Henry Chauncey

INTRODUCTION

As industrialization permeated daily life in the United States after the Civil War, the advantages of systems were everywhere apparent. Americans had too long listened to the courtly muses of Europe, Ralph Waldo Emerson had cautioned, and must now hasten to end their long apprenticeship to learning from other lands.¹ Emerson claimed for his congregation a desire for public, civic-minded thoughts in 1837, and the national spirit he evoked was applauded.

Under the grind of the McCormick Reaper, the land fell into orderly rows. Nationalism and commerce seemed inseparable. By the time that Ulysses S. Grant finished his memoir in 1885, he acknowledged that slavery was, naturally, one cause of the War of Rebellion. He also made the point that it was not moral outrage alone that drove the republic toward suffering. Before railroads, telegraphs, and steamboats—when America was still in search of the national character defined by Emerson—each state was almost a separate nationality. “At that time,” he wrote, “the subject of slavery caused but little or no disturbance to the public mind.” “But,” he continued, “the country grew, rapid transit was established, and trade and commerce between the States got to be so much greater than before, that the power of the National government became more felt and recognized and, therefore, ← 1 | 2 → had to be enlisted in the cause of this institution.” Truth be told, he concluded, the Civil War was probably better fought sooner than later, and the nation was better off. Because of the war, Grant explained, commerce, trade, and travel with Europe had increased. Before the war, monarchial Europe believed that the republic was “a rope of sand that would part the moment the slightest strain was brought upon it.”² Grant was at one with Emerson. Now that America had proven itself formidable, he concluded, it was clear that its citizens “made more rapid progress than we otherwise should have made.” The development of a central government facilitating commerce, Grant believed, led to a national identity.

Howard Mumford Jones called the period between 1865 and 1915 the Age of Energy. The nation emerged from the Civil War, the Harvard University historian wrote, as a great military and economic power, “strong, unpredictable, and parochial.” We were led by a succession of presidents who “at the best were high-class mediocrities and at worst U.S. Grant.”³ We became industrialized. We gave birth to dreamers and reformers driven by a force that was amoral and ambiguous, proven good or bad only in application. But the verdict was not yet in on the project. Believing that immediacy must be weighed against long-run social good, Jones postponed moral judgment regarding the impact of industrialization. He stuck to the aesthetics of the Illustrated Catalogue of the Centennial Exhibition, Philadelphia, 1876 and to the poetry of Walt Whitman. He avoided the sharp edges of the relationship of energy—defined as force, or vigor—to industrialism, corporate management, and “other devices for the quick creation of wealth.”

Although he attempted to default to aesthetics, Jones would not escape so easily. He published his book in 1971, and some of his readers already had copies of Michel Foucault’s Historie de la Folie on their bookshelves.⁴ Receding in the national consciousness was the idea of an essence, whether aesthetic or metaphysical, that floated above the drag of daily life; advancing was constructivism and its sweaty commitment to interpreting everything under the sun as the ubiquitous manifestation of power. Interpretative perspective in 1970, perhaps especially of the period between 1865 and 1915, was materialist.

Whether the observer is an aesthete or constructivist, it is nevertheless true that America was reimagined during the nineteenth century. While death by guns and germs carried the horrors of colonialism abroad, it was steel that transformed the visual landscape at home. The expansion of railroads—Jones records as epic the growth from little more than 9,000 miles of ← 2 | 3 → rail in 1850 to 175,000 miles of rail by 1900—demanded a new manufacturing process.⁵ Coal-smelted iron used to fashion the rails wore out too rapidly. English inventor Henry Bessemer’s controlled combustion process yielded a more durable rail from steel and launched large-scale alloy production. In the building trades, concrete reinforced by steel allowed architects to move away from the earth-bound limits of load-bearing walls to the heaven-vaulted buildings called skyscrapers.⁶ Whether riding on the Transcontinental Railroad between Omaha and Sacramento in 1869 or entering the front door of the Produce Exchange Building in Manhattan in 1884, uniformity—the consequence of the energy driving industrialism—was around every corner.

