Other Voices

New SAT has some significant improvements, some problems

A teacher examines a student’s worksheet during a SAT preparation class.
A teacher examines a student’s worksheet during a SAT preparation class. AP

David Coleman, president of the College Board, wants everyone to know that the new SAT, which students recently took for the first time, is just as good as the old test at predicting who would do well in college.

Of course, he also wanted to be clear, in introducing the SAT to a conference of the Education Writers Association, that his test was new and improved as well.

For one thing, it makes the silly essay portion of the test optional; it was both gameable and, in terms of the way it was scored, hardly an indicator of who can write well.

The new SAT also reformats the testing of vocabulary, eliminating the $4 words that required weeks of drill-and-kill memorization and then would never be used again.

Plus, there’s no longer an extra penalty for guessing wrong.

Also to its credit, the College Board has added services to help the students who can’t afford thousands of dollars’ worth of private test prep. Free online prep and practice tests are available through the nonprofit Khan Academy.

And students whose income is low enough to qualify them for free test taking also automatically qualify for waivers of college application fees, which normally cost about $80 per college, a significant sum for working families.

But most important is that the new SAT is supposed to align with the Common Core standards that have been adopted to one extent or another in 40-plus states (not Texas).

This includes a heavier emphasis on reading — even in the math problems — and more critical thinking skills. That’s what colleges say they want, and what students are lacking.

What Coleman didn’t spend much time discussing are SAT problems that haven’t been solved.

The test may require more critical thinking skills, but it is still coachable; it won’t put an end to the big and growing high-end test-preparation industry that gives affluent kids a leg up on the system.

Poor kids get two free shots at taking the SAT; kids with more money can take the test five to 10 times, and some of them do. Then many of the colleges allow them to “superscore” — report only their best scores on each section.

I recently met a sophomore who’s taken the test five times. His mother said she had spent $10,000 on test preparation, and his scores had risen by 300 points.

And what about Coleman’s assertion that the test is useful for college admissions officers?

If the SAT is a reflection of the Common Core lessons, and those lessons reflect the skills that colleges need to see in students, why isn’t the new test a better predictor of freshman college success than the old one?

Studies have shown that the SAT is almost as good as a student’s grades at predicting college success during freshman year.

Using the standardized test in addition to grades gave admissions officers a better picture than grades alone.

Beyond freshman year, however, research on the SAT’s predictive value gets mixed reviews.

A study of colleges that have gone test-optional — applicants can report their scores or not — found that students who didn’t submit their scores fared just as well throughout college as those who did, though they might have opted for easier courses.

A recent report by the Harvard Graduate School of Education suggested that at some schools the SAT might be a good predictor of success — for instance, a mediocre math score probably indicates a kid who would struggle at MIT or Caltech — but at others, it might not make much difference.

Karin Klein writes about education for the Los Angeles Times editorial board.