<![CDATA[Poverty remains a complex socioeconomic challenge in East Java Province, largely due to strong inter-regional dependencies. This study aims to model poverty rates across 38 regencies/cities in East Java using a Spatial Autoregressive (SAR) approach while identifying its key driving factors. Secondary data were collected from the Central Bureau of Statistics (BPS) and analyzed using a cross-sectional spatial model with a Queen Contiguity weight matrix. The Global Moran’s I test confirms a significant spatial autocorrelation in poverty rates across regions (I = 0.2258, p = 0.0032). Based on Lagrange Multiplier tests, the SAR model performs better than the standard Ordinary Least Squares (OLS) model, achieving an R2 of 0.7837 and a lower AIC value (170.72 compared to 174.46 for OLS). Open Unemployment Rate and Per Capita Expenditure significantly affect poverty levels. Furthermore, the spatial autoregressive parameter ( = 0.2944, p-value = 0.0185) confirms a positive spatial spillover effect from neighboring areas. Practically, these findings suggest that local governments should shift toward collaborative cross-border policies rather than handling poverty in isolation.]]>
Copyrights © 2026
