College admissions will never be fair

Some parents will do anything to get their kids into the college of their dreams. They think it’s normal to spend thousands of dollars on tutors, SAT prep and college-essay advisers, and to use whatever influence they have to get an edge. In some circles, not trying to game the system is even seen as parental failure.

Understandable as such behavior may be, it also works against one of the primary purposes of education and college in particular: to improve the lot of kids who aren’t lucky enough to have rich and influential parents. So how do we strike a better balance? How do we build a system that fosters social mobility while accepting the reality of parental gaming?

It’s an issue that has sparked some heated debates. In a recent lawsuit, an organization called Students for Fair Admissions excoriates Harvard for bias against Asian applicants. It claims that “an Asian-American applicant with 25 percent chance of admission, for example, would have a 35 percent chance if he were white, 75 percent if he were Hispanic, and 95 percent chance if he were African-American.”

In New York, Mayor Bill de Blasio wants to reform the city’s specialized high school system, where less than a tenth of slots went to the blacks and Latinos who comprise two-thirds of the city’s student population. He sees one culprit in the admissions exam, which parents strive to game by subjecting kids to months of expensive test prep courses. (Disclosure: Two of my three white sons are in specialized New York City high schools, and they attended an elite primary school.)

Colleges are increasingly souring on standardized admissions tests: Some 180 no longer require an SAT score, for example. Once seen as an important leveler, scores on such tests correlate with family income. This renders them not particularly useful for schools hoping to diversify student bodies.

So how to assess applicants? There’s no general agreement about what makes one candidate more qualified than another. Supporters of the Harvard suit dismiss “personal ratings,” which are subjective and probably do enable anti-Asian prejudice. But let’s face it, no measures are truly objective. SATs, GPAs, extracurricular ratings — they’re all prone to bias.

I’d suggest thinking more systematically about the longer-term mission of education, and how to create incentives to fulfill it. Perhaps parents’ urge to game the system can even be turned to its advantage.

Consider the University of Texas at Austin, the state’s top public education institution. It automatically admits all Texas students who graduate in the top 6 percent of their high school class — irrespective of the school’s location or reputation. As a result, some poorly prepared kids inevitably take the places of “better qualified” students. But the system fulfills the broader goal of promoting social mobility, by providing a subsidized education for students willing to make the effort.

This approach creates some interesting incentives. Given the stiff competition to make it into the 6 percent in the most privileged neighborhoods, gaming-inclined parents might move elsewhere. Just imagine what would happen in New York City. This, in turn, promotes economic and racial diversity — the opposite of the tendency toward segregation that we’re seeing now. The average local middle school would likely be better off.

There are other ways to make good schools accessible to everyone, and we should try them. Education shouldn’t be a competition with only a few predetermined winners.

Cathy O’Neil, a mathematician, is a Bloomberg Opinion columnist.