A Courageous Look at the American High School

The legacy of James Coleman

Video: Paul Peterson talks with Nathan Glazer

In Saving Schools: From Horace Mann to Virtual Learning, scheduled for release by Harvard University Press this spring, Paul E. Peterson tells how five individuals—Horace Mann, John Dewey, Martin Luther King Jr., Al Shanker, and William Bennett—shaped American education in ways they never expected. Peterson chronicles how education became ever more centralized and bureaucratized, creating the monolithic system in place in the early 21st century. The story nonetheless ends on a hopeful note, as Peterson characterizes Julie Young’s innovative work at Florida Virtual School as a harbinger of an educational future in which learning finally becomes customized to each student’s circumstances.

In Peterson’s account, sociologist James Coleman plays a pivotal role. In the following excerpts from the book, Peterson recalls the bitterness of the controversy provoked by Coleman’s writings and reveals, for the first time, how Coleman’s insights were rooted in his own high-school experiences.

Adapted from Saving Schools: From Horace Mann to Virtual Learning by Paul E. Peterson, to be published March 2010 by Harvard University Press. Copyright © 2010 by the President and Fellows of Harvard College. All rights reserved.

Excellence was seldom to be found in 2006, when David Ferrero, an officer of the Bill and Melinda Gates Foundation, reviewed five firsthand, book-length accounts of teaching and learning at individual high schools. In one account, a rookie teacher, telling her own story, “struggles to establish authority in her classes and generally fails;…her students ritually defy her, going so far as to openly declare their intention to get her fired for the sheer sport of it.” At another school, “numerous attempts” by well-meaning, hardworking teachers fail “to coax students out of their shells, engage them in important issues, and motivate them to perform on tests.” On and on such tales go. A powerful but hostile peer group seemed in charge of the learning process.

According to Cornell economist John Bishop, the problem begins in middle school, where “nerds” are harassed. “Studiousness is denigrated…in part because it shifts up the grading curve and forces others to work harder to get good grades…. Victims of nerd harassment hardly ever tell their parents, their siblings, or their friends. Most accept the proposition that…acting like a dork is bad…. Complaining to a teacher is self-defeating. Squealing on classmates only exacerbates [the situation].”

The problem did not appear suddenly at the beginning of the twenty-first century. Fifty years earlier James Coleman, reflecting on his own adolescence, had detected something quite similar and then provided a sociological explanation for the phenomenon.

James S. Coleman

Born in 1926, Coleman began his graduate studies in sociology at Columbia University in 1951, one year before [John] Dewey died at the age of ninety-two. The two intellectuals had much in common. Both came from ordinary, small-town families, but they both had entrepreneurial spirit, tremendous energy, and personal fortitude that belied their surface modesty. Neither was a brilliant lecturer, but both were kind, gentle, supportive mentors, surrounded by devoted graduate students. Like most Americans, both were pragmatists—concerned less about systematic theory than about learning what worked in practice. Neither saw his work on education as the centerpiece of his life’s work. Dewey was a philosopher, Coleman a social theorist and mathematical model-builder. Yet neither man would have made as lasting a contribution were it not for his work on schools.

Despite the similarities, Dewey and Coleman walked in contrasting intellectual worlds. If Dewey’s thinking was shaped by Rousseau, Hegel, and the Romantic tradition more generally, Coleman’s owed more to two Scottish empiricists: David Hume and Adam Smith. The “Emile” of significance to Coleman was not Rousseau’s mythical child but Emile Durkheim, a sociologist whose point of departure was not the state of nature but a well-defined community context. Coleman’s work was more disciplined than was Dewey’s. Trained in survey research and modern analytic techniques—random sampling, systematic data collection, rigorous comparisons—taking hold at Columbia, Coleman was able to test his ideas in ways unavailable to Dewey. Most important, Dewey and Coleman had separate agendas: Dewey’s ideas shaped the public schools of the twentieth century; Coleman deconstructed what Dewey had built.

