Asian Journal of Language, Literature and Culture Studies

This study examined whether a short, teacher-guided ChatGPT-supported writing module was associated with pre–post improvements in Grade 12 students’ professional email writing performance in an English for Academic and Professional Purposes (EAPP) context. Using a one-group pretest–posttest quasi-experimental design, 75 Senior High School students completed a professional email writing task before and after the intervention. The intervention was delivered across the class sessions and trained students to (a) generate and refine appropriate email content using structured prompts, (b) evaluate ChatGPT outputs critically against audience and purpose, and (c) revise drafts with teacher guidance and rubric-aligned criteria. To assess independent performance/transfer, the posttest writing task was completed without AI assistance. Student emails were rated using an analytic rubric covering email structure, clarity, tone, and accuracy. Paired-samples results indicated a statistically significant increase in overall writing scores from pretest (M = 11.97, SD = 2.07) to posttest (M = 14.93, SD = 1.93), t(74) = 11.28, p < .001, reflecting a large pre–post improvement (Cohen’s dz = 1.30, paired). Component-level descriptives showed gains across all rubric dimensions, with the largest improvements observed in email structure and tone. Given the absence of a control/comparison group, the findings should be interpreted as preliminary evidence of association rather than definitive causal effects. Implications for teacher-mediated integration of generative AI tools in writing instruction and directions for stronger comparative designs are discussed.