The United States’ economy desperately needs all the scientific, engineering, and IT geniuses it can find. One of the most important functions that the SAT can serve is to identify young Americans with that kind of intellectual potential.
For many years, the scholarly literature has indicated that we have been missing a lot of that talent because one of its key components, spatial ability, is not identified by the verbal component of the SAT and only partially identified by the math component. The current best guess is that we’re failing to identify about half of students within the top one percent of spatial ability. That estimate comes from an important new study by scholars at Vanderbilt University about to be published in Psychological Science and already summarized in the New York Times.
The good news is that IQ tests have accurately measured spatial ability for decades and the items to do so could easily be incorporated into the SAT. The bad news is that it’s extremely unlikely that the College Board, which administers the SAT, will have the nerve to do so. Why? Because the largest gender differences and the largest ethnic differences are found in the subtests that measure spatial skills. Here’s the dilemma facing the top brass at the College Board: if they add a spatial component to go with their math and verbal components, they will indeed identify lots of extremely talented students whose potential is underestimated by the existing components of the SAT. But that spatial component will also show larger gender and ethnic differences than the other components (if you’re curious, the big winners from such a revision of the SAT would be Asians and males). What do you suppose the chances are that the College Board will be willing to take the heat for such a result? If you want to make a bet, I’ll take zero and you can have everything else.