Comment

The paper by Michael T. Nettles, Catherine M. Millett, and Douglas D. Ready makes an important contribution to the ongoing conversation about the racial gap in standardized test scores. Among other findings, the paper documents that blacks on average score lower on the SAT, American College Test (ACT), and National Assessment of Educational Progress (NAEP) than whites and that the disparities exist at all income and socioeconomic status (SES) levels. While I have no quarrel with the paper's findings, as a preliminary matter I do wonder whether a mistake is being made by focusing on income instead of wealth. Nettles, Millett, and Ready's research, like all the research in this field, is based on income comparisons. But as the work of Melvin Oliver, Thomas Shapiro, and Dalton Conley has made clear, income disparities can mask even greater wealth disparities. The real point may be that reductions in income disparities—as more African American families enter the middle class—mask large and continuing disparities in wealth. In other words, black families, on average, do not have the same household wealth as white families with comparable incomes (factoring in savings, home value, and so on). Because wealth, even more than income, allows families to purchase important educational goods at every stage of a child's development, research on racial gaps needs to begin to take account of differences in wealth.

Nettles, Millett, and Ready's research also raises a more fundamental question: Is it a mistake to speak of the racial gap in standardized test scores and to overlook the greater consequences of the gap when disadvantaged kids are the ones on the losing side?

I would suggest that more than one racial gap is at work and that the racial gap has more dire consequences for kids at the bottom end of the social class hierarchy. My position is based in part on personal experience, because I have been on the wrong side of the racialachievement gap in both contexts. In seventh grade, I went to Hunter College High School in New York City, which is a school for high achievers and is, at least in a relative sense, a high socioeconomic status school. The racial test score gap was obvious, and as a result of it, most of the black and Latino students were required to go to summer school before seventh grade started. Though I left Hunter after one year, I stayed in touch with some of the minority students I entered with. Upon graduation from Hunter, they went off en masse to college and university.

After leaving Hunter I moved to Atlanta and went to Franklin Roosevelt High School—a mostly black, low socioeconomic status public school. The differences between the educational opportunities available at Hunter and Roosevelt were stark. I loved my school, my classmates, and my teachers, but I am not proud of the reality that fewer than ten of the kids in my graduating class went on to finish college.

Though I have not seen the research, I would imagine, based on the work of Nettles and others, that the African American students at Hunter and Roosevelt both performed at a lower rate than white students of comparable socioeconomic status. And I guess that this reflects a black-white test score gap. But the consequences of the test score gap for the students at Hunter and the students at Roosevelt are dramatically different. And I personally believe that the more pressing issue today is the Roosevelts of the inner cities—the schools where only a few kids are going to college and more are dropping out, getting pregnant, or, in many cases, getting locked up.

The importance of the question of educating disadvantaged students is reflected in some of the papers presented at the Brookings conference. David P. Baker's excellent paper discusses how the achievement gap between advantaged students and disadvantaged students in this country is much greater than in many others around...