Unlike Dewey, Coleman never became a household name, yet his impact on American education has been immense. At his memorial service in 1995, New York senator Daniel Patrick Moynihan observed that the man they were remembering was among “a very small number of people who end up defining a major part of the intellectual agenda for their times. Their work is both so powerful and so well argued…that others are inspired to focus on these same issues.” Coleman’s impact was not without its ironies, however. His research served the civil rights movement King had begun but also the reaction that was to follow. His studies first accelerated and then helped put the brakes on school desegregation. A part of his work has been taken to mean that schools are insignificant, while another part suggests they are decisive. Coleman himself saw no contradictions.

We know few details about Coleman’s early educational experiences, in part because Coleman himself wanted us to believe that at age twenty-five he had sprung directly from the head of—well, not Zeus, but Robert Merton and Paul Lazarsfeld, two men in Columbia’s sociology department whom many students thought had godlike qualities. Reflecting back on what seems to have been something like a conversion experience, Coleman said: “I left a job as a chemist…and took on a new life…. The transformation was nearly complete. Except for my wife (and other kin who lived far away in the Midwest and South), I shed all prior associations…. [After] the resocialization I underwent at Columbia from 1951 to 1955… I was a different person.” It was Merton’s social-theory course that did the trick, “a conversion experience for those of us eager for conversion.”

The grandson of an evangelical preacher, Coleman certainly knew the religious meaning of the concept he was invoking. But his first twenty-five years left more of a mark on him than he was willing to acknowledge. Born in Bedford, Indiana, he began high school in Greenhills, Ohio, a place he wrote about almost wistfully: “School life had, for a few of us, a more academic focus, in retrospect surprisingly so.” Shortly thereafter his father took a job as a factory foreman in Louisville, Kentucky, a city that had two public high schools for boys: “Male (with a college preparatory curriculum) and Manual (with vocational and pre-engineering curricula).”

Coleman adjusted to his new school [Manual High] by becoming a member of the school’s football team. The “boys who counted in the school,” he writes, “were the first-string varsity football players,” because “Male and Manual were locked in a fierce football rivalry that culminated every Thanksgiving Day but flavored the whole school year.” He was quickly drawn in. “[The] environment had shaped [his] own investment of time and effort, intensely focused on football, although arguably [his] comparative advantage lay elsewhere.” Otherwise, high school “failed” him. Apart from an eleventh-grade algebra class, he could not find anything “to excite my interest and capture my full attention.” One day, while hitchhiking to football practice, he thought longingly: “If only they would not destroy in us the interest with which we came to school, I would ask for nothing more.” Only when Coleman arrived at Columbia did he find faculty members with a “personal (that is, selfish) interest in some of their students. They seemed to be interested in those students in a way I had never felt since the ninth grade,” perhaps because “graduate students help bring professors closer to immortality.”

He nonetheless attended a small college before joining the Navy in the middle of World War II. After his discharge, he used his benefits under the voucher-like GI Bill to earn a B.S. degree in chemical engineering from Purdue University. Though he was then hired by Eastman Kodak in Rochester, New York, Coleman was still a frustrated product of Manual High, a technician who wanted a more intellectual challenge. Despite his limited resources, he made a dramatic career decision to pursue a Ph.D. in sociology. Rejected by Harvard and Michigan, he won admission to the overcrowded program at Columbia.

He could not have been more fortunate. In 1951, Paul F. Lazarsfeld was using newly developed quantitative techniques to look at practical topics: mass media, advertising and political campaigns. At the same time, Robert K. Merton was systematizing his sense of the ironic—unexpected things happen for reasons no one anticipates—to which he gave the rather pompous label “latent-function theory.” Coleman drank from both professorial wellsprings, but it was Merton who “provided the inspiration for it all.” In his italicized words: “I worked with Lipset, worked for Lazarsfeld, and worked to be like Merton.” Like Merton, Coleman viewed the world with an outsider’s irony: things are not as they seem, and consequences differ from what is expected. At a personal level, Merton endeared himself to Coleman the day he asked the young man about his dissertation plans. Told that none had been devised, Merton suggested that Coleman simply use the chapters he had drafted for a study of trade unions he was writing in collaboration with Seymour Martin Lipset, the department’s up-and-coming assistant professor. Acting on this advice, Coleman had his thesis completed just three years after matriculation. Shortly thereafter, he submitted a research proposal to the U.S. Office of Education’s new Cooperative Research Program.