Measured Lives

As an archetype of national progress, efficient design yielded consistent delivery in commerce as well as education. The antebellum philosophy of Horace Mann in Massachusetts, emphasizing public support for schools that would result in an informed citizenry, was supported by John Dewey in Illinois. During the period from the surrender of the Confederacy at Appomattox courthouse to the sinking of the Lusitania off the coast of Ireland, the United States emerged as a global power because of its ability to implement and deliver, in both manufacturing and education, a new form of capitalism that unified mass production and individual gain in the service of reliability and, therefore, of prediction. The logic was remorseless: If a reliable sewing machine manufactured by I. M. Singer and Company could enable a mother to make clothes for her husband and children with time-saving precision, then the future would be bright.⁷ With assembly lines made efficient in the first decades of the twentieth century by scientific management experts such as mechanical engineer Frederick W. Taylor and psychologist Hugo Münsterberg, systemization in education was the next logical step.⁸ With the inevitable embrace of the public intellectual imagined by Emerson sure to come, it could not have been otherwise. More than a movement, progressivism was a romance—exactly the self-conscious, pluralistic, comprehensive educational movement that historian of education Lawrence Cremin said it was.

While the spirit of progressive was poetic, the daily operating routine was journeyman prose with the occasional rhetorical flourish. The desire for method was becoming profoundly evident in the field of psychology. From its 1892 origin in Worcester, Massachusetts, the American Psychological Association ← 3 | 4 → served as the hub of attempts to measure the mind. At that first fin de siècle meeting were the founders: G. Stanley Hall of Clark University, George S. Fullerton of the University of Pennsylvania, Joseph Jastrow of the University of Wisconsin, William James of Harvard, George T. Ladd of Yale University, James McKeen Cattell of Columbia University, and J. Mark Baldwin of the University of Toronto. By the numbers, the new discipline was small: There were nineteen laboratories for the study of psychology and two journals, both edited by Hall. To advance psychology as an area of scientific study, the new organization met at the University of Pennsylvania on December 27, 1892, with a total of thirty-one members. By the time Samuel W. Fernberger published the first history of the organization on 1932, it had been transformed into a big business. By 1947, the organization adopted a unit-based structure to manage its members and their research specializations, with Division 5—Evaluation, Measurement, and Statistics—as one of the nineteen charter units.⁹

In the early twenty-first century, it is difficult to envision America without educational measurement. Measurement may be used as a way to understand developed ability and attitudes toward those abilities. In application, the field provides a framework for the study of biological, experimental, personality, social, developmental, clinical, health, forensic, industrial, and other applications. As Denny Borsboom has noted in his philosophical study of conceptual issues in contemporary psychometrics, teachers test students on a range of performance, parents have their children evaluated for their capability, nations have their schools examined for proficiency, and corporations and nonprofits have their personnel evaluated.¹⁰ The sheer volume of tests and their purported ability to measure psychological constructs of interest is staggering. Knockout blows are regularly delivered by controversies ranging from curricular representation in a test to consequences of test use on student learning.

Yet no history of educational measurement exists. Defined as a field dedicated to research of cognitive and personality factors associated with learning—and complete with its own classification as an instructional program code (13.0604) by the National Center for Education Statistics under the title Educational Assessment, Testing, and Measurement—educational measurement certainly deserves its own history. This absence is indeed curious, especially when we consider the role that evaluation has played in American education since the physician Joseph Meyer Rice reported on his five-month tour of public schools in 1893 and found them wanting, with some so poor as to resemble “a stage of civilization before the age of steam ← 4 | 5 → and electricity.”¹¹ The publication of The Public-School System of the United States showcased top management in action, with Rice serving as the visible hand. If we adopt the periodization offered by Alfred D. Chandler in his history of the managerial revolution in American business, we see that Rice, a self-appointed top manager, was reviewing the nation’s organizational system of education. He would give advice to determine the long-term objectives of the enterprise and how best to allocate resources, under conditions of scarcity, to meet these objectives.¹²

