The Social Consequences of Choice: Why It Matters Where PoorChildren Go to School


The Social Consequences of Choice: Why It Matters Where PoorChildren Go to School

July 25, 1996 18 min read Download Report
Denis Doyle
Senior Research Fellow


Congress soon will be considering major legislation that would enable poor children to attend the schools of their choice. Representatives J. C. Watts (R-OK) and James Talent (R-MO) have introduced a comprehensive legislative initiative, Saving Our Children: The American Community Renewal Act of 1996 (H.R. 3467).1

Title II, the Low Income Educational Opportunity Act of 1996, would establish an education choice scholarship program that includes both private and religious schools in "renewal communities." Each locality would determine the value of the scholarship for poor children, but the maximum value of the scholarship would not exceed the per capita cost of educating children in a local public school. The scholarships available to poor children would have a minimum value not less than 66 percent of the per capita cost of educating children in the local public schools or the tuition normally charged by the private school. Parents would not be forbidden from using scholarships to pay for tuition in religious schools, and local authorities would not be able to discriminate against parents or children who do so. The scholarship program would be supported from existing federal funds.

Where their children go to school is a serious question for parents. President and Mrs. Clinton are willing to pay an annual tuition of more than $11,000 to send their daughter Chelsea to the exclusive Sidwell Friends School. Vice President and Mrs. Gore chose St. Albans School, a few blocks from Sidwell, for their son, as did D.C.'s Shadow Senator Jesse Jackson. The Clintons, like the Gores and many other parents among the nation's social and political elite, have made a conscious decision that the Washington, D.C., public schools, recently the subject of failed congressional reform efforts, are not good enough for their children. They also have decided that a school with religious affiliation is important.

While most Americans intuitively understand what is best for their children, public policy has made it difficult for all but the well-to-do to choose the school they prefer. Indeed, implicit in public policy is the assumption that where a child goes to school is unimportant. But the evidence is in: Where a child goes to school matters a good deal -- better a good school than a bad school.2

More important, the empirical evidence clearly reveals that most private schools do a better job than most public schools, particularly with poor and minority youngsters. Yet for more than a century, public policies in the United States have made it virtually impossible to provide public funds for children to attend private schools.3 In this regard, the United States stands virtually alone among the world's democracies.4 With enactment of the Watts-Talent bill, this dubious distinction would come to an end.

The Changing Intellectual Climate

For many years -- in the 1960s and 1970s in particular -- a body of thought developed in this country which emphasized a child's "background" characteristics, which were assumed to be paramount. Success in school was the luck of the socioeconomic draw. It made little or no difference where children went to school, whether urban or rural, north or south, east or west, public or private; what was important was what they brought to school with them.5 If they were poor and dispossessed, they probably would do poorly. If they were poor and members of a minority group, they were almost certain to do poorly. By way of contrast, advantaged youngsters from enriched backgrounds did well. That was -- and is -- the argument. That it flies in the face of experience and common sense has done nothing to temper its enthusiastic reception among policymaking elites.

The modern incarnation of this general argument emerged from both the incomplete interpretation and the substantial misinterpretation of early work by James Coleman, a nationally prominent sociological researcher, then at Johns Hopkins.6

In research conducted under the auspices of the U.S. Office of Education (predecessor to the U.S. Department of Education), Coleman and his colleagues initially found little in the way of school effects; factors like race, income, and parental level of education were stronger than school characteristics as predictors of school performance. What did seem to make a modest difference was who you went to school with; poor black youngsters appeared to do better in integrated settings. But as any competent policy analyst knows, absence of proof is not proof of absence. Because "school effects" were elusive or hard to find and measure did not mean they did not exist.

To the contrary, if "school effects" really were only weak (or worse yet, if they did not exist at all), the whole rationale of public education would come tumbling down. Indeed, so might that of private schooling. If nothing matters, there is no point in worrying about the right way to educate children.

There was, however, a widespread desire to believe a variation of such an unlikely interpretation. Among liberals who believed in vigorous, interventionist government, the lack of "school effects" was interpreted to mean that these effects were not sufficiently robust. What was needed was "more" impact: more money for schools. If race and socioeconomic status (or SES in the jargon of sociologists) were more powerful determinants of school performance than schools themselves, the schools must be weak. Other larger, even more ambitious government programs would be in order.

