SPATIAL MODELING OF CHILD MALNUTRITION IN INDONESIA USING GEOGRAPHICALLY WEIGHTED MULTIVARIATE REGRESSION (GWMR)

Abstrak

In Indonesia aspires to become a developed nation by 2045, with one of its key pillars being the improvement of human resource quality through the achievement of Sustainable Development Goal (SDG) 2: ending hunger and ensuring access to adequate nutrition. However, the prevalence of stunting, wasting, and underweight among children under five remains a critical challenge that hampers these efforts. This study aims to simultaneously analyze the determinants influencing these three forms of malnutrition among Indonesian children by incorporating spatial aspects through the Geographically Weighted Multivariate Regression (GWMR) approach. The analysis employs nine predictor variables representing socioeconomic, demographic, and environmental factors across all provinces in Indonesia. The findings reveal that Complete Basic Immunization, Knowledge of Stunting Prevention, and Lower-Middle Economic Status consistently have significant effects on stunting and underweight. Meanwhile, Complete Basic Immunization and Complementary Feeding Practices play major roles in influencing wasting across provinces.Spatial analysis highlights varying patterns of determinants across regions. Western Indonesia (Java, Sumatra, and western Kalimantan) is more influenced by community behavior (mothers without a MCH Book,Children receiving complete basic immunizations receiving and children recheived complementary feeding), access to adequate sanitation, and lower-middle economic status. In contrast, Eastern Indonesia (Maluku and Papua) is more affected by structural conditions such as preterm births, low immunization coverage, knowledge of stunting prevention, and economic limitations. Central Indonesia demonstrates a more complex and varied combination of influencing factors. Furthermore, the GWMR model exhibits substantially better performance compared to the global (multivariate linear regression) model, as indicated by a significantly lower AIC value (Global AIC = 287.537; GWMR AIC = 44.956). These findings underscore the importance of spatially adaptive and decentralized nutrition policies to ensure more targeted and context-specific interventions.