Schooling, Self-Selection, and Health
检验教育与健康正相关的两种解释：教育对健康的直接效应，或未观测变量（如时间贴现率）的间接效应。使用四种健康指标估计模型，结果一致表明教育的直接效应更重要。
Economists have long realized that schooling and good health are strongly positively correlated. Some conclude that schooling has a direct positive effect on the production of good health while others argue that some unobserved variable such as rate of time discount positively affects both health and schooling. This study investigates the validity of alternative explanations for the observed schooling-health correlation. Models are estimated using four different measures of overall health: disability, functional limitations, and systolic and diastolic blood pressures. The results uniformly indicate that the direct effect of schooling on health is more important than the effect of unobservables.