Just as geography was destiny in the 19th century, SES became destiny in the 20th. The ironies in this development were many. In liberal hands, it became an argument designed to bolster the theory that the only good school is a rich school (and a rich public one at that). At the same time, it became an argument that absolves the existing school of its obligations to its students. It "blames the victim." The school is not at fault if students do not learn. Schools have little or no effect; it all depends on the raw material.7 This sentiment is nicely encapsulated in a song ("Gee, Officer Krupke!") from Leonard Bernstein's classic West Side Story: "Hey, I'm depraved on accounta I'm deprived."

Why Schools Matter

Such conventional liberal theory is bad pedagogy and worse public policy. Indeed, it fell to James Coleman, then a professor at the University of Chicago, to reopen the argument with extraordinarily powerful research that demonstrated the strong effect of Catholic schools, particularly on poor black children.8

This research, widely reported in 1981 when it appeared in summary form in The Public Interest, is as valid today as it was then. Coleman also noted the embarrassing fact, still a stark feature of today's school choice debate, that "there are many who vigorously oppose making private school attendance easier and at the same time have their own children enrolled in private school."9 Coleman set out to answer three vital questions:

  1. Do private schools provide better education?
  2. If they do, what is the public policy interest in denying poor children access to them?
  3. Is it true, as opponents of public funding allege, that private schools racially segregate children?

Coleman's first hurdle was methodological: how to distinguish "the effects of selection from the effects of the school itself" to control, "even to the extent of overcompensating," for selection into the schools. This he did. His academic findings are hardly surprising: "higher learning in the Catholic schools than in the public school."

The literature in this area is not restricted to Coleman's seminal work, as John Convey's 218-page Catholic Schools Make a Difference reveals.10 Convey reports that in a series of studies based on National Assessment of Educational Progress (NAEP) data, Lee and Stewart "documented consistently higher scores of Catholic school students compared with public school students." For example, the average Catholic school student in grades 4 and 8 scored significantly higher in reading and writing.11 In addition, Andrew Greeley's analysis of data from the High School and Beyond longitudinal study showed superior performance among whites, blacks, and Hispanics in Catholic school by every single category (Vocabulary, Reading, Math 1, Math 2, Science, and Writing), and Marks and Lee, writing in 1989, found higher scores in Catholic schools in grades 3, 7, and 11 among whites, blacks, and Hispanics in every single category for math, science, and reading.12

The most imposing studies, however, are those of Coleman and his colleagues. Catholic school sophomores, for example, scored 10 percent higher in science and 12 percent higher in civics, and from 17 percent to 21 percent higher in mathematics, writing, reading, and vocabulary. Catholic school seniors also consistently outscored public high school students: 10 percent to 17 percent higher in reading, mathematics, and vocabulary, and from 3 percent lower to 7 percent higher on three tests that measure ability more than achievement.13

The last finding strongly suggests that the explanation for higher scores lies not in some "ability" differential, but in what the schools do with what they have. Catholic schools expect much of their students, whatever their race and background, and expecting much, get much.

Although Coleman found even higher scores in non-Catholic private schools, the more important sample is Catholic schools, which are the dominant force in the non-government school market.

Coleman also was concerned with other issues, most notably the impact of choice on racial integration. Here the findings were equally unambiguous: A child is more likely to attend school with a child of another race in the private sector than in the public sector. The importance of this early finding would be hard to overstate; "suburban schools within the public sector are used as a haven to a much greater extent than in the private sector." Counterbalancing "bright" flight to the suburbs is the behavior of students in Catholic school. Coleman's figures on dropouts are striking. Among white students in public school, 13.1 percent dropped out compared with 2.6 percent in Catholic schools; among Hispanic students, the difference was two to one (19.1 percent to 9.3 percent); and among black students, the difference was more than 3.5 to one (17.2 percent to 4.6 percent).14

Coleman's first finding concerning school effects should have been received with rejoicing by the professional educators who had endured their share of bad press. Their magazines and journals should have featured articles on why schools make a difference. So also should the nation's civil rights leadership have reacted positively to Coleman's finding about racial integration in the private sector. Indeed, the refusal of the U.S. Supreme Court to establish "metropolitan" remedies for past cases of racial discrimination makes the Coleman finding all the more important, because private school attendance is almost never geographically based. Private schools, already better integrated than public schools, offer more fertile fields for expanded racial integration over time than do the public schools.