Until this point, nothing in Coleman’s early career indicated he would become the premier education sociologist of the twentieth century. No one at Columbia specialized in educational sociology, a field Coleman disparaged as languishing in the cellar of the discipline. But as he was ruminating over possible topics for a federal grant proposal, Manual High came up one night at a dinner party the Colemans were hosting for Martin Trow (coauthor, with Coleman and Lipset, of the trade union study) and his wife. The Trows had attended elite schools where sports were subservient to academics, not only in the schools’ official focus but also in the students’ interests and social relationships. How different from Manual High!

Turning the conversation into a research proposal, Coleman laid out a plan to study several schools in Illinois, near the University of Chicago, where Coleman had been hired as an assistant professor. The book that emerged, The Adolescent Society [1961], which is as much a theoretical commentary on Manual High as an analysis of ten schools in Illinois, remains Coleman’s masterpiece. According to Coleman, the focus at these schools was on sports stars, cheerleaders, and other members of the leading crowd, known more for smart dressing than for smarts per se. Those who studied hard and got good grades were edged to the social sidelines. For those who excelled scholastically, success must appear to have been “gained without special efforts, without doing anything beyond the required work.” Otherwise, one is socially isolated by “the crowd.” Ostensibly, schools are educational institutions, but their latent function is social and quite inimical to educational purposes. It is the way in which U.S. schools are organized that is the problem, Coleman says. They resemble jails, the military, and factories: all of these institutions are run by an “administrative corps” that makes demands upon a larger group (students, prisoners, soldiers, workers). In response, the larger group develops a set of norms that govern the choices individuals make. “The same process which occurs among prisoners in a jail and among workers in a factory is found among students in a school. The institution is different, but the demands are there, and the students develop a collective response to these demands. This response takes a similar form to that of workers in industry—holding down effort to a level which can be maintained by all. The students’ name for the rate-buster is the ‘curveraiser,’…and their methods of enforcing the work-restricting norms are similar to those of workers—ridicule, kidding, exclusion from the group.” With his typical irony, Coleman dedicated the book “To my own high school, du Pont Manual Training High School, Louisville, Kentucky.”


The occasion for his contribution was provided by a little-noticed clause buried in the Civil Rights Act of 1964, which called for a “survey concerning the lack…of equal opportunities…by reason of race, color, religion or national origin in public education.” Though not a prominent public figure, James S. Coleman was the logical choice for directing the survey. He had been trained in survey research, was an acknowledged expert on high schools, and was sympathetic to the civil rights movement—he and his son had been arrested at a demonstration in Baltimore. Coleman…agreed to take on the assignment only after “some hesitation” and “extensive discussion” that transformed what at first seemed to be nothing more than a collection of racial-segregation statistics into the first nationwide study of the factors that affect student achievement. Students at 4,000 randomly selected schools across the country were tested in various subjects. The study also collected information on characteristics of the schools the students attended: racial composition, per-pupil expenditures, the college degrees teachers had earned, teacher ability (as measured by performance on a test), the number of books in the school library, and much more. Family background information was collected as well.

The study was to go forward with more-than-deliberate speed, as results were expected to reveal a need for federal action to equalize educational opportunity, the keystone of Lyndon Johnson’s “Great Society.” Imagine, then, the shock inside the White House when a draft of the report began circulating inside the administration. Daniel Patrick Moynihan, one of Johnson’s top domestic advisers, gave a sense of the reaction when he recalled being greeted in the spring of 1966 by Harvard professor Martin Lipset with the query: “You know what Coleman is finding, don’t you?” “I said, ‘What?’ He said: ‘All family.’ I said, ‘Oh, Lord.’” The next day Moynihan informed the secretary of Health, Education, and Welfare (HEW) to get ready, as the research project was about to produce findings the administration “was not going to like.” The project report [Equality of Educational Opportunity, 1966], later known as “Coleman I” after two additional reports appeared, was released on Independence Day weekend, 1966. That was thought to be a good time to announce negative news, since much of the press was on holiday. The strategy worked: few but academics paid attention, and only gradually did its message sink in.

