Preliminary Draft. Please do not quote or circulate without author’s permission. The Black-White Test Score Gap: Lessons from the Panel Study of Income Dynamics

Based on two waves of panel data for three age cohorts of children from the Panel Study of Income Dynamics Child Development Supplements, we find large black-white test score differences among children of all ages. Even before children start formal schooling, black children score.8 and.5 of a standard deviation lower than whites in Applied Problem and Letter-Word tests respectively. Except for the oldest cohort, the gaps for all tests widened when children’s cognitive skills were assessed again six years later. All achievement gaps before grade three can be accounted for when we control for the child’s characteristics, home environment (both structural and cultural factors), and a proxy for mother’s cognitive skills. As children advance to higher grades, there is a diminishing role of these covariates in explaining the achievement gap. In preschool years, the gaps are reduced to less than.2 of a standard deviation when all covariates are controlled for. In the first three years of school, the gaps are reduced to about.3 of a standard deviation, whereas in high school years, the gaps remain a statistically significant.5 and.65 of a standard deviation for Applied Problem and Letter-Word scores respectively after all the covariates are controlled for (representing 63 % an