DINAMIKA KEPUTUSAN PERNIKAHAN USIA DINI PADA PEREMPUAN: PENDEKATAN ANALISIS FAKTOR SOSIAL EKONOMI BERBASIS DATA PANEL DI INDONESIA

Abstract

Early marriage among women remains an unresolved human development challenge in Indonesia, characterized by fluctuating prevalence rates and pronounced regional disparities. This study examines the effects of female education level, the Gender Empowerment Index (GEI), Female Labour Force Participation Rate (FLFPR), and female per capita expenditure on early marriage rates across 10 Indonesian provinces during 2020–2024. Using a balanced panel of 50 observations, model selection through Chow, Hausman, and Lagrange Multiplier tests identified the Random Effects Model (REM) as the most appropriate estimator. Results show that female education (β = −0.464; p = 0.0000), GEI (β = −0.260; p = 0.0414), and FLFPR (β = −0.266; p = 0.0021) each exert a significant negative effect on early marriage prevalence, while female per capita expenditure is statistically insignificant (p = 0.2649). The model explains 62.51% of inter-provincial variation (Adjusted R² = 0.6251), and a high intra-class correlation (ρ = 0.8085) highlights the need for region-specific policy responses. These findings underscore that expanding access to education, advancing gender empowerment, and broadening formal employment opportunities for women are the most effective levers for reducing early marriage in Indonesia.