Research on English Writing Teaching Reform Based on Multi-subject Collaborative Evaluation: A Case Study of the Intelligent English Writing Course

Driven by the new-era educational evaluation reform emphasizing process-based and developmental assessment, traditional college English writing teaching is constrained by teacher-dominated single evaluation, delayed feedback, and low student participation, which hinder the sustainable improvement of students’ writing competence. While artificial intelligence (AI) has facilitated innovative writing evaluation approaches, most existing studies merely adopt single AI or conventional teacher-peer evaluation, with few empirical explorations on progressive multi-subject evaluation models suitable for application-oriented universities. To fill this gap, this study takes the Intelligent English Writing course as the research object and constructs a four-stage progressive collaborative evaluation model integrating student self-evaluation, peer evaluation, critical AI optimization, and whole-process teacher diagnosis. A one-semester teaching intervention was conducted on four parallel classes. Pre-test and post-test writing scores as well as valid questionnaire data (N=59, Cronbach’s α=0.89) were used to examine its effectiveness. The results demonstrate that the proposed model significantly improves students’ writing performance, learning engagement, metacognitive reflection, and critical feedback screening abilities, while effectively reducing their writing anxiety and optimizing the traditional classroom ecology. This study clarifies the auxiliary role and critical application principles of AI tools, balances technological empowerment and humanistic teaching, and establishes a replicable practical paradigm for intelligent writing instruction. It provides credible evidence and actionable references for college English writing teaching reform and optimized educational evaluation.