Inequality in Child Development and Public Policy in Australia, Canada, United Kingdom, and United States

This project is a four-country study of socio-economic differences in child outcomes as children move through the school years. Parents have strong, perhaps even primordial motivations to invest in the social and economic well-being of their children. But as economic inequality rises and the gap between the rich and the poor grows increasingly larger, the ability of rich and poor families to make such investments diverges significantly. One consequence of the differential ability of rich and poor parents to invest in their children is larger gaps in child outcomes by socio-economic status. The evidence to date suggests that the correlation between parental SES and children's cognitive and developmental outcomes (often called the SES gradient) is solidly established by the time children enter school. It has appeared, without fail, as early in childhood as it can be measured in every country so far studied. But what happens after that is less well understood. Nor is the story necessarily the same everywhere. In countries with high levels of inequality, like the U.S., early SES gaps may be particularly wide, and as the process of cumulative advantage plays out over time, these gaps may grow as children age. In other countries, where inequality is less pronounced, initial SES gradients may be weaker and public investments in institutions such as universal education and health care may serve to moderate the effects of private advantage. Bruce Bradbury, Miles Corak, Jane Waldfogel, and Elizabeth Washbrook propose to study the development of socio-economic differences in child outcomes in Australia, Canada, the U.K. and the U.S. The four Anglophone countries provide an interesting contrast. The U.S. has both the highest levels of economic inequality and the highest levels of intergenerational persistence of economic advantage. The other three countries show more moderate levels of inequality and intergenerational persistence. Bradbury and his colleagues will frame their study by addressing two central questions. First, they ask whether SES gradients in outcomes such as cognitive ability are present at school entry in all four countries and whether there are significant differences in the size of these gradients across countries. And second, they ask whether the gradients widen, hold constant, or diminish as children move through the school years and how those developmental patterns differ by country. They propose to focus on three key outcomes of developmental interest: cognitive performance, socio-emotional skills, and child health. Data for the study will come from contemporary, high-quality, and nationally representative longitudinal data sets in each of the four countries. The most unique aspect of this study is the ability to trace the development of the same individuals from an early age over an extended period with comparable measures of key life outcomes at similar ages in nearly contemporaneous cohorts in all four countries. The Longitudinal Study of Australian Children (LSAC; N=4,983) follows children from age 4 in 2004 through about age 12 in 2012. In the National Longitudinal Study of Canadian Youth (NLSCY; N=8,522) children are selected at age 4-5 in 1998-2004 and followed through age 13. The U.K.'s Millennium Cohort Study (MCS; N=19,517) follows children born in 2000 to age 11. And in the U.S., the Early Childhood Longitudinal Study-Kindergarten (ECLS-K; N=21,400) cohort follows children entering kindergarten in 1998 through to age 13. All data sources contain a rich variety of information on family and child background as well as assessments of children's cognitive development, emotional and behavioral well-being, and health at three distinct time points in children's development: school entry (age 4-5), primary school (age 6-10), and middle school (age 11-13). Data for these analyses are publicly available (LSAC and MCS), or in the case of restricted datasets, the PIs already have licenses for data access (NLSCY and ECLS-K). All analyses will be conducted using two primary measures of parental socio-economic status: parental education measured as the highest education of either parent in the home and parental income quintiles. The primary dependent variable of interest is children's cognitive development, but substantial attention will also be paid to emotional and behavioral development, and general measures of health. In order to extend their analyses of cognitive development to cover a slightly longer age span, they will rely on data from the PISA studies, the international comparisons of educational achievement for 15-year olds, to bolster the analyses of cognitive differences.