MINIMUM WAGE DIFFERENCES AND ECONOMIC INEQUALITY IN INDONESIA: A MIXED -METHODS ANALYSIS INTEGRATING MACHINE LEARNING

Abstract

Economic inequality remains a persistent challenge in Indonesia, as reflected in disparities in income distribution across provinces. This study aims to analyze the effect of minimum wage differences on economic inequality by incorporating poverty as a supporting variable using a mixed-method approach. The quantitative analysis utilizes provincial-level data obtained from the Badan Pusat Statistik, including minimum wage, poverty rate, and Gini Ratio, which is classified into high and low inequality categories. The data are processed using Orange Data Mining by applying Logistic Regression, Decision Tree, and Random Forest models, with performance evaluated using AUC, accuracy, precision, recall, and F1-score. The results show that Logistic Regression achieves the best overall performance, indicating that the relationship between minimum wage, poverty, and inequality can be effectively captured through a relatively stable model, while feature importance analysis reveals that minimum wage has a more dominant influence compared to poverty. The qualitative findings based on the PRISMA framework indicate that the impact of minimum wage policies on inequality is conditional and varies across regions, particularly depending on labor market structures and the prevalence of informal employment. The integration of findings suggests that economic inequality is shaped by complex interactions between wage policy, poverty, and regional conditions, implying that minimum wage policy alone is insufficient and should be complemented by broader socio-economic strategie.