At high income levels, cognitive abilities are less predictive of school grades among white compared to minoritized students.

Socioeconomic disparities in academic attainment – termed socioeconomic achievement gaps – are persistent and undermine equitable occupational and health trajectories across the lifespan. However, it’s not well understood whether the extent to which cognitive abilities are reflected in students’ school grades varies across socioeconomic and demographic characteristics. In this longitudinal study, we assessed whether prospective associations between cognitive abilities and school grades differ by income-to-needs ratio, parental education, and area deprivation in 11,868 youth from the Adolescent Brain Cognitive Development study – a diverse national pediatric sample from the U.S. We additionally examined further stratification by sex assigned at birth and racial and ethnic identity. Socioeconomic and cognitive variables were measured at baseline (2016-2018; mean age 9.9 years (SD=0.7)). School grades were reported three years post baseline. We show that reading decoding ability and overall crystallized cognition were less predictive of school grades at higher levels of income-to-needs ratio. Stratification by race/ethnicity indicated that the weaker associations between crystallized cognitive functions and school grades at high incomes were only evident in non-Hispanic White students. Higher average academic achievement in this group may be more reliant on factors like greater parental expectations, cognitive stimulation at home, and attendance of schools with grade inflation practices. Conversely, students from less affluent backgrounds may face greater pressures to demonstrate cognitive abilities to receive high grades. In sum, cognitive abilities are not equitably reflected in school grades across economic and racial student strata and potential structural advantages availed to affluent White students in the U.S. require further examination.