基于CSE的智能评估在中职高考英语写作教学中的实证研究

An Empirical Study on Intelligent Assessment Based on CSE in the English Writing Teaching of Vocational High School Students for College Entrance Examination

摘要: 本研究旨在解决传统中职英语写作教学中教师批改负担重、反馈不及时、缺乏个性化指导的问题。为此，研究者构建并实践了一种融合《中国英语能力等级量表》(CSE)与生成式人工智能的“四阶段两动态介入”智能动态评估模式。将该模式应用于一个为期一学期的中本贯通二年级班级的高考英语写作教学实验中，采用单一组前后测设计，通过写作任务、问卷调查等方式收集数据，运用配对样本t检验等方法进行分析。研究结果表明，该教学模式显著提升了学生的整体写作成绩以及在内容、语言、结构等维度的单项得分，尤其对学业后进生的促进作用更为明显。研究表明，该模式为实现写作教学的智能评价与“以评促学”提供了有效路径。

Abstract: This study aims to address the problems of heavy teacher workload in correcting, delayed feedback, and lack of personalized guidance in traditional vocational high school English writing teaching. To this end, the researchers have constructed and implemented a “four-stage two-dynamic-intervention” intelligent dynamic assessment model that integrates the “China’s Standards of English Language Ability” (CSE) and generative artificial intelligence. This model was applied to a one-semester teaching experiment of College Entrance Examination English writing for a second-year class in a secondary vocational school. A single-group pretest-posttest design was adopted, and data were collected through writing tasks and questionnaires. Paired sample t-tests and other methods were used for analysis. The research results show that this teaching model significantly improved the overall writing scores of students and their scores in content, language, and structure. The improvement was particularly significant for students with lower academic performance. Research shows that this model provides an effective path for achieving intelligent evaluation in writing teaching and promoting learning through evaluation.