<![CDATA[Poverty is one of the problems still faced by Indonesia. The problem of poverty is a development priority because poverty is a complex and multidimensional problem. Therefore, to reduce poverty, it is necessary to know the factors that influence the number of people living in poverty. The influencing factors in each region are different due to the effects of spatial heterogeneity between regions such as geographical, economic, and socio-cultural conditions. This research considers spatial factors by using the Geographically Weighted Negative Binomial Regression (GWNBR) method on poverty-based regions in Kalimantan Island. This research uses eleven independent variables. The weighting function used is the Adaptive gaussian kernel because the adaptive kernel can produce the number of weights that adjust to the distribution of observations. The stage starts with descriptive statistics and checking multicollinearity. Then proceed with the formation of Poisson Regression, because the data used is enumerated data. Then check for overdispersion. If overdispersion is detected where the variance is bigger than the mean, then Negative Binomial Regression is continued. After that, it is tested for the presence or absence of spatial heterogeneity. If there is, proceed to find the bandwidth and Euclidean distance. After that, the graphical weighting matrix is searched. Then proceed with GWNBR modeling. The results of the analysis show that there are seven significant variables, including the percentage of households with the main source of lighting is non-state electricity company (PLN), average monthly net income of informal workers, population density for every square kilometer, monthly per capita expense on food and non-food essentials, percentage of people who have a health complaint and do not treat it because there is no money and percentage of population 15 years and above who do not have a diploma. Based on the categories of significant variables, six groups were formed in 56 districts/cities in Kalimantan Island.]]>
