Impacts of a ChatGPT-supported concept mapping approach on students’ database programming achievement and their problem-solving and critical thinking awareness
This study proposes a ChatGPT-supported concept mapping-based database programming (GPT-CMDP) approach to address challenges in teaching abstract programming concepts. Using ChatGPT as a virtual teaching assistant integrated with concept mapping, the model assists university students in system analysis, database design, and query syntax debugging. A quasi-experimental evaluation was conducted with an experimental group using GPT-CMDP and a control group using conventional concept mapping. The empirical results demonstrate that students using the GPT-CMDP approach achieved significantly higher academic performance, better critical thinking awareness, and enhanced problem-solving skills compared to the conventional method.
Figure 1. Programming instruction model.
Technology Overview
The technology features a structured instructional model integrating generative AI with graphic knowledge representations. It implements a six-step framework: organizing structured units, micro-step syntax breakdown, active exploration, real-time feedback, reflective practice, and self-paced learning. Effectively, ChatGPT serves as an interactive conversational agent providing instant SQL syntax generation and debugging feedback.
Applications & Benefits
This framework applies to computer programming education, specifically advanced SQL database instruction in higher education. Strategically, it compensates for the shortage of teaching assistant resources by providing personalized, real-time guidance. Ultimately, the method fosters student autonomy, reduces dependency on teachers, and systematically improves programming learning efficiency and cognitive outcomes.
Abstract：
In the era of artificial intelligence (AI), there is growing interest in applying generative AI to education, particularly in programming instruction. Traditional instruction often challenges students to grasp abstract programming concepts. This study explored using ChatGPT as a virtual teaching assistant to enhance learning support. Teachers integrated ChatGPT with concept mapping to assist students in system analysis, design, and question-solving. We proposed a ChatGPT-supported concept mapping-based database programming (GPT-CMDP) approach and evaluated its effectiveness. Participants were university students in northern Taiwan, with 30 in the GPT-CMDP group and 26 in a control group using the conventional concept mapping-based approach (C-CMDP). Results showed that GPT-CMDP students had significantly higher academic performance and better critical thinking and problem-solving skills than the C-CMDP group. These findings highlight the potential of generative AI in enhancing programming education and offer valuable directions for future research.
Impacts of a ChatGPT-supported concept mapping approach on students’ database programming achievement and their problem-solving and critical thinking awareness
Author：Lee Yen-Fen, Hwang Gwo-Jen, Cheng Li-Chen
Year：2026
Source publication： Interactive Learning Environments, Volume 34, 2026 - Issue 3
Subfield Highest percentage： 99% Education #16 / 1698