The Effectiveness of AI in Assessing Esl University Students’ Argumentative Writing in Comparison to Rubric-Based Assessment and Toulmin’s Model of Argumentation

Mastering argumentative writing is essential for ESL university students, yet many continue to face challenges in structuring logical arguments, providing evidence, and articulating ideas effectively. This study examines their writing proficiency through rubric-based assessment and Toulmin’s Model of Argumentation, while also comparing AI-assisted grading with human evaluation. A total of 72 student essays were analysed to assess argument structure, reasoning, and language proficiency. Results indicate that while students consistently present claims and grounds, they often struggle with incorporating warrants, backing, and rebuttals. Additional issues include weak organisation, repetitive reasoning, and frequent grammatical errors that impair clarity. AI-based grading demonstrated efficiency in evaluating grammar and structure but tended to be stricter than human evaluators. The findings highlight the importance of explicit instruction in argumentation, the integration of Toulmin’s Model into ESL pedagogy, and targeted grammar interventions. AI assessment tools, while operationally beneficial, should augment rather than supplant human evaluators to maintain holistic judgment. Strengthening these areas can enhance students’ critical thinking and overall writing performance.