To everyone’s surprise, Coleman I found that within regions and types of communities (urban, suburban, and rural), expenditures per pupil were about the same in black and white schools. Even more remarkable, students did not learn more just because more was spent on their education. Nor did any other material resource of a school have much of an effect on how well Johnny and Suzy read—not the number of students in the class, nor the teacher’s credentials, nor the newness of the textbooks, nor the number of books in the library, nor anything physical or material that schools had for years considered important. What did count were a host of family-background characteristics: mother’s education, father’s education, family income, having fewer siblings, the number of books in the home, and other factors—all of which together explained more of the variation among students in their reading achievement than any school-related factor.

One finding in Coleman I saved the day for the Johnson administration. The authors found that student achievement was affected by the social composition of the pupils at a school. If a low-income African American child had fellow students who were white or from a higher socioeconomic status, the child did better at reading. The converse was not true, however: a white child did not suffer educationally from having black classmates. In other words, the influence of peers was asymmetrical. Desegregation helped blacks without hurting whites. Many years later, the Nobel Prize–winning econometrician James Heckman and his colleague Derek Neal called that asymmetrical result Coleman’s “least robust” finding. But Coleman never doubted it. Testifying before the Senate Select Committee on Equal Educational Opportunity, he said black students at segregated schools were “deprived of the most effective educational resources contained in the schools: those brought by other children as a result of their home environment.” Whatever regrets the Johnson administration might have had about some parts of Coleman I, it was pleased by the ammunition the report provided for the ongoing desegregation campaign.

So it was truly ironic that Coleman, the very academic whose work provided the clearest educational justification for school desegregation, would in his next major study [Trends in School Segregation, 1968–73, 1975], the “white flight” study (known as Coleman II), produce findings that called into question many of the policies being used to desegregate the schools. Using data collected by the newly established Civil Rights Commission, Coleman II tracked trends in black and white school enrollments in cities across the United States. He and his colleagues found that white families were moving outward more rapidly from those central cities where racial desegregation plans were being implemented.

Coleman expressed concern that, as a practical matter, busing of students within districts was self-defeating. Within school districts, to be sure, the segregation index fell from 0.63 to 0.37 in the years 1968–1972. But that only intensified segregation between districts. Said Coleman, “The emerging problem with regard to school desegregation is the problem of segregation between central city and suburbs.” Schools were at risk of being as segregated as they had ever been, exactly as Justice [Thurgood] Marshall had predicted.

Not since Cleopatra heard about Antony’s dalliances has a messenger come so close to being poisoned. Scholars turned on Coleman with an unexpected vengeance that introduced a more virulent tone into the world of education policy research. Well-known Harvard psychology professor Thomas F. Pettigrew claimed that Coleman II “should not be taken seriously.” The NAACP general counsel called the Chicago sociologist “without a doubt, a first-class fraud…. He is not entitled to any credence or any reliability or any belief with respect to the things he says he has found.” A Washington Post columnist questioned whether Coleman was mixing research with advocacy, quoting then deputy director of the National Science Foundation (and future president of the University of California) Richard Atkinson as saying, “A lot of what goes under the name of social science is just junk…. Too often [when] speaking on issues of education [scholars use] research evidence as a disguise for advocating a particular policy.” Atkinson was careful not to mention Coleman by name, but such innuendo by distinguished leaders fed the anti-Coleman fire. It flamed into an effort, led by the sociologist Alfred McClung Lee, then the president of the American Sociological Association (ASA), to censure or expel Coleman from the organization’s membership for having spread “flammable propaganda.” Though that blaze was contained, “few sociologists ever had to endure the high profile public controversy which swirled around him.” Years later, Coleman recalled the ASA plenary session held to debate the report: “The passions generated at that session are hard to reconstruct now, but I still have the posters that were plastered at the entrance to the ballroom and behind the podium, covered with Nazi swastikas, epithets, and my name.”


