Reframing the Achievement Gap

This article was written by Robert Evans, Ed.D.

The achievement gap, the persistent disparity between the performance of African American and Hispanic students and that of white and Asian American students, is perhaps the most stubborn, perplexing issue confronting American schools.  Closing the gap is widely seen as important not just to our educational system but ultimately to our economy, our social stability, and our moral health as a nation.

The conventional wisdom has it that the achievement gap is a school problem. This belief is invitingly simple, allowing a narrow focus on schools that suits the current passion for accountability through testing. But it is fatally shortsighted. It misunderstands and mistreats schools and, more important, black and Hispanic students.

When we set the achievement gap and schooling itself in the broader context of how children grow up, it becomes clear that the issue far transcends the classroom.  Its roots lie well beyond the reach of schools, and so the underlying dilemma will require much, much more than school-based strategies and programs.  Educators must do all they can to pursue promising approaches for reducing the gap.  But holding them, almost alone, accountable for closing it is a doomed strategy that can only disserve our most vulnerable children.


Two facts about the gap are clear: its origins lie neither in students nor in schools. Skin color, ethnic status, poverty—none of these, by themselves, determine a student’s performance. There are black and Hispanic students everywhere, including those whose families are poor, who succeed impressively. Nor, for their part, do schools create the disparity. Substantial numbers of black and Hispanic students begin kindergarten well behind other students in academic readiness. Both sets of facts are equally important, but most “achievement gap critics” emphasize the former and minimize the latter. They blame educators for failing to eliminate the gap and indeed for enlarging it. Reduced to its core, their logic is: all children are created equal, but all children are not performing equally in school; the gap typically worsens as children advance through the grades; the fault must therefore be the schools’, so the solution must lie in school; the necessary knowledge and tools are available, and schools must be pressed to apply them.

This critique actually targets two achievement gaps, urban and suburban, and operates at two levels, structure and practice. The most obvious issues are found in the schools with the largest minority enrollments. Predominantly urban, many of these schools are, by almost any measure, less congenial to learning than others, because, proportionally, they have more teachers who are inexperienced, poorly trained, and uncertified; more textbooks that are outdated; fewer computers; larger class sizes; and buildings that are in worse repair and more marked by violence. The Education Trust has detailed egregious practices by incompetent teachers in these schools.1Other advocacy groups have concentrated on inequities in the resources, materials, and physical conditions of high-minority, low-performing schools.[i] The collective case is indisputable: the students who most need our best teachers and best learning environments rarely have access to either.

At bottom, these problems really involve the structural characteristics of urban schools, and they reflect economic and political realities that are mostly beyond the power of those schools to remedy.  For example, high-minority, low-performing schools hire fewer top quality teachers than others and have greater turnover—not because they want to but because they can’t attract and retain better candidates.  It is technically accurate, but largely pointless, to blame the substandard practice of underqualified teachers for the poor outcomes of their students; these teachers shouldn’t be teaching in the first place.

However, the achievement gap has a second front. It persists among middle-class African American and Hispanic students in suburban communities, even among many who do not live in poverty, whose parents are professionals, and who attend schools that are well staffed and have ample resources. Yet here, too, the disparity in achievement is distressing. To many critics, this is unmistakable proof that the gap stems from the way students are treated and taught in school. Teachers, they argue, are too often racist, even if subtly and unconsciously, and too often parochial in their pedagogy. Teachers both expect too little of black and Hispanic students and give them too little outreach and support. Their methods fail to address individual differences and cultural and other factors that affect the learning styles, motivation, and behavior of Hispanic and black students (and, for that matter, of students with special needs).

What is needed, say many critics, is a primary focus on equity that makes the school a welcoming, comforting, safe environment for Hispanic and African American students and their parents, and that reaches out to them in ways that engage them in the school and in the process of learning.  This requires, among other things, that teachers build personal relationships with students and that they communicate a consistent message that emphasizes high expectations for achievement and a commitment to help students reach these goals.  It calls for a curriculum that fully integrates key elements of black and Hispanic life and culture.

