Child labour is closely associated with poverty. However, the direction of causality is an empirical question. There is need to control for potential endogeneity in order to be able to adequately estimate the factors that determine child labour. This study proposed a model of an agricultural household to explain the factors that affect the household's decision to involve their children in child labour and the type of influence each factor has on the household. These factors include household resources, child characteristics, community characteristics, school availability, etc. The data was analysed using both Tobit and Logit models. The Tobit model was used to find the relationship between the factors and duration of child work while the Logit model was used for the participation of the child in farm work. The outcome of the analysis showed that among agricultural households in Ethiopia, child labour is a normal good increasing with income. However, the impact on the male child was different from that of the female child, suggesting that gender bias with respect to child labour might exist in Ethiopia. The male child is made to participate more in farm work than the female child, though the females responded more to household land holding (size). This can be attributed to the need for the household decision maker to substitute household chores performed by the female child for farm work. The substitution effect of increase in income on household decision on child farm work is higher than the income effect, irrespective of the gender of the child, although the effect was significant for the male child but not significant for the female child. Also, school availability is a very important factor for both the male and the female child. The impact of household size in this analysis suggests the presence of division of labour, and the significance of the mother's education on the female child's response suggests that the effect of cultural belief system changes with the mother's education.
Dissertation (MSc (Agric))--University of Pretoria, 2017.