The premise of your answer is that MIT wants good "students" and "test takers". What if that's not what they want so much anymore? A test is supposed to be a metric, an indicator of something. But metrics tend to become target as soon as they're exposed. Maybe then they still remain an indicator of something, but not necessarily what they were originally intended to filter for.
I understand that it might feel like the goalposts are being moved for people who optimize to score high on such metrics, but such is the nature of this type of games. That's also why search engine companies have to keep refining their algorithms.
MIT has an obvious incentive to admit rich, well-connected people, and an obvious incentive to maintain a reputation for admitting the most intelligent on merit. We should be sceptical of any changes they make that purport to be about getting more meritorious students, especially if those changes result in the admission of more rich, well-connected people.
Good test takers are people who have learned things and know them well enough to pass a test on it. There are many kinds of test. The SAT is the kind designed by psychometricians, experts with Ph.D.s building on millions of man hours of research to make tests that are both valid and reliable, that predict success in college and are as close to ungameable as possible.
MIT could drop the real SAT tomorrow and still fill its freshman class with valedictorians who captained their high school sports team and have ten AP 5s. Schools with fewer applicants have the SAT.
>Good test takers are people who have learned things and know them well enough to pass a test on it.
Tell that to the massive test prep industry. Many many years ago I took a trial SAT and scored around 1000. Then I took a Kaplan SAT prep course, and lo and behold after learning all the tricks scored a 1300 on a test SAT. Back then the test prep course cost > $1000. I had a car in high school, so I could drive to the test prep school in the evenings.
Seems to me that people who prep for the tests (not necessarily learned things through high school) and are willing to spend time and money have a great margin of benefit.
It's the same thing for tech interview prep industry. The ones who prep well for the interview, do well in the interview.
> The SAT is the kind designed by psychometricians, experts with Ph.D.s building on millions of man hours of research to make tests that are both valid and reliable, that predict success in college and are as close to ungameable as possible.
This is what the CollegeBoard wants you think. You don’t have to be all that great to realize that the people writing the test are hardly better than you are at both the subject and writing test, and in many case significantly worse.
These are positions for which a Ph.D. is an absolute requirement. Compare Assistant or Associate Professor.
> About you
> Education/Experience:
> Doctorate in psychometrics, educational research, educational measurement or a related field is required. New PhD level applicants will be considered, as will applicants with additional experience. A minimum of 3 years relevant experience is required for the Associate level.
I understand that it might feel like the goalposts are being moved for people who optimize to score high on such metrics, but such is the nature of this type of games. That's also why search engine companies have to keep refining their algorithms.