영어논술평가에서의 일반 척도, 세부과업별 척도, 이분식 척도의 비교 연구

초록

The purpose of this study is to examine how forty student raters and two expert ones utilize each of three rating scales-general, task-specific, and binary-choice scale-in an L2 argumentative essay test. The study compared indices of inter-rater reliability across three rating scales, conducted survey analysis, and made interviews with students and experts in order to understand their perceptions on the rating scales. The results suggested that a task-specific scale yielded highest inter-rater reliability specifically in the content area. A large majority of students and all raters pointed out the task-specific scale as one which is able to measure writing competence most accurately. This paper ended up with the comment that we need to deeply explore the features and functions that a particular rating scale has in nature in order to develop the rating scale geared for the testing context for a valid and reliable writing test.

키워드