Poverty is still one of the major problems in East Java, even though this province has an important role in supporting the national economy. This condition shows that development in each district/city has not been evenly distributed, so a data-based analysis is needed to determine the factors that influence the poverty rate. This study aims to analyze the influence of socioeconomic variables on the poverty rate category in East Java using a multinomial logistic regression model. The data used is secondary data from the Central Bureau of Statistics (BPS) in 2023 which covers 38 districts/cities. The independent variables analyzed consisted of life expectancy, average years of schooling, open unemployment rate, labor force participation rate, expenditure per capita, human development index, and gross regional domestic product (GRDP) per capita. The analysis process involved data exploration, multicollinearity test, multinomial logistic regression modelling, simultaneous and partial parameter significance test, and model performance evaluation. The results show that per capita expenditure is the only variable that has a significant effect on poverty level classification. The model is able to classify the data with an accuracy of 81% and a McFadden R² value of 0.6483, which means the model has a fairly good performance. This finding shows the importance of increasing people's purchasing power as an effort to reduce poverty. This research is expected to be a reference for local governments in formulating more targeted and data-based policies.
Isyanto et al. (Sun,) studied this question.