In 1981, Coleman wrote his third major report, identified here as Coleman III. Two years previously, Coleman and his colleagues at the University of Chicago had been asked by the National Center for Educational Statistics to extend the work begun in Coleman I. The study was to be more than a single-shot survey along the lines of Coleman’s earlier work. Instead, several rounds of data were to be collected. A nationally representative sample of high schoolers was to be tested as sophomores and then again as seniors, after which they would be followed into college and the labor force. In this way, Coleman expected to find out how much students learned between their sophomore and senior years, as well as the impact of schooling on college attendance and labor force participation. Coleman also convinced the U.S. Department of Education, which was funding the study, to look at private schools as well as public ones. He now got his chance to see if private and public schools across the country were as different from one another as Manual High differed from those elite schools his friends at Columbia had attended.

The survey of some 70,000 students at more than 1,000 high schools was conducted in the spring of 1980. Working at his usual extraordinary pace, Coleman reported his team’s findings back to the government that same September, even as a presidential election campaign was in full swing. After the election was over and the Reagan administration had assumed office, the results from the first round of data collection were released. Coleman reported that sophomores in Catholic schools performed at higher levels than those in public schools, apparently showing in practice what [Milton] Friedman had argued in theory. In education circles, it was about as dramatic as the first proof of Einstein’s theory of relativity. Coleman explained his findings by claiming that students at Catholic schools benefited from the “social capital” surrounding the religious school: parents knew and supported one another as they attended Mass and participated together in other religious activities. As another group of sociologists put it, “Catholic schools benefit from a network of social relations, characterized by trust, that constitute a form of ‘social capital.’… Trust accrues because school participants, both students and faculty, choose to be there.”

The attacks on Coleman III were no more polite and detached than the attacks on Coleman II. The day it was released, “people entering the auditorium were handed leaflets attacking the study.” The executive director of the National Association of Secondary School Principals insisted that the study used “incomplete data inappropriately applied.” The New York Times chided Coleman for publicizing his results, saying that “sociologists invite trouble” when they seek “the stardom of advocacy based on their fallible predictions.” Its news reports quoted Coleman out of context in order to give the impression that he himself thought “the study was deeply flawed and that [he] was retreating from his conclusions,” though Coleman had said nothing of the sort. A number of professors and education experts denounced the report. One called it a “premature” report of “an ax-grinding nature.” Fumed one Harvard faculty member, “While the findings are wrapped in a mantle of social science research, the report is inconsistent with the notion of disciplined inquiry,” curiously objecting to the fact that “the findings are presented quite plainly.” Another set of critics opened their essay with: “The methods and interpretations used by [Coleman and his colleagues] fall below the minimum standards for social-scientific research.”

A good deal of the rhetoric can safely be ignored, but two criticisms were valid. (1) Students at fee-charging private schools cannot easily be compared to those attending free public schools, because they come from families who are willing to pay for their children’s education. Although Coleman III adjusted for parental education and many other family background characteristics, that adjustment did not necessarily take into account the greater educational commitment of parents who were willing to pay for their children’s education. (2) The study showed that sophomores in private school performed at a higher level, but it did not prove that they had learned more there. It was possible that the children who were being sent to private school were, to begin with, more capable students.

Coleman and his colleagues replied to these criticisms two years later when the second round of “High School and Beyond” data became available. This time, they were able to show that students in private schools had learned more between their sophomore and senior years than their counterparts in public school had. The findings calmed the skepticism of the more reasonable of their critics.

Coleman and his colleagues made some errors. They might have decided to withhold their results until they’d gathered information on student gains in achievement in high school, not just the initial sophomore scores. And they made various methodological errors, as frequently happens when one is undertaking an innovative project. But the biggest tactical errors were made by Coleman’s opponents. By relentlessly attacking Coleman III, they helped to place school choice on the national political agenda. What had been an academic debating point during the 1970s became, in the 1980s, a part of the national conversation.

Last Updated


Notify Me When Education Next

Posts a Big Story

Business + Editorial Office

Program on Education Policy and Governance
Harvard Kennedy School
79 JFK Street, Cambridge, MA 02138
Phone (617) 496-5488
Fax (617) 496-4428
Email Education_Next@hks.harvard.edu

For subscription service to the printed journal
Phone (617) 496-5488
Email subscriptions@educationnext.org

Copyright © 2024 President & Fellows of Harvard College