Developing Critical Thinking in Second Language Learners: Exploring Generative AI like ChatGPT as a Tool for Argumentative Essay Writing
Project Overview
The document investigates the application of generative AI, particularly ChatGPT, in improving critical thinking and argumentative writing skills among second language learners through the development of the ChatGPT Guideline for Critical Argumentative Writing (CGCAW). This structured framework combines critical thinking models with the capabilities of ChatGPT to guide students in their writing endeavors. An experiment was conducted involving two groups: one utilizing the CGCAW framework and the other employing ChatGPT without any specific guidelines. Results indicated that while the group using the CGCAW framework showed enhancements in certain aspects of writing, the control group outperformed them in overall language mechanics. These findings highlight the effectiveness of integrating AI tools in education but also underscore the need for refinement within the CGCAW framework and suggest that further research is necessary to optimize AI-enhanced educational practices. Overall, the study emphasizes the potential of generative AI in fostering critical skills in language learners while pointing to existing challenges that warrant additional exploration.
Key Applications
ChatGPT Guideline for Critical Argumentative Writing (CGCAW)
Context: Second language learning environment for university students
Implementation: An experimental study with two groups of students, where one group used the CGCAW framework to guide their use of ChatGPT while writing argumentative essays.
Outcomes: The experimental group showed improvements in clarity, logical coherence, and use of evidence, indicating the potential of ChatGPT to enhance specific aspects of writing.
Challenges: The control group performed better in overall language mechanics and articulation of main arguments, suggesting the CGCAW framework may need refinement.
Implementation Barriers
Technical Barrier
Participants' unfamiliarity with using ChatGPT may hinder effective utilization of the tool.
Proposed Solutions: Providing training sessions for students on how to effectively use ChatGPT could mitigate this issue.
Time Constraint
The 40-minute time limit for writing the essays was inadequate for some participants.
Proposed Solutions: Allowing longer time frames for essay completion or conducting multiple sessions could alleviate this barrier.
Individual Differences
Differences in participants' prior knowledge and individual capacities may have affected the results.
Proposed Solutions: Matching participants based on their prior knowledge of writing and generative AI tools could enhance the experiment's validity.
Project Team
Simon Suh
Researcher
Jihyuk Bang
Researcher
Ji Woo Han
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Simon Suh, Jihyuk Bang, Ji Woo Han
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai