Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
ChatGPT-Empowered Writing Strategies in EFL Students’ Academic Writing: Calibre, Challenges and Chances
11
Zitationen
2
Autoren
2024
Jahr
Abstract
ChatGPT’s remarkable ability to produce academic texts has generated significant interest in educational and academic circles. This study provides a specific overview of ChatGPT’s current usage and explores its potential applications, limitations, and implications in English academic writing for EFL students, who often face challenges in language proficiency, content organization, and critical thinking. Using a mixed-methods research approach, this study employed a CSE-based questionnaire and focus group interviews to investigate how ChatGPT can empower academic writing strategies (WS) and how respondents perceive its assistance. Data was collected from 60 Chinese university juniors majoring in English. Quantitative data were analyzed using descriptive statistics and regression analysis, while qualitative data underwent thematic analysis. Findings indicate that ChatGPT can help students apply academic WS more effectively by comprehending research trends, generating writing outlines, enriching writing content, synthesizing literature, and refining papers. However, issues such as potential plagiarism, inaccurate output, improper citations, and the digital gap between users and non-users must be addressed. The study suggests that while ChatGPT-empowered writing can better equip academic WS in planning, composing, and revising, respectively, it is crucial to scrutinize the quality of AI-generated texts. Further research will be urgently expected regarding ChatGPT’s long-term impact on academic integrity, the development of educational policies for ethical AI use, and the integration of ChatGPT into pedagogical approaches to enhance EFL students’ writing and critical thinking.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.485 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.371 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 7.827 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.781 Zit.
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI)
2018 · 5.549 Zit.