Analisis Eksploitasi Pekerja Anak dari Sisi Pendidikan di Pulau Jawa dengan Regresi Logistik Biner Multilevel

Abstract

Education gives children the opportunity to have a better life in the future, but unfavorable economic conditions forced children to work and caused some children not to go to school. Child laborers are vulnerable to exploitation because they are considered weak and have no bargaining position in the work world. This study aims to analyze the variables that affect the exploitation of child labor in terms of education. Child laborers are said to be exploited in terms of education if they have never attended school or are no longer in school. The results of the Susenas 2019 show that exploitation is more common in boys, being older, working in the formal sector, working longer hours, living in a rural area, having a smaller household size, having a male, married, not working, low education, and older age household head. The random effect test shows that the variation in the exploitation of child labor in terms of education is significant between groups so that the 3-level  model is better than the 1-level binary logistic regression model. The exploitation of child labor in terms of education at the individual level is influenced by age, sector of work, and working hours of children; at the household level it is influenced by the gender and education of household head, classification of residence, level of welfare; and at the regional level it is affected by regency/municipality poverty.