A Four-Country Study on the Relationship Between Parental Educational Homogamy and Children's Health from Infancy to Adolescence

Popul Res Policy Rev. 2022 Feb:41:251-284. doi: 10.1007/s11113-020-09627-2. Epub 2021 Jan 2.

Abstract

This study explores the relationship between parental educational similarity - educational concordance (homogamy) or discordance (heterogamy) - and children's health outcomes. Its contribution is threefold. First and foremost, I use longitudinal data on children's health outcomes tracking children from age 1 to 15, thus being able to assess whether the relationship changes at key life-course and developmental stages of children. This is an important addition to the relevant literature, where the focus is solely on outcomes at birth. Second, I look at different health outcomes, namely height-for-age (HFA) and BMI-for-age (BFA) z-scores, alongside their dichotomized counterparts, stunting and thinness. Third, I conduct the same set of analyses in Ethiopia, India, Peru, and Vietnam, thus providing multi-context evidence from countries at different levels of development and with different socio-economic characteristics and gender dynamics. Results reveal important heterogeneity across contexts. In Ethiopia and India, parental educational homogamy is associated with worse health outcomes in infancy and childhood, while associations are positive in Peru and, foremost, Vietnam. Complementary estimates from matching techniques show that these associations tend to fade after age 1, except in Vietnam, where the positive relationship persists through adolescence, thus supporting the homogamy-benefit hypothesis not only at birth, but also across the early life course. Insights from this study contribute to the inequality debate on the intergenerational transmission of advantage and disadvantage and shed additional light on the relationship between early-life conditions and later-life outcomes in critical periods of children's lives.

Keywords: adolescence; child health; educational homogamy; longitudinal data; low- and middle-income countries.