Liying Luo, Penn State
Jiahui Xu, Penn State
John Robert Warren, University of Minnesota
Kenneth M Langa, University of Michigan
Recent research indicates that adult children’s education may influence their parents’ health at older ages. However, empirical evidence about this upstream effect from children to parents is mixed likely due in part to differing methodological approaches. Moreover, a critical question remains understudied: Would increasing children’s education increase or reduce observed educational inequalities in health among parents? To answer this cross-generation question, we first explicate a two-generation upstream framework that accounts for intergenerational selection and interdependent pathways. We then use multilevel marginal structural models and well-defined casual estimands to analyze longitudinal data from the Health and Retirement Study. We find that educational disparities in five health conditions at the population level would be significantly reduced if younger generations obtained more education. While the health advantages for college completing parents would remain unchanged, disadvantages for the least educated parents (less than 12 years of schooling) would decrease by 27 to 43 percent. This reduction in inequality is evident when education increases among either the least or the most educated child.
Keywords: Health and Morbidity, Longitudinal studies , Data and Methods, Population Ageing