<![CDATA[Educational quality in Indonesia faces disparities in infrastructure and instructional innovation, necessitating data- driven school profiling to support targeted policies. This quantitative study aims to analyze school quality profile clusters using an integrated Principal Component Analysis (PCA) and K-Means Clustering approach. Utilizing secondary data from the 2025 National Assessment Public Report, 13,931 schools were retained from an initial dataset of 23,225 following data cleaning and outlier elimination. PCA reduced seven Z-score standardized indicators into two principal components, accounting for 52.94% of total variance (PC1 = 30.89% and PC2 = 22.05%). Comparative evaluation using the Elbow method, Silhouette Coefficient (0.43), Davies-Bouldin Index (0.805), and Calinski-Harabasz Index (11,530) consistently established k=3 as the optimal partition. The analysis identified: Cluster 1 (3,224 schools) with strong infrastructure but weak innovation (infrastructure-usage gap); Cluster 2 (7,777 schools) with dual deficiencies across both dimensions; and Cluster 3 (2,930 schools) demonstrating superior pedagogical practices despite facility constraints. These findings highlight the critical need to shift from a one-size-fits-all policy approach toward cluster-targeted interventions.]]>
Copyrights © 2026
