计算机化语言测试的效度研究——浅析计算机能力与测试构念的关系

Validating Computer-Based Language Tests:Computer Literacy and the Construct of a Computer-Based Language Test

随着信息技术运用领域的不断拓展,以计算机为媒介的语言交际已成为现实,计算机化语言教学和测试也迅速发展。在测试领域,技术的使用已不仅仅是为了提高测试效率,更重要的是提高测试效度。本文根据Messick(1989)的整体效度理论,运用Chalhoub-Deville(2003)的局部构念定义,分析计算机化语言测试中计算机能力与测试构念的关系及其对效度的影响。本文指出,当计算机成为语言交际的重要媒介时,计算机能力成为与构念密切相关的语境因素,以计算机为媒介的交际语境与语言能力交互影响,共同构成计算机化语言测试的构念。

With the increasingly extensive application of information technology, computer-mediated language communication has become a fact of life. Computer-assisted language teaching and testing have also entered a stage of fast development. In the field of testing, technology is not only employed for improving test efficiency, but more importantly, for enhancing test validity. This article addresses the validity issue .of the relationship between computer literacy and the construct of a computer-based language test. Based on a review of Messiek＇s（1989） unitary construct definition and Chalhoub-Deville＇s （2003） local context-bound view of language ability, the article argues that, when computer has become an important contextual facet in language communication, computer literacy should be considered as a contextual factor closely related to the construct being measured in a computer-based language test. In other words, individual language ability and the computer-mediated communicative context interact in ways that change them both.