Such reactions, however, were rare, and the reason is not hard to fathom. Had Professor Coleman's rigorous social science work revealed stronger public than private school effects, there would have been joy in the great halls of the education establishment.15 But because one of the bastions of "good schooling" is the private sector,16 public school educators could not -- would not -- accept his findings. Coleman, on the other hand, did no such disservice to the authorities who run the public schools. He followed the evidence where it led and found that all schools that exhibit similar characteristics have strong school effects. All schools, whether public or private, that exhibit the same cluster of attributes -- high expectations, a focused curriculum, good order and discipline, caring teachers, enterprising students -- have relatively robust "school effects." What is important is that these characteristics are found more readily and more often in private schools.

The Impact of Social Capital

In addition to his school effects work, Professor Coleman returned to the questions that had intrigued him originally: Even if some schools are better than others, what difference does what the student brings to school make?17

What are the cultural and social traits that make a difference? The potential for synergy is obvious, as are the public policy implications.

"Social capital" is the cluster of skills, attitudes, and (to a lesser extent) knowledge that the student acquires in the non-school environment. Clearly, a child who is disruptive and unruly and who has only weakly developed academic interests is less likely to do well in any school, "good" or "bad," while youngsters who bring with them a love of learning, disciplined and orderly habits, and respect for themselves and other students, as well as for the teacher, are much more likely to do well, all other things being equal. These traits -- what might be called pro-academic traits -- are the product of caring and supportive communities, voluntarily created and maintained. Though they may include people who look alike, in the final analysis they are not defined by race, socioeconomic status, or ethnicity; they bridge all those gaps. They are associative communities made up of superficially similar or different people who share common values and virtues. In a school setting, to use a quaint term, they are "communities of scholarship."

Members of Congress and state and local legislators alike should recognize the public policy implications of "social capital." The most important single thing they can do in this regard is to encourage its formation and continuance by underwriting education choice. Not only does choice permit the expression of "social capital;" it builds and reinforces the habits of mind that are essential to voluntary association -- the same habits of mind that are essential to civil society. In a school setting, choice imposes the burden of choosing; there is no escape. One must make one's bed and lie in it; no one else does it for you. School choice proposals, including the school choice component of the Watts-Talent legislation, foster democratic habits of mind that are essential to a free society.

The High Stakes in Schools

The reasons for the high quality of private schools are several and have to do largely with organization, incentives, and rewards. Most simply, voluntary association makes a difference. In the public sector, the organizational climate conspires against dedicated teachers and hard-working students; in the private sector, it encourages and rewards high levels of performance. In short, monopolies are no better in the public sector than in the private sector. They serve only the interests of the "owners," never the interests of workers or clients.

In the public sector, of course, the "owners" do not literally own the schools. Rather, they are managers and bureaucrats, and the schools are run for them, not for teachers and students. This rather unappealing picture is not restricted to the public sector; managerial capitalism also looks a good deal like this, which is why it is under fire from liberal and conservative critics alike. The Watts-Talent legislation would permit parents of poor children to "own" their own schools instead of having to submit to the bureaucracy that runs government schools.


America's policymakers seem willing to tolerate failures and weaknesses in education that they would not tolerate in any other area of public policy. Coleman's findings are so powerful that were he dealing with any other area of public policy -- health, welfare, housing, transportation, or juvenile justice -- shifts in public policy would begin to reflect the new knowledge he contributed. But public schooling is a nearly immovable object; public schools are run for the convenience of, and at the pleasure of, their managers, not for the benefit of students and their families.

When they debate the issue of school choice, Members of Congress should consider the "school effects" literature in terms of academic outcomes; its conclusions are wholly unsurprising. In light of their "environment," one would expect private schools to "do better." They must if they are to survive. All parents who enroll their children in private school go through at least a two-step process.

First, they must decide to enroll their children in a non-government school and find the financial wherewithal to deliver on this decision. Almost without exception, this involves the payment of substantial tuition and fees out of pocket. Indeed, because people value what they pay for, "full-ride" scholarships are almost unknown in private elementary and secondary education. Even -- perhaps particularly -- schools like Father Clemens's Our Lady of Angels in Chicago charge at least a modest tuition.18

Second, the financial commitment families make to send their children to private school is a proxy for a personal commitment to education, the importance of which is hard to overestimate. It makes schools and their families partners in a way that the a public sector monopoly can never hope to achieve.

