Areas of Expertise


Achievement Gaps
Literature Library


Can the Achievement Gaps be Overcome?

  1. How to Close the Achievement Gaps: Research and Policy

    1. Social and Cultural Factors

Don’t Treat the Symptom: Cure the Disease

Claude M. Steele argues that efforts to remedy the black-white test score gap need to aim at the root causes, and not just at the tests themselves.

This reports some of the ideas and findings from the following source:

Steele, C. M. (1998, November/December). The black-white test score gap.The American Prospect,9. Retrieved August 8, 2002, from

To see other reports that originated from this same citation, go to the bibliography.

Steele criticizes the research of Christopher Jencks and Meredith Phillips on the test score gap for attributing too much importance to test scores as a means of righting racial inequality.

  • Tests are poor predictors. First of all, Steele shows, standardized tests do not accurately predict students’ future achievement. The SAT, for example, measures only about 18% of the factors that influence college freshmen’s grades. So Steele doubts that the skills measured by such tests have much of a direct influence on things such as the earning gap between black and white workers. The real causes of this gap lie elsewhere, and the test score disparity is only a symptom, not a cause.
  • Negative stereotypes are one major cause. Many scholars have suggested that blacks suffer from historically rooted stereotypes that negatively affect their achievement. Blacks come to doubt their own abilities, to feel intimidated by situations where they feel they may underperform whites. Such stereotypes foster mistrust and alienation that, Steele believes, help account for blacks’ poor test performance.
  • Don’t overlook the social context. While Steele does not deny the importance of public policy to remedy the test score gap, he urges that such policies be directed at the root causes of the gap. "Focusing on [the test score gap] now could seem like treating a symptom while ignoring the disease," he says. "Shouldn't we focus first on righting the structures of racial inequity and let test scores take care of themselves?" (Claude M. Steele section, ¶ 1). To this end, he recommends providing greater resources to parents, homes, and communities, in addition to schools.

Jencks and Phillips Respond

Jencks and Phillips admit that it is not the test scores that matter but rather the skills. These skills, they insist, are not just a product of cultural or economic context, however. While black students who grow up in more affluent families and attend predominately white schools do tend to perform at a level closer to whites, conclusive evidence for this effect is lacking.

For example, a comparison of five- and six-year-olds whose parents had the same average annual income showed that the black-white gap was reduced by only a fifth when taking into account the income factor. Hence, better scores are not merely a proxy for higher socioeconomic status.

On the other hand, the skills such tests measure are actually good predictors of the black-white gap in job earnings, Jencks and Phillips claim. This is important, they say, because if skills and not cultural context are the most important factors in the earnings gap, then skills are relatively easy to improve. If, however, it is the cultural context that is the most important factor, then improving blacks’ test scores and earnings will be much more difficult. It is significantly easier to build students’ and workers’ skills than it is to alter centuries-old patterns of discrimination and stereotyping.

Research Design: This article consists of a series of responses from prominent experts to Christopher Jencks and Meredith Phillips’ article America’s Next Achievement Test, The American Prospect, 9, September-October 1998.


Copyright © 2008 Learning Point Associates. All rights reserved.