<![CDATA[Stunting remains a major public health issue in Indonesia, including in Kabupaten Bone Bolango, which recorded a high prevalence in 2023. Variations in social, economic, and environmental conditions across observation locations indicate the need for analyses that account for spatial differences between areas. This study aims to identify the spatial variation in the effects of significant stunting risk factors using the Geographically Weighted Logistic Regression (GWLR) method. The data were obtained from the Health Office of Kabupaten Bone Bolango in 2019. The independent variables included complete basic immunization (X1), the percentage of low-birth-weight infants (X2), and the percentage of exclusive breastfeeding (X3), with the response variable defined as high stunting prevalence (1) and low stunting prevalence (0). The analysis comprised multicollinearity testing, the Breusch-Pagan spatial heterogeneity test, bandwidth selection using cross-validation, construction of an adaptive Gaussian kernel weighting matrix, and parameter estimation via maximum likelihood with the Newton-Raphson method. The multicollinearity test indicated that all variables were free from collinearity (VIF 10). The Breusch-Pagan test revealed the presence of spatial heterogeneity (p 0.10), confirming the appropriateness of the GWLR model. The results showed that the percentage of exclusive breastfeeding was significantly higher in Bone Raya, Bulawa, Bone, Bone Pantai, and Kabila Bone, whereas complete basic immunization and the percentage of low-birth-weight infants were not significantly different. These findings indicate that exclusive breastfeeding is a risk factor for stunting, with significant spatial variation, suggesting that stunting intervention strategies should be designed on a point-by-point, location-specific basis, taking into account the local characteristics of each observation point.]]>
