The College Board is an organization whose entire reason for existence is to quantify the unquantifiable. For its first 120 years, the board largely limited this alchemy to the translation of scholastic aptitude into a single number. But last week, the College Board announced that it will introduce a new “adversity score” aimed at measuring the hardships that high school students have faced in their still-young lives.
The adversity score — which will be shared with colleges and universities along with a student’s SAT performance — is based on 31 data points that, according to research, are correlated with academic achievement. They include factors associated with an applicant’s neighborhood (such as the percentage of adults with less than a high school diploma) and those tied to a student’s high school (such as the number of Advanced Placement courses offered). College admissions officers are already placing great weight on the measure. At Yale College, which has had access to adversity scores for several years as part of a pilot program, the admissions dean says the new metric “is literally affecting every application we look at.”
Critics have complained that the adversity score will undermine meritocracy. But there is nothing “meritocratic” about ignoring the hurdles that a student has jumped on the road to college. The College Board’s decision to acknowledge adversity in the admissions process could have been an important step toward a more comprehensive assessment of aptitude and achievement.
Alas, the adversity score reflects at best a half-hearted effort to measure hardship. The board has conspicuously omitted a central factor shaping the lives of college applicants: race. As a result, a metric designed to guide admissions officers in their consideration of adversity threatens to mislead instead.
Differences in the life experiences of white and African-American children — even children who live in the same neighborhood and attend the same school — start early. African-American children are 3.6 times more likely than white students to be suspended from pre-school — a disparity that, according to Yale researchers, is linked to implicit bias among early childhood educators.
Those disparities persist long beyond the pre-school years. Researchers at Texas A&M University found that white children in first grade have more supportive relationships with their teachers than African-American classmates do. Researchers at the University of Virginia and Johns Hopkins University discovered similar results in high school. Given that these same high school teachers assign grades and write recommendation letters, it would be surprising if these differences did not manifest in college admissions.
Racial inequalities inside the classroom are replicated outside it, too. Job seekers with white-sounding names are more likely to receive callbacks than applicants with identifiably African-American names—suggesting that it will be harder for African-American high schoolers to compile résumés with fancy internships. White students, moreover, are likelier to have social and familial connections to friends and relatives who occupy positions of influence with extracurricular opportunities to dispense. The effects of race ricochet throughout the college application.
These effects extend to the SAT itself. A study in the Harvard Educational Review finds that the pre-screening of questions on the verbal portion of the exam favors the inclusion of items that generate larger white-black disparities. This finding is backed up by other research indicating that standardized test scores underestimate minority academic achievement.
And then there are the day-to-day differences between life as an African-American and life as a white American that the College Board’s race-blind measure simply cannot capture. Nobody has ever been stopped and frisked by police for attending a high school with a low number of Advanced Placement offerings. Race colors our most mundane and meaningful interactions in ways that the adversity score’s 31 factors do not.
The best defense of the College Board’s adversity score is that it is a backdoor way to integrate race into college admissions for schools that are barred by state law from considering race explicitly, like the University of California. But that would be allowing a heckler’s veto to dictate outcomes nationwide. The fact that voters in some states have mandated “color-blind” admissions to public colleges and universities should not lead admissions officers elsewhere to ignore the role of race in applicants’ lives.
No college admissions process is actually color-blind anyway. When high school teachers are affected by the same implicit racial biases that permeate the population, ostensibly “meritocratic” criteria such as grades and recommendation letters already have race baked into them. A student’s extracurricular record — and even her SAT score — already reflect her racial background. This is why defenders of affirmative action should not cede the mantle of “meritocracy”: the one we’ve built is an illusion.
So how should the College Board move forward? One possibility would be to identify the variables associated with adversity — race, neighborhood poverty rate, number of high school AP offerings, and so on — and then to calculate the influence of each factor on SAT score. A student’s adversity index would then tell admissions officers: how do we think this applicant would have scored if, instead of her actual background, she was born white and privileged? Other demographic characteristics — such as a Latinx background — could be integrated into the adversity index in the same way.
There may be other sensible ways to approach the challenge. But instead of starting a frank conversation about the consideration of adversity in college admissions, the College Board has selected a measure that obfuscates the relationship between adversity and race. Rather than puncturing the myth of color-blindness, the board has chosen to perpetuate it.
Accounting for adversity in testing was itself a test. Unfortunately, the College Board has so far failed.