As Cremin has written, the nascent and disconnected efforts to estimate educational effectiveness in the 1870s and 1880s coalesced in the 1890s into a national evaluation framework. During this period—a golden age to some and a gilded one to others—the founders of the American Psychological Association published their conceptual approaches in books such as Talks to Teachers on Psychology by William James.¹³ Based on a series of lectures given at the request of the Harvard Corporation to Cambridge teachers in 1892, the book was the first of many, although the reticence of James to make grand promises for the field of psychology was unique. Educational researchers, a new breed of psychologist intent on field-testing assumptions about learning, began to promote their experimentalism through publication, as Dewey had done during his tenure at the University of Chicago between 1894 and 1904.¹⁴ The findings of such research, with scales that could be used to measure student ability across time and circumstance, appeared in the curricula of state normal schools and university departments of education. The new science of education supported the nation’s faith in efficacy leading to progress. With a birth at the beginning of the American century and a history largely unexamined in the world of ideas, here is a tale worth telling in an education biography.

Time

A global history of educational measurement could find its international roots in the civil service examinations, high-stakes testing situations that identified individuals holding skills important to the imperial bureaucracy, given during the Ming (1368–1644) and Ch’ing (1644–1911) periods. As Benjamin A. Elman has documented, thousands of examination candidates congregated biennially in counties and triennially at national capitals for these examinations, and their social, economic, cultural, and political impact were substantial. Books of examination essays appeared, along with charms from temples for good ← 5 | 6 → luck. “Distrusted by the dynasty, which created and maintained an architecture of surveillance housing thousands of guards in the examination compound in addition to the candidates and their examiners,” Elman writes, “the candidates themselves were a hodgepodge of high and low, young sons of the famous and old men on their last try, savvy urban southerners and country bumpkins from the northern small towns and villages. What they shared was years of preparation to compete for the few places that would separate them in their disparate futures. Success was alluring; failure was humiliating.”¹⁵ If a history of educational measurement were to begin in a search for international origins, such a study would perhaps focus on the measurement practices of these examinations, on the way the examiners designed the tests to place emphasis on the mastery of a common classical language by the candidates, including a memorized canon. Evolution of the examination would focus on the prescribed writing form, the Eight-Legged Essay, begun in the early Ming dynasty—a highly stylized writing genre in which the writer broke open the topic, received it, began the discussion, proceeded with the initial, transition, middle, and latter leg, and concluded the exposition.¹⁶ That history of educational measurement would certainly focus on the endurance of this expository form and its ability to capture expertise, with attention to the assessment of the essay by examiners who literally counted the number of characters, syntactic development (with special emphasis on parallelism), and main ideas. Attention would also be placed on the impulse to modernize the nation in 1904 as the government ended five hundred years of an empire-wide examination system in order to achieve Westernization throughout the nation, subordinating, as Elman notes, the old examination to new forms of education requiring public school and college entrance examinations.

Focusing on the Western tradition, a history of educational measurement would certainly highlight the philosophical assumptions and methodological practices of British psychologist Charles Spearman. In two seminal articles published in The American Journal of Psychology in 1904 as England advanced modernization through the example of science, Spearman originated both the idea of a singular “intellective saturation” and a means to evaluate it. His concept of general intelligence, and his measurement of the factors comprising it, supported the same social aim of modernism that decanonized the Chinese rhetorical form begun in the fourteenth century. The identification of general intelligence, Spearman noted, would support “the long awaited general rational basis for public examinations. Instead of continuing ineffectively to protest that high marks in Greek syntax are no ← 6 | 7 → test as to the capacity of men to command troops or to administer provinces, we shall at last actually determine the precise accuracy of the various means of measuring General Intelligence, and then we shall in an equally positive objective manner ascertain the exact relative importance of this General Intelligence as compared with the other characteristics desirable for the particular post which a candidate is to assume.”¹⁷ In Spearman is found the brash language that reflected a movement against convention evident in the Demoiselles d’Avignon by the young Spanish painter Pablo Picasso in 1907 and in the model of atomic structure proposed by Danish physicist Niels Bohr in 1913. Modernism was a calculated offense against conventionality, as Peter Gay defined its chief attribute.¹⁸ Part of a new cultural movement supporting progress through rigorous query of convention, intellectuals such as Spearman desired to make the world new.

