“Even the Best Schools Can’t Close the Race Achievement Gap” by Richard Rothstein (Septemebr-October 2004 P&R Issue)

By Richard Rothsten (Click here to view the entire P&R issue)

The achievement gap between poor and middle-class, black and white children is an educational challenge, but we prevent ourselves from solving it because of a commonplace belief that poverty and race can’t “cause” low achievement and that therefore schools must be failing to teach disadvantaged children adequately.

The achievement gap between poor and middle-class, black and white children is an educational challenge, but we prevent ourselves from solving it because of a commonplace belief that poverty and race can’t “cause” low achievement and that therefore schools must be failing to teach disadvantaged children adequately. After all, we see many highly successful students from lower-class backgrounds. Their success seems to prove that social class cannot be what impedes most disadvantaged students.

Yet the success of some lower-class students proves nothing about schools’ power to close the achievement gap. There is a distribution of achievement in every social group. These distributions overlap. While average achievement of low-income students is below average achievement of middle-class students, there are always some middle-class students who achieve below typical low-income levels. Some low-income students achieve above typical middle-class levels. “Demography is not destiny,” but students’ family characteristics are a powerful influence on their relative average achievement, even in the best of schools.

Widely repeated accounts of schools that somehow elicit consistently high achievement from lower-class children almost always turn out, upon examination, to be flawed. In some cases, “schools that beat the odds” are highly selective, enrolling only the most able or most motivated lower-class children. Some are not truly lower-class schools – for example, schools enrolling children who qualify for subsidized lunches because their parents are poorly paid but highly educated. Some schools “succeed” with lower-class children by defining high achievement at such a low level that all students can reach it, despite big gaps that remain at higher levels. And some schools’ successes are statistical flukes – their high test scores last for only one year, in only one grade and in only one subject.

While the idea that “if some children can defy the demographic odds, all can” seems plausible, it reflects a reasoning whose naiveté we easily recognize in other policy areas. In human affairs, where multiple causation is typical, causes are not disproved by exceptions. Tobacco firms once claimed that smoking does not cause cancer because we all know people who smoked without getting cancer. We now consider such reasoning specious. We understand that because no single cause is rigidly deterministic, some people can smoke without harm, but we also understand that, on average, smoking is dangerous. Yet despite such understanding, quite sophisticated people often proclaim that success of some poor children proves that social disadvantage does not cause low achievement.

Social Class and Learning

Partly, our confusion stems from failing to examine the concrete ways that social class actually affects learning. Describing these may help to make their influence more obvious.

Overall, lower-income children are in poorer health, and poor health depresses student achievement, no matter how effective a school may be. Low-income children have poorer vision, partly because of prenatal conditions, partly because, even as toddlers, they watch too much television both at home and in low-quality daycare settings, so their eyes are more poorly trained. Trying to read, their eyes may wander or have difficulty tracking print or focusing. A good part of the over-identification of learning disabilities for lower-class children is probably attributable simply to undiagnosed vision problems for which therapy is available and for which special education placement should be unnecessary.

Lower-class children have poorer oral hygiene, more lead poisoning, more asthma, poorer nutrition, less adequate pediatric care, more exposure to smoke, and a host of other health problems – on average. Because, for example, lower-class children typically have less adequate dental care, they are more likely to have toothaches and resulting discomfort that affects concentration.

Because low-income children are more likely to live in communities where landlords use high-sulfur home heating oil, and where diesel trucks frequently pass en route to industrial and commercial sites, such children are more likely to suffer from asthma, leading to more absences from school and drowsiness (from lying awake wheezing at night) when present. Recent surveys of black children in Chicago and in New York City’s Harlem community found one of every four children suffering from asthma, a rate six times as great as that for all children. Asthma is now the single biggest cause of chronic school absence.

Because primary care physicians are few in low-income communities (the physician to population ratio is less than a third the rate in middle-class communities), disadvantaged children (even those with health insurance) are also more likely to miss school for relatively minor problems, like common ear infections, for which middle-class children are treated promptly. If in attendance, children with earaches have more difficulty paying attention.

Each of these well-documented social class differences in health is likely to have a palpable effect on academic achievement. The influence of each may be small, but combined, the influence of all is probably huge.

The growing unaffordability of adequate housing for low-income families also affects achievement – children whose families have difficulty finding stable housing are more likely to be mobile, and student mobility is an important cause of failing student performance. [See “High Classroom Turnover: How Children Get Left Behind, Poverty & Race, May/June 2002.] A 1994 government report found that 30% of the poorest children had attended at least three different schools by third grade, while only 10% of middle-class children did so. Blacks were more than twice as likely as whites to change schools this much. It is hard to imagine how teachers, no matter how well trained, can be as effective for children who move in and out of their classrooms.

