Don't Blame the Scantron
Why the SAT’s limitations are a good thing.
By Ezra Klein
Tuesday October 31, 2006
The recent debate between Asheesh Siddique and Tim Fernholz on the SAT was smart, fun, and cutting. But a few points deserve further exploration. Siddique’s attack appeared to rest on the criticism that the SAT doesn’t test “interpretive arguments.” All due respect for Siddique, but we should hope it doesn’t try to.
Interpretive arguments are subjective things. The SAT is an attempt to measure, if not objective things, then at least quantifiable things to compare, say, kids who took easy algebra classes with those who had hardass instructors. A “B” in the latter may signify more effort and understanding than an “A” in the former, but that wouldn’t show up on a transcript. The SAT attempts to flatten those nuances—it is a necessarily blunt instrument. But, in the vein of Siddique, some folks have an aversion to bubbles, and think the test should quantify more relevant skills. Recently, to silence these critics, like the University of California, the SAT added an essay section. That was dumb.
Over 1.2 million students took the SAT in 2006. Over 1.2 million essays were scored. And not to cast aspersions on our fine corps of weekend graders, but the correlation between an essay’s grade and its length is substantial. To try to grade 1.2 million papers not merely by the clarity of their prose, but the quality of their critical thought, would be impossibly subjective. Let the teachers who have full semesters with students grade their argumentative capacity; the SAT is for questions that can be answered in a bubble.
More worrisome than the type of questions is the type of test-taker. As Siddique and Fernholz both point out, the cost of the test is nothing to laugh at. And many go much farther than that, arguing that the test proves little other than what income bracket you hail from. See Students Against Testing or FairTest.org for powerful versions of the economic bias critique. And the inherent bias is worsened by the test-prep industry that, as Siddique points out, skews the results for kids whose parents can purchase an advantage.
But that effect, while real, is less of a distortion than some seem to think. The study Siddique links to suggests that the median increase from paid preparation is 60 points, or 3.7 percent (at least on the old, 1600-scale SAT). Those 60 points, while not insignificant, are a second or third order problem. Few admissions decisions have been made because someone got a 1,070 rather than a 1,010.
More to the point, a wealthy student’s advantage on the test doesn’t begin in the spring quarter of her junior year. It begins when she’s read to as a child, when she attends a good school in a safe neighborhood, and when she can study in her own room in a house with books and educated parents. In short, her advantages begin when she’s born well-off. The distorting effects of class, peer groups, and environment begin far, far before a Princeton Review session is scheduled.
And that is why it is foolish to heap blame on the SAT. By junior year of high school, the advantages of the well-off, and the disadvantages of the less-well-off have already worked their Marxist magic. The SAT is a sorting mechanism for kids on their way out of high school; whatever it’s picking up, or amplifying, is hardly its fault. Indeed, part of its value is as a (somewhat) objective metric for how deeply we’re disadvantaging kids in the years preceding the exam.
Schools, which all grade relatively to their own students, can obscure the difference in preparation between a 4.0 at a private school and one at a dilapidated inner-city degree-mill. Progressives have a tendency to look at the test and the disparities it exposes, and demand its abolition. But to rid the world of the SAT wouldn’t bridge the class chasm—it would hide it. Yes, the SAT currently keeps no end of kids out of good schools—that’s the point. It’s supposed to test preparedness, deductive abilities, and reasoning—kids who weren’t taught that will be wrecked by a demanding college. And without the ability to weed those kids out, the students from bad schools who could still excel at Princeton would never have the opportunity to vie for admission. The inevitable lack of preparation and poor retention rates of “high-performing” kids in poverty-ravaged districts would spur admissions officers to cease green-lighting any applicants at all from those areas, just as they did before the SAT was invented.
None of this, however, is a call to inaction. Rather, it’s a cry for realism and reform. The SAT’s hated vice—its class bias—is its great virtue. We need a sorting mechanism, if only to see how deeply disenfranchised large swaths of the population are. The grotesque inequalities laid bare in the cold columns of each year’s results can’t help next year’s class much—but they force us to help the next generation. At the very least, so long as the test remains a magnifying glass for the pernicious impacts of inequality and class, neither liberals nor anyone else will have plausible deniability for turning a blind eye to the mockery we’re making of our meritocracy—not when kids are juniors in high school, but when they’re shopping in the juniors section of K-Mart.
Oh, and the test should be free.
Ezra Klein is the Writing Fellow at The American Prospect. His blog is at www.EzraKlein.com.
--------
Comments
|
!
