Of course, it’s easy to shake heads and wag fingers from a comfortable distance. But it’s also easy to understand why people engage in this form of self-harm. From time to time, all of us are tempted to ignore unhappy realities. And unfortunately, reality usually comes calling anyway, bill in hand.
The pandemic provided several vivid illustrations of this principle, including the fallout from the decision many colleges made during the pandemic to relax their requirements for standardized test scores.
This was a quite reasonable thing to do in 2020, when, through no fault of their own, many kids had difficulty taking the SAT or the ACT — their scheduled test was canceled, or they or someone they lived with was immunocompromised. But the colleges’ policies continued long after we had excellent vaccines, in part because those tests gave us a lot of very unwelcome information.
They told us, for example, that academic ability is unequally distributed. Some people are better at math, some people are better at English and some people aren’t terrific at either. And with that information came an even more painful fact: Many of those differences mirror other inequalities in our society, including the most pernicious ones. Very generally: Rich kids do better than poor kids. White and Asian kids do better than Black and Hispanic kids. On the math sections, boys perform better than girls.
Despite decades of attempts to narrow those gaps, they’ve stubbornly refused to close. Eventually, people decided that the problem was the tests themselves. A sizeable cottage industry sprung up to provide critics with dubious research supposedly showing that the tests don’t predict college performance very well.
Demand for research suggesting tests don’t mean much was fueled by another uncomfortable fact: Test scores gave critics of affirmative action a way to quantify the boost (or detriment) various groups were getting in admissions. This problem became urgent as lawsuits filed by Students for Fair Admissions wended their way toward a Supreme Court that seemed eager to end affirmative action as we know it.
If the pandemic gave grateful admissions offices the excuse they needed to go test-optional, the court’s gutting of affirmative action gave them every reason to stay that way — or ditch the tests entirely. After the decision was handed down in June, I heard a lot of surprisingly glum conservatives predict that it wouldn’t matter, because colleges would just keep practicing affirmative action under another name, and vanishing test requirements would make it hard to draw the direct comparisons among groups that could unmask what admissions offices were doing.
But MIT bucked the trend in 2022 by announcing its return to mandatory testing. Last month, Dartmouth followed suit, becoming the first Ivy League school to do so. Yale and Brown soon followed. On Monday, the University of Texas at Austin became the latest to join the parade — and gave us a peek at the numbers driving its decision.
Last year, UT received 73,000 applications to join its freshman class. The 42 percent of applicants who submitted test results had a median score of 1420 on the SAT, while those who opted out of submitting scores had a median of 1160. Enrolled students who had submitted scores also performed significantly better during their first semester of college: Controlling for a wide range of factors, their grade-point average was nearly a full letter grade higher than that of students who didn’t submit, and they were 55 percent less likely to end up in the sub-2.0 danger zone.
We can’t know whether this is exactly what other schools were seeing, but we can suspect. And if we’re honest, everyone should have suspected it even before we got this data. The SAT is not a measurement of innate human value, but it is a measurement of whether a person can do the kinds of things people have to do in college courses, from performing basic mathematical operations to quickly gleaning meaning from written passages. It would be shocking if results on this test weren’t correlated with college performance.
It’s a major problem that, as things stand, the acquisition of those skills is also correlated with factors such as race and parental wealth. But we cannot fix that problem by simply throwing away the messages that reality is sending us.
Credit: Source link
