The Origins and Evolution of the SAT
Although the first SAT test was administered in 1926, if you are looking for the roots of the test, you have to look all the way back to the beginning of the 20th Century to the development of the first standardized intelligence tests. Alfred Binet, a French psychologist, teamed up with French physician Theodore Simon to develop an instrument that would give psychologists and educators the ability to assess mental retardation among French School children. Between 1905 and 1908 Binet and Simon worked in concert to develop an assessment that would allow them to quickly and effectively compare the different levels of psychological and cognitive functioning between people. Using a series of increasingly difficult questions, the Binet-Simon tests were used to measure attention, memory and verbal skills. In 1916, Stanford University psychologist Lewis Terman integrated the findings of the Binet-Simon research and released the “Stanford Revision of the Binet-Simon Scale” AKA the “Stanford-Binet” IQ test which today is on its 5th revision and remains a popular assessment in the field of Intelligence Testing.
These early IQ tests gained the attention of the US Military, which saw in them a means to rapidly evaluate and effectively classify the abilities of its broad applicant pool. One of the first enlisted men to administer these Army IQ tests, Carl Campbell Brigham, became the creator of the SAT. When Brigham returned from WWI, he returned to the safe-haven of his Alma Mater to serve as a professor of psychology at Princeton College. At Princeton, Brigham delved further into IQ testing and authored A Study of American Intelligence. In addition to his faculty duties, Brigham was asked to serve on the admissions committee for Princeton College. On this committee Brigham came in contact with members of the College Board, the association responsible for administering the then lengthy, essay-based admission tests known as the “college boards.”
The College Board had been in existence since 1900 when a group of elite private prep schools banded together to cement the marriage between the prestigious Northeastern prep schools and the Ivy League colleges. Boarding schools such as Andover and Exeter wanted a uniform admissions test that the colleges would honor and the Ivies wanted to admit students who had achieved a certain level of academic competency. The “college boards” were designed to do just this.
While serving on the Princeton admissions committee, Brigham was greatly turned off by the inherent subjectivity and likely inaccuracy of the essay-based “college boards;” he was confident he could design a more efficient and more objective test using his psychometric testing experience he had gained during the First World War. Brigham modified the army IQ test to assess the abilities of a collegiate-bound applicant pool. Like the Army IQ tests, his new test, the SAT, relied heavily on vocabulary and word familiarity through the use of synonyms, antonyms and analogies; it also included shape identification and recognition of facial expressions. On June 23, 1926, under the auspices of the College Board, Brigham administered the first ever SAT test to 8,040 high school students. During the first 6-7 years of its administration, the SAT tests were not used to determine admissions. Rather, these early tests were used to determine the validity of the SAT, to help better understand the correlation of the SAT to academic performance as measured by Freshman GPA. By 1933, Brigham had amassed enough data to make the case that the SAT was a useful predictor of academic performance and a valuable tool for university admissions departments. Enter Harvey Chauncey.
Though Brigham invented the test, Harvey Chauncey would inextricably link himself to this test and would transform the SAT into the academic force that it has become today. The son of an Episcopal minister, Chauncey graduated from Harvard College and eventually became the Assistant Dean at Harvard. During his tenure as Assistant Dean, Chauncey grew close to the president of Harvard College, James Bryant Conant. Conant had a vision of transforming Harvard into a more meritocratic institution–one guided by the principles of Thomas Jefferson’s “natural aristocracy.” Conant looked to his young Assistant Dean to take the first steps towards achieving this Jeffersonian vision through the creation of a new merit-based scholarship program.
Conant charged Chauncey with the task of developing a more appropriate assessment tool for his new merit-based scholarship. The “college boards,” though useful for distinguishing potentially successful students among the elite academies, would be far less effective in identifying talented students in a mid-western public school setting. Familiar with Brigham’s experimental testing, Chauncey saw the SAT as the perfect mechanism to identify potential scholarship applicants from non-New England settings. Chauncey took the fledgling SAT and adapted the test to emphasize verbal and mathematical abilities.
