Modeling of child labour exploitation status in Indonesia using multilevel binary logistic regression

The Indonesian constitution recognizes guarantees the right of child to rest and leisure, engage in play, recreational activities appropriate age so that they should not be working. Employers are also prohibited employ children. However, many children come work because poverty, even though labour is close exploitation. Theoretically, individual contextual factors affect exploitation status labour. This study aims analyze variables influence Indonesia based on data from National Socio-Economic Survey (Susenas) 2018. random effect test shows there differences between regency/municipality multilevel binary logistic regression performs better than one level regression. More 80 percent labourers exploited terms education working hours. Variables significantly at gender, occupation sector labour, household head. Meanwhile, poverty rates mean years schooling regional level.