Sittie Hainieh Taratingan
Mindanao State University, Marawi City

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Effects of Chat-GPT (Generative Pre-Trained Transformer) in the Writing Skills of Undergraduate Students: A Scoping Review Hamdoni Pangandaman; Mirella Veras; Samiel Macalaba; Abolbashar Mangontawar; Sittie Ainah Mai-Alauya; Saliha Janine Gubaten; Maslainie Aliola; Norhanie Lininding; Paida Abdulmalik; Sittie Hainieh Taratingan
Journal of Artificial Intelligence in Education & Learning Innovation Vol. 2 No. 1 (2026): Journal of Artificial Intelligence in Education & Learning Innovation
Publisher : CV Rezki Media

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56003/jaieli.v2i1.712

Abstract

Background: The integration of artificial intelligence (AI) in academic settings has gained increasing attention, particularly in improving students' writing skills. ChatGPT, a generative AI tool, has been explored as a potential aid in academic writing, providing instant feedback and enhancing students' ability to structure, revise, and refine their written work. However, there remains a need to systematically assess its effectiveness in undergraduate education Objectives: This study aimed to evaluate the impact of ChatGPT on the writing skills of undergraduate students through a scoping review, analyzing its effectiveness in enhancing grammar, vocabulary, coherence, and overall writing proficiency. Methods: A scoping review methodology was employed to identify relevant studies published between 2020 and 2024. A comprehensive search was conducted across seven academic databases: ScienceDirect, Scopus, ProQuest, EBSCOhost, Sage Journals, Taylor & Francis, and PubMed, using Boolean operators and predefined inclusion criteria. The studies were appraised using the Joanna Briggs Institute (JBI) Critical Appraisal Tool for quasi-experimental research and the ROBVIS checklist for randomized controlled trials (RCTs) to assess the risk of bias. The synthesis was conducted following the Synthesis Without Meta-Analysis (SWiM) guidelines, with PRISMA used for article selection and appraisal. Results: A total of nine studies were included in the review, conducted in Saudi Arabia, China, Ecuador, Indonesia, Iran, Pakistan, and India, involving a total of 839 undergraduate students. The duration of interventions varied, with most lasting between 10 to 16 weeks. The overall risk of bias was low, indicating strong methodological reliability. The findings revealed that ChatGPT significantly improved students' writing proficiency, particularly in areas of grammatical accuracy, vocabulary use, organization, and argument structure. Also, students using ChatGPT exhibited increased engagement and motivation in writing tasks compared to traditional methods. However, some studies cautioned against over-reliance on AI-generated text and raised concerns about academic integrity and originality. Conclusion: ChatGPT demonstrates promising potential as an AI-assisted writing tool, offering immediate feedback, structured learning support, and improved writing outcomes for undergraduate students. While its effectiveness is evident, further research is needed to establish best practices for integrating AI into academic writing instruction, ensuring that students develop critical thinking and independent writing skills while leveraging AI’s benefits responsibly.