Scaffolding Computational Thinking With ChatGPT

ChatGPT has received considerable attention in education, particularly in programming education because of its capabilities in automated code generation and program repairing and scoring. However, few empirical studies have investigated the use of ChatGPT to customize a learning system for scaffoldi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on learning technologies 2024, Vol.17, p.1668-1682
Hauptverfasser: Liao, Jian, Zhong, Linrong, Zhe, Longting, Xu, Handan, Liu, Ming, Xie, Tao
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:ChatGPT has received considerable attention in education, particularly in programming education because of its capabilities in automated code generation and program repairing and scoring. However, few empirical studies have investigated the use of ChatGPT to customize a learning system for scaffolding students' computational thinking. Therefore, this article proposes an intelligent programming scaffolding system using ChatGPT following the theoretical framework of computational thinking and scaffolding. A mixed-method study was conducted to investigate the affordance of the scaffolding system using ChatGPT, and the findings show that most students had positive attitudes about the proposed system, and it was effective in improving their computational thinking generally but not their problem-solving skills. Therefore, more scaffolding strategies are discussed with the aim of improving student computational thinking, especially regarding problem-solving skills. The findings of this study are expected to guide future designs of generative artificial intelligence tools embedded in intelligent learning systems to foster students' computational thinking and programming learning.
ISSN:1939-1382
2372-0050
DOI:10.1109/TLT.2024.3392896