Towards a better understanding of the development of non-cognitive skills in children. Final report: DFG grant no. SCHI 1377/1-1 and SCHI 1377/1-2

Non-cognitive skills are key predictors of central life outcomes such as educational attainment, earnings and health outcomes. Despite their fundamental importance, we know surprisingly little about how these skills form. This project advances our understanding of the formation of non-cognitive skills in childhood and adolescence. In defining non-cognitive skills, we adopt an interdisciplinary approach that encompasses both economic preferences and personality traits. Skill measurement relies on incentivized experiments and validated survey scales. By combining the collection of four waves of panel data on non-cognitive skills of 3,000 whole families with a randomized controlled trial, we provide causal evidence on investments as possible drivers of skill formation on top of cutting-edge descriptive evidence. Children in 135 elementary schools in Bangladesh were randomly assigned to participation in the social and emotional learning program Lions Quest (LQ) that is designed to enhance children's non-cognitive skills. Using the model of skill formation as a common underlying theoretical framework, our findings on skill formation between age 6 and 18 include the following: Participation in LQ increases children's self-control and prosociality. Our results indicate sensitive periods in the formation of self-control and patience around ages 7-9; prosociality is similarly malleable throughout ages 7-11. Participation in LQ also increases children's educational attainment, which in part seems to operate through improving LQ teachers' teaching style. Using our panel data, we go beyond previous cross-sectional evidence by studying the dynamic, within individual development of children's preferences over time. We provide first evidence on self productivity and cross-fertilization of children's preferences. We demonstrate that parents' mental health, their parenting style and investments into children are important sources of the substantial heterogeneity in children's preference trajectories. Based on a novel experimental measure of parental paternalism, we show that most parents interfere paternalistically in their children's intertemporal decision-making to (effectively) mitigate their present bias. Finally, our results highlight the importance of the local environment beyond the family for the formation of children's preferences. We find that adverse shocks such as natural catastrophes can reverse the typical age trajectory of patience, challenging the common notion that patience universally increases as children grow. Using spatial autoregressive models and Kriging, we show that models with spatial components explain a considerable part of so far unexplained variation in children's preferences. In sum, our findings promote basic research on the formation of non-cognitive skills and offer advice to parents, teachers and policy makers on how to foster the development of children's skills.