Higher Admissions, page 2
Left unaddressed by the Court, however, is the question posed by Lemann and further considered by Marvin Krislov in his commentary. Tests have been used as a proxy for merit. They have been believed to be a good predictor of academic success, as a recent commentary in the New York Times by columnist David Leonhardt argues,16 although studies have long shown that they better approximate first-semester college GPA than they predict leadership, creativity, or other markers of a successful collegiate experience. But as Krislov notes, they also track with socioeconomic background, blurring the lines between meritocracy and aristocracy.
This book, and the series of which it is a part, assumes that talent is distributed across the nation and world but access to opportunity is not. It invites the reader to understand the history of standardized testing and the creation of a testing industry that began with hopes of expanding opportunity and democratizing access at elite colleges, and it shows how, rather than shattering class privileges, the exams reinforced the relation between doing well on the tests and coming from families and neighborhoods with considerable resources. At its core the book asks us to think deeply about what is meant by merit. Can one test, taken over a few hours, tell us all we need to know about a potential candidate? It also cautions us that finding suitable alternatives to tests that have been validated over decades may take more than a minute. As important, it invites us to probe our commitment to equal opportunity in the United States by asking, what is the purpose of access to education? This question is always important to revisit. The answers we offer have deep importance in a world shaped by technological change, violent geopolitical conflicts, growing distrust of institutions, and an ever-widening gap between educational achievers and those who never get the chance to show their talents.
CHAPTER 1
The Birth of the American Meritocracy
IN THE SUMMER OF 1948, Henry Chauncey, the president of the brand-new Educational Testing Service (ETS), read an article that he found quite disturbing. It was in a journal called the Scientific Monthly; its title was “The Measurement of Mental Systems (Can Intelligence Be Measured?),” and its authors were W. Allison Davis, an anthropologist who was at that moment possibly the most prominent Black academic in the United States, and Robert Havighurst, a colleague of Davis’s at the University of Chicago. They argued that intelligence tests were a fraud, a way of wrapping the fortunate children of the middle and upper classes in a mantle of scientifically demonstrated superiority. The tests, they wrote, measured only “a very narrow range of mental activities” and carried “a strong cultural handicap for pupils of lower socioeconomic groups.” They assessed “academic or linguistic activities” and then submitted themselves to a kind of circular validation process, because “a teacher’s rating of a pupil is an estimate of the pupil’s performance on the same kind of problems as those in the standard tests.”1
Chauncey was the scion of a New England Puritan family—he was a direct descendant of Charles Chauncy, the second president of Harvard College—who was born in 1905, the same year that the first intelligence test was administered by Alfred Binet in Paris. As a college freshman he had taken a course on mental tests and become entranced with them, believing them to be a scientific miracle with unlimited potential to better humanity. Although he was one of the founders of what some people now call “meritocracy” in America (the word didn’t exist at the time ETS was founded), Chauncey himself had the virtues prized by a previous system of elite selection: he was tall, handsome, athletic, energetic, public spirited, optimistic, and a natural leader, but not especially intellectual. He believed that mental tests were a great technological innovation, like the telephone or the light bulb, and so could only be beneficial, but he didn’t have a particular social vision that he wanted tests to serve. He simply believed in them. Indeed, through the years of ETS’s greatest growth, driven by its devising and administering aptitude-based tests for admission to selective colleges and universities, Chauncey tried constantly to find other kinds of tests for ETS to offer, mainly without success.
As a young man, Chauncey had become an assistant dean at Harvard, a job that he came to find frustratingly minor. In 1933, he acquired a new boss: James Bryant Conant, a chemist who was Harvard’s first non–Boston Brahmin president in more than sixty years. Conant shared Chauncey’s interest in testing, but he had far more definite ideas than Chauncey did about what testing’s purpose should be. The Harvard College that Conant took over was dominated by boarding school graduates from prosperous families in the Northeast. Its admissions test was a battery of essay exams administered by the College Entrance Examination Board (known as the College Board), then a small organization that had been created in 1900 to align the curricula of a few dozen mostly private schools with a handful of elite private colleges and universities. It was quite difficult for public school students, and students from the middle of the country, even to take these exams let alone to excel on them, not only because of their content but also because the number of test sites was limited.
Conant told Chauncey that he wanted to start a new scholarship program at Harvard, aimed at students who wouldn’t ordinarily take the College Board tests. He assigned Chauncey to find a test that was as close as possible to an IQ test—that is, one that measured what Conant saw as innate intellectual ability, unconnected to the quality of the applicant’s high school education. After some searching, Chauncey met a Princeton professor named Carl Brigham, who had worked on the first mass administration of an IQ test, to U.S. Army inductees in the First World War, and had gone on to produce an adaptation of that test for college admissions purposes. Brigham called it the Scholastic Aptitude Test. He administered it experimentally for the first time in 1926. And now Harvard adopted it for its new scholarship program.
