This paper investigates the distributional characteristics of racial differences in mathematics achievement, with particular attention to the potential influence of unexplained, and possibly unwarranted, racial differentials in rates of school suspension. It is well known that black students consistently score lower than whites on achievement tests, on average, even after controlling for family and schooling factors. Using data from the National Longitudinal Survey of Youth 1997 cohort, we decompose the racial gap in mathematics test scores from the Peabody Individual Achievement Test-Revised (PIAT-R) into a component due to racial differences in underlying characteristics and another component that is unexplained by differences in measured characteristics. We account for the possible endogeneity of suspensions in our analysis and show that the portion of the racial gap that is unexplained differs between the top and the bottom of the test score distribution. Our results suggest that researchers should pay more attention to the problem of concentration of black students among those at the bottom of the distribution and their scarcity among those at the top of the test score distribution to better understand the factors that account for the observed disparities.
- Achievement tests
- Distributional characteristics
- Mathematics achievement
- National Longitudinal Survey of Youth (NLSY)
- Peabody Individual Achievement Test-Revised (PIAT-R)
- Racial disparity
- Test scores