Educational Time Use and the Cognitive Development of Children: Evidence from the Longitudinal Study of Australian Children

This thesis uses data from the Longitudinal Study of Australian Children to study the relationship between educational time at home and the cognitive development of children. After an introduction, chapter 2 reviews the literature on home learning activities and the cognitive development of children. It finds that the widely used home environment index (or home activities index) attempts to proxy for a few inputs, mainly time inputs. More recently, research has begun to examine the effect of parental time investment. However, economic research has largely not taken into account a child's own time investment, even though other fields of study have shown that children can self learn. The paper then provides descriptive evidence that shows the strong association between time investments and children's cognitive outcomes for a wide range of age groups. Furthermore, the association is stronger for child's own time investment than for parental time investment. Finally, it discusses and uses the data to illustrate sources of endogeneity that arise when examining the relationship between time investments and children's cognition and how not controlling for covariates can confound this relationship. Chapter 3 studies the importance of maternal/paternal and child's own educational time investment for childhood cognitive development. It estimates the cognitive development production function under alternative specifications. Two key findings emerge from the analysis: when children are 2-3 years old, the most important activity for current cognitive skills is educational time spent with the mother. When children are 6-7 years old, the child's own time investment in educational activities is more important for current cognitive skill development than maternal/paternal educational time. Using a recently developed exogeneity test, I fail to reject exogeneity for most of the specifications used in this paper. Chapter 4 extends the analysis from chapter 3 to later waves. We find that for 10-15 year olds, alone educational time is more important for children's cognitive development than parental educational time. We also find that the return to alone educational time is largest when children are 6-7 years old, and it is decreasing as children get older. However, differing returns to parental and child's own time investment can be generated by two mechanisms: 1) differing returns at the disaggregated level (i.e. differing returns for the same activities or the same types of activities) and/or 2) differing time allocation at the disaggregated level (i.e. differing time allocation within parental and child's own time investment). We derive an Oaxaca style decomposition to determine the contribution of each channel. Our decomposition results show that 'differing returns' is the largest component behind the differing coefficients on alone and parental educational time, accounting for 82%-88% of the difference. Therefore, we find that doing the same activity alone is more productive than when it is done with a parent