Does ChatGPT Write Like a Student? Engagement Markers in Argumentative Essays.

This article examines the differences in reader engagement between argumentative essays generated by ChatGPT, a large language model (LLM), and those written by British university students. By analyzing 145 essays from each group, the study finds that student essays contain significantly more and a greater variety of engagement markers—such as questions and personal asides—that foster interaction and persuasion, whereas ChatGPT essays rely more on appeals to shared knowledge and exhibit fewer interactive features. These differences are attributed to ChatGPT's statistical, data-driven generation process, which lacks genuine audience awareness and the ability to adapt rhetorically to readers. The findings highlight important pedagogical implications, suggesting that while ChatGPT can assist in producing coherent academic texts, explicit instruction in rhetorical engagement remains essential for developing students' critical thinking and persuasive writing skills.