Please use this identifier to cite or link to this item: https://repository.isls.org/handle/1/11165
Title: Using Learning Analytics in Understanding College Students’ Behavior in ChatGPT-Facilitated Programming Learning
Authors: Xu, Jie
Sun, Dan
Li, Yan
Keywords: Learning Sciences
Issue Date: 2024
Publisher: International Society of the Learning Sciences
Citation: Xu, J., Sun, D., & Li, Y. (2024). Using Learning Analytics in Understanding College Students’ Behavior in ChatGPT-Facilitated Programming Learning. In Lindgren, R., Asino, T. I., Kyza, E. A., Looi, C. K., Keifert, D. T., & Suárez, E. (Eds.), Proceedings of the 18th International Conference of the Learning Sciences - ICLS 2024 (pp. 75-82). International Society of the Learning Sciences.
Abstract: As a typical generative artificial intelligence tool, ChatGPT has gained wide attention in its application to programming. However, few researchers have conducted empirical studies to explore how learners use ChatGPT for programming learning. This study analyzes learners' programming learning behaviors to explore the characteristics of different performance learners using ChatGPT-assisted programming learning. The study analyzes the programming process of 15 learners adopting the fine-grained programming behavior method. The results indicate that: (1) Learners regard ChatGPT as a useful programming learning resource and rely on it to guide their learning process. (2) Learners tend to copy debug error messages to ChatGPT and copy its feedback, which may potentially weaken their deep understanding of the code. (3) High-performing group learners use ChatGPT more frequently for programming assistance while low-performing group learners use ChatGPT the least. Based on the findings, this study proposes suggestions for using ChatGPT to assist college students in programming learning.
Description: Long Paper
URI: https://doi.org/10.22318/icls2024.110098
https://repository.isls.org//handle/1/11165
Appears in Collections:ISLS Annual Meeting 2024

Files in This Item:
File SizeFormat 
ICLS2024_75-82.pdf474.53 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.