The Black-White Test Score Gap: Why It Persists and What Can Be Done
African Americans score lower than European Americans on vocabulary, reading, and math tests, as well as on tests that claim to measure scholastic aptitude and intelligence. The gap appears before children enter kindergarten and it persists into adulthood. It has narrowed since 1970, but the typical American black still scores below 75 percent of American whites on almost every standardized test. This statistic does not imply, of course, that all blacks score below all whites. There is a lot of overlap between the two groups. Nonetheless, the test score gap is large enough to have significant social and economic consequences.
Closing the black-white test score gap would probably do more to promote racial equality in the United States than any other strategy now under serious discussion. Judging by the currently available statistical evidence, eliminating the test score gap would sharply increase black college graduation rates, making them nearly equal to white rates. Such a change would also allow selective colleges to phase out racial preferences in admission, which have long been a flashpoint for racial conflict. Eliminating the test score gap would also reduce racial disparities in men’s earnings and would probably eliminate the racial disparities in women’s earnings.
Narrowing the test score gap would require continuous effort by both blacks and whites, and it would probably take more than one generation. But we think it can be done. This conviction rests on three facts. First, black-white differences in academic achievement have narrowed since 1970. The National Assessment of Educational Progress (NAEP) data on 17-year-olds show that the reading gap narrowed more than two-fifths between 1971 and 1994. The math gap has also narrowed, though not as much. Five major national surveys of high school seniors conducted since 1965 show the same trend. So do surveys of younger students. The gap narrowed because black children’s scores rose, not because white children’s scores fell.
Second, even IQ scores clearly respond to changes in the environment. IQ scores, for example, have risen dramatically throughout the world since the 1930s. In America, 82 percent of those who took the Stanford-Binet test in 1978 scored above the 1932 average for individuals of the same age. The average black did about as well on the Stanford-Binet test in 1978 as the average white did in 1932.
Third, when black or mixed-race children are raised in white rather than black homes, their pre-adolescent test scores rise dramatically. These adoptees’ scores seem to fall in adolescence, but this could easily be because their social and cultural environment comes to resemble that of other black teenagers.
Explaining the Gap
Traditional explanations for the black-white test score gap have not stood up well to the test of time. During the 1960s, most liberals blamed the gap on some combination of black poverty, racial segregation, and inadequate funding of black schools. Since then, the number of affluent black families has grown dramatically, but their children’s test scores still lag far behind those of white children from equally affluent families. School desegregation may have played some role in reducing the black-white test score gap in the South, but school desegregation also seems to have costs for blacks, and when we compare initially similar students in today’s schools, those who attend desegregated schools learn only slightly more than those in segregated schools.
Recent evidence suggests that disparities in school resources do affect achievement, but resource disparities between black and white children have shrunk steadily over time. The average black child now attends school in a district that spends as much per pupil as the average white child’s district. Black children’s schools also have about the same number of teachers per pupil as white schools. Predominantly white schools seem to attract more skilled teachers than black schools, but while black students who attend predominantly white schools probably benefit from having better teachers, this advantage seems to be offset by the social costs of being in an overwhelmingly white environment. In any event, schools cannot be the main reason for the black-white test score gap, because it appears before children enter school and persists even when black and white children attend the same schools. If schools play an important role in perpetuating the gap, either desegregated schools must be treating black and white children very differently or else black and white children must react very differently to the same treatment.
The three most common “conservative” explanations for the black-white gap-genes, the culture of poverty, and single motherhood-are also hard to reconcile with the available evidence. There is no direct genetic evidence for or against the theory that the black-white gap is innate, because we have not yet identified the genes that affect skills like reading, math, and abstract reasoning. Studies of mixed-race children and black children adopted by white parents suggest, however, that racial differences in test performance are largely if not entirely environmental in origin.
Cultural differences associated with chronic poverty may account for some of the black-white test score gap, but they cannot be the main explanation, since the gap persists among affluent children. And while children raised by single mothers score lower on most standardized tests than children raised by married couples, this difference almost disappears once we take account of the fact that women who become single mothers come from less advantaged families, have lower test scores, and complete less schooling than women with husbands.
We suspect that successful new explanations for the test score gap will differ from their predecessors in several ways.
First, instead of emphasizing the kinds of racial differences that economists and sociologists usually study (parents’ economic resources, parents’ position in the occupational hierarchy, parents’ exposure to formal education, and parents’ living arrangements), successful theories will take more account of the factors that psychologists have traditionally emphasized (the way family members interact with one another and with the outside world, for example). A good explanation of why white five-year-olds have bigger vocabularies than black five-year-olds is likely to focus on how much the parents talk to their children, how they deal with their children’s questions, and how they react when their children either learn or fail to learn something, not on how much money the parents have.
Second, instead of looking mainly for resource differences between predominantly black and predominantly white schools, successful theories will probably have to look more carefully at the way black and white children respond to the same classroom experiences, such as being in a smaller classroom, having a more competent teacher, having a teacher of their own race, or having a teacher with high expectations for those who perform below the norm for their age group.
