Birth Weight and Cognitive Development in Adolescence

Bridget K. Gorman, University of North Carolina at Chapel Hill

Using data from the National Longitudinal Survey of Adolescent Health (Add Health), I investigate the relationship between birth weight and cognitive development among adolescents aged 12-17. Bivariate OLS regression models reveal a significant, positive relationship between birth weight and PPVT score (a measure of verbal intelligence). Controlling for demographic, socioeconomic, and other adolescent characteristics modifies, but does not eliminate, this relationship. However, as traditional regression models cannot fully account for family background characteristics (i.e., the relationship between birth weight and PPVT may be spurious, due in part to the poorer SES profile of low birth weight infants), I exploit the sibling subsample of the Add Health data to more adequately control family background effects. Fixed-effects models of non-twin full siblings indicate no significant association between birth weight and PPVT, suggesting that 'traditional' cross-sectional models overstate the influence of birth weight for cognitive development in adolescence.