Studies examining the effects of socio-economic background on health outcomes is largely based on estimation of mean effects. This study provides evidence that these effects differ across health distributions in important ways, and thus offer a new methodological framework.
In this DIVERGE study I offer a new methodological framework based on quantile regression to be used when examining effects of family environment circumstances on health outcomes. To test this new framework, I examine the effects socio-economic origin on children's health with regard to mental health and physical wellbeing.
"The methodological framework put forward in this study reveals significant heterogeneity in the impact of Socio-Economic Origin across health outcome distributions. For future research to be informative, this heterogeneity must be considered."
I use data from the Danish Longitudinal Survey of Children, and propose a new approach to investigate the impact of family environment circumstances. I employ quantile regression as proposed by Koenker (2001), which is based on using the minimization of weighted absolute deviations to estimate conditional quantile functions. I test this approach as a new framework by investigating how socio-economic origin impact gradients in children's health outcomes across three distinct age-stages: age 7, 11, and 15.
Socio-economic origin is defined as mothers education and parental disposable income. I examine two dimensions of children's health outcomes, namely mental health (measured by means of SDQ) and physical fitness (measured by means of BMI).
Gradients in Children's Health Outcomes
My findings show that by employing a new methodological approach that allows the investigation to move beyond the estimation of mean effects, there are strong and significant distributional effects of family environment circumstances on children’s health outcomes.
With regard to children’s mental health, children in less affluent circumstances experience much larger effects (up to 2.096) relative to their peers in the most affluent circumstances. This trend is prevalent throughout the childhood years and into the children’s teens (age 7 to 15).
Whereas a similar trend is present with regard to physical wellbeing, this trend is much less pronounced and persistent across the included age-stages. Children in less affluent circumstances experience larger effects (up to 0.845) relative to their peers in the most affluent circumstances. In contrast to mental health, however, the effects on physical wellbeing are mainly present in children’s early teen-years (age 15).
Research and Policy Implications
Due to the extensive heterogeneity in effects with regard to magnitude and duration that this study showed, this research has important research and policy implications too. My findings suggest a need to apply a more nuanced methodological approach when evaluating interventions related to children’s health outcomes.