People are fired up about the June SAT.
When the College Board released SAT scores last week, many students noticed something off: they missed the same number of questions as on previous tests, but their scores were lower. In some cases, students answered more questions correctly only to see their scores go down.
How is this possible? Well, unlike most high school grades, the SAT scores that get reported to colleges are not "raw" scores that directly reflect the percentage of questions answered correctly. Instead, they are “scaled” scores out of 800 in each section. This scaled score is what allows colleges to compare a score that a student received in May in Brazil with a score received in October in Florida, or a score received by an applicant this year with a score received by an applicant last year.
There are minor variations in difficulty between one test form and another, so the scale used -- commonly called the test's "curve" -- changes slightly from test to test. In one typical test, missing 5 questions across the two math sections might lower a student's score from a "perfect" 800 to a 760. In another, the same number of errors might only lower the score to a 780. The College Board calls this process test equating, and it's based on test statistics, not on individual performance.
This kind of "equating" -- and the frustration it can cause students -- is nothing new. So why are people so upset about the June SAT? Two reasons: First, the curve was the steepest by far since the new SAT was first administered in 2016. According to some students' score reports, missing the same number of questions resulted in scores as many as 90 points lower than in previous test administrations. A gap that large raises alarms: while equating works well between tests with only a slight variation in difficulty, it doesn't work well when some tests are substantially "easier" than others. The frustration over the curve is felt particularly strongly among high-scoring students since a test with a steep curve fails to distinguish meaningfully between content-based errors and a small number of careless mistakes.
Second, four items were removed from the Verbal side of the test (two from the Reading section, and two from the Writing and Language section). The test is designed to be able to provide an accurate score even when some questions are deemed flawed after the fact and removed from consideration; still, having four items removed on top of a steep curve for an "easy" test has left many students feeling like the College Board's June test was simply unfair.
What can be done?
The College Board has released a statement affirming that scores are accurate, so don’t expect any adjustments to the test's scale or scoring. Instead of focusing on this one test, students should remember that standardized testing is a long game. Most of our students take the test multiple times in part because the tests -- as well as individual performance -- can vary from day-to-day. The ultimate goal of testing is to present colleges with a score that reflects your hard work and your abilities, and it's difficult for any one test to indicate more than just your performance on a single day. This is a big part of why many colleges "superscore" the test, taking the highest score from each section across multiple tests. Even rising seniors still have several opportunities to take the test again, including August 25 (registration deadline July 27, late registration deadline August 15), October 6, and November 3.
The June test was unusual -- and with all the outcry, you can be sure the College Board will be working even harder to ensure consistent test difficulty in the future. But under normal circumstances, "equating" makes the tests more fair, not less. If you're well prepared and one test section feels harder than usual on test day, you should expect the curve to reflect that. And if you unexpectedly finish a section earlier than usual, be sure to take that opportunity to double check for careless mistakes.