School Quality and Health at Midlife: Evidence from High School and Beyond

school desks

Research on the association between education and later-life cognitive outcomes usually only considers the quantity of schooling that people complete. However, people with the same level of completed schooling are highly diverse with respect to the quality of schooling they experienced. Using a unique new data resource, we ask how secondary school quality is associated with cognitive functioning at age ~60—net of school quantity and confounders. We use data from the High School and Beyond cohort—a large, diverse, nationally representative probability sample of American high school sophomores and seniors in 1980. HSB panelists were interviewed many times between 1980 and 2022, with high rates of participation; HSB has been supplemented with administrative data (e.g., school transcripts, mortality records) and the 2021/2022 follow-up gathered fluid biomarkers useful for studying markers of risk for cognitive impairment. Measures from the 1980s provide extraordinarily detailed information about educational contexts, opportunities, and outcomes. We measure secondary school quality using value-added models, which estimate each school’s contribution to student academic achievement conditional on students’ baseline achievement and family socioeconomic background. Our analyses assess the degree to which attending a high value-added secondary school is associated with better cognitive outcomes at age ~60 net of educational attainment. Our main dependent variables—assessed in 2021/2022—include self-reported memory complaints and self-reported conditions of relevance to risk of cognitive impairment (e.g., diabetes, hypertension). Net of educational attainment and demographic factors, we find that individuals who attended a high value-added secondary school experience fewer cognitive complaints (as assessed using the AD8). They are also less likely to have been diagnosed with diabetes or hypertension. In the next few months, we will have assessed heterogeneity in results across demographic and socioeconomic groups.

H. Chung, University of Minnesota; R. Warren, University of Minnesota; E. Grodsky, University of Wisconsin

Friday, July 7, 2023
RC28, Paris, France