Experimental Design
The project will survey a randomly selected subset of approximately 7,000 children aged 3-8 of 7,500 adult respondents in the Kenya Life Panel Survey (KLPS), creating the new KLPS-Kids dataset. The original KLPS sample contains Kenyans who participated in one or more of three earlier interventions: a health study known as the Primary School Deworming Program (PSDP; described in detail in Miguel and Kremer, 2004), a vocational training voucher study, or a cash grant intervention (described in detail in Hicks et al., 2015c).
Previous results indicate that primary school deworming led to health, schooling, and labor market gains in young adulthood, even 10 years after the launch of the program (Baird et al., 2016). Although existing work does not find much evidence of substantial labor market gains due to vocational training (Hicks et al., 2015c), findings do suggest substantial self-employment profit gains, at least in the short run, due to unconditional cash grants (Hicks et al., 2015b). These sizeable direct impacts on parents provide a potential channel for the intergenerational impacts on their children that we will study in the current project, although it is possible that gains for parents in unmeasured dimensions might also influence child outcomes.
In order to measure the impacts on the recipients’ children, this project will create locally appropriate versions of both standard and innovative survey instruments designed to measure various domains of development among children aged 3-5 and 6-8. Since the selection of beneficiaries for the PSDP, vocational training voucher, and cash grants were randomized, the data will enable the estimation of causal impacts of these programs on recipients’ children’s outcomes, overcoming the key methodological problem of confounding.