And in pedagogy it demands differentiated instruction that emphasizes variety and flexibility in methods and goals.  Teachers must stop assuming that a whole class should learn the same material in the same way and reach the same benchmarks at the same time.  Instead, they must foster mastery of key concepts more than particular facts.  Individualization and adaptation must become the cornerstones of instruction.  Teachers must emphasize options, such as multiple texts and supplementary materials, group projects, assignments pitched at different levels of complexity, and assessments that measure student performance with respect to a range of abilities.[ii]

Most of this case is compelling.  Conditions in the under-resourced, poorly-staffed schools attended by most African American and Hispanic students are fundamentally unfair—“savage inequalities,” Jonathon Kozol has called them—and cry out for correction.[iii] Many low income black and Hispanic parents do need extra encouragement to engage themselves in their children’s school work.  Since all learning is personal, strong teacher-pupil relationships that combine high expectations with strong support are essential.  And the critique of traditional, one-size-fits-all curriculum and instruction has obvious merit.  But the core diagnosis— the school is the primary cause of and must be the primary cure for the achievement gap—is deeply flawed.  It exaggerates the influence of schooling and underestimates the impact of the major contributors to the achievement gap, which occur outside of school.


Achievement gap critics assume that schooling can exert a powerful, transformative impact on large numbers of students.  The truth, alas, is that schooling has much less leverage on children than commonly thought.  Not just on Hispanic and black students but on all students.  In our national debate about school accountability we have come to equate “education” with “schooling.”  This is a serious error.  Formal schooling plays an important role in a child’s overall life and learning, but not the most influential role.  In fact, as the psychologist Mihaly Csikszentmihalyi points out, whether children learn “does not depend primarily on what happens in school, but on the experiences, habits, values, and ideas they acquire from the environment in which they live.”[iv]

Schooling is a part of that environment, but only a part.  By the time seniors graduate at age 18 they have spent only ten percent of their lives, including none of their formative first years, in school.  For most children, the nature of their schooling is not nearly as significant as the nature of the parenting, they receive, or of their socioeconomic status, or, for that matter, of the media culture that surrounds them.  Increasingly, the 90 percent of their lives that students spend outside of school affects their “experiences, habits, values, and ideas” in ways that undermine academic achievement.[v]

Viewed in the context of a child’s whole growing up, schooling is a relatively “weak treatment,” responsible, in most cases, for no more than 25 percent of the child’s total outcome, if that.[vi] Of course, schooling has a deep impact on some students and can be life-saving for others.  But not for most.  For example, nearly 90 percent of the difference among students’ math scores on some tests can be predicted without knowing anything about their schools; one only needs to know the number of parents in the home, the level of the parents’ education, the type of community in which the family lives, and the state’s poverty rate.[vii] As Howard Gardner has observed, we can accurately project a child’s chances of completing college and her eventual income by knowing only her ZIP code.[viii] And all of the experiential factors that exert an enduring effect on a child’s I.Q. appear to stem from the family environment—the presence of two parents, the cognitive stimulation and nurturance they provide, their income and education levels, and so on—not one stems from school.[ix]

Evidence of this kind about the limits of schooling’s leverage is not new.  Indeed, it dates back 40 years.  In the 1960s, James Coleman offered disconcerting evidence that schooling had relatively little effect on the ultimate equality of students’ life outcomes, that parents’ involvement in their children’s lives affected achievement and eventual success much more powerfully.[x] These findings caused considerable controversy and led to a thorough reanalysis of Coleman’s data by Christopher Jencks and a team at Harvard.  Their conclusion was even sharper.  The data, they said, confirmed that the school’s influence was “marginal,” that children were indeed affected far more by “what happens at home [and also perhaps] by what happens on the streets and by what they see on television.”  A school’s output, they found, depends almost entirely on “the characteristics of the entering children.  Everything else—the school budget, its policies, the characteristics of the teachers—is either secondary or completely irrelevant.”[xi]

