Abstract
BACKGROUND: Learning engagement refers to the time and effort students invest in effective learning practices, and it serves as a crucial indicator of their learning abilities as well as a significant proximal factor in predicting their academic performance. However, undergraduate students in Chinese universities have demonstrated only moderately high levels of learning engagement in the context of blended learning, which may be enhanced through the introduction of knowledge graph. This study aimed to examine whether students’ learning engagement could be enhanced through knowledge graph-based blended learning. The findings are intended to provide a basis for implementing knowledge graph-based blended learning in future medical education. METHODS: A quasi-experimental design was employed, involving a control group (n = 36) of medical students from the 2020 cohort who received traditional blended learning and an experimental group (n = 50) from the 2021 cohort. Knowledge graph-based blended learning was implemented in a core clinical medicine course, Diagnostics. The course was developed and implemented using the ADDIE (Analysis, Design, Development, Implementation, and Evaluation) instructional design model. Post-intervention, course grades and levels of learning engagement were compared between the two groups, and the learning satisfaction of the experimental group was analyzed. RESULTS: The results indicated that the experimental group achieved significantly higher scores in the learning absorption dimension (P = 0.035) and overall course grades. There was no statistically significant difference between the two groups in the total scores of learning engagement, dedication, and vigor (P > 0.05). The average online learning time for students in the experimental group was 43.81 h, which demonstrated a positive correlation with their level of learning engagement (r = 0.843, P < 0.001). The overall learning satisfaction was high (94.69 ± 12.15) during the knowledge graph-based blended learning process. CONCLUSION: The application of knowledge graph-based blended learning in the Diagnostics course facilitates the improvement of medical students’ learning absorption, academic performance, and teaching effectiveness. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12909-025-08428-5.