College Attendance, Degree Completion, and Health: Decomposing the Education-Health Gradient

• In Event: Sociology of Education Section Roundtables
  In Refereed Roundtable (60 minute): Table 24. Equity, Justice, and Health in Shaping Educational Outcomes

Abstract

Despite extensive research linking educational attainment to health outcomes, the mechanisms through which college education improves health remain poorly understood. A particularly puzzling pattern is the marginal health benefit observed among individuals with "some college" education—those who attended college but did not complete a bachelor's degree—who exhibit health outcomes similar to high school graduates despite substantial postsecondary investment. Understanding this anomaly requires disentangling whether health benefits derive from the college experience itself or from degree completion. This study addresses this question by applying a causal mediation framework that decomposes the total effect of college education on health into two distinct pathways: the direct effect of college attendance independent of degree completion, and the indirect effect operating through degree completion itself.
Using data from the National Longitudinal Survey of Youth 1997 Cohort (NLSY97, n = 4,619), I employ the Regression-with-Residuals (RWR) approach to adjust for both pre-college confounders—including family socioeconomic status, cognitive ability, and demographic characteristics—and post-treatment confounders such as college type and field of study. This design enables identification of natural direct and indirect effects while exploiting the deterministic constraint that degree completion requires prior college attendance, thereby relaxing the otherwise untestable cross-world independence assumption.
Results indicate that the total effect of college education on self-rated health is statistically significant, but this benefit operates almost entirely through degree completion. The natural direct effect of college attendance alone is negligible and nonsignificant, whereas the natural indirect effect through graduation accounts for approximately 76 percent of the total effect. These findings suggest that the credential itself, rather than the college experience, is the primary mechanism linking higher education to better health outcomes, with direct implications for policies targeting completion rather than access alone.

Author

• Seungyeon Jung, Brown University