Probably the most important "school effects" issue is not the measured academic "outcomes" of schooling, but the school's impact in the market. In a market, a school must "sell" and parents must "buy." That is the marketpar excellence. Goods and services that do not sell have no effect; indeed, they do not stay in the market. Only monopolists -- particularly government monopolists -- stay in markets when no rational consumer wants their product or service. Poor children, for all practical purposes, today have no options.

This issue came to a head this year when Congress considered school choice as part of the Washington, D.C., FY 1996 appropriations bill. Representative Steve Gunderson (R-WI) offered an amendment that would have given scholarships of up to $3,000 for students whose family incomes were at or below the poverty level and up to $1,500 for students whose family incomes were not more than 185 percent above the poverty level. The Senate, however, when presented with this opportunity to expand choice for children in Washington's failing public school system, refused to do so. Only four members of the Senate had sent their children to D.C.'s public schools.19

The obvious assumption, of course, is that D.C.'s public school system is tolerable for poor children but not a serious option for members of the United States Senate.

On the wisdom of these official restrictions, let the late Professor James Coleman answer the critics. His is an eloquent summation of the case for justice and dignity:

There may be a rationale for some protective barriers to encourage participation in the public schools, but certainly not those that exist now, which harm most the interests of those least well-off and protect most those public schools that are the worst. In short, the tuition barrier to private schooling as it exists now is almost certainly harmful to the public interest, and especially harmful to the interests of those least well-off.20

As Members of Congress consider the Watts-Talent school choice provisions, they have another chance to do the right thing.