Strong stuff, and even more disturbing than Coleman’s assessment.  Analyses such as these helped provoke a powerful counterattack.  This was just prior to the heyday of education’s so-called romantic critics—John Holt, Charles Silberman, George Dennison, and others, joined later by Kozol—who began exposing the shortcomings of urban public schools and of traditional approaches to teaching.  Perhaps the strongest opposition came from Ron Edmonds, who founded the effective schools movement.  He insisted that schooling could be powerfully influential in students’ lives and that the failures of poor and minority students were really failures of educators.  “We can,” he declared, “whenever and wherever we choose, successfully teach all children whose schooling is of interest to us. We already know more than we need to do that.”[xii] This claim has been taken as a beacon by a generation of school reformers, in part because later research into what came to be called efficacy suggested that, when a teacher has high expectations for students and provides high levels of support, learning typically improves.  In the end, these arguments trounced those of Coleman and Jencks and became the dominant orthodoxy in education.  Several decades later, they lie at the core of President George W. Bush’s No Child Left Behind legislation, which assigns almost full responsibility for closing the achievement gap to our schools.

Though the victory of Edmonds and others over Coleman and Jencks was sweeping, it hasn’t produced anything like the success its advocates envisioned.  Schools have not been able to overcome the effects of social and economic changes that have weakened the foundations of family and community life and thus of childrearing in America.  And in recent decades, evidence has continued to underscore the limitations of schooling’s effect on students.  These findings have been demonstrated in research that has been ignored by achievement gap critics and school improvement advocates alike.

Nearly 15 years ago, Paul Barton and Richard Coley summarized a whole range of studies showing that student achievement depends heavily on whether there are two parents in the home, whether children feel secure there, whether the household is intellectually stimulating, whether parents model and foster self-regulation and perseverance, and whether they limit television watching, monitor homework, and make certain the child attends school regularly.[xiii] And research into the achievement gap itself underscores the importance of non-school influences on children’s school performance.


More and more data indicate that the gap begins well before kindergarten, that it is a universal phenomenon, not a purely American one, and that it is compounded outside of school as children move through the grades.  What follows is a partial list of relevant non-school factors identified in recent research.  Most are well known but rarely accorded the weight they deserve:

• there are substantial inequalities in children’s school readiness right from the beginning.  Low-income kindergarteners (a group that includes large numbers of black and Hispanic children) typically start school at least a full year behind others in reading, and with a vocabulary of 5,000 words (vs. 20,000 for their middle-class peers), in part because many don’t attend pre-school, in part because low-income parents speak, on average, much less to their children than do parents who are professionals (600 words per hour vs. 2,100), and in part because low-income parents tend to read to their children much less than other parents do;[xiv]

• similar differences in school readiness occur worldwide.  As Richard Rothstein notes, class background influences student achievement around the world, and nowhere are schools able to overcome this influence.  In virtually every country studied, there is a strong correlation between students’ literacy and the number of books in their homes.  In some countries the literacy gap between the children of high- and low-status workers is even larger than it is in the U.S.[xv]

• much of the achievement gap appears to occur not in school but over the summer.  As Gerald Bracey has argued, “summer loss” merits much more attention than it has received.  Some researchers have reported that when elementary students’ progress is measured between September and May, most are found to be advancing at a roughly similar rate, but that when they are tested the following September, those from low-income families have regressed, while their middle-class and upper middle-class peers have continued to make progress.  By the start of middle school, their cumulative summer loss can amount to more than two full years in verbal achievement and nearly as much in math, even if their teachers help them advance at the same pace as others during the school year;[xvi]

• black and Hispanic students change schools much more often than others.  Between first and third grade 27% of black and 25% of Hispanic students change school three or more times (vs. 13% of whites); in many urban classrooms the turnover rate of students approaches 50% per year, which significantly affects the learning of students who move and complicates the efforts of teachers to maintain continuity of instruction for those who don’t;[xvii]

• black and Hispanic students watch much more TV than other students.  Indeed, their viewing tends to be well above the levels that correlate with lower school performance, especially in reading.  More than 40% of black students and more than 20% of Hispanic students watch more than six hours per day, while just 13% of white students watch as much.[xviii]

• black students have lower levels of parent availability than white and Asian American students.  Only 38% live with two parents (vs. 75% of white students).  Many black students are living with single mothers and in poverty, a combination that puts children from any ethnic group at risk for, among other problems, poor attendance and achievement and behavior problems; [xix]