Differences in wealth are also likely to affect achievement, but these are usually overlooked because most analysts focus only on annual family income to indicate disadvantage. This makes it hard to understand why black students, on average, score lower than whites whose family incomes are the same. It is easier to understand this pattern when we recognize that children can have similar family incomes but be of different economic classes: black families with low income in any year are likely to have been poor for longer than white families with similar income in that year. White families are likely to own far more assets that support their children’s achievement than are black families at the same income level, partly because black middle-class parents are more likely to be the first generation in their families to have middle-class status. Although median black family income is now nearly 2/3 of white income, black family assets are still only 12% of whites’. This difference means that, among white and black families with the same middle-class incomes, the whites are more likely to have savings for college. This makes white children’s college aspirations more practical, and therefore more commonplace.

Child Rearing/Personality Traits

Social class differences however, amount to more than these quantifiable differences in health, housing, income and assets. There are powerful social class differences in child rearing habits and personality traits, and these too cause average differences in academic achievement by social class.

Consider how parents of different social classes tend to raise children. Young children of more educated parents are read to more consistently, and are encouraged to read more by themselves when they are older. Most children whose parents have college degrees are read to daily before they begin kindergarten; few children whose parents have only a high school diploma or less benefit from daily reading. White children are more likely than blacks to be read to in pre-kindergarten years.

A five-year-old who enters school recognizing some words and who has turned pages of many stories will be easier to teach than one who has rarely held a book. The latter can be taught, but the child with a stronger home literacy background will typically post higher scores on reading tests than one for whom book reading is unfamiliar — even if both children benefit from high expectations and effective teaching. So, the achievement gap begins.

If a society with such differences wants children, irrespective of social class, to have the same chance to achieve academic goals, it should find ways to help lower-class children enter school having the same familiarity with books as middle-class children have. This requires re-thinking the institutional settings in which we provide early childhood care, beginning in infancy.

Some people acknowledge the impact of such differences but find it hard to accept that good schools should have so difficult a time overcoming them. This would be easier to understand if Americans had a broader international perspective on education. Class backgrounds influence relative achievement everywhere. The inability of schools to overcome the disadvantage of less literate homes is not a peculiar American failure but a universal reality. Turkish immigrant students suffer from an achievement gap in Germany, as do Algerians in France, as do Caribbean, African, Pakistani and Bangladeshi pupils in Great Britain, and as do Okinawans and low-caste Buraku in Japan.

An international survey of 15-year-olds, conducted in 2000, found a strong relationship in almost every nation between parental occupation and student literacy. The gap between literacy of children of the highest status workers (like doctors, professors, lawyers) and the lowest status workers (like waiters and waitresses, taxi drivers, mechanics) was even greater in Germany and in the United Kingdom than it was in the United States. After reviewing these results, a U.S. Department of Education summary concluded that “most participating countries do not differ significantly from the United States in terms of the strength of the relationship between socioeconomic status and literacy in any subject.” Remarkably, the Department published this conclusion at the very time it was guiding a bill through Congress — “No Child Left Behind” — that demanded every school in the nation abolish social class differences in achievement within 12 years.

Urging less educated parents to read to children can’t fully compensate for differences in school readiness. If children see parents read to solve their own problems or for entertainment, children are more likely to want to read themselves. Parents who bring reading material home from work demonstrate by example to children that reading is not a segmented burden but a seamless activity that bridges work and leisure. Parents who read to children but don’t read for themselves send a different message.

How parents read to children is as important as whether they do; more educated parents read aloud differently. When working-class parents read aloud, they are more likely to tell children to pay attention without interruptions or to sound out words or name letters. When they ask children about a story, questions are more likely to be factual, asking for names of objects or memory of events.

Parents who are more literate are more likely to ask questions that are creative, interpretive or connective, like “what do you think will happen next?,” “does that remind you of what we did yesterday?” Middle-class parents are more likely to read aloud, to have fun, to start conversations, as an entree to the world outside. Their children learn that reading is enjoyable and are more motivated to read in school.

There are stark class differences not only in how parents read but in how they converse. Explaining events in the broader world to children, in dinner talk, for example, may have as much of an influence on test scores as early reading itself. Through such conversations, children develop vocabularies and become familiar with contexts for reading in school. Educated parents are more likely to engage in such talk and to begin it with infants and toddlers, conducting pretend conversations long before infants can understand the language. Typically, middle-class parents “ask” infants about their needs, then provide answers for the children (“Are you ready for a nap, now? Yes, you are, aren’t you?”). Instructions are more likely to be given indirectly (“You don’t want to make so much noise, do you?”). Such instruction is really an invitation for a child to work through the reasoning behind an order and to internalize it. Middle-class parents implicitly begin academic instruction for infants with such indirect guidance.

