To determine whether education has an intergenerational effect on physical and mental health of the next generation, we separately model the relationship between the average years of parental education and self-report scores of their children's physical and mental health as measured through the SF12 questionnaire. We find that there is no significant relationship between one’s SF12 mental or physical health scores and the average years of parental education, especially after holding confounding variables like the individual’s own education, race, family income and access to a healthcare provider constant. We do however see repeated evidence of an individual’s own education being correlated with their health scores, which emphasizes the importance of providing adequate access to education to every generation for optimal societal health outcomes, especially since these effects are not intergenerationally persistent. Finally, the paper discusses the potential limitations of our model in terms of external validity and internal validity - namely omitted variable bias, imprecise variables and potentially misspecified functional form, and suggests modifications for future replications of the study.
You may find the dataset created using the NLSY97(https://www.nlsinfo.org/investigator/pages/login?p=timeout) panel survey data, named Chevalier_Mehdiyar-1.dta, as well as our do-file and log-file in this repository for replication purposes. The final paper is also available.