The University of California has decided to stop using tests to decide who gets into college. That is a bad idea
On May 21st, 2020, the University of California (UC) Office of the President issued a press release announcing that UC, one of the largest public universities in the US, would no longer require applicants to submit SAT or ACT scores to be considered for admission. UC said that fall enrollment in 2021 and 2022 would be possible without taking a test. Starting in the fall of 2023 and 2024, UC would be "test blind." This means that neither the SAT nor the ACT would be taken into account in the admissions process. However, the scores from these tests could still be used for things like course placement and figuring out scholarships after a student had been accepted. UC also said it would "begin a process to find or make a new test that matches the content UC expects students to have mastered to show that California freshmen are ready for college."
By becoming "test optional" and then "test blind," UC has joined a growing number of US colleges and universities that do not require SAT or ACT scores for admission. The website, fairtest.org, now lists more than 1,700 “test optional” US colleges and universities and over 80 that are “test free.” The number of schools in each category has been rising yearly.
People who have been paying attention to UC's discussions about whether to keep or get rid of standardized admissions tests may be surprised by the school's decision. After all, in July 2018, UC President Janet Napolitano "asked the Academic Senate to do a study to "examine the current use of standardized testing for UC undergraduate admission; review the testing principles developed in 2002 and revised in 2010; and figure out if any changes in admission testing policies or practices are needed to make sure the University keeps using standardized tests in the right way." The 225-page report that resulted was released on February 3rd, 2020, and it completely undermined UC’s decision to phase out and eliminate consideration of college admission test scores.
Before a decision was made about whether or not to let a student into a UC campus, 14 things were looked at. Some of these were traditional academic factors like high school grade point average (HSGPA) and SAT or ACT scores. Several non-academic factors were also taken into account, such as "background, opportunities, and socioeconomic status (such as first-generation, family income, and academic opportunities in their local school), as well as factors that are harder to measure, such as a student's trajectory of improvement and the strength of their extracurricular activities." As part of a "comprehensive review" of each applicant, this was done.
The UC Academic Council Standardized Testing Task Force (STTF) looked at the UC admissions process and "found that standardized college admission test scores help predict important parts of student success, such as the undergraduate grade point average (UGPA), staying in school, and graduating." STTF found that test scores were better predictors of GPA in the first year of college than HSGPA.
This is because, over time, high school grades have steadily gotten better, making it harder to use HSGPA to tell the difference between excellent and good students. Statisticians call this effect the "ceiling effect" or "restriction of range." To see this, think about the fact that grade point averages in the US usually go up to 4.0. (A). The best score you can get on the SAT is 1600, and the best score you can get on the ACT is 36. There are a lot more students who apply to college with a 4.0 GPA than with a 1600 on the SAT or a 36 on the ACT. So, STTF found that since 2007, grade inflation has made it harder for HSGPA to predict college success, but college admission test scores have kept their predictive power. In the US, high school GPAs have been going up in recent years, while SAT and ACT scores have stayed the same.
One reason UC President Napolitano asked for a study of the admissions process was to see if the use of admissions test scores unfairly hurt any group. STTF found, though, that:
Test scores are predictive for all demographic groups and disciplines, even after controlling for HSGPA. In fact, test scores are better predictors of success for students who are Underrepresented Minority students (URMs), who are first-generation, or whose families are low-income: that is, test scores explain more of the variance in UGPA and completion rates for students in these groups. One consequence of dropping test scores would be increased reliance on HSGPA in admissions. The STTF found that California high schools vary greatly in grading standards, and that grade inflation is part of why the predictive power of HSGPA has decreased since the last UC study.
STTF also found that the UC admissions process put a lot more weight on HSGPA than on admissions tests, even though HSGPA has become less useful over time (which had retained their predictive value). In its report, STTF said, "As a result of comprehensive review, applicants from less advantaged demographic groups are accepted at a higher rate for any given test score. The point wasn't what the STTF thought it would be before we started looking at the data."
STTF's report also showed that the University of California's admissions system does something called "race norming." For example, a Latino student with a SAT score of 1000 has a 50% chance of getting into UC, while a white student with the same score has a 30% chance. A Latino student with a SAT score of 1000 has a 50 percent chance of getting into college. A white student with a SAT score of 1200 has the same chance. The report also said, "Students with low test scores are more likely to drop out of UC within a year" and "Students with low test scores are more likely to drop out of UC without a degree."
In its report, STTF listed other ways that admission test scores helped with the process of getting into UC. Every year during the admissions season, UC sends offers of admission to high school seniors. The number of offers it makes depends on how many spots it has for new students and how many of the students it thinks will accept the offers. UC found that using admission test scores to predict the rate of acceptance was helpful. It had also found that admissions tests were helpful for figuring out how smart students from outside of California are who want to join the UC system. For example, STTF found that out-of-state high school students had slightly higher HSGPAs than California high school students, but they had much better scores on admissions tests.
After reviewing and analyzing the data, STTF's report said, "Some people who want to change the way admissions work will call for making SAT and ACT scores optional, but we think that doing so could have big, unexpected, and bad effects on the makeup of incoming classes." "The Task Force does not recommend that UC make standardized tests optional for applicants at this time," said the report's conclusion.
Still, three months later, UC's administration stopped using the SAT and ACT to decide who gets in. The reasons for dropping the SAT and ACT were not made clear, but some UC officials said that they thought the SAT was "racist," but they didn't give any proof to back up that claim.
In a two-part article (here and here) for Psychology Today, cognitive psychologist Andrew R.A. Conway criticized UC’s decision. “In sum,” he concluded, “there is simply no evidence in the task force report to support the claim that the SAT is systematically biased in favor of certain racial/ethnic groups.” By moving toward “test free” admissions without consideration of scores on standardized college admissions tests, UC is moving away from what most of the industrial world’s colleges and universities have been requiring.
More than 20 years ago, psychologist Douglas Detterman wrote, “It is somewhat ironic, and a fact often lost in the debate over the role of tests, that standardized tests were adopted by colleges and universities to increase fairness in the admissions process.” He added, “Test scores indicate a problem that needs to be corrected, but the problem will not be corrected by eliminating tests.”