In 1934 Chauncey administered the reconfigured SAT to a pool of potential scholarship candidates from mid-western public high schools. Using the SAT, high school transcripts and recommendations, Chauncey selected 10 young men to attend Harvard for the class of 1938. Eight of these ten Harvard National Scholarship recipients would go on to graduate Phi Beta Kappa. In 1937 Harvey Chauncey convinced the other member institutions of the Ivy League to use the SAT as the admissions device for all prospective scholarship students. In April 24, 1937 the Scholarship Examinations were administered at 150 sites all over the country to 2,005 high school seniors. After the success of the SAT and the Scholarship Examinations, in 1941 Conant agreed to make the SAT an admissions requirement for all Harvard College applicants.
Harvey Chauncey’s lucky stars were shining when the Army and Navy contracted him to conduct a standardized test to 300,000 new recruits to help with officer selection during the Second World War. On April 2, 1943 the Army-Navy Qualification Test, adapted from the SAT, was administered to 316,000 high school seniors. This was a revolutionary event: a true logistical triumph. This assessment involved 40 times the number of members as had any other test in the history of standardized testing. The era of national standardized testing was born.
Through much political maneuvering, Harvey Chauncey, with the support of Conant and the College Board, set off to create a new institution to oversee all standardized testing in the country. With the financial backing of the Carnegie Foundation, the Educational Testing Service was officially chartered on December 18, 1947. Conant was appointed chairman of the board, and Chauncey was appointed the president of ETS. The College Board, in a surprising move, funded the new ETS and gave it exclusive rights to administer its standardized tests. In exchange, the College Board would receive a cut from every test administered. ETS would administer the tests at a national level, the College Board would get kickbacks, and Chauncey would oversee the testing monopoly of the US.
The ETS opened its doors in Princeton, New Jersey, and within a year, opened a satellite branch in Berkeley, CA. The ETS invested deeply into new test development, releasing the LSAT in 1948 and the MCAT in 1949. ETS’ inception was timed perfectly with the dramatic increase in admissions for American higher education. Post WWII the Army GI Bill went into effect, and President Truman, under the advice of the George Zook Commission on Higher Education, committed to greatly expand the American University system. Higher Education enrollment skyrocketed from 2 million students in 1951 to 3 million students in 1957 to 4 million students in 1961. ETS was becoming more and more entrenched in the collegiate admissions process. During the1950s 100 colleges made the SAT an admission requirement, and 1957 was the first year that half a million students took the SAT. It was also the first year students were informed of their scores as the veil of secrecy shrouding ETS lifted ever so slightly.
As more colleges considered using the SAT in their admissions decisions, the first thing these schools wanted to know was: how useful is this test as a predictor of academic performance? Proving test validity, that the test assesses what it is supposed to assess, was, and remains, one of the principal tasks of ETS. The predictive validity of the old “college boards” was somewhere around .20. The predictive validity of the SAT is around .4 (indicating a 40% correlation between SAT and freshman GPA). When you add in high school GPA, the validity coefficient rises to .5, indicating a 50% chance of predicting freshman GPA from SAT plus high school GPA. We now know that the SAT correlates as strongly with parental income as with freshman GPA, but ETS tends to downplay this information.
Another meaningful measure of any assessment is its reliability. How likely will a student perform consistently on multiple iterations of the same test? It turns out the SAT is an incredibly reliable assessment with a reliability coefficient of over .9.
As more and more colleges began to accept and even require the SAT, another force was stirring in the academic community. Many people questioned the premise that an IQ test was the best way to determine who deserved admission into college. They felt that academic achievement in high school was a more useful metric than any “innate” ability or aptitude predicted by the SAT. One of the greatest minds in testing, a psychology professor at the University of Iowa, E.F. Lindquist, was pursuing a different track from ETS. Lindquist created a contest in 1929 called the Iowa Academic Meet, which eventually led to the creation of the Iowa Every Pupil tests in1931. The “Iowa tests” were soon administered to public school students across the country and continue to play a major role in academic assessment to this day.
