Construction and Application of a Dynamic Assessment Model for College English Writing Based on CSE Empowered by Digital Intelligence

Abstract

Aiming at the problems existing in current college English writing assessment, such as ambiguous dimensions, delayed feedback, inaccurate competence positioning, and excessive subjectivity, this study constructs a digital-intelligent empowered dynamic assessment model for college English writing based on the China’s Standards of English Language Ability (CSE), integrating the requirements of CSE and the trend of digital-intelligent education reform. The model fuses digital-intelligent technologies including Natural Language Processing (NLP) and big data analysis, with CSE's writing proficiency grading standards as the core, to achieve quantitative assessment dimensions, dynamic assessment processes, personalized feedback, and precise interventions. A one-year teaching experiment was conducted in the experimental class, comparing writing scores, competence improvement rates, and learning satisfaction between the experimental group and the control group to verify the model's effectiveness and practicality. The results show that the model can significantly improve students' core literacy in English writing, optimize teachers' assessment efficiency, standardize assessment standards, and provide theoretical support and practical paths for the reform of college English writing teaching and digital-intelligent assessment practice in institutions of higher education.