Narrowing the focus to the United States, a history of educational measurement would certainly narrate the great modernist systemization during World War I. While that story has been partially told by Stephen J. Gould, the firsthand account by Robert M. Yerkes, Psychological Examining in the United States Army, remains only selectively explored.¹⁹ In that eight-hundred-and-ninety-page memoir, we find a record of the first large-scale effort to measure individual differences in the nation’s history. Nearly everyone who would take a leadership role in psychological measurement would become part of the war effort. The new president of the American Psychological Association, Yerkes, caught the attention of the Surgeon General of the Army, William C. Gorgas. As a leader in the battle against malaria during the building of the Panama Canal, Gorgas had come to appreciate the value of scientific method and the knowledge it yielded. With Stanford University’s Lewis M. Terman, the psychologist most responsible for ushering the concept of general intelligence into America in his revision of the mental measurement test developed by Alfred Benet and Theodore Simon, Yerkes produced examinations to test the intelligence of both literate and illiterate recruits. Examining some 200,000 men a month during its height, the testing program administered the Army Alpha and Beta tests to 1,750, 000 recruits. The Committee on Classification of Personnel, with Yerkes as leader, grouped and distributed soldiers through test results so that each company would have its share of superior, average, and inferior men. Daniel T. Kevles, who remains the best historian of the role of psychologists during World War I, concludes that the widespread use of examinations during the war dramatized intelligence testing and, in the process, incorporated psychological measurement into mainstream American culture.²⁰ ← 7 | 8 →

With the publication of Carl Campbell Brigham’s A Study of American Intelligence, the resistance began. Educated at Princeton University and a lieutenant in Yerkes’s First Company of Commissioned Psychologists during the war, Brigham brought forward selected findings from Psychological Examining in the United States Army to argue that the average intelligence of immigrants was declining. An empirical justification of the superiority of the Nordic race and its debilitation through intermarriage with inferior races, Brigham’s 1923 Princeton University Press book was praised by Yerkes as “a notable service to psychologists, to sociology, and above all to our law-makers” who would now understand the significance of eugenics in the way desired by Sir Francis Galton: “the science which deals with all the influences that improve the inborn qualities of a race.”²¹ In the pages of The Century Magazine, the journalist Walter Lippmann, a Harvard graduate who led the protest, made it clear that Brigham was out to “prove that the foreign born who have come in the last twenty years are an inferior stock.” “Apparently,” Lippmann wryly observed, “it never occurred to him, however, to compare scores in the intelligence tests with the percentage of foreign born in the different states.” The tests did not measure native ability; the results simply reflected the time immigrants had been in the country. Sensing that the variables of literacy and culture had haunted the entire examination process of Army recruits, Lippmann drenched his analysis in irony: “If anyone thinks this ascending curve of the scores of the foreign born in accordance with the time spent in our American environment is a sign of what we call roughly Americanization, he has not reckoned with the Psychological Battalion of Death.”²² The phrase stuck.

On the scientific front, a former fellow officer, E. G. Boring, also distanced himself from Brigham’s book in the pages of The New Republic. It is unsettling, the Harvard associate professor wrote, that Brigham discarded the effects of knowledge of English in stating his differences among immigrant groups. The judgment was harsh: “The trouble is with the data,” and hence “we are by no means ready definitely to recommend legislation.” The mountain of Army statistics assembled by Yerkes, Terman, Brigham, and Boring himself was simply not collected under scientific conditions. “That in this case the mountain could bring forth only a timid mouse may be due to the fact that mountains for all their size do not necessarily have leviathans in them.”²³ Everybody could write.

Rhetorical flourishes did little to stop the momentum, however, and educational measurement became part of the American landscape after the Armistice. While we may wonder whether World War I and its aftermath was a success story for empirical science or a horror story of social impact, it is clear ← 8 | 9 → that measurement had an impact on the nation’s educational system. Brigham himself became the key researcher in the college admission program conducted by the College Entrance Examination Board and ushered in the age of multiple-choice testing with the Scholastic Aptitude Test, first administered on June 23, 1926, to 8,040 candidates.²⁴ While Spearman may have inferred a relationship between psychological measurement (termed “correlational psychology” in his 1904 article on general intelligence) and testing (understood as the present efficiency of examinations),²⁵ it was Brigham who first brought the core values of early twentieth-century psychological measurement—rigor of method resulting in empirically justified results—into education.

