Civil rights lawyers recently sent a demand letter to the University of California stating that they will sue unless the school stops considering SAT and ACT scores in its admissions process. In their letter, they allege that the SAT and ACT are little more than a proxy for race and wealth and should play no role in college admissions, particularly because high school grades are a better predictor of college success than standardized tests. Use of these scores, the letter goes on to say, is discriminatory and violates the equal protection clause of the California Constitution.
Standardized testing has racist origins. Carl Brigham, the inventor of the original SAT, was a eugenicist and wrote that the test would help prove the superiority of the white race and prevent “the infiltration of white blood into the Negro.” Because the test purported to test “aptitude” for learning rather than, say, memorization of information, schools saw it as a good tool for finding the students with the most capacity. The 1930s saw a rapid expansion of the use of such tests, even as critics pointed out the racism in the content of the questions and argued that the multiple-choice format encouraged memorization rather than measuring natural intelligence. Reliance on this biased testing consistently excluded communities of color and English Language Learners from educational opportunities.
The racial disparities in standardized test outcomes persist today. A Brookings Institute study examined math scores among students who took the SAT in 2015 and found that, while the average score was 511 (out of 800), on average, white students scored 534, Asian students scored 598, black students scored 428, and Latinx students scored 457. The study found a similar racial breakdown in ACT scores. While the standard explanation for these differing racial outcomes is the ongoing segregation of American schools and the unequal access to resources, some studies claim there is a bias in the actual questions on the verbal section.
These studies looked at “differential item functioning” (DIF) – questions where students with similar academic backgrounds and achievements gave answers differentiated by race, i.e., black and white students predicted to get similar scores answered the questions differently along racial lines. One rationale for the existence of DIF is that certain expressions are used more commonly among different cultures, so white students may gain an advantage simply by belonging to the dominant culture where expressions being tested on the SAT are regularly used.
In 2016, College Board, which produces the SAT, made sweeping changes to the test in an attempt to make it less memorization-based and more focused on preparing students for college and their careers. It placed emphasis on reasoning through answers and understanding words in context rather than using test taking strategies and remembering obscure vocabulary terms. Nevertheless, recent data shows that the racial gaps found in the old version of the SAT persist in the new version.
College Board announced and then abandoned another major change this year: the “adversity score.” Designed to counterbalance the economic and social factors thought to drive the achievement gap, the so-called “adversity score” (College Board called it the “Environmental Context Dashboard”) would have given colleges a number based on 15 factors thought to measure the amount of disadvantage a student has faced. Facing criticism, College Board decided to provide schools with a bulleted list of information about each student’s neighborhood and high school rather than a single score.
Given that standardized testing has been so problematic for nearly a century, and high school grades are a far better predictor of college success than test scores, what explains the enduring popularity of tests like the SAT and ACT? It may be because standardized testing is big business. The National Board of Education on Testing and Public Policy at Boston College places the value of the testing market as between $400 million and $700 million. In addition to the companies that actually produce the tests, there is a collateral market – the test prep industry. In 2015, parents spent $13.1 billion on test preparation, tutoring, and counseling. Small wonder that in 2001, when the president of the University of California system announced he was considering dropping the SAT requirement, College Board negotiated, and the school ended up staying with a revised version of the test.
Assuming that the lawsuit threatened in the above demand letter is filed, this case will be an important one to watch. California banned affirmative action by referendum in 1996, and the U.C. System’s African American and Latinx enrollment fell drastically. Despite recent Supreme Court precedent upholding affirmative action policies, there is good reason to think that the court will soon revisit the issue and, given the current makeup of the Court, race-conscious admissions policies may be banned altogether across the United States. Should this occur, schools will need to find so-called “race-neutral” means of ensuring diversity in their incoming classes. Given the experience of schools in California, reliance on standardized testing in a “race-blind” admissions process will likely negatively impact diversity. This case could provide a model for challenging the use of standardized tests in a world without affirmative action. Notably, the letter only brings up the state constitution and state anti-discrimination statutes, likely due to the changing makeup of the federal judiciary; this too may be the future of civil rights litigation.
School administrators would be wise to begin rethinking their dependence on standardized testing now. Even the SAT’s founder, Carl Brigham, came to regret inventing the test, acknowledging that scores had little to do with natural intelligence and more to do with “a composite including schooling, family background, familiarity with English and everything else, relevant and irrelevant.” Nearly a hundred years of revisions and experimentation have done little to change the racial and class bias inherent in this test – it may be time to give it up altogether.