4 ms·

Indeed. SAT is basically "how good are you at cramming a bunch of relatively arbitrary information into your brain and regurgitating it in a high pressure, time

by kortex 5y ago

Indeed. SAT is basically "how good are you at cramming a bunch of relatively arbitrary information into your brain and regurgitating it in a high pressure, timed scenario". That is a kind of intelligence, but there are very few occupations where that kind of intelligence is useful. Lawyers and doctors. Both the "intensely study something rather uninteresting" and "perform under high stress and time demand" are specific kinds of intelligence, and I happen to know many intelligent folks who are expressly bad at it. They don't "test well". Anecdotally, diligent study and fast-testing are anti-correlated in many of these smart folks.

ex_amazon_sde 5y ago
SAT is more "how good AND MOTIVATED are you at cramming a bunch of relatively arbitrary information into your brain"
mycologos 5y ago
> cramming a bunch of relatively arbitrary information into your brain What? Most of the SAT is evaluating high school-level math, writing, and reading skills. It's not like they're asking you to memorize historical dates or names or poems or formulas. Maybe the relatively narrow part of the test on vocabulary falls into this category, but even those words are almost entirely things you'll just pick up if you like reading books. And, now that I'm looking into it, only 10 out of 52 questions in the reading section are even about vocabulary [1]. You can skip 9 of them and still get a 700 on reading/writing [2]. Assuming you get an 800 on math, that's a 98th percentile total score and a 90th percentile on reading/writing alone [3], with almost no vocabulary prep. I agree about the high pressure aspect, there doesn't seem to be a good way around that. [1] https://www.kaptest.com/study/sat/whats-tested-on-the-sat-vocabulary/ https://www.kaptest.com/study/sat/whats-tested-on-the-sat-vo... [2] https://blog.prepscholar.com/how-many-questions-can-you-skip-for-a-good-sat-score https://blog.prepscholar.com/how-many-questions-can-you-skip... [3] https://blog.prepscholar.com/sat-percentiles-and-score-rankings https://blog.prepscholar.com/sat-percentiles-and-score-ranki...
fighterpilot 5y ago
But they are linked. Studies have shown a very high correlation. One outlier doesn't change the high correlation.
- austincheney 5y ago
  If those studies don’t separate preparation from raw performance they only suggest a self-reinforcing bias different from capability. Any correlation then is purely anecdotal.
  - fighterpilot 5y ago
    The high correlation suggests that both preparation/study and raw intelligence are important in getting a high SAT score. If you sacrifice one of those two things, you'll get a low SAT score. If you have both, you'll get a high score.
    - austincheney 5y ago
      Which of those is more important and by how much though? If both are important factors it would necessary to determine that weighted measure before falsely drawing a correlation that is incomplete or biased.
    - wutbrodo 5y ago
      > If you sacrifice one of those two things, you'll get a low SAT score. If you have both, you'll get a high score. Even this is saying too much. I know someone who literally did two or three practice tests and got just shy of a 2400 (he got every q right and lost ten pts for his essay). There's nothing on the SAT that's difficult enough to need prep for a high school junior, with the possible exception of the permutation/combination qs. Though note that this leaves a massive hole for those who are doing math at middle-school levels, which is true of a shocking fraction of the student population.

by kortex 5y ago

ex_amazon_sde 5y ago

SAT is more "how good AND MOTIVATED are you at cramming a bunch of relatively arbitrary information into your brain"

mycologos 5y ago

> cramming a bunch of relatively arbitrary information into your brain What? Most of the SAT is evaluating high school-level math, writing, and reading skills. It's not like they're asking you to memorize historical dates or names or poems or formulas. Maybe the relatively narrow part of the test on vocabulary falls into this category, but even those words are almost entirely things you'll just pick up if you like reading books. And, now that I'm looking into it, only 10 out of 52 questions in the reading section are even about vocabulary [1]. You can skip 9 of them and still get a 700 on reading/writing [2]. Assuming you get an 800 on math, that's a 98th percentile total score and a 90th percentile on reading/writing alone [3], with almost no vocabulary prep. I agree about the high pressure aspect, there doesn't seem to be a good way around that. [1] https://www.kaptest.com/study/sat/whats-tested-on-the-sat-vocabulary/ https://www.kaptest.com/study/sat/whats-tested-on-the-sat-vo... [2] https://blog.prepscholar.com/how-many-questions-can-you-skip-for-a-good-sat-score https://blog.prepscholar.com/how-many-questions-can-you-skip... [3] https://blog.prepscholar.com/sat-percentiles-and-score-rankings https://blog.prepscholar.com/sat-percentiles-and-score-ranki...

fighterpilot 5y ago

But they are linked. Studies have shown a very high correlation. One outlier doesn't change the high correlation.

austincheney 5y ago
If those studies don’t separate preparation from raw performance they only suggest a self-reinforcing bias different from capability. Any correlation then is purely anecdotal.
- fighterpilot 5y ago
  The high correlation suggests that both preparation/study and raw intelligence are important in getting a high SAT score. If you sacrifice one of those two things, you'll get a low SAT score. If you have both, you'll get a high score.
  - austincheney 5y ago
    Which of those is more important and by how much though? If both are important factors it would necessary to determine that weighted measure before falsely drawing a correlation that is incomplete or biased.
  - wutbrodo 5y ago
    > If you sacrifice one of those two things, you'll get a low SAT score. If you have both, you'll get a high score. Even this is saying too much. I know someone who literally did two or three practice tests and got just shy of a 2400 (he got every q right and lost ten pts for his essay). There's nothing on the SAT that's difficult enough to need prep for a high school junior, with the possible exception of the permutation/combination qs. Though note that this leaves a massive hole for those who are doing math at middle-school levels, which is true of a shocking fraction of the student population.