A Better SAT Starts With a Better College Board

It’s so easy to criticize the SAT that most observers overlook the weaknesses of its architect, the College Board. Until we replace the latter, however, we will never fix the former. The College Board has every incentive to create a complex, stressful, expensive college admissions system. And because it is accountable to no one, it has done just that.
The College Board and ACT add over $500 million a year to the cost of applying to college. They generate $100 million in profits annually. Since the utility of the tests is largely limited to the 800,000 high-achieving students headed to selective schools, these organizations are costing each one of those students a $1,200 fee—and that’s before the additional expense of test prep.

Imagine an organization responsible for devising and maintaining a meaningful admissions ecosystem, which would be evaluated annually by its ability to improve the fit between students and schools (as measured by high graduation and low transfer rates) while lowering the cost and stress of the admissions process. This new organization would neither administer tests nor profit from them; it would simply be a coordinating body funded by participating schools.

Unlike many SAT detractors, I am not against tests. Common metrics help us avoid other problems (for example, when New York dropped the SAT as part of its Regents Scholarship competition, it saw immediate and large-scale grade manipulation). I’m just against bad tests. There are several ways we could do better.

First, we should move away from the notion of a standard, one-size-fits-all high school curriculum leading to a single test of math and English, and simply encourage curricula providers to attach secure tests to their curricula. Their tests would not drive the curricula—they would be the tail on the dog, following them in both content and spirit. We would report scores, then, from multiple tests and approved providers.

Schools and students would have the freedom to choose the courses and tests most interesting to them, and their scores would reflect their strengths in the subjects they found most compelling. The Advanced Placement (AP) tests are a good example of this approach. These tests are rigorous, and each is consistent with the curriculum. Importantly, none of these tests is for everyone. Some students might take calculus, while others might take American History; some might choose another advanced curriculum and test, like the International Baccalaureate (IB); while others, not ready for that level of work, would take less rigorous curricula and tests.

This flexibility would create some logistical challenges, but they are solvable. First, the approved tests would all use a common curve (perhaps with a familiar score range of 60-100). Colleges would receive three or four scores from each student, and would note the subjects as well as the scores, much as they do now with SAT subject tests.

The larger challenge would be score manipulation—curriculum providers dumbing down their tests to make their students look more attractive. Without undue cost, we must make sure an 85 on any one test means the same thing as an 85 on any other.

To do this in the short term, our new organization would require testing companies to validate their curves against matrix-based international tests like the NAEP (National Assessment of Educational Progress) or PISA (Programme for International Student Assessment). In the long term, this equating will get easier—we will simply look at longitudinal data for various colleges. An 85 on each test should predict the same GPAs in appropriate subjects at the same colleges.

As problems go, this one is entrenched and expensive, but surprisingly fixable. Any number of entities could, and should, step forward to create that organization (including the New England Board of Higher Education?), and save college-bound students hundreds of millions of dollars each year. If no one is ready to take on the challenge, perhaps NEBHE might convene a two-day summit to think through the solution and nominate a champion.

Every 12 years, the president of the College Board announces critical changes to the SAT, necessary to make the test relevant and useful. He does this while assuring schools that the test will be similar enough that new and old scores will be entirely comparable. And bizarrely, each identical announcement is accompanied by a breathless front-page article in the New York Times. It’s time to stop believing.

Letting go of the College Board and the SAT will be like leaving a bad relationship: unthinkable at first, and then painful, but soon enough, you wonder why you stuck around as long as you did. I look forward to the day that we can start cutting the SAT out of all the family portraits.

John Katzman is the founder and CEO of Noodle, the world’s largest education site. He has been an expert on college admissions testing since founding The Princeton Review in 1981. He also founded and led 2U, an education technology company that partners with prestigious universities to offer online degree programs.


Comments are closed.