ChatGPT-assisted language learning: Effects on Vietnamese English majors' writing skills and motivation

Downloads

Published

2025-06-23

Section: Regular Articles


Received: 17 December, 2024
Accepted: 11 June, 2025

Authors

  • Nguyen Huu Hoang Email ORCiD Academy of Journalism and Communication, Vietnam
DOI: https://doi.org/10.29140/jaltcall.v21n2.102429

Abstract

This study investigated the effects of ChatGPT-assisted instruction on Vietnamese English majors' writing skills and motivation. Using a mixed-methods sequential explanatory design, the study compared two groups of first-year English majors (N = 76) over a 15-week semester. The experimental group received ChatGPT-assisted writing instruction, while the control group followed traditional instruction. Data were collected through pre- and post-tests using IELTS writing assessments, motivation questionnaires, and semi-structured interviews. ANCOVA results revealed significant improvements in the experimental group's writing performance across all domains, with particularly strong effects on task achievement (ηp² = .30) and coherence (ηp² = .27). Notable enhancements were also observed in writing motivation dimensions, especially self-efficacy (ηp² = .31) and engagement (ηp² = .29). Qualitative findings indicated that ChatGPT's immediate feedback and scaffolded support facilitated an iterative learning process, transforming students' approach from task completion to active engagement. The study demonstrates that ChatGPT-assisted instruction can effectively enhance both writing skills and motivation when implemented within a structured pedagogical framework. However, successful integration requires a sufficient adjustment period and careful attention to developing balanced approaches that promote learner autonomy.


Keywords: ChatGPT, EFL writing skills, writing motivation, English majors, Vietnam

Suggested Citation:

Huu Hoang, N. (2025). ChatGPT-assisted language learning: Effects on Vietnamese English majors’ writing skills and motivation. The JALT CALL Journal, 21(2), 102429. https://doi.org/10.29140/jaltcall.v21n2.102429

Downloads

Download data is not yet available.