Lindquist believed in the Iowa model for testing and vehemently opposed the elitist nature of the SAT. Lindquist rejected Chauncey’s job offers to work for ETS, and in 1959 planted his own flag on the testing field, creating his own organization: American College Testing. The ACT, based in Iowa, would take on a different market than the ETS. The ETS had a monopoly in the East and a strong stake in California. ACT positioned itself with the non-selective public universities of the mid-west. ETS was rooted in IQ testing; ACT had its foundations in Iowa standardized public school achievement testing. ETS focused on the elite; the ACT focused on the masses. The battle lines were drawn, and the marketing wars began as the two companies fought, school by school, to achieve testing dominance. Another test company of merit joined the fray: the National Merit Scholarship Corporation. John Stalnaker, Chauncey’s old boss from the College Board, headed up the National Merit Corp. He contracted ETS to develop and eventually administer the National Merit Test, which ultimately became tied to the PSAT test.
While other rivals were jumping into the lucrative testing administration business, a young Stanley Kaplan from Flatbush (Brooklyn), New York saw a phenomenal opportunity emerging. In 1938, Stanley K. opened the first SAT prep class in his basement in Brooklyn. After each test, Kaplan threw a party with root beer and kosher franks and asked each one of his students to recall a particular SAT question that he or she had memorized. Within a few years, Kaplan had amassed a substantial battery of previously administered SAT questions, and these formed the core of his curriculum. The ETS became aware of Stanley’s growing operation and investigated, unsuccessfully, legal means of putting him out of business. Kaplan’s fame grew, and students flocked to his operation from around the globe. In the early 1970s Kaplan opened satellite offices outside of New York, and by 1975, he had opened 70 centers. In 1985 Kaplan sold his business to the Washington Post, who would turn Kaplan into an international Brand.
As the ETS grew, Chauncey always kept his eye on the California University System. Chauncey understood how much revenue an exclusive contract with the California system would bring his company. Chauncey opened his first branch office in Berkeley, CA for this reason. In 1958 ETS offered to run all SAT tests for the California system completely free of charge. By 1967, the University of California required all of its applicants to take the SAT. The University of California became ETS’ largest and most profitable client.
When Chauncey retired in 1970, he had built ETS into an empire. Today ETS employs more than 2,000 employees, and administers over 11 million tests each year in 180 countries.
But by the time Chauncey left ETS, the organization had already begun to draw heavy criticism from its many detractors. In the 1950’s and 1960’s the “non-profit” status of ETS had come under investigation by the IRS. Beyond the economics of ETS, many began to cry fowl at the very principles underlying the SAT. In 1962, Banesh Hoffman, unleashed a pointed criticism of aptitude and achievement tests in his book, the Tyranny of Testing. Minority groups and feminist organizations were quick to point to the disparities in SAT scores rising from gender and race. In 1974 Ralph Nader, corporate-abuse crusader, and Allen Nairn began a full-scale investigation of the ETS. Taking Nader’s lead, in 1975 the Federal Trade Commission opened an investigation into the ETS. The FTC found that, to the chagrin of the ETS, against its fiercest howls of dissent and protest, test-preparation does in fact raise SAT scores. Allen Nairn’s Report The Reign of ETS attacked the organization from every possible angle. The “truth in testing” movement sprung up across the country, and legislation was brought against the ETS in 37 states.
New York pushed the “truth in testing” movement through the state legislature, and in 1979 the Truth in Testing Law passed, mandating that ETS release old copies of SAT tests. The Legislature determined that people taking Kaplan courses had an unfair advantage on the SAT with access to Stanley K’s bank of old SAT questions. They found that the only way to level the playing field would be to allow everyone to have equal access. Additionally, the Truth in Testing Law demanded that customers have the right to receive score reports and have a breakdown of their performance. Stanley Kaplan was vindicated, and the test prep business was given the green light. In 1981 with all the ire aimed at ETS and the smashing success of Stanley Kaplan’s organization, a young Princeton student, John Katzman, opened a small tutoring shop in Manhattan called the Princeton Review. Katzman’s shop took an aggressive anti-ETS stance, riding the wave of anger at the SAT. He lambasted the ETS as a greedy, antiquated organization, and marketed his courses as a means to “stick it to the man.” Young people loved the anti-establishment message. They took on the task of thumbing their nose at the ETS, and “beating” the test using clever tricks and techniques. Eventually test prep shops began popping all over the landscape as colleges grew more expensive and simultaneously more competitive. Suddenly the SAT was more important than anyone would have previously imagined.
