<p>What’s the purpose of scaling the tests so that the majority of AP kids in AP Chinese get 5’s (70%), while in Compsci it’s (25%) and in APES it’s (8%)?</p>
<p>It’s based on how college students perform. The questions give to college students and those that get a certain grade in their college class get a certain number of questions right and that becomes the cutoff for the various scores. There is a lot of self selection in regards to which students takes the exams. For example, lots of people taking AP Chinese are native speakers (language exams are always based on nonnative speakers).</p>
<p>I don’t think you understand what scaling the test means. It isn’t supposed to fit a perfect bell curve. As aldfig0 said, it is based on statistics related to how college students score. That is why one year’s score distributions varies with the next year’s for the same exam. I believe APES’s was a few percent points lower than usual last year.</p>
<p>I think self-selection is the most important factor. The AP Calculus BC and AP Physics C student pool is a lot smaller (and, arguably, more intellectually gifted) than the overall pool for APES or AP Psych. With fewer, smarter kids taking math/science tests, there’s a seemingly disproportionate number of 5’s on the hardest AP exams to pass.</p>
<p>Calculus BC tends to be self-selected students who are great at math and like it.</p>
<p>Students who are merely good at math and less confident in their ability at math tend to choose calculus AB.</p>