模糊评分:外语口语测试评分新思路

Fuzzy Marking: A New Approach to Marking Oral Proficiency

外语口试主要采用两种评分方法:分项评分和整体评分。这两种评分方法在原理和操作上都存在明显问题。在分析这些问题的基础上,本文提出了“模糊评分”的概念,指出将模糊数学的基本原理应用于口语测试的评分实践具有科学性和可行性。模糊数学的方法适合于描述模糊现象,作者由此推论,基于模糊数学原理的模糊评分在基本理念和操作上比现行评分方法具有明显优势,可以较好解决现行评分模式下使用精确数学的方法所遇到的一些问题。

Two marking schemes, namely, analytic marking and holistic marking are currently employed to assess oral proficiency in the field of language testing. Though apparently different, these approaches represent an orientation for precision in scorers' judgment by assigning precise scores to testees' oral performance, either in the name of the various components of oral communicative competence, or of oral communicative competence as a whole. The assumption which underlies precise scoring is that one's oral ability can be indicated by a single point on a measure of oral ability. However, there are obvious problems with precise scoring, which supposedly reflects so-called precise judgment. This paper attempts to outline a framework which we prefer to call Fuzzy Marking, and which provides an alternative to the current approaches to the marking of oral proficiency. Arguing that oral ability should be viewed as a broad surface rather than a single point, we first analyze some basic problems with current approaches, then propose the Fuzzy Marking Model by introducing the core concept of Membership in Fuzzy Mathematics. By comparing and contrasting the Fuzzy Marking Model and the current approaches, we draw the conclusion that marking practice within our framework will be able to solve some practical problems scorers encounter in their current practice and provide more objective, thus more precise judgments of testees' oral proficiency.