Background and Objectives
Child work is assumed to negatively affect children’s health and development [11] including cognitive development. This paper explores the relationship between the daily hours a child spends working at age 11-12 years and their cognitive abilities at age 11-12 and 14-15. It considers an inclusive and a narrow definition of child work in order to compare the results that each produces.   
Methods  
The main outcome variables were PPVT score in round 2 (cross-sectional analyses) and PPVT score in round 3 (prospective analyses). The main exposure variables were hours spent in all types of work, and hours spent in economic work in round 2.   Linear regression techniques were used to explore the associations between daily hours spent working at age 11-12, and cognitive test score at age 11-12 and 14-15. Crude analyses were carried out, before being adjusted for covariates identified from existing literature.  
Results
Crude models suggested a negative association between the hours spent in both economic work and all work at age 11-12 years, and cognitive abilities at ages 11-12 and 14-15. However once adjusted for all covariates these associations become non-significant in three of the regression models, with the exception of the observed negative association between spending 1-2 hours per day in economic
work in round 2 and PPVT score in round 3.  
Conclusions
In general, the relationship between work and PPVT score exists because the children who work greater hours are poorer, and are therefore more likely to suffer from lower PPVT regardless of their work. Poverty therefore seems to be the driver of both child work and lower cognitive ability. However the observed association between spending 1-2 hours in economic work and lower PPVT suggests that there is some association between economic work and PPVT score that is not explained by socioeconomic covariates.