Exploring the Effectiveness of the IDEA Instructional Model in Integrating Generative AI into University Statistics Education
List of Authors
Qian-Qian Lin, Yuan-Horng Lin
Keyword
Perceptions of Statistics; Generative AI; Statistics Education; Statistics Literacy
Abstract
The purpose of this study is to apply generative AI in university statistics instruction. Based on the purpose of this study, this study proposes the IDEA instructional model and employs thematic analysis to examine students’ perceptions of their learning outcomes. Big data and statistical literacy are important research topics in statistics education. The focus of statistics education lies in the formation of meaningful concepts, while complex computations can be supported by technological tools. The rapid development of generative AI is influencing higher education; however, research on the integration of generative AI into the teaching and learning of statistics remains limited. In this study, the content of statistics instruction of higher education covers five topics, which are estimation of confidence intervals for one population mean, hypothesis testing for one population mean, the relationship between confidence interval estimation, one-way analysis of variance, and simple linear regression. This study proposes the IDEA (Instruction, Drilling, Experimentation, Abstraction) instructional model in statistics instruction. The main instructional focus is on learning statistical theories and practicing problem-solving, followed by using generative AI for data simulation and validation. The research data is analyzed using thematic analysis .According to the results of thematic analysis, these perceptions include: 1. Generative AI enhances learning efficiency; 2. Generative AI promotes the deepening of statistical concepts and supports visual understanding; 3. Generative AI serves as an interactive and assistive learning tool; 4. Usage of generative AI involves risks and experience limitations; 5. The use of generative AI leads to frustration and cognitive gaps; 6. Generative AI fosters the growth of statistical thinking and the shift in learning attitudes. Based on these perceptions, students generally hold positive perceptions toward the integration of generative AI into university statistics learning. Based on the findings, this paper offers instructional implications and suggestions for future research in university statistics education.