Ignoring race in the college admissions process lowers diversity outcomes but has no effect on the academic standards of an admitted class, according to a new study from Cornell researchers.
The study, published Thursday, used data from an unnamed university to build an artificial intelligence–powered ranking algorithm that could simulate the impact of the affirmative action ban on racial diversity and academic merit. It found that when race was removed from the equation, the number of underrepresented minority students in the top-ranked list of applicants fell by 62 percent, from 53 percent of the pool to just 20 percent. At the same time, the average test scores of the top applicants did not change significantly.
“We see no evidence that would support the narrative that Black and Hispanic applicants are admitted even though there are more qualified applicants in the pool,” René Kizilcec, associate professor of information science at Cornell and a co-author of the report, said in a statement.
At the majority of selective colleges that have released demographic class profiles, the share of matriculating minority students fell this fall, though those results varied by institutions and the data is still largely inconclusive.
The researchers also said the study was an important test of the use of AI to review college applications, which they predict will be normalized over the next several years.