Yet such instruction is quite different from what policymakers nowadays consider “academic” for young children: explicit training in letter and number recognition, letter-sound correspondence, and so on. Such drill in basic skills can be helpful but is unlikely to close the social class gap in learning.

Soon after middle-class children become verbal, parents typically draw them into adult conversations so children can practice expressing their own opinions. Lower-class children are more likely to be expected to be seen and not heard. Inclusion this early in adult conversations develops a sense of entitlement in middle-class children; they feel comfortable addressing adults as equals and without deference. Children who want reasons rather than being willing to accept assertions on adult authority develop intellectual skills upon which later academic success in school will rely. Certainly, some lower-class children have such skills and some middle-class children lack them. But, on average, a sense of entitlement is social class-based.

Parents whose professional occupations entail authority and responsibility typically believe more strongly that they can affect their environments and solve problems. At work, they explore alternatives and negotiate compromises. They naturally express these personality traits at home when they design activities where children figure out solutions for themselves. Even the youngest middle-class children practice traits that make academic success more likely when they negotiate what to wear or to eat. When middle-class parents give orders, they are more likely to explain why the rules are reasonable.

But parents whose jobs entail following orders or doing routine tasks exude a lesser sense of efficacy. Their children are less likely to be encouraged to negotiate clothing or food. Lower-class parents are more likely to instruct children by giving directions without extended discussion. Following orders, after all, is how they themselves behave at work. So their children are also more likely to be fatalistic about obstacles they face, in and out of school.

Middle-class children’s self-assurance is enhanced in after-school activities that sometimes require large fees for enrollment and almost always require parents to have enough free time and resources to provide transportation. Organized sports, music, drama and dance programs build self-confidence (with both trophies and admiring adult spectators) and discipline in middle-class children. Lower-class parents find the fees for such activities more daunting, and transportation may also be more of a problem. In many cases, such organized athletic and artistic activities are not available anywhere in lower-class neighborhoods. So lower-class children’s sports are more informal and less confidence-building, with less opportunity to learn teamwork and self-discipline. For children with greater self-confidence, unfamiliar school challenges can be exciting; such children, who are more likely to be from middle-class homes, are more likely to succeed than those who are less self-confident.

Homework exacerbates academic differences between middle- and working-class children because middle-class parents are more likely to assist with homework. Yet homework would increase the achievement gap even if all parents were able to assist. Parents from different social classes supervise homework differently. Consistent with overall patterns of language use, middle-class parents — particularly those whose own occupational habits require problem solving — are more likely to assist by posing questions that decompose problems and that help children figure out correct answers. Lower-class parents are more likely to guide children with direct instructions. Children from both strata may go to school with completed homework, but middle-class children gain more in intellectual power from the exercise than do lower-class children.

Twenty years ago, Betty Hart and Todd Risley, researchers from the University of Kansas, visited families from different social classes to monitor the conversations between parents and toddlers. Hart and Risley found that, on average, professional parents spoke over 2,000 words per hour to their children, working-class parents spoke about 1,300, and welfare mothers spoke about 600. So, by age three, children of professionals had vocabularies that were nearly 50% greater than those of working-class children and twice as large as those of welfare children.

Deficits like these cannot be made up by schools alone, no matter how high the teachers’ expectations. For all children to achieve the same goals, the less advantaged would have to enter school with verbal fluency similar to the fluency of middle-class children.

The Kansas researchers also tracked how often parents verbally encouraged children’s behavior, and how often parents reprimanded their children. Toddlers of professionals got an average of six encouragements per reprimand. Working-class children had two. For welfare children, the ratio was reversed, an average of one encouragement for two reprimands. Children whose initiative was encouraged from a very early age are probably more likely, on average, to take responsibility for their own learning.

Social class differences in role modeling also make an achievement gap almost inevitable. Not surprisingly, middle-class professional parents tend to associate with, and be friends with, similarly educated professionals. Working-class parents have fewer professional friends. If parents and their friends perform jobs requiring little academic skill, their children’s images of their own futures are influenced. On average, these children must struggle harder to motivate themselves to achieve than children who assume that, as in their parents’ social circle, the only roles are doctor, lawyer, teacher, social worker, manager, administrator or businessperson.

Even disadvantaged children now usually say they plan to attend college. College has become such a broad rhetorical goal that black eighth graders tell surveyors they expect to earn college degrees as often as white eighth graders respond in this way. But despite these intentions to pursue education, fewer black than white eighth graders actually graduate from high school four years later, fewer eventually enroll in college the year after high school graduation, and even fewer persist to get bachelor’s degrees.