  1. For a discussion of the bill, see Robert Rector, "God and the Underclass," National Review, July 15, 1996, pp. 30-33; see also Christine Olson, "Saving the Children: How Congress Can Help Poor Kids and Revitalize America's Inner-City Neighborhoods," Heritage Foundation Issue Bulletin, forthcoming.
  2. The most important work in this field is that of the late sociologist James Coleman. A researcher of towering intellect and erudition who also was devoid of ideological predisposition, Coleman followed the data wherever they led, often at the cost of personal peace of mind. He was vilified by the American Sociological Association for taking the politically incorrect position that public support of private schooling was a defensible public policy option (although, to its credit, the ASA retracted its condemnation to award him a medal shortly before his death). Coleman never wavered in his quiet and determined defense of the truth as he understood it. While his prose style and the heavy quantitative emphasis of his major works are a bit daunting for the generalist, his articles are clear, concise, and lively. A perfect example -- as current today as when it was published -- is "Public Schools, Private Schools, and the Public Interest," The Public Interest, No. 64 (Summer 1981). Father Andrew M. Greeley's work is on point as well. In particular, see Catholic High Schools and Minority Students (New Brunswick, N.J.: Transaction Books, 1982).
  3. In particular, see "Foreword," in Thomas Vitullo-Martin and Bruce S. Cooper, Separation of Church and Child: The Constitution and Federal Aid to Religious Schools (Indianapolis, Ind.: Hudson Institute, 1987).
  4. See Charles L. Glenn, Choice of Schools in Six Nations (Washington, D.C.: U.S. Department of Education, 1989), and Educational Freedom in Eastern Europe (Washington, D.C.: Cato Institute, 1995). Glenn, one of the nation's leading authorities on public funding for attendance at private schools, for 21 years was the Massachusetts State Department of Education official responsible for civil rights and urban education opportunity. Recognizing the failure of busing, Glenn pioneered education choice programs.
  5. One long-standing trend, the concentration of high test scores in the Northern tier of states from Maine to Montana, led Senator Daniel Patrick Moynihan (D-NY) to propose facetiously the following policy remedy to states concerned about low test scores: Move your state north.
  6. Coleman's first major study which permitted a systematic examination of this and related questions was prepared under the terms of a major grant from the U.S. Office of Education, part of the U.S. Department of Health, Education and Welfare (now the Department of Health and Human Services). Titled Equality of Educational Opportunity, it was released in 1966 and caused an immediate sensation. Coleman and his colleagues could find no school effects of consequence; indeed, any school effect seemed to be overwhelmed by student background characteristics. Policymakers were led in two different directions by this surprising "non-finding." Many analysts concluded that school effects were minimal (at least from a public policy standpoint; they still put their own children in "good" schools). Needless to say, their public policy conclusion was a counsel of despair; after all, the whole American experiment in self-governance was based on the Founders' conviction that an educated citizenry is essential to democracy. The other direction, equally misleading, fueled the notion that black children could learn only if they were integrated with white children. While racial integration is a laudable objective, it is patronizing in the extreme to argue that black children must be integrated to learn. More to the point is the quality of facilities, books, teacher preparation, and the like.
  7. As the Manhattan Institute's Sy Fliegel wryly notes, it is like the excuse offered by the swimming teacher who says, "I taught them to swim but they sank."
  8. The most complete source on this and related work is John J. Convey, Catholic Schools Make a Difference: Twenty-Five Years of Research (Washington, D.C.: National Catholic Education Association, 1992).
  9. Coleman, "Public Schools, Private Schools, and the Public Interest," p. 20; all subsequent Coleman quotes are drawn from the same source.
  10. Convey, a professor at Catholic University, studied "the studies" on Catholic education to provide an overview and synthesis of Catholic school research published between 1965 and 1991. His book is an invaluable resource.
  11. Convey, Catholic Schools Make a Difference, p. 61.
  12. Ibid., pp. 174-175.
  13. Ibid., p. 62. The data in question are part of the ongoing National Longitudinal Sample (NLS), collected by the U.S. Department of Education and available for independent analysis from the Department.
  14. Ibid., p. 176.
  15. Although most Americans are convinced that private schools are simply better than public schools, this is not an altogether implausible proposition. It was true in Australia, leading the Australian Labour Party (similar to the Democratic Party in the United States) to support public aid to private schools. Sparked by the work of Father James Carroll, who later became Bishop of Sydney, public funding was made available on the basis of the following argument: Catholic schools are inferior (by almost any academic or fiscal measure) and will remain so unless public funding is provided; Catholics (and non-Catholics who attend these schools) will not abandon their schools; and no good public end is served by denying poor Australian children a quality education in the schools in which they chose to enroll. For more on this interesting situation, see Denis P. Doyle, "Family Choice in Education: The Case of Denmark, Holland and Australia," in Private Schools and Public Policy: International Perspectives, ed. William Lowe Boyd and James G. Cibulka (New York, N.Y.: Falmer Press, 1989), and "Why Congress Should Give Poor Children Choice of Religious Schools," Heritage Foundation Backgrounder, forthcoming.
  16. Not to put too fine a point on it, but private schools are a red flag to the public school bull. The intensity of the hostility most public school leaders feel for private schools is understood more easily in psychological terms than in logical terms. Private schools, for example, are accused of being "elitist," a sin as serious as racism or sexism in the modern pantheon of bad attitudes. But the facts, as Coleman and others have revealed, are otherwise. For a longer disquisition on this general subject, see "A Din of Inequity? Private Schools Reconsidered," Teachers College Record, Vol. 82, No. 4 (Summer 198l).
  17. For an accessible article on the subject, see James S. Coleman, "Schools and the Communities They Serve," Phi Delta Kappa Vol. 66, No. 8 (April 1985), pp. 527-532. For a more scholarly treatment, see James S. Coleman, "Social Capital in the Creation of Human Capital," American Journal of Sociology, Vol. 94, Supplement (1988), pp. S95-S120.
  18. An example of more than passing importance, Father Clemens's school operates on an extended day and year, is in the middle of the ghetto, has a waiting list, accepts only children from families on welfare, and still charges tuition. His reasoning? If they can afford a phone, a TV, sneakers, and Bulls jackets, they can scrape together some tuition. Moreover, all families are expected to give time to school projects -- not just the PTA, but painting and fix-up activities as well. Finally, families that really cannot afford any tuition are allowed to "work off" the obligation: Families that pay something are families that take school seriously.
  19. Lisa Nevins, "Gingrich Rips Democrats on School Vouchers," The Washington Times, April 18, 1996, p. A10.
  20. Coleman, "Public Schools, Private Schools, and the Public Interest," p. 30.


Denis Doyle

Senior Research Fellow