering that one of the current goals of English education is to raise the ability to grasp the essential information from long dialogues or passages, multiple questions for one passage should be better than a single question for one passage. This is also evidenced by the fact that internationally accepted English tests, such as TOEFL and IELTS, adopt this model. I hope this study will make a contribution to the development of English testing and education in Korean high schools.