“Just because the bureaucrats who set the levels are bad at math doesn’t mean you are.”

If you’re a Rhode Island student (or parent) just getting back your SAT results, and you’re upset that the state says you aren’t “meeting expectations,” don’t be. The SAT is to a large extent set on a curve, and the national average tends to fluctuate around 500. In fact, when the SAT was originally instituted, the idea was that the mean would be 500, and the standard deviation would be 100. Later, the minimum score was set at 200, and the maximum score was set at 800.

When you think of what “meeting expectations” or “proficency” mean, you would think that actually that standard should be lower than the average. That to be not proficient should mean that you’re really doing badly, that you’re in the bottom of the class, not just below average. In fact, the proficiency level gets set well above the average. The required English score is 610, and the required math score is 590.

When Rhode Island’s educational bureaucrats release numbers saying that most students in our state aren’t proficient on the SAT, that’s arbitrary and meaningless. All that means is that they’ve set the proficiency level well above the average.

It gets worse. The way the SAT is set up, 500 is supposed to be the average for the students who take it. Nationwide, it’s mostly students looking for admission into competitive schools who take the SAT. SAT test-takers are overwhelmingly above-average students. But in Rhode Island, every student is required to take the SAT to graduate, even if they don’t intend on going to a competitive college or even a college at all. So our educational bureaucrats are expecting that Rhode Island students should be performing well above the average of how above-average students perform. It’s unrealistic, and they know it’s unrealistic. These numbers are chosen, deliberately, to create a narrative that our students are dumb and failing.

Objective tests, like the NAEP, show that Rhode Island underperforms our demographics, but not in a catastrophic way. It’s consistent with the picture most of us already know: bad schools but pretty resilient students who manage to learn anyway, despite the schools.

Now, since the SAT rewrite, scores have been elevated nationwide, with the averages around 530 instead of 500. This means that RI students are performing pretty well on the proficiency metric. If the SAT does a score correction to bring the mean down to 500 next year, then we’ll see a sudden drop in our percent proficient. But that will be meaningless. It would be wrong to conclude that, say, the new Education Commissioner would be responsible for such a drop.

So if you’re not “meeting expectations,” don’t worry. You still might be well above average. Certainly, don’t convince yourself that you’re a bad student, and school is not for you. Especially don’t convince yourself that you’re bad at math. Just because the bureaucrats who set the levels are bad at math doesn’t mean you are.

Newest Oldest Best
Notify of
Paul Klinkman
Paul Klinkman

I’ll expand upon Doug Tarnopol’s comment. The SAT is partly a trivia test. It quizzes students on rare words and on mathematical anomalies. Great numbers of students study long lists of uncommon words that are likely to show up on the test. I once heard that the Educational Testing Service paid its secretaries $20 per question to take old SAT tests home and write new questions by changing a few numbers. Hiring professors or even graduate students to write brand new questions costs too much. Some of the questions use psychological tricks. For example, a question will contain the number 13 and one of the five answers is also “13”. A number of students won’t want to choose the “13” answer simply because that same number is already in the question. Hundreds of students would probably make reasonably good physicians for every student selected. To eliminate 99% of the candidates,… Read more »

Doug Tarnopol
Doug Tarnopol

It’s actually worse than this, Senator Bell. As you can see here — https://collegereadiness.collegeboard.org/pdf/understanding-sat-scores.pdf on page 6 in the document — the percentile ranks for the scores marked “proficient” are: 610 in Verbal: 81st for a nationally representative sample; 75th for an SAT user 590 in Math: 79th for a nationally representative sample; 72nd for an SAT user For those who don’t know, that means that to be labeled “proficient” a RIer has to score better than 81% of anyone who takes the test and 75% of anyone who takes the SAT for Verbal and 79% and 72%, respectively, for Math. I tutor this stuff; I know how it works (have done it for 25 years at Kaplan and other companies and on my own; have written books, online courses, you name it). The larger issue is that these tests are pretty much useless, as even the U of Chicago… Read more »


[…] RI Senator thinks Rhode Island’s SAT proficiency levels are preposterously high […]

Greg Gerritt

RI bureaucrats use phony math and unrealistic expectations in many areas. Thanks Sam for this explanation on the SAT. I will publish one soon on the weird expectations they have on econoimc development.