Analyzing Poverty Across Indonesian Provinces Using Panel Data Regression

Abstract

This study aims to analyze the factors that influence poverty levels in provinces in Indonesia using panel data regression. This study uses secondary data from the Central Statistics Agency (BPS). The data used includes variables such as provincial poverty rate (TKP), gross regional product per capita (PPK), open unemployment rate (TPT), and average length of schooling (RLS). The results of the analysis show that the fixed effect model is the most appropriate model for this data, as confirmed by the Chow test and the Hausman test. RLS has a negative and significant effect on poverty rates, indicating that increased education correlates with a decrease in poverty. Meanwhile, TPT has a positive but insignificant effect at the 5% level, although it is close to significant at the 10% level. This suggests that increased unemployment tends to increase poverty. PPK has a negative but insignificant effect on poverty levels, reflecting the phenomenon of “growth without justice” where the benefits of growth are not felt equally by all segments of society. This model has a very high level of suitability, with an R-squared value of 0.996810, which means that 99.68% of the variation in poverty levels can be explained by the independent variables