Since its founding in 1900, suspicions had surrounded the content of the College Board examinations, especially their adherence to a British literacy canon as Mandarin, some felt, as the content mastery of Chinese literature that had driven the Eight-Legged Essay. In the first instance of an independent researcher attacking a large-scale testing consortium, Edward L. Thorndike had publically criticized the College Board examinations a 1906 issue of Educational Review. From his post at Teachers College, where he would train the new superintendents required to systematize elementary and secondary education according to scientific principles, he reported in 1906 that the tests were the very embodiment of a range of error, from their flawed design (resulting in a failure to sort for admissions) to the unintended consequences of their use (causing an “intolerable injustice” to the individual student).²⁶ Such criticisms were obviated, at least in the beginning, by the Scholastic Aptitude Test in its ever-efficient design. With nine subtests in verbal and mathematical content, the test was radically dissimilar from those historically used by the Board. Absent were essay questions scored by readers who often failed to achieve reliability in their judgment of the examination booklets; present were multiple choice questions scored mechanically by clerks.

Standardized in design and yielding efficient analysis, the new limited response tests performed so admirably for eighteen years that, following the bombing of Pearl Harbor on December 7, 1941, leaders of the College Board pressed for their adoption.²⁷ As the use of essay tests declined, the rise of objective tests was rapid. Just before the war, in June of 1941, 12,455 students wrote detailed answers in pursuit of admission to college; by April of 1942, 16,626 students had taken the Scholastic Aptitude Test. The cost for reading the answer booklets in 1941 had been $40,744.96, while the cost of scoring the multiple choice examinations had been only $7,659.86.²⁸ The promise of the psychological examination used in World War I—methodological rigor ← 9 | 10 → yielding empirically justified results—had now been realized in an educational setting. Part of the field of educational measurement, large-scale admissions testing had begun.

By the end of World War II, the business of testing was becoming aligned with the new vision of science that had won the war. The depiction of science as a uniform process had changed. In 1910, John Dewey had offered five distinct steps in How We Think: observation of a problem, identification of its location and definition, proposal for a possible solution, development by reason of that possible solution, and further observation leading to rejection or acceptance.²⁹ Following World War II, that process seemed terribly reductionist. In similar fashion, receding was the image of the isolated intellectual alone in his workshop—the shadow of Thomas Edison prowling around Menlo Park to create lightbulbs. Men like Harvard University president James B. Conant, the public intellectual imagined by Emerson, had been placed in charge of government supported war efforts. In 1940, Conant had been appointed to the newly formed National Defense Research Committee, later the Office of Scientific Research and Development, created to mobilize the nation’s contributions to military research. He had been appointed to that post by Vannevar Bush, a former engineer at Massachusetts Institute of Technology who would, with Conant, coordinate the Manhattan Project. At Los Alamos, under the direction of J. Robert Oppenheimer, 4,000 civilians and 2,000 men in uniform would develop the atomic bomb at a cost of $143.7 million.³⁰ As work at Los Alamos had demonstrated, science was not a simple, staged event. Science was a big, phased process.

In 1951 James B. Conant attacked the received view of the scientific method in Science and Common Sense. In his review of the book, Vannevar Bush, president of the Carnegie Institution for Science, praised Conant for making it “crystal clear that there is no such thing as the scientific method.”³¹ There could be no doubt that Dewey and his classroom-based worldview of science was the target here. As John W. Rudolph keenly observed, those who had experienced the large-scale research and development projects in the United States from 1941 to 1945 found the facile, five-step method “difficult to swallow as an accurate picture of their work,” especially at a time when the public support of scientific research was viewed a crucial tool in the Cold War battle against communism.³² Books such as Bush’s Science—The Endless Frontier in 1945 and the founding of the National Science Foundation in 1950 attested to the rejection of an elementalist depiction of science.³³ Required was a vision, faceted and nuanced, to inspire the postwar world. Required was a new theory that embraced complexity. ← 10 | 11 →