• black children have significantly higher rates of low birth weight and lead poisoning than white and Asian American children.  Both conditions can seriously impair cognitive and academic functioning;[xx]

• black and Hispanic adolescent peer cultures in some schools appear to exert a negative influence on performance.  Specifically, these youth cultures can foster the attitude that using Standard English, being smart, and working hard constitute a kind of sellout—acting “too white.”  Where this attitude prevails, it can discourage minority students, even those from professional, middle-class homes with strong parental support in good suburban schools, from enrolling in challenging courses and investing in their work.[xxi]

This last nonschool factor brings us back to the question of racism and its chicken-and-egg intricacies. Do minority students perform and behave as they do because of the ways they are treated and taught in school or are they treated and taught as they are because of the ways they perform and behave? The answer seems to be both. The concept of “stereotype threat” advanced by Claude Steele is relevant to the first view. This is the fear of “being viewed through the lens of a negative stereotype, or . . . of doing something that would . . . confirm that stereotype.”22 In a series of ingenious experiments, Steele has shown that we are all vulnerable to this threat, and he has traced ways that it can interfere with academic performance. He argues that it leads black students to protect against the pain and humiliation of being seen as less capable by not investing themselves and their energies in schooling and its norms, a reaction he calls “disidentification.” Teachers as individuals and schools as institutions, Steele asserts, perpetuate stereotype threat and disidentification, even without intending to, which, of course, contributes to the achievement gap.

While Steele’s research is powerful, vulnerability to stereotype threat and disidentification are universal characteristics.  They are not created by teachers and schools. And charges by other critics that racism is a primary explanation for the achievement gap are not persuasive. Racism surely persists among white (and other) teachers, but it is vastly diminished since the days when schools were formally segregated. Subtle and unconscious racism still constitutes racism, but it is a long way from blatant devaluation. Moreover, the late John Ogbu and other researchers have pointed out that teachers’ expectations for students appear to be shaped at least as much by students’ actual behavior as by the teachers’ prejudice. Teachers who expect comparatively lower academic engagement by black and Hispanic students may well be reflecting, as Ogbu found, their real-life experiences, not only their preexisting racist expectations or behavior they themselves have caused.[xxii] Without denying the existence of teacher prejudice or its negative impact, we need to realize that, as the preceding list indicates, with respect to the achievement gap, it is not just black and Hispanic students who face real obstacles; so, too, do their teachers.


Most achievement gap critics I meet believe that schools can be instruments of social change. They discount decades of evidence that schools reflect society much more than they shape it. Some are educators I greatly respect who have committed their careers to equity and social justice for all students, and they reject as a cop-out any hesitation about schooling’s potential to reduce the achievement gap. Even when they acknowledge the impact of students’ out-of-school lives, they worry that dwelling on these conditions will reduce educators’ sense of urgency and responsibility for the gap. And they cling to the conviction that schooling, properly provided, is powerful enough to overcome the outside factors. “These are the kids and families we have,” they say. “It’s up to us to teach them.” Two phrases I often hear them use are, “No blaming, no excuses,” and, “If we don’t do it, who will?”

These responses leave me deeply ambivalent: I admire the passion and devotion; I am dismayed by the hubris and futility.  It makes perfect sense to me to wish that schooling could be decisive in the lives of all black and Hispanic children—indeed, of all children—no matter what their life circumstances.  It makes no sense to me to believe this to be true, since doing so ignores the fundamentals of child development.  Worse, it contributes to a larger cop-out in which scapegoating schools substitutes for addressing the primary causes of the achievement gap.

The danger of letting schools off the hook must be weighed against a greater danger, that of letting everyone else off the hook, which is unfair to schools and, most important, harmful to Hispanic and black children. What we need is not an either/or dichotomy of the kind that so often distorts debates in education but a larger, more honest acknowledgment of the complexities of the gap and of the realities of schooling. And we need a reconsideration of what is reasonable to expect of educators and what is a realistic strategy for helping African American and Hispanic students. Such a perspective would begin not by asking how schools should address the achievement gap but by asking what conditions need to be in place so that schools can do so. First, it would look beyond schools in seeking ways to prevent the gap. Next, it would consider what must be done for schools so that they can do their part once children arrive. Then it would turn to what must be done by schools themselves. At every step, it would balance wishes against realities.

