3 ms·

Doesn't that enhance the effect of the gate rather than detract from it? If, ceteris paribus, being around the shiniest of gold stars helps you shimmer a littl

Doesn't that enhance the effect of the gate rather than detract from it? If, ceteris paribus, being around the shiniest of gold stars helps you shimmer a little brighter, that seems likely to make the gating-function effect more pronounced, not less. My criticism of the article is that there's no alternative solution given. You want to abolish all admissions? That's cool. What do we replace it with? A lottery seems like a lead balloon and anything else seems to have an effective gatekeeping function . . .

marcinzm 3y ago
One of the points of the article seemed to be to do gating latter rather than earlier. I'm saying that this is detrimental as being around the dullest stars will make you a dull star more often than not.
- joelfried 3y ago
  Perhaps I missed it but I didn't see the author put forward a mechanism for how the "gate later" approach was supposed to work. Did I miss it? To your point though, absent an implementation of "gate later", it's not clear to me that what the author would want would put you near the "dullest stars". There's no way Harvard (or any other Ivy+) would accept a "gate later" implementation that would give them the dullest of stars. The differentiators between folks with any non-zero chance of acceptance are already pretty small, and I don't think "allow everyone in" would be feasible as they would not have room to put all of the people who want to be there.
cosmojg 3y ago
> You want to abolish all admissions? That's cool. What do we replace it with? A lottery seems like a lead balloon and anything else seems to have an effective gatekeeping function If an institution doesn't want to simply accept the top N performers on one or more standardized tests, they could go one step further and perform an SAT/ACT-weighted lottery followed by a culling. This can be per-major or per-department, but many universities already enforce a set of near-universal required courses. Set a minimum performance requirement on those courses and bada bing bada boom, the institution is getting the cream of the crop for whatever quantitative definition of "cream" they want to set. People will be naturally disincentivized from applying to overly competitive programs in which they're likely to fail. There are probably many effective variations of this idea, too, like using a geographically distributed population-weighted lottery for applicants meeting some threshold score instead. What are the failure modes of such quantitative approaches? Could the outcomes be worse than those of our current system? What are some better alternatives?

marcinzm 3y ago

One of the points of the article seemed to be to do gating latter rather than earlier. I'm saying that this is detrimental as being around the dullest stars will make you a dull star more often than not.

joelfried 3y ago
Perhaps I missed it but I didn't see the author put forward a mechanism for how the "gate later" approach was supposed to work. Did I miss it? To your point though, absent an implementation of "gate later", it's not clear to me that what the author would want would put you near the "dullest stars". There's no way Harvard (or any other Ivy+) would accept a "gate later" implementation that would give them the dullest of stars. The differentiators between folks with any non-zero chance of acceptance are already pretty small, and I don't think "allow everyone in" would be feasible as they would not have room to put all of the people who want to be there.

cosmojg 3y ago

> You want to abolish all admissions? That's cool. What do we replace it with? A lottery seems like a lead balloon and anything else seems to have an effective gatekeeping function If an institution doesn't want to simply accept the top N performers on one or more standardized tests, they could go one step further and perform an SAT/ACT-weighted lottery followed by a culling. This can be per-major or per-department, but many universities already enforce a set of near-universal required courses. Set a minimum performance requirement on those courses and bada bