The paper examines the linkages between child labour and schooling of children ages 5-17 years using Cameroon based on National Household Survey (ECAM 3, 2007). An econometric specification that explicitly takes this interdependency into account enables us to characterize the child’s schooling and work decisions jointly is the Bivariate Probit. In rural areas, the results show that male children are more likely to attend school. They also indicate the existence of a gender bias in children’s allocation. On the other hand, the nature of the relationship to the head has insignificant impact on the likelihood of school attendance as well as market work participation. The educational level of the head of the household significantly improves child education and decrease the likelihood of child labour. Then, children from households headed by a person with at least primary education are more likely to attend child work participation and head education. In the same time, in urban Cameroon, we find supportive evidence of the luxury axiom that poverty drives child labour. Our result shows that poverty appears to be the main culprit of the prevalence of child labour in urban areas. Household educational level variables significantly reduce the probability of child labor and improve the likelihood of children being in school in all cases