Quantiles of the gain distribution of an early childhood intervention

Abstract

SummaryWe investigate the distribution of gains among participants in the Infant Health and Development Program, an understudied randomized controlled trial that targets infants with low birth weight. Our primary focus is on assessing the effects in cognitive and health outcomes within distinct subgroups, which we define based on the outcomes that would occur in the absence of program participation. We propose a strategy to estimate the distribution of gains from the program by using anthropometrics measurements taken at birth, under the assumption that potential outcomes depend on underlying latent factors explaining neonatal health. Our findings reveal that the enhancements in cognitive and health outcomes at 36 months are not uniformly distributed among program participants. The variability in these effects can be attributed to several factors, including neonatal health, post‐natal shocks, and family income.