Unpacking Skills at the Cradle: A Machine Learning Approach to Construct Infant Skill Measures

ended

A growing body of cross-disciplinary research highlights the importance of a child’s environment in the first years of life for skill development and outcomes over the life course. This period is thought to be important for human capital accumulation both because very young children are sensitive to their environment and because deprivation during this period can have long-term consequences. Long-term follow-up studies of early childhood development (ECD) interventions to improve nutrition and create stimulating environments have found large and wide-ranging effects into adulthood: increased college attendance, employment, and earnings and reductions in teen pregnancy and criminal activity. These promising long-term effects of ECD interventions have led to calls from policy-makers and academics to develop large-scale programs that integrate ECD interventions into existing public service infrastructure. Despite this recent call to action, evidence of medium-term impacts of more recent integrated ECD intervention shows that there is substantial fade-out of short-term treatment effects. The puzzling persistence and fade-out patterns of ECD interventions requires more evidence on follow-up studies of existing ECD interventions. Of equal importance is a more detailed analysis of which early skills interventions should target to achieve medium-and long-term improvements in human capital. The proposed research agenda aims to address those concerns in two ways. First, by providing new evidence of medium-term effects on a wide range of child development outcomes of a clustered randomized controlled ECD intervention. Secondly, by developing a new methodology to measure infant skills using innovative tools from the field of machine learning. Both research projects would bring more clarity to the academic literature and policy-makers on which type of infant skills are worth investing in.