A bigger reason than affordability is that while disadvantaged students say they plan on college, they don’t feel as much parental, community or peer pressure to take the courses or to get the grades to qualify and to study hard to become more attractive to college admission officers. Lower-class parents say they expect children to perform well, but are less likely to enforce these expectations, for example with rewards or punishments for report card grades. Teachers and counselors can stress doing well in school to lower-class children, but such lessons compete with children’s own self-images, formed early in life and reinforced daily at home.

Culture and Expectations

Partly, there may be a black community culture of underachievement that helps to explain why even middle-class black children often don’t do as well in school as white children from seemingly similar socioeconomic backgrounds. Middle-class black students don’t study as hard as white middle-class students, and blacks are more disruptive in class than whites from similar income strata. This culture of underachievement is easier to understand than to cure. Throughout American history, many black students who excelled in school were not rewarded in the labor market for that effort. Many black college graduates could only find work as servants, as Pullman car porters or, in white-collar fields, as assistants to less qualified whites. Many Americans believe that these practices have disappeared and that blacks and whites with similar test scores now have similar earnings and occupational status. But labor market discrimination, even for blacks whose test scores are comparable to whites, continues to play an important role. Especially for black males with high school educations, discrimination continues to be a big factor.

Evidence for this comes from the continued success of employment discrimination cases — for example, a prominent 1996 case in which Texaco settled for a payment of $176 million to black employees after taped conversations of executives revealed pervasive racist attitudes, presumably not restricted to executives of this corporation. Other evidence comes from studies finding that black workers with darker complexions have less labor market success than those with lighter complexions but identical education, age and criminal records. Still more evidence comes from studies in which blacks and whites with similar qualifications are sent to apply for job vacancies; the whites are typically more successful than the blacks. One recent study trained young, well-groomed and articulate black and white college graduates to pose as high school graduates with otherwise identical qualifications except that some reported convictions for drug possession. When these youths submitted applications for entry level jobs, the applications of whites with criminal records got positive responses more often than the applications of blacks with no criminal records.

So the expectation of black students that their academic efforts will be less rewarded than efforts of their white peers is rational for the majority of black students who do not expect to complete college. Some will reduce their academic effort as a result. We can say that they should not do so and, instead, should redouble their efforts in response to the greater obstacles they face. But as long as racial discrimination persists, the average achievement of black students will be lower than the average achievement of whites, simply because many blacks (especially males) who see that academic effort has less of a payoff will respond rationally by reducing their effort.

Helpful Policies

If we properly identify the actual social class characteristics that produce differences in average achievement, we should be able to design policies that narrow the achievement gap. Certainly, improvement of instructional practices is among these, but alone, a focus on school reform is bound to be frustrating and ultimately unsuccessful. To work, school improvement must combine with policies that narrow the social and economic differences among children. Where these differences cannot easily be narrowed, school should be redefined to cover more of the early childhood, after-school and summer times when the disparate influences of families and communities are most powerful.

Because the gap is already huge at age three, the most important new investment should probably be in early childhood programs. Pre-kindergarten classes for four-year-olds are needed, but barely begin to address the problem. The quality of early childhood programs is as important as the existence of programs themselves. Too many low-income children are parked before television sets in low-quality daycare settings. To narrow the gap, care for infants and toddlers should be provided by adults who can create the kind of intellectual environment that is typically experienced by middle-class infants and toddlers. This requires professional care-givers and low child:adult ratios.

After-school and summer experiences for lower-class children, similar to programs middle-class children take for granted, would also likely be needed to narrow the gap. This does not mean remedial programs where lower-class children get added drill in math and reading. Certainly, remediation should be part of an adequate after-school and summer program, but only a part. The advantage that middle-class children gain after school and in summer likely comes from self-confidence they acquire and awareness they develop of the world outside, from organized athletics, dance, drama, museum visits, recreational reading and other activities that develop inquisitiveness, creativity, self-discipline and organizational skills. After-school and summer programs can be expected to have a chance to narrow the achievement gap only by attempting to duplicate such experiences.

Provision of health care services to lower-class children and their families is also required to narrow the achievement gap. Some health care services are relatively inexpensive, like school vision and dental clinics that cost less than schools typically spend on many less effective reforms. A full array of health services will cost more, but likely can’t be avoided if there is a true intent to raise the achievement of lower-class children.

Policies to make stable housing affordable to low-income working families with children and policies to support the earnings of such families should also be thought of as educational policies — they can have a big impact on student achievement, irrespective of school quality.

The association of social and economic disadvantage with an achievement gap has long been well known to educators. Most, however, have avoided the obvious implication: To improve lower-class children’s learning, amelioration of the social and economic conditions of their lives is also needed. Calling attention to this link is not to make excuses for poor school performance. It is, rather, to be honest about the social support schools require if they are to fulfill the public’s expectation that the achievement gap disappear. Only if school improvement proceeds simultaneously with social and economic reform can this expectation be fulfilled.