The SAT was never meant to be a fixed or static entity. Throughout the years it has undergone numerous changes. In 1994 antonyms and synonyms were stricken from the test. And in 1994 the test was officially “recentered:” the mean score was adjusted to reflect the scores of the current population of students. The SAT was previously “normed” in the 1960’s, and the mean score was set at 1000 (to populate the normal curve). As raw and scaled scores dropped over several decades, the test had to be recalibrated or “recentered” creating a new relationship between the raw scores and the scaled scores. Scores were effectively inflated. While a perfect 1600 was a true rarity before 1994, after the recentering a student hitting a perfect 1600 was no longer a source of surprise. Most parents who took the SAT prior to 1994 would score 70 to 100 points higher on the post 1994 SAT.
The next major change to the SAT was driven by the largest and most profitable client in ETS’ portfolio: the University of California system. Trouble was brewing in the 1990s: the validity of the test was again under fire, and the preparedness of high school students was in question. New studies revealed that the SAT was a weaker predictor of collegiate performance than the SAT 2 subject tests. SAT 2 + GPA was the best means of determining Freshman GPA. The question was raised: “so why do we need the SAT 1 at all?” The University of California president, Richard C. Atkinson, lay down the gauntlet. He gave ETS an ultimatum: reform this test or we will no longer require it for admissions. The College Board flipped, and rightfully so. It could not afford to lose an account as profitable as the University of California system. Meetings were called, committees were formed, and investigations began.
By 2004 the groundwork for the new SAT was laid. The new SAT would absorb the SAT 2 Writing test (including grammar and the writing sample), add harder math and more reading comprehension, and get rid of the quantitative comparisons and analogies.
On March 12, 2005 the new SAT was administered nationally, Writing section and all. Though colleges continue to embrace the time-honored Math and Verbal sections of the SAT, many colleges are giving a lukewarm reception to the new Writing section. They are not certain if the new Writing section will add any predictive validity to their admissions’ algorithm. Will grammar and speedy essay writing have any relevance to performing well in college? Is the skill too narrowly defined? Is it a good practice to completely disregard factual errors in the essay? Is the Writing test too easily coached and will therefore lead to a greater disparity between the haves and have-nots? Will it punish those from lower Socio-economic-Status?
Many schools are refusing to even look at the new Writing test: University of Chicago, MIT, GA Tech to name a few. A study by Kaplan indicates that 47% of schools plan to discount the Writing score entirely for the next few years. But that leaves a good portion of schools that will look at the new Writing scores. 22% of schools, especially the more competitive schools, will use the Writing section in the admissions equation, as they have used the SAT 2 subject tests in the equation for several decades. Schools in this category include U Penn, UVA, Stanford, University of Wisconsin, Madison, Dartmouth, etc? These schools will count the Writing section, but will assign it less weight than the Critical Reading and Math sections. Some schools will only use the Writing score as a “tie-breaker.” Other schools will use the Writing assessment to sniff out students who may not have written their own college admission essays or personal statements.
The ACT, in a predictable competitive response, rolled out its own optional essay at the end of each ACT administration. An ACT survey reveals that 20% of colleges require the writing section, 20% recommend it but don’t require it, and 60% neither require nor recommend it.
With all of these changes, the two competing assessments resemble each other more than ever, and they are becoming increasingly interchangeable in the world of admissions.
In 2004, 1.48 million students took the SAT and 1.19 million students took the ACT. There was some speculation that the changes in the SAT would push more students to sit for the ACT, but we will have to wait to get the new numbers for 2005.
For better or for worse Harvey Chauncey’s little test has transformed the landscape of academia, and some would say our society. There has always been and will always be controversy surrounding this assessment. It has advantages for educators, primarily cost and efficiency, but it seems to widen the divide between the “haves” and the “have-nots” of our society. The fact that the SAT is ultimately coachable violates the essential meritocratic principles of the test. Eventually we will find a superior instrument and a more valid predictor of future academic performance. Until that day the key is to be informed, be prepared, and keep this little test in proper perspective.
- Lemann, Nicholas. The Big Test: The Secret of the American Meritocracy
- Atlanta Journal Constitution
- New York Times