The Effect of Chatbot Use on Students’ Expectations and Achievement in STEM Flipped Learning Activities: A Pilot Study

Abstract

The core of STEM education lies in enhancing teaching strategies to stimulate students’ self-learning ability and cultivate critical thinking and problem-solving skills. However, there is little research exploring the impact of students’ pre-class learning on their classroom performance, including learning intentions and emotional expression, which also affects the effectiveness of STEM learning activities in flipped classrooms. Therefore, this study applied natural language processing technology (NLP) to establish a text-based chatbot for guiding and accompanying students in pre-class learning in a flipped classroom. To understand whether students’ pre-class learning with the chatbot has a positive impact on their learning effectiveness and expectations, we recruited 30 high school students for a quasi-experimental design and randomly assigned them to two groups: an experimental group with the chatbot during pre-class learning and a control group with traditional flipped classroom learning methods. The results showed that the experimental group performed better in learning achievement and expectations. These findings can provide insights for educators committed to implementing flipped learning, including the design of teaching activities and the needs of students beyond the learning materials in the learning process.