The new Harvard scholarship program was a great success by Conant’s and Chauncey’s lights. Soon Harvard was using the SAT as a selection device for all its scholarship students. Then it persuaded all the other College Board schools to use the SAT as their scholarship selection device. During the Second World War, Chauncey supervised the administration of an adapted SAT called the Army-Navy College Qualification Test—used to select soldiers for specialized, technical roles in the service—to more than three hundred thousand high school seniors on a single day. This proved it would be possible for a testing agency to scan an entire age cohort simultaneously, as a prelude to assigning them to the roles for which they were supposedly best suited. Also during the war, the College Board dropped its old essay exams and made the SAT its sole college admissions test. During the years after the war, a series of adept maneuvers by Conant and Chauncey led to the establishment of ETS in Princeton, New Jersey, to maintain the SAT and to develop a broad range of other exams for the College Board and other clients. Chauncey, in becoming ETS’s president, was able to take on a role commensurate with his ambitions.
So when he came across the article by Davis and Havighurst, he was annoyed: here was a sentiment that he hadn’t encountered during the preceding fifteen years of uninterrupted ascension for the SAT. “They take the extreme and, I believe, radical point of view that any test items showing different difficulties for different socioeconomic groups are inappropriate,” Chauncey wrote in a diary that he kept at the time. He continued: “If ability has any relation to success in life parents in upper socio-economic groups should have more ability than those in lower socio-economic groups. And if there is anything in heredity (such as tall parents having tall children) one would expect children of high socio-economic group parents to have more ability than children of low socio-economic group parents.”2
We are conditioned to think of the moment when Chauncey was writing this entry in his journal—the post–Second World War moment—as one of a great democratization of higher education, and perhaps of American life generally. The GI Bill opened up to millions of veterans who would not otherwise have gone past high school the possibility of getting college degrees. Their successful experience convinced the country that many more Americans could benefit from higher education, and this conviction underlay a historically unprecedented expansion of American colleges and universities (and, perhaps as a consequence, of the size and well-being of the American middle class). I will go into more detail about this expansion in the next chapter. For now, it’s important to keep in mind that Chauncey and Conant’s project to establish the SAT as the standard admissions device for higher education was a separate endeavor, though also a highly consequential one. It aimed to change the student population, and by extension the overall nature, of a limited number of highly selective colleges and universities. It was driven by a vision of the American future, but we should be precise about what that vision was: a more democratically selected educated elite, not greatly enhanced opportunity for the majority of Americans, and not the advancement of historically marginalized people. A great deal of trouble has come from our tendency to conflate elite selection with mass opportunity. My aim in this chapter is to disentangle them. That is a necessary precondition to thinking clearly about testing.
Chauncey, whom I got to know well in the final years of his very long life (he died in 2002, a few weeks short of his ninety-eighth birthday), would have been amazed, at least back in 1948, to learn that the objections Davis and Havighurst were raising would one day become widespread and potent, so much so as to threaten the SAT existentially. How could he have failed to see this? He and his colleagues inhabited a small, tightly enclosed world—Chauncey and many of his close friends and professional associates were fellow graduates of a single small New England boarding school, Groton—in which everything implied by the terms “diversity,” “equity,” and “inclusion,” which educators use constantly today, simply wasn’t part of the conversation. The moral wrong represented by the pervasive American system of racial rank ordering—calling attention to which was the central theme of Davis’s career—was strikingly absent from the consciousness of most of America’s elite white liberals, even at a time when the early stirrings of the civil rights movement should have been clearly visible to them. During the early years of ETS, the global competition between the United States and the Soviet Union was the overwhelming preoccupation of the American establishment, and most of its members believed that the central task of educators should be to play their part in this struggle.
That isn’t to say the SAT did not have a social vision behind it, as all widely used tests must. That vision, powerfully motivating in the minds of its early champions, is so different from the way admissions tests function now, and from the grounds on which they are defended, that it requires explanation.
Carl Brigham, the SAT’s inventor, was as a young man an enthusiastic member of the eugenics movement—which, in the early twentieth century, captured the imagination of much of the American patrician class with its lurid warnings about the country’s being overwhelmed by high-breeding, genetically inferior members of the lower orders. The eugenicists were definitely racists, though the primary targets of their racism were people who would now be considered white, like Catholics and Jews who had recently come to the United States from southern and eastern Europe. Brigham wrote a book based on the results of the U.S. Army intelligence tests, called A Study of American Intelligence, affirming these prejudices. “American intelligence is declining, and will proceed with an accelerating rate as the racial admixture becomes more and more extensive,” he wrote. “These are the plain, if somewhat ugly, facts that our study shows.”3
But, almost alone among prominent eugenicists, Brigham had a change of heart. In 1928, only two years after the debut of the SAT, he renounced his former views publicly at a convention of eugenicists, and in 1930 he issued a formal retraction of A Study of American Intelligence, calling it “pretentious” and “without foundation.”4 In 1932 he published a book pointedly titled A Study of Error. In 1935, in an unpublished manuscript, Brigham wrote, “The test movement came into this country some twenty-five or thirty years ago accompanied by one of the most glorious fallacies in the history of science, namely, that the tests measured native intelligence purely and simply without regard to training or schooling. I hope nobody believes that now. The test scores definitely are a composite including schooling, family background, familiarity with English, and everything else, relevant and irrelevant. The ‘native intelligence’ hypothesis is dead.”
