Skripsi
PENGELOMPOKAN PROVINSI DI INDONESIA MENGGUNAKAN ALGORITMA K-MEANS BERDASARKAN PERSENTASE ANGKA PUTUS SEKOLAH SMA NEGERI
The high school dropout rate in public high schools remains a significant problem in Indonesia due to various factors, such as socioeconomic conditions, educational facilities, and the ratio of teachers to students. This study aims to group provinces in Indonesia based on the indicator of high school dropout rates in 2024/2025 using the K-Means Clustering algorithm. The data used were obtained from the Ministry of Primary and Secondary Education in 2025, covering 38 provinces and 6 variables, namely the percentage of dropouts (X_1), the ratio of educational units to students (X_2), the ratio of educators to students (X_3), the percentage of certified educators (X_4), the percentage of educational facilities in good condition (X_5), and the percentage of students with parents earning less than IDR 1.000.000 (X_6). The data was standardized and analyzed using K-Means Clustering, with the optimal number of clusters determined using the Elbow and Silhouette methods. The results showed that both methods resulted the same optimal number of clusters, namely K = 3, with 12 provinces in cluster 1, 19 provinces in cluster 2, and 7 provinces in cluster 3. These three clusters respectively represent groups of provinces with distinct characteristics. Cluster 1 shows a low dropout rate with better educational quality but limited service capacity, cluster 2 reflects moderate conditions or is close to the national average, while cluster 3 shows a higher dropout risk associated with lower socioeconomic conditions and educational quality. These differences are mainly observed in the ratio of educators to students, the percentage of certified educators, the percentage of educational facilities in good condition, and the percentage of students with parents earning less than IDR 1.000.000. These results indicate educational disparities between provinces and provide a basis for the formulation of educational policies. Keywords: K-Means Clustering, Dropout Rates, Education, Elbow, Silhouette, Public High Schools