A meaningful effort to address the achievement gap would start not in school but where the gap starts—in early childhood. Consider a public health analogy. If a serious, life-altering illness apparently caused by environmental factors were striking a group of children (as asthma currently appears to be doing in some urban areas), we would surely want to develop treatments for those afflicted, but we would especially want to prevent the condition in the first place. So too with the achievement gap: we should be concentrating not simply on classroom-based treatments, which can only leave schools and students playing catch-up, but on prevention and early intervention. No matter what we might hope, I cannot envision any scenario in which schools— on a broad scale and on a consistent basis— can overcome major academic deficits any group of children bring with them to kindergarten, especially if ongoing conditions in the children’s lives continue to deepen those deficits. (Claims by the Education Trust that several thousand “high-flying” schools accomplish precisely this miracle, enrolling large proportions of low-income Hispanic and black students and showing impressive results, have been thoroughly discredited.[xxiii])  Minimizing the influence of outside factors and insisting that the gap be closed by schools encourages us to discount conditions that are critical to the development and education of children.  It creates, as Paul Barton observes, excuses for bad public policies that ignore what is required to prevent learning gaps in the first place.[xxiv]


Real help for most urban Hispanic and African American students would begin with a national commitment to improve their school readiness during their first five years of life.  This would include a wide array of steps that I can only begin to list here, ranging from the expansion of high-quality pre-school and after-school programs that include support and training for parents[xxv] to a campaign to slash the amount of television Hispanic and African American children watch and encourage their parents to read to them regularly.  This effort would go all the way to a reconsideration of policies and laws that have helped to increase the proportion of children living in poverty and lacking adequate health care and that have driven welfare mothers into low-paying jobs and thus forced their children into low-quality child care.

Schools have an important role to play in any campaign to close the achievement gap.  But the primary question involving most of them is what we must do for them so that they can help their black and Hispanic students make maximum academic progress.  The immediate answer is that we should try to equalize the structural conditions of schooling across the nation.  As a matter of simple fairness, we should stop denying our neediest students the advantages other students enjoy.  This would require, among other things, being able to attract large numbers of strong teachers into our lowest-performing schools, which would depend on having the necessary resources to make these schools congenial to learning, their class sizes teachable, and so on.  Often, this would mean starting with infrastructure—replacing and repairing decrepit buildings, guaranteeing the most basic supplies, and the like.  Merely beginning to list these improvements suggests the daunting cost and complexity involved.  (According to a court-appointed panel, New York City alone requires more than $9 billion in facilities improvements and more than $5 billion per year in additional operating funds.)

The matter of what schools themselves should do about the learning of African American and Hispanic students—the question that dominates the achievement gap debate among educators—is really of immediate relevance only to those schools that are potentially able to make a meaningful impact because they have adequate resources and staffing.  Even these schools must face the fact that we don’t yet have an instructional methodology that is guaranteed to work with all students at all levels.  When it comes to differentiated instruction and related initiatives we need both modesty and perseverance.

Modesty first.  Because educators I respect believe that differentiated instruction is a real key to closing the achievement gap, I wish I did, too.  But I think an objective observer would have to be doubtful, not about whether the approach offers exciting possibilities but about its overall potential.  It is unlikely to work wonders for students who move frequently from school to school, say, or who watch more than six hours of TV per day. Moreover, differentiated instruction greatly increases the scope and complexity of teachers’ work— the planning, the actual instruction—and thus demands extra sophistication, time, and energy.  And it becomes more challenging as class size grows, as heterogeneity increases, and especially as students move to the upper grades, by which time the cumulative gaps in their performance have widened considerably and the curriculum is innately more content-driven and less amenable to individualization.