One of Brigham’s complaints was that the people conducting research on intelligence testing were the same people who were promoting the idea of native intelligence as a single, all-important inherited trait. They assumed that test results measured an immutable property of the brain, and their research always confirmed their theory. The most important of these promoters was Lewis Terman, a Stanford professor who had adapted Binet’s test for American use and who devised the ubiquitous concept of IQ , or intelligence quotient. Binet had developed his test to identify children who might need special help as they entered France’s newly universal public elementary school system, so that they could succeed there. Terman had an opposite concern, with identifying young people whose scores were unusually high so that they could develop their talents through special educational opportunities. His main concern about students other than the highest scorers was that they would hold back the super-brainy by being put in classes alongside them. The formerly widespread use of IQ test–based tracking systems in American public schools owed a lot to Terman’s influence.
Conant, from what we know, was not a eugenicist, but he was a believer in Terman’s ideas about what IQ tests measured and the uses to which their results should be put. During and after his twenty-year reign as president of Harvard, he exercised a broad national influence beyond what any one university president does today; he played a large part not only in the rise of the SAT but also in the development of the atomic bomb, the creation of the National Science Foundation, and many other initiatives. So while his interest in the SAT began as part of a crusade to remake the student body of Harvard College, it quickly became part of a much larger project. Conant wanted to remake American society as a whole, and it wasn’t implausible for him to believe he could have a significant influence there.
In a series of writings during the Second World War, Conant laid out his ideas. He had two main concerns: that the country’s vast, decentralized public school system would permit the talents of potential top scientists and technocrats to go undetected and therefore unused in the great struggles of the postwar era; and that America would develop debilitating class tensions that would leave it unable to compete successfully with rivals who claimed to have classless societies. Conant certainly hadn’t dropped his concern with the composition of the student body of the university he led, but he was not, as so many people are now, primarily concerned with making admissions a fair contest so that precious slots in the Harvard student body would go to the people who most deserved them. (And anyway, Harvard at the time accepted two-thirds of its applicants.) His sights were set higher than that. In an article in the Atlantic Monthly in 1943 titled “Wanted: American Radicals,” Conant called for the creation of a new ideal American type, a kind of frontiersman of the modern age,5 who would be “a fanatical believer in equality,” committed to “wielding the axe against the root of inherited privilege.”6 What this hotblooded rhetoric obscures is that Conant’s aim was to confer large opportunities on a small group of people; the devotion to equality and hatred of privilege he envisioned pertained to the selection method, not the overall character of American society.
From the vantage point of today, it’s hard to imagine how Conant could have believed that instantiating the SAT as one of society’s basic sorting mechanisms would lead to this kind of result. It helps explain his thinking to understand that he assumed the SAT did not measure education, family background, or preparation, only innate mental ability. Therefore it would be class neutral because—another of Conant’s assumptions—the highest scorers would be scattered almost randomly around the country. Conant further assumed that whatever opportunity a very high SAT score opened up for somebody, it would not be heritable: on the model of those chosen to be guardians in Plato’s imaginary republic, that person’s children would return to the ranks of ordinary people. Unlike members of anointed elite groups in the past, they would not try to turn themselves into a hereditary aristocracy. (In his article, Conant proposed that all personal wealth be legally confiscated at death.) Conant, who had Europe’s elite universities in mind as a model, thought of the SAT as a device to select people for roles that were important but not especially remunerative. Those who got high scores and were tracked for special roles would become, essentially, public servants, working for large government agencies or universities, indifferent to money and status.
To clarify Conant’s very non-twenty-first-century idea of what constituted a radically democratized United States, it may help to consider three public campaigns he waged in the years after the war: against the GI Bill; in favor of mandatory universal national service; and in favor of the creation of large comprehensive public high schools in the burgeoning suburbs. What tied these together was Conant’s fervent belief in elite selection, combined with his equally fervent belief in national cohesion. National service and public school were meant to bring all Americans together into a shared, government-managed, character-shaping experience while they were young. But this would happen before they arrived in the higher education system, which Conant believed should be an experience for the few—the right few—and not the many. Along with other prominent presidents of private universities (Robert Maynard Hutchins of the University of Chicago wrote an article about the GI Bill called “The Threat to American Education”), he was actively opposed to the great historic expansion of higher education after the war.
