Education Opinion

SAT’s Predictive Value Is Minimal

By Walt Gardner — September 07, 2015 1 min read
  • Save to favorites
  • Print

If the past is any indication, the next batch of SAT scores will once again generate heated comments by both supporters and detractors (“California SAT scores show 41% of test-takers are ready for college,” Los Angeles Times, Sep. 3). But what stands out in my mind is the College Board’s claim that scoring 1,550 or higher means students are ready for college.

The evidence calls that assertion into question. In 1984, Bates College decided to engage in a pioneering experiment. It made submission of SAT scores optional for consideration for admission. In 2004, Bates announced that its 20-year study found virtually no difference in the four-year academic performance record and on-time graduation rates of 7,000 submitters and nonsubmitters.

Since then, more than 850 colleges and universities have dropped the requirement for SAT or ACT results as part of the admissions process. Since the spring of 2014 alone, more than two dozen schools have followed suit. If there were any credibility to the College Board’s claim, then why would so many institutions of higher learning do so?

Realizing that its flagship - and highly profitable - exam is under increasing fire, the chief of assessment of the College Board attempted to make a case by arguing that the new SAT now measures classroom content first (“The New SAT: Will It Be Better?” The New York Times, May 8). But because the SAT remains a norm-referenced test, it is still designed primarily to allow ranking of test-takers. If its designers loaded up the exam with items measuring the most important content that teachers effectively taught, scores would likely be bunched together, making comparisons hard to accomplish. In that case, the College Board would not be able to deliver on its promise to its clients.

To avoid that possibility, the test-makers have to engineer score spread. They’ve learned from experience that the best way to do so is to include many items that are affected by socioeconomic factors. It’s a matter of survival, and the College Board does not want to lose customers. That’s why I continue to have little confidence in the predictive value of the SAT one way or another. Yet I expect little to change, except for the format of the test.

The opinions expressed in Walt Gardner’s Reality Check are strictly those of the author(s) and do not reflect the opinions or endorsement of Editorial Projects in Education, or any of its publications.