Correlation Analysis of Achievement Patterns among Subjects in the Korean National Licensing Examination for Korean Medicine Doctors

Objectives: This study aimed to analyze the structural relationships among subjects in the Korean national licensing examination for Korean medicine doctors using correlations of percent correct scores.Methods: This study analyzed national examination data from 2019 to 2025 (N = 5,412). Percent correct scores for each subject were calculated and standardized within each year using z-scores to adjust for variations in examination difficulty. Pearson correlation coefficients were used to construct an inter-subject correlation matrix and the sum of intersubject correlations to evaluate the centrality of each subject. Hierarchical clustering and non-metric multidimensional scaling were applied to visualize the relationships among subjects.Results: All subject pairs showed positive correlations. Internal medicine 1 demonstrated the highest percent correct score and total correlation with the lowest variability while herbology showed the lowest achievement and the highest variability. Hierarchical clustering and non-metric multidimensional scaling identified a cluster of clinical subjects including internal medicine, gynecology, and pediatrics. In contrast, basic subjects such as herbology and physiology showed relatively independent patterns from other subjects.Conclusions: The performance patterns suggest a potential dual structure consisting of an integrated clinical subject cluster and relatively independent basic subjects. These findings provide a basis for effective learning strategies and underscore the need for an integrated curriculum to bridge the gap between basic and clinical subjects. In particular, clinical subjects synergistically reinforce each other, while basic subjects such as physiology and herbology may require more distinct, targeted approaches and longitudinal integration with clinical subjects.