<![CDATA[This study presents a quantitative structural evaluation of undergraduate data science curricula across Indonesian universities using a compositional modeling approach. Existing competency frameworks identify essential domains in data science education, yet limited empirical research evaluates how these competencies are proportionally structured across institutions. Using course-level curriculum data collected from 37 undergraduate programs, courses were mapped into five competency domains: programming, statistical foundations, machine learning, applied practice, and ethics. Each curriculum was operationalized as a normalized compositional competency vector and analyzed using the Curriculum Structural Balance Index (CSBI), a distance-based metric that measures proportional structural imbalance across competency domains. Hierarchical clustering was further applied to identify recurring curriculum typologies. The findings reveal a consistent structural pattern characterized by strong concentration in programming and statistics, moderate variation in machine learning, and limited integration of applied practice and ethics. This recurring configuration, identified as a Foundational-Dominant model, indicates that current programs prioritize technical consolidation over interdisciplinary integration. Unlike conventional curriculum evaluation approaches that primarily assess competency presence or coverage, the proposed CSBI framework quantitatively evaluates proportional competency balance and enables systematic comparison of curriculum structures across institutions. The study contributes a measurable and replicable analytic approach for evaluating interdisciplinary curriculum architecture in higher education.]]>
Copyrights © 2026
