<p>I was looking at an old SAT that my oldest son took 6 1/2 years ago. He missed 5 math questions.</p>
<p>My middle son took the SAT in June and missed three math questions and didn't answer one, but <em>he got a lower score</em> by 10 points!!</p>
<p>How can that be??</p>
<p>The curve was probably different.</p>
<p>different tests have different scores. -1 can be anywhere from a 760 to an 800. I’m sure there’s that much of a variance for getting more wrong as well.</p>
<p>So if more kids do better on a specific test, the curve is steeper? I wonder if, because more kids are prepping for the SAT as time goes on, the curve is just getting steeper and steeper…bummer for my middle son.</p>
<p>The curve has been getting steeper for the past few years. All we can do is deal with it.</p>
<p>The curve is set before the test, fwiw, so it has nothing to do with who takes the test that particular day.</p>
<p>Marvin is correct. Theoretically everyone could get a perfect score, and the belief that it’s better to take a test in January (when it’s generally less crowded) is false.</p>
<p>marvin,</p>
<p>I’m a bit dense. Can you explain what that means and why it’s a different curve for each test? It does sound like it’s a bit of the luck of the draw. Or do they make tests of varying difficulties and set the curve to be a little more generous on a test that they deem harder?</p>
<p>It’s impossible to make all the tests the exact same difficulty. There are different versions of the test because of safety reasons (even different versions within each test date) so the curve is there to balance that. Yes, the curve is more generous for harder tests.</p>
<p>The CB tests their tests in a number of ways, one of which is the dummy sections (or “experimental” sections). The curve is based on their research. One way to prove this is that when tests have been re-used (internationally) the curve always exactly matches the curve of the previous sitting.</p>
<p>(as for “luck of the draw,” well, I guess, but the way curves work is that harder tests have more charitable curves and easier tests have less forgiving curves; if a student gets a “hard” test with a “friendly” curve is she lucky or unlucky? Data shows that scores are damn consistent across test sittings, so I’d argue it all comes out in the wash. Really well prepared kids get very high scores regardless of the curve.)</p>
<p>Good point. My son who only missed three hadn’t taken Alg. II yet, so he will be more prepared the next time. My son who missed more but got a higher score had begun calculus but obviously still had holes in his knowledge at that time; his retake was without mistakes.</p>