Hierarchical Cluster Analysis Based on Stunting-Related Indicators in Maluku and North Maluku

Abstract

Stunting remains one of the major public health challenges in Indonesia because it affects children's physical growth, cognitive development, and future productivity. Identifying regional characteristics based on stunting-related indicators is important to support more targeted intervention policies. This study aims to classify regions in Maluku and North Maluku Provinces based on stunting-related indicators using Hierarchical Cluster Analysis. The study employed secondary data from 21 districts/cities in 2023 obtained from the Ministry of Health of the Republic of Indonesia, the National Socio-Economic Survey (SUSENAS), Statistics Indonesia (BPS), and the National Food Agency. The variables analyzed were expected years of schooling, food security index, households with access to proper sanitation, poverty rate, access to safe drinking water, complete basic immunization coverage, and the percentage of children under two years who received breastfeeding. Prior to clustering, all variables were standardized using Z-scores. Hierarchical Cluster Analysis was performed using Ward’s linkage method and Squared Euclidean Distance. The results indicated that the regions could be classified into two clusters. The agglomeration schedule and dendrogram supported a two-cluster solution, which was further confirmed by K-Means validation. The first cluster consisted of four regions characterized by higher expected years of schooling, food security, sanitation coverage, access to safe drinking water, and complete basic immunization coverage. The second cluster consisted of seventeen regions with relatively less favorable characteristics. Although the average prevalence of stunting was relatively similar between the two clusters, the results suggest that regions with comparable stunting prevalence may possess different socioeconomic, environmental, and health-related profiles. These findings provide useful information for identifying regional characteristics and may support the formulation of more targeted stunting prevention and control strategies.