Successful theories will therefore have to pay more attention to psychological and cultural influences, which are much harder to measure than income, education, and living arrangements. Collecting accurate data on black and white parents’ habits, values, behavior, and ideas is not easy, and it would take time. It might well require an investment of time and effort comparable to the effort that went into developing cognitive tests during the first half of the 20th century. But without such work, we are in constant danger of seeing black-white differences as an inevitable byproduct of people’s genes or of “cultural” factors that nobody can change.
Our argument that reducing the black-white test score gap would do more to move America toward racial equality than any politically plausible alternative rests on two problematic premises: that policies aimed at reducing the test score gap are in fact politically feasible and that such policies can in fact reduce the gap.
Public support for almost any policy depends partly on whether the beneficiaries are perceived as deserving or undeserving. One obvious advantage of programs directed at children is that hardly anyone blames first graders’ ignorance on their lack of motivation. First graders of every race seem eager to please. Both black and white adults often think that older black children lack academic motivation, but most adults still blame this on the children s parents or schools, not on the children themselves. That was why Lyndon Johnson emphasized helping children in his original war on poverty.
Policies that reduce the black-white gap will not, of course, be politically popular if they improve black children’s test scores at white children’s expense. Both school desegregation and eliminating academically selective classes at desegregated schools have aroused strong white resistance because of the perceived cost to white children. But these policies would not do blacks much good even if whites were willing to adopt them. The most promising school-related strategies for reducing the black-white test score gap seem to involve changes like reducing class size, setting minimum standards of academic competency for teachers, and raising teachers’ expectations for low-performing students. All these changes would benefit both blacks and whites, but all appear to be especially beneficial for blacks.
An experiment carried out by the state of Tennessee during 1985-89 found, for example, that cutting class size in the early grades raised both black and white children’s test scores and that these gains were sustained even after children moved on to larger classes. The experiment also found that gains were much larger for blacks than for whites. Historical evidence also seems to support the hypothesis that the black-white test score gap falls when class size falls. When low birth rates reduced school enrollment in the 1970s, the teacher-pupil ratio rose and classes shrank. Independent analyses by Ronald Ferguson and David Grissmer suggest that this change in class size was followed by a marked decline in the black-white test score gap.
Although measuring teachers’ competence is harder than counting the number of children in a classroom, teachers’ test scores show a stronger association with how much students learn than any other widely used measure. Teacher competency exams are thus likely to boost children’s performance. Since the teachers who fail such tests are concentrated in black schools, such exams would probably prove especially beneficial to black students, although this benefit may be partially offset by the fact that the teachers who fail such tests are also disproportionately black.
Ferguson’s review of the literature on teachers’ expectations concludes that teachers do have lower expectations for blacks than for whites, but that this is largely because blacks enter school with weaker cognitive skills than whites and learn a bit less after entering. But Ferguson also finds some evidence that low teacher expectations have a more negative effect on black children than on their white classmates.
Research also suggests that black-white differences in parenting practices contribute to the test score gap. Improving parenting skills may therefore be as important as improving schools. The puzzle is how to proceed. Like teachers, parents are usually suspicious of unsolicited advice about how to deal with their children. But once parents become convinced that a particular practice really helps their children, many adopt it. As a practical political matter, whites cannot tell black parents to change their parenting practices without provoking charges of ethnocentrism, racism, and much else. But blacks are hardly the only parents who need help. We should be promoting better parenting practices for all parents in every way we can, including television, which reaches both blacks and whites.
Finally, conservatives who want to improve academic achievement should stop emphasizing the relationship between heredity and achievement and play up the importance of another conservative virtue—namely, hard work. Americans seem to be unusually likely to attribute academic failure to low ability rather than inadequate effort. When Harold Stevenson and James Stigler asked American, Japanese, and Taiwanese parents and teachers why some children did better than others in school, the Americans were more likely to emphasize ability whereas the Japanese and Taiwanese were more likely to emphasize effort. This difference does not seem to reflect a difference in fundamental beliefs about causation. Children all over the world recognize that both ability and effort affect achievement, and the same is probably true for their parents as well. But attributing failure to inadequate effort implies that if you work harder, you will learn more. Attributing it to ability serves as an excuse for doing nothing.
Americans’ emphasis on innate ability is likely to have especially negative consequences for African Americans, whose anxiety about racial stereotypes and intellectual competence can even depress their performance on standardized tests. Claude Steele and Joshua Aronson, for example, have shown that black Stanford undergraduates, unlike their white classmates, do measurably worse on tests when they are asked to record their race before taking the test or told that the test measures intellectual ability.
Time for Renewed Attention
Psychologists, sociologists, and educational researchers have devoted far less attention to the black-white test score gap over the past quarter-century than they should have. Cowed by the hostile reaction to Daniel Patrick Moynihan’s 1965 report on the status of the black family and to Arthur Jensen’s 1969 article arguing that racial differences in test performance were likely to be partly innate, most social scientists have chosen safer topics and hoped the problem would go away. We can do better.