4 ms·

If you think that's what happened, why don't you make that argument? What is that thing unrelated to test scores which still has value? (Similarly, what is the

If you think that's what happened, why don't you make that argument? What is that thing unrelated to test scores which still has value? (Similarly, what is the hidden thing in transit costs, in health care, etc?) "Maybe there's a hidden thing this guy missed, and I can't stand people like him" isn't really a counterargument.

hiddencost 10y ago
SSC does this regularly: he'll take a problem, choose an arbitrary metric, and then run the whole way. It's a cultural rejection of expertise (especially expertise from the social sciences) which is endemic to tech and mathematics. It leads to terrible policy recommendations (c.f., the effect on education by the Gates Foundation and Zuckerberg's 100M to NJ). ("The Best and the Brightest" is a good book that talks about this phenomena in another context.) As a starting point, we can talk about issues related to: -- increasing prevalence of mental health issues -- the expanding nature of school populations (a lot more students, coming often from much more disadvantaged backgrounds) -- shifts from off-campus housing to on-campus housing which have necessitated a massive expansion of facilities staff, security, dining halls, etc. -- changes in the legal burden that schools need to reach (e.g., title 9)
- yummyfajitas 10y ago
  You are still not arguing that he's wrong. You're merely arguing that if he were wrong, he might be wrong at other times for similar reasons.
- metaphorm 10y ago
  author of SSC is an M.D. (psychiatry resident), not specifically a math or tech guy. I agree with you, however, that too much faith in dubious metrics is a serious problem. I don't particularly think the main thesis of today's post is based on dubious metrics though, but perhaps some specific claims in it (about education outcomes, for example) are not well supported. that's beside the point though. he wasn't examining whether or not we got our money's worth on all this increased education spending. he was pointing out that the cost of education, on a per student basis, has risen massively over the last several decades, and there is not particularly a good reason why this should have occurred.

hiddencost 10y ago

SSC does this regularly: he'll take a problem, choose an arbitrary metric, and then run the whole way. It's a cultural rejection of expertise (especially expertise from the social sciences) which is endemic to tech and mathematics. It leads to terrible policy recommendations (c.f., the effect on education by the Gates Foundation and Zuckerberg's 100M to NJ). ("The Best and the Brightest" is a good book that talks about this phenomena in another context.) As a starting point, we can talk about issues related to: -- increasing prevalence of mental health issues -- the expanding nature of school populations (a lot more students, coming often from much more disadvantaged backgrounds) -- shifts from off-campus housing to on-campus housing which have necessitated a massive expansion of facilities staff, security, dining halls, etc. -- changes in the legal burden that schools need to reach (e.g., title 9)

yummyfajitas 10y ago
You are still not arguing that he's wrong. You're merely arguing that if he were wrong, he might be wrong at other times for similar reasons.
metaphorm 10y ago
author of SSC is an M.D. (psychiatry resident), not specifically a math or tech guy. I agree with you, however, that too much faith in dubious metrics is a serious problem. I don't particularly think the main thesis of today's post is based on dubious metrics though, but perhaps some specific claims in it (about education outcomes, for example) are not well supported. that's beside the point though. he wasn't examining whether or not we got our money's worth on all this increased education spending. he was pointing out that the cost of education, on a per student basis, has risen massively over the last several decades, and there is not particularly a good reason why this should have occurred.