Schools and socioeconomic inequality in achievement – Revisiting the ‘school equalization’ hypothesis in the United States

Schooling is often seen as equalizing learning outcomes among children from different socioeconomic backgrounds. However, empirical support for the ‘school equalization’ hypothesis in the United States remains inconclusive, despite decades of research relying on summer breaks to estimate schooling effects. This study applies, for the first time in the U.S., the differential exposure approach – a recently developed causal identification strategy – as an alternative to the seasonal comparison design. We unify both approaches within a single analytical framework, enabling direct comparison of their estimated effects under consistent conditions. Using nationally representative data from the Early Childhood Longitudinal Study (ECLS-K:2011), we show that the seasonal comparison design may substantially inflate estimated school effects on learning, potentially misattributing to schools what are, in fact, consequences of changes in non-school environments during summer. Equalization effects estimated via the differential exposure approach were about 50% smaller than those inferred through seasonal comparisons. These findings suggest a tempered, grade-specific perspective: while schools reduce socioeconomic disparities in learning, their equalizing impact appears smaller than previously believed and diminishes across grades. The article challenges foundational assumptions in prior research and, by establishing a methodological basis for cross-national comparison, suggests that school equalization is contingent on national institutional contexts.