Last week, University of California (UC) provost Michael Brown brought to a close not only a two-year but a five decade “experiment” into the use of aptitude and achievement testing in the admission process. According to the Public Policy and Higher Education Center at UC Berkeley, in 1960 the university began “a series of experiments” to determine whether “(1) the test improved prediction of freshman grades; (2) it could be used to assess grade inflation; and (3) it could be used to help manage enrollment growth.” After 50 years, the University seems to have concluded that the answer to at least two of those is no.
There is no question that the test has been useful in restricting and managing enrollment, one of the initial goals defined by the 1960s experiment, but this decision reinforces that the testing experiment hasn’t achieved the other goals. The use of testing didn’t do much to improve prediction of first year GPA. The use of testing hasn’t substantially helped assess grade inflation. The use of testing hasn’t furthered the university’s mission of providing an equal chance at a quality education for all students.
During this admissions testing experiment period, the importance of the tests have varied. From 1968 to 1977, the SAT and three SAT Subject Tests (then called Achievement Tests) were irrelevant for any applicant with a GPA above a 3.1 but required for those with a lower GPAs. This policy gave way to an eligibility index which had a sliding requirement for those with GPAs between 2.78 and 3.29. In 1992, the Subject Tests were removed from the eligibility index. Subject Tests were added back in 1999, this time for all applicants, and weighed twice as heavily in the index as the SAT or ACT.
The constant changes to not only the eligibility index but also to the tests themselves, which had at least 5 major revisions since 1968, support the decision to conclude that the experiment has been tried . . . and failed. No matter what version or what weighting is given the test, the most consistent benefit has been for the children of the wealthy, college educated parents. California public universities overall do a better than average job admitting and enrolling low income, first generation, and under-represented students, however those students are consistently channelled into the least resourced campuses of the system. First generation college students comprise 74% of UC Merced students while UCLA, which rejects 88% of its applicants, is only 31% first gen. UC Berkeley and UCLA enroll the lowest percent of Pell-eligible students, while UC Riverside and Davis enroll the highest. This outcome seems antithetical to a public institution focused on creating opportunities for all Californians. Ending the use of testing will not only reduce one obstacle to equitable access but will influence students and colleges around the country.
Since college admission policy drives student application and preparation behavior, this change will likely reduce the number of SAT/ACT test takers each year. California is the most populous state, exports the largest number of college students, and accounts for about one-eighth of SAT test-takers each year. If a significant number of Californians stop taking the SAT and ACT at all, then colleges who rely on enrolling those students will have to make adjustments.
The UCs testing policy has long been a driver of industry behavior. The adoption of the SAT by the system in 1960 was followed by other public universities. In 2001, when then-President Richard Atkinson proposed that the UCs no longer require the ACT or SAT, the College Board responded by redesigning the test. That time the tests earned a reprieve. This time, the UCs have not just indicted the SAT and ACT but admission testing in general. If the most influential public university system not only found the SAT and ACT lacking but also did not see any viable alternative test, what universities will be able to logically justify requiring them?
The response from those who believe in testing has already followed the predictable course. Cries of the “death of merit” and claims that the “fairest” part (though there is objectively proof of this) of the admissions process is being eliminated have started. Those who would have benefited from testing and test preparation will argue that removing one of thirteen factors from the UCs admissions review will disadvantage low-income, black, and Latinx students, groups who have never reaped the benefits that testing provides.
The publisher of the ACT responded to the announcement by stating, “Abandoning the use of longstanding, trustworthy, and objective assessments like the ACT test introduces greater subjectivity and uncertainty into the admissions process.” The test publisher claimed that not using their test will “worsen entrenched inequalities and dim the prospects of students from underrepresented populations,” contrary to their own research that shows that those groups benefit the least from testing.
Despite the protests of those who believe in tests or have vested interests in propping up testing, the last year has given early indications of the benefit of removing the tests. In California and across the country, many colleges that became test-optional or test-free in the last year saw their largest, most diverse, and most qualified applicant pool.
Fifty years ago, the UCs began an experiment and after all that time doesn’t have definitive proof that testing has added valuable information to the prediction of who’ll succeed. Twenty years ago, Dr. Atkinson said, ”I concluded what many others have concluded — that America’s overemphasis on the SAT is compromising our educational system.”
Dr. Atkinson added that the SAT was ”not compatible with the American view on how merit should be defined and opportunities distributed.”
Ending the SAT experiment is not only the right thing, it’s long overdue.