DOI: 10.1177/09287329261495504 ISSN: 0928-7329

Anthropometric classification of foot morphology in Korean adults using K-means cluster analysis

Jun-hee Kim

Background

Foot morphology reflects genetic, environmental, and lifestyle influences and affects biomechanics, footwear design, and healthcare outcomes.

Objective

This study aimed to derive practical anthropometric profiles of adult foot morphology using anthropometric measurements and K-means clustering as a basis for ergonomic and clinical research.

Methods

Foot measurement data from 4045 adults aged 20–59 years were obtained from the 8th Size Korea Anthropometric Survey (2020–2023) using standardized three-dimensional scanning. Twelve variables, including nine height-normalized linear dimensions and three angular measures, were analyzed using K-means clustering separately by sex. The number of clusters was selected based on complementary validity criteria, subsampling-based stability, morphological granularity, and interpretability. Cluster profiles were compared using one-way analysis of variance (ANOVA) and post-hoc tests.

Results

Four clusters were identified for each sex. Although the validation indices did not indicate a single optimal solution, k = 4 was selected because it captured an elevated navicular profile absent at k = 3, whereas k = 5 showed lower stability without revealing an additional morphological pattern. The profiles differed primarily in foot size, dorsal dimensions, and navicular height, while forefoot width and hallux alignment further distinguished the female profiles. Overall ANOVA results were significant for all 12 variables (p < 0.001); however, some pairwise post-hoc comparisons were not significant.

Conclusion

Four sex-specific profiles provided a practical representation of continuous foot morphological variation among Korean adults. These profiles may support shoe-last and orthotic design, but their functional and clinical relevance requires external validation using footwear fit, plantar pressure, and comfort outcomes.