AI-Driven ChatGPT vs. traditional methods: A case study on Year 5 Pupils’ writing skills in Penang
Abstract
The integration of AI-driven ChatGPT platforms in English as a Second Language (ESL) education has shown promise in enhancing writing skills. This study explored the effectiveness of these digital tools in improving ESL learners’ writing competencies. The objective was to investigate the impact of AI-driven ChatGPT platforms on ESL learners’ writing skills, specifically focusing on grammar accuracy in terms of sentence structures and words/phrases usage. A qualitative method was employed, involving 25 Year 5 pupils who participated in writing classes using either ChatGPT-assisted tools or traditional learning methods. The data was analyzed using Language Competence Theory. The findings revealed that AI-driven ChatGPT was predominantly helpful in producing essays with proper sentence structures, primarily using compound-complex sentences (independent clause+coordinating conjunction+independent clause+dependent clause). In contrast, traditional writing methods resulted in predominantly simple sentences (subject+verb), demonstrating limited English language proficiency among students. In terms of words/phrases, ChatGPT utilised advanced words/phrases, whereas traditional method writing employed basic words/phrases. The year 5 pupils using ChatGPT often had to refer to the dictionary to understand the meanings of words/phrases, compared to the traditional method where they used familiar words/phrases from their daily words/phrases. This contrast highlighted the difference in language proficiency levels between the two methods. The use of AI-driven tools like ChatGPT in ESL writing instruction showed potential for enhancing students’ English language proficiency. However, it also raised questions about the balance between AI assistance and authentic language usage. Educators should consider how to leverage AI tools to scaffold learning while ensuring students develop independent writing skills. Further research could explore the long-term effects of AI use on language acquisition and strategies for integrating AI tools with traditional teaching methods to optimize learning outcomes.
Fields of Study & Subjects
Artificial Intelligence in Education
Writing and Handwriting Education
EFL/ESL Teaching and Learning
Publication details
- Journal / source
- AIP conference proceedings
- Volume
- 3393
- Pages
- 020003–020003
- Publication date
- 1 January 2026
- Language
- EN