And so it was that, in 1955, educational psychologist Lee J. Cronbach and philosopher of science Paul E. Meehl reconceptualized psychological measurement in general, and educational measurement in particular, to reflect a postwar vision of complexity. At the center of their new system was the construct, “a postulated attribute of people, assumed to be reflected in test performance.” “In test validation,” they further explained, “the attribute about which we make statements in interpreting a test is a construct. We expect a person at any time to possess or not possess a qualitative attribute (amnesia) or structure, or to possess some degree of a quantitative attribute (cheerfulness). A construct has certain associated meanings carried in statements of this general character: Persons who possess this attribute will, in situation X, act in manner Y (with a stated probability). The logic of construct validation is invoked whether the construct is highly systematized or loose, used in ramified theory or a few simple propositions, used in absolute prepositions or probability statements. We seek to specify how one is to defend a proposed interpretation of a test: we are not recommending any one type of interpretation.”³⁴ In a single passage, Cronbach and Meehl had formulated an agenda for measurement that, in its quest for authentication, would fuel the measurement project. The object of inquiry was a postulated attribute that might not be observed—would, in fact, float just above the routine of daily observation—unless glimpsed in test performance. Shadows cast on wall, these attributes would be studied by scientists to confirm the construct. With sufficient attributes confirmed over time by empirical methods, the existence of the construct, with all its complexities, might be brought out of the shadows. Because evidence of the construct was conveyed in a proposed interpretation, revelation was itself complex: Construct identification was intertwined with the proposed interpretation of the test results. Further, there was no one type of interpretation. The shadows were endless. The pursuit of construct validity, faceted and nuanced, was a dream come true.

Here was a system of inquiry worthy of the National Science Foundation itself. By the following year, the first grant for basic research was given to Stanford University’s Leon Festinger for his experimental studies in cognitive dissonance, a construct based on the desire to resolve conflict and the attributes involved in doing so. As Otto N. Larsen observed in his history of the social sciences at the National Science Foundation, the social sciences—psychology and measurement among them—had earned a place, although a modest one, in the big science agendas of the middle of the twentieth century.³⁵ ← 11 | 12 →

The study of constructs would serve educational measurement well, and the related area of prediction would be the foundation for much of educational measurement in the second half of the twentieth century. Meehl had claimed in 1954—and was still demonstrating the endurance of his claim in 1986—that statistical prediction was preferable to clinical judgment. As Meehl noted in a telling example, if a four-variable regression equation or a Glueck actuarial table tells the criminal court judge that a delinquent will probably commit a felony in the next three years, and if a social worker says that he will probably not, a controversy exists.³⁶ Statistical prediction, the key tool of validity, should be used to make behavioral predictions. Without a review of the literature informing judgment, ideographic analysis was invalid.

As it had earlier in the century, debate would continue stemming from the mid-century publications of Cronbach and Meehl regarding the nature of construct validity and capability of statistical prediction. Those calculating actuarial tables clashed with clinicians who spent their lives with those in the sample. Many felt that the Laplace demon of 1814 had been unleashed.³⁷ The dream of the nineteenth century had come to pass. Given for one instant an intelligence which could comprehend all the forces by which nature is animated and the respective situations of all who populate it, it would seem almost natural that self-same intelligence would embrace in a formula the movements of all creation. For such singular intelligence, nothing would be uncertain. The future, as the past, would be present to its eyes.

By 1970, educational measurement, and its particular use in testing, had become for many an embodiment of the Laplace demon. In 2013, the unleashed demon continues to prowl the pages of Bruce D. Baker’s blog, School Finance 101. Posts with titles such as “Gaming Adequacy by Creating a Veneer of Empirical Validity” attest to the enduring distrust of statistical surrogation.³⁸

Theory

Narrative in design, this biography acknowledges the presence of demons in the history it presents. More expository than argumentative, however, mine is a project designed to write back into history the life of a balanced man—someone who leveraged Cronbach’s contributions to empiricism and who would agree with Baker that those who use statistical techniques to justify predetermined agendas are scoundrels. ← 12 | 13 →

In order to write fairly, I have been deeply mindful of the role of theory in histories of education. A 2011 special issue of History of Education Quarterly was devoted to just that topic. Identification of theory, as the contributors demonstrate, is helpful in allowing differentiation of identities, actions, events, and rationales. Marxist, feminist, critical race, queer, and social constructivist theories were identified in the special issue, as well as postmodern, poststructural, and postcolonial theories.³⁹ Absent from the list was genre theory.