Meanwhile, while all these changes are going on, everyone is still citing statistics of SAT/IQ correlation as of 2003.
But we only have the rear view mirror!
These massive compilations of data take time to assemble and filter through the system, ultimately incorporating them into the entire corporate world system so that adequate adjustments to the college student processing and sorting system can be made in order to improve efficiency, and ultimately, competitive profitability.
That's not the only factor. I frequently defend the SAT, but the College Board, which produces it, is increasingly dishonest or at least willfully uninformed about why the SAT is predictive. The SAT used to stand for "Scholastic Aptitude Test", and "aptitude" sounds a lot like intelligence. But they dropped the term "aptitude", because retaining it raises the question of whether groups have different average scholastic aptitudes, since their SAT average scores differ. Since the College Board does *not* want to talk about that, it will *not* do any research correlating SAT scores with IQ. At the same time, if it eliminated the g loading of the SAT, its ability to predict college grades would decline substantially. So the College Board tries to make the SAT look like less of an IQ test (dropping analogies, and soon, "obscure words") while still being enough of one to retain value.