Perseverance second.  There is no reason to stop experimenting with new approaches.  Quite the contrary.  Differentiated instruction has real promise, especially at the elementary level.  Common sense calls for continued exploration of this and other alternatives to traditional pedagogy.  For example, many districts have begun concentrating on literacy initiatives in elementary grades, since reading and writing are so crucial to later success.  A number have also begun to divide their large high schools into smaller units which can offer, among other benefits, improved student-teacher ratios and closer student-teacher relationships, as well as greater individualization.  There is also intriguing evidence that when classes are small enough for teachers to forge strong connections with their students, the performance of minority students improves, just as Steele and others have predicted.[xxvi] Within the limits dictated by their resources and the other priorities pressed upon them, schools must keep pursuing these and other ways to engage all minority students in the enterprise of learning.


Needless to say, there seems little likelihood that, as a matter of national policy, America will soon tackle the out-of-school causes of the achievement gap in any sustained, preventive way.  Doing so would demand a fundamental shift in the current accountability mindset that holds schools almost solely responsible for student outcomes.  It would also demand an aggressive increase in funding for K-12 education, for pre-school and after-school programs, and for medical, social, and other services.  All of this runs directly counter to the current market-worshipping, tax-cutting philosophy in Washington and would surely meet fierce political resistance from conservatives.  At the same time, a serious effort to enhance students’ school readiness wouldn’t just press government to do more for children; it would press their parents to do so, as well, which would spark resistance at the grass-roots level.  (Witness the criticism directed by African Americans at comedian Bill Cosby over his call for greater responsibility on the part of black parents.)  No wonder so many people prefer to concentrate solely on the schools.

While it seems doubtful that the necessary changes will materialize, I think the evidence is overwhelming: without real preventive investment in the pre-school experience of most Hispanic and black children, and in the ongoing circumstances in which they grow up, the achievement gap is unlikely to narrow appreciably.  We can expect essentially what we’ve had: small incremental improvements overall and isolated stories of “turnaround” schools sparked by heroic leaders and dedicated teachers.  These turnarounds, though truly inspiring, are literally exceptional and prove temporary and unreplicable on any large scale.  Making the investment may not guarantee that we will succeed in closing the gap; not making the investment virtually guarantees that the gap we won’t.

How, in such a context, can we at least stop scapegoating schools and reset realistic expectations for them?  Researchers Valerie Lee and David Burkam, having studied the challenges African American and Hispanic children bring to kindergarten, point us in the right direction.  “We should,” they argue, “expect schools to increase achievement for all students, regardless of race, income, class, and prior achievement.  But it is unreasonable to expect schools to completely eliminate any large pre-existing inequalities… especially if those schools are under-funded and over-challenged.”[xxvii] We need to help bad schools get better, we need to promote high standards for all, and we need to demand the very best from all our educators.  But we also need to acknowledge even small gains made against long odds.  And we need to celebrate the dedication and commitment of those who keep pursuing these gains.  Above all, we must remember that schools must be—but can only be—a part of the solution.  We cannot close the achievement gap without them.  They cannot close it by themselves.

This article appeared in slightly different form in the Phi Delta Kappan, May, 20005, pp. 582-589.  Robert Evans, a psychologist and school consultant, is the Director of The Human Relations Service in Wellesley, MA, and the author of Family Matters: How Schools Can Cope with The Crisis in Childrearing, from which parts of this article are adapted, and The Human Side of School Change.  He can be reached at


[i] Kati Haycock, “Helping All Students Schieve,” Educational Leadership, March 2001, pp. 6-11.

[ii] See Carol Ann Tomlinson, How to Differentiate Instruction in Mixed-Ability Classrooms, 2nd Ed. (Alexandria, VA: ASCD, 2001).

[iii] Jonathon Kozol, Savage Inequalities: Children in America’s Schools (New York: Crown, 1991).

[iv] Mihalyi Csikszentmihalyi, “Education for The Twenty-First Century,” Daedalus, Fall, 1995, p. 107.

[v] Robert Evans, Family Matters: How Schools Can Cope with The Crisis in Childrearing (San Francisco: Jossey-Bass, 2004).

[vi] James Gallagher, “Education, Alone, Is A Weak Treatment,” Education Week, July 8, 1998, pp. 60, 43.

[vii] Glen E. Robinson, and David P. Brandon, NAEP Test Scores: Should They Be Used to Compare and Rank State Educational Quality? (Arlington VA: Educational Research Service, 1994).

[viii] Howard Gardner, “Paroxysms of Choice,” New York Review of Books, 19 October, 2000, pp. 44-49.

[ix] David J. Armor, “Environmental Effects on IQ: From the Family or From Schools?”  Education Week, 19 November, 2003, pp. 32-33.

[x] James Coleman, Equality of Educational Opportunity, (Washington, D.C.: U.S. Government Printing Office, 1969).

[xi] Christopher Jencks, et al., Inequality: A Reassessment of the Effect of Family and Schooling in America (New York: Basic Books, 1972), pp. 255-256.

[xii] Ron Edmonds, “Effective Schools for the Urban Poor,” Educational Leadership, October, 1979, p. 23.

[xiii] Paul Barton and Richard J. Coley, America’s Smallest School: The Family (Princeton, NJ: Educational Testing Service Policy Information Center, 1992).

[xiv] Valerie E. Lee and David T. Burkam, Inequality at the starting gate: Social background differences in achievement as children begin school (Washington, D.C.: Economic Policy Institute, 2002), pp. 42-44; Paul E. Barton, The Achievement Gap: Baselines for Tracking Progress.  (Princeton, N.J.: Educational Testing Service Policy Information Center, 2003), pp. 30-31; and John Merrow, “The ‘Failure’ of Head Start,” Education Week, 25 September, 2002, p. 52.  See also Harold Hodgkinson, Leaving Too Many Children Behind (Washington, D.C., The Institute for Educational Leadership, April, 2003), available at: (access date: June 3, 2004).

[xv] Richard Rothstein, “Class and The Classroom,” American School Board Journal, October, 2004, pp. 16-21.  Emphasis in original.

[xvi] Doris R. Entwisle and Karl L. Alexander, “Summer Setback:  Race, Poverty, School Composition, and Mathematics Achievement in The First Two Years Of School,” American Sociological Review, 57, 1992, pp. 72-84; Richard Allington, and Anne McGill-Franzen, “The Impact of Summer Setback on The Reading Achievement Gap”  Phi Delta Kappan, September, 2003, pp. 68-75.; and Gerald W. Bracey, “Summer Loss: The Phenomenon No One Wants to Deal With,” Phi Delta Kappan, 2002, pp. 12-13.

[xvii] Barton, pp. 22-23; Sam Dillon, “When Students Are in Flux, Schools Are in Crisis,” New York Times, July 21, .2004, p. B9.

[xviii] Barton,  pp. 32-33; Lee and Burkam, pp. 36-39.

[xix] Barton, pp. 34-35; Sara McLanahan and Gary Sandefur, Growing Up with A Single Parent (Cambridge, MA: Harvard University Press, 1994), pp. 2; 45.

[xx] Barton, pp. 26-27; and Dalton Conley and Neil G. Bennett, “Is Biology Destiny? Birth Weight and Life Chances.”  American Sociological Review, June, 2000, pp. 458-467.

[xxi] Laurence Steinberg, Beyond the Classroom (New York: Simon and Schuster, 1996) pp. 159-162; and John Ogbu, Black American Students in an Affluent Suburb (Hillsdale, NJ: Lawrence Erlbaum, 2003), pp. 198-210.

[xxii] Ogbu, p. 37.

[xxiii] Gerald W. Bracey, “The 12th Bracey Report On the Condition of Public Education” Phi Delta Kappan, 2002, p. 148; and Stephen Krashen, “Don’t Trust Ed. Trust,” Substance, February, 2002, p. 3.

[xxiv] Barton, p. 36.

[xxv] We already have models that have demonstrated enduring positive effects. The most notable is the High/Scope Perry Preschool Project.  See Lawrence J. Schweinhart, Helen V. Barnes, and David P. Weikart, Significant benefits: The High/Scope Perry Preschool study through age 27 (Monographs of the High/Scope Educational Research Foundation, 10, Ypsilanti: High/Scope Press, 1993).  See also Merrow.

[xxvi] Phil Smith, Alex Molnar, and John Zahorik, “Class-Size Reduction: A Fresh Look at The Data.”  Educational Leadership, September, 2003, pp. 72-74.

[xxvii] Lee and Burkam, p. 1.