Areas of Expertise


Achievement Gaps
Literature Library


Can the Achievement Gaps be Overcome?

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

    1. Public Policy

What Can Be Done to Narrow the Black-White Test Score Gap?

Two researchers look at what we know about reducing the black-white test score gap. They conclude that there is still much to learn. They also comment on what schools can do and what the society can do to address the issue.

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

Jencks, C., & Phillips, M. (1998). The black-white test score gap: An introduction. In C. Jencks and M. Phillips (Eds.),The black-white test score gap(pp. 1-51). Washington, DC: Brookings Institution. Pp. 43-46

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

Researchers Christopher Jencks and Meredith Phillips say that reducing the black-white test score gap "would do more to move America toward racial equality than any politically plausible alternative" (p. 43).

However, the question is, do we know enough about the achievement gap to effectively reduce it? Jencks and Phillips say that there is much still to be explained. There is much we do not know.

Jencks and Phillips look at the research on efforts to reduce the achievement gap. What has worked? Can we tell why it has worked? The authors do not come up with a confident list of strategies for closing the gap. We still do not know quite how to close the gap, they say. Also, say the authors, it is still not clear that the policies required to institute the remedies would be politically viable.

Still, we are not completely in the dark. Jencks and Phillips provide an overview of the research on school strategies to reduce the achievement gap.

What Can Schools Do to Narrow the Gap?

Jencks and Phillips say that researcher Ronald Ferguson focuses more on what schools can do to reduce the achievement gap than whether they cause it. Based on his survey of the research on addressing the achievement gaps, Ferguson makes five conclusions:

  1. Teachers have lower expectations of black students than white students.
  2. Teacher expectations have a greater impact on black students' performance than on white students' performance.
  3. Teachers expect less of black students because their past performance has been worse than white students.
  4. Teachers perpetuate racial disparities by basing their expectations on past performance.
  5. Simply encouraging teachers to change their expectations of black students is not likely to have much effect.

Rather than focusing on the host of possible social causes of the achievement gap, Jencks and Phillips say that Ferguson focuses more on schools and, in particular, teachers. Based on his research, Ferguson does make several observations regarding strategies that schools might employ to reduce the achievement gap.



Class Size

Based on a review of the research, a substantial number of randomized experiments indicate that reducing class size raises test scores. Reducing class size in the early grades (kindergarten&–3rd grade) by about one-third (from about 23 students to 15 students) increases the math and reading scores for all students, but particularly for black students. And, while these benefits decrease somewhat after 3rd grade, the benefits were still apparent even as late as the7th grade.

Based on historical observation of state efforts to reduce class size, Ferguson finds that while smaller class sizes did increase reading scores, math scores did not rise as much.

It is not clear why experiments would show a rise in all scores while large-scale implementation of smaller class sizes would not have had the same effect. Jencks and Phillips speculate that other influences (such as less-demanding curricula) may have offset increases in math scores.

Teacher Competence

Jencks and Phillips say that Ferguson argues that a teacher's test scores are the best predictor of the teacher's ability to raise students' test scores. While there have been no randomized experiments to support this finding, a large body of non-experimental research indicates that high-scoring teachers are more effective at raising test scores. The studies also find that black students benefit more from these high-scoring teachers.

The answer would seem straightforward: simply use competency exams to screen out low-scoring teachers. Jencks and Phillips say that this approach is problematic for a couple of reasons:

  • Cut-off levels for competency exams would have to be set sufficiently high to screen out poorly performing teachers, and currently few states or school districts require very high scores.
  • Raising the performance levels for competency exams would have the effect of reducing the proportion of qualifying black teachers. Jencks and Phillips point out that there is weak evidence that black students may learn better from black teachers.


Jencks and Phillips say that studies show that school desegregation can raise black students' test scores under some circumstances. One study found that desegregation helped raise Southern black students' scores in the 1970s. A more recent study found that attending a "whiter" school probably helped black students' scores in the early grades, but not in middle school and high school.

Jencks and Phillips say that since teachers in racially mixed schools tend to have higher scoring teachers, the fact that this has no effect is puzzling.

Ability Grouping

Jencks and Phillips say that the evidence on ability grouping (that is, tracking higher and lower performing students into homogeneous classrooms) does not fit the preconceptions of either proponents or opponents. Based on research, there is no significant effectpositive or negativefor the lowest scoring students by placing them in low-ability classes. While there may be some benefit to the highest scoring students by placing them in high-ability classes, these effects are very small.

Another strategygrouping students with similar abilities within a heterogeneous classroomappears to raise the math scores for all students. However, because there have been only a few randomized studies of this kind of strategy, the effects on reading are not known, and it is difficult to be sure what the real effect on math scores is.

Finally, there is evidence that white students are more likely to take academically challenging classes in high school. However, it appears that previous grades, test scores, and socioeconomic status are more important than race for determining this pattern. Why would socioeconomic status matter? Jencks and Phillips say that we do not know. They say that, at any rate, schools could probably do more to push low socioeconomic level students into more-demanding courses.

What Can the U.S. Do to Narrow the Gap?

Jencks and Phillips lay out a number of strategies for addressing the test score gap. They say that some are easy, but expensive, while others are inexpensive, but difficult. Some are politically realistic, while othersregardless of their promiseare politically unrealistic. All strategies have challenges.


Challenges and Issues

Cut class size

The authors argue that this would be an easy, but expensive option.

Screen out incompetent teachers

While screening out low-performing teachers would not be expensive for many school districts, it could be expensive for many urban school districts that have fewer applicants.

School desegregation

This option may arouse racial fears and prejudices. While school desegregation probably raises black students' reading scores, it is not clear whether it causes a decrease in white students' scores. Also, once black families begin to make up more than 20% of a neighborhood, white families become reluctant to move into the neighborhood. This makes establishing stable, desegregated schools difficult.

Group students according to ability

Because grouping students by ability at the elementary level is largely symbolic, eliminating ability grouping would probably not have much effect for either blacks or whites. At the secondary school level it would be ridiculous, say the authors, to eliminate demanding courses. The strategy should not be to eliminate challenging classes for whites, but rather to encourage more black students to take these classes.

Change the structure of educational finance; give more to disadvantaged schools

One option would be to give significantly more educational dollars to disadvantaged schools. Jencks and Phillips say that while giving 10%&–20% more money to disadvantaged schools might be politically realistic, it would be unrealistic to expect voters to approve giving disadvantaged schools 50%&–100% more money than other schools. A system in which black schools get far more money than white schools is politically inconceivable, say the authors.

Change children's preschool experiences

The authors point out that black and white children start school in very different positions: the typical 3-or 4-year-old black child's vocabulary falls below the twentieth percentile of the national distribution. So, waiting until a child is in school to address the gap is too late. Getting black kids into preschool is probably less important than changing their home experiences, say the authors. Black preschoolers are already concentrated in Head Start programs. Also, research indicates that cognitively oriented preschool programs can improve black children's scores. However,few Head Start teachers are trained to teach cognitive skills.

Change young children's home experiences

The authors say that "changing the way parents deal with their children may be the single most important thing we can do to improve children's cognitive skills" (p. 46).Getting parents to change their practices, however, is very difficult. Parents are often suspicious of unsolicited advice. Additionally, the authors say that we cannot simply tell black parents to change without provoking charges of racism. In order to change parents' behaviors, the authors say that we need to focus on all parents and use as many tools as we have at our disposal.

Convince the public that the test score gap is not genetic in origin

The authors say that as long as there is a public feeling that the test score gaps are ultimately hereditary in origin, then teachers, policy makers, and others can use this as an excuse for not adequately wrestling with the problem. Even if training low-achieving children to gain adequate cognitive skills is "hard work," that does not excuse us as a society from the task.

The authors admit that the above strategies do not amount to a prescription for success in addressing the test score gap. However, Jencks and Phillips say that no one has a detailed blueprint for solving the problem. The authors think that by being honest about what we know and what we do not know we can, perhaps, do better than we have done in the past.

Research Design: Authors Christopher Jencks and Meredith Phillips review theories and research on the black-white test score gap. They also comment on effects of the gap and what things might be done to narrow it.


Copyright © 2008 Learning Point Associates. All rights reserved.