评分量表对口语考试构念效度的影响:交互构念理论视角

Effects of rating scales on the test construct validity in the speaking assessment:From the perspective of the interactionalist construct theory

本研究依据交互构念理论中的"极简型交互观"和"温和型交互观",参考中国英语能力等级量表的口语量表,设计了"按能力分项评分量表"与"按任务评分量表",并比较了两种评分量表对英语口语考试分数、分数差异来源以及分数解释的影响。研究立足语言构念与交际语境的交互性,从口语评分角度拓展了定义行为表现评估考试构念的视野。

Based on the "minimal interactionalist claim" and "moderate interactionalist claim" of the interactionalist construct theory, the present study develops test-based analytic scales and task-based analytic scales for English speaking performance rating, using the speaking scales of China’s Standards of English(CSE) as a source of reference. It examines the effects of the rating scales on test scores, sources of score variation and score interpretation. Focusing on the interaction between the language construct and context of communication, the study expands the vision of defining test construct for performance assessment from the perspective of speaking performance rating.