The causal effect of education on health: A consideration of methodological challenges, mediating effects, and variation by age

Abstract

Individuals with lower levels of education show significantly lower levels of life expectancy, are more likely to be physical or mental ill, have higher probabilities of suffering from chronic diseases, and show higher rates of functional limitations and disabilities. However, although there is a wide range of empirical literature showing these associations, the causality of the relationship between education and health is still under discussion. Methodological challenges, scarce evidence for some of the mechanisms linking education to health, and a strong focus on the adult population still render it unclear whether education in general influences health and which mechanisms are predominantly responsible. Hence, this cumulative dissertation project examines the causal effect of education on health in different age groups, considers rarely observed mechanisms, and takes into account methodological challenges in three self-contained articles. Despite the empirical chapters of this thesis, it also provides an overview article (Chapter 1), which derives the main research questions of the project, presents a detailed theoretical model bringing together economic, sociological, and psychological theories, contains a comprehensive overview of the state of research, highlights the main contribution of the conducted research, describes the research designs, and summarizes the main findings and draw a conclusion. The first empirical article (Chapter 2) examines whether education in terms of its impact on individual skills and knowledge and education-related feedback affects children and adolescent mental health. Using fixed-effects panel regression models, this article firstly eliminates time-constant confounding factors and shows that school grades are highly important for children’s and adolescents’ life satisfaction approximating their mental health, but academic competence levels only affect specific subgroups. In the second empirical article (Chapter 3), my coauthor and I investigate whether social relationships contribute as a mechanism to the effect of education on health in adults aged between 30 and 60. By implementing a causally driven mediation analysis, we show that the overall level of social integration especially the number of close friends and relatives outside the household and active membership in a voluntary association explain some part of the effect. This article also highlights that although social relationships appear to mediate the effect of education on mental health, they do not play a similar role in the relationship between education and individual global health status. The third empirical article (Chapter 4) analyzes the effect of education on various health-related outcomes in older adults. Using an instrumental variable approach following a more sophisticated implementation strategy compared to prior research, this study addresses the effect of the expansion of compulsory schooling on individuals’ global health status, physical constitution, smoking behavior, alcohol consumption patterns, and physical activity levels. Results, however, indicate that, except for regular alcohol consumption, education does not causally affect the health outcomes considered. In summary, the results of this thesis contribute to the wide literature addressing the relationship between education and health by highlighting the relevance of education-related aspects to children’s and adolescents mental health, the mediating role of social relationships in middle-aged adults, and the lack of a causal effect of the expansion of compulsory schooling on health-related behavior and health status in older adults.

Universal Solvent or catch-all Proxy? : Causal Evidence on the Effect of Education on Political Participation and Attitudes in Germany

“…The shaded cells mark the period of time we cover in our sample prior to, and after the reform. Source: Cygan-Rehm (2022); Cygan-Rehm and Maeder (2013); Helbig and Nikolai (2015), illustration based on Kroh (2021).…”

“…More recently, research interests expanded to other educational returns like health and health behavior (Galama et al (2017); Kroh (2021)), deviance (Lochner (2011); Machin et al (2011)) or social and political engagement (Huang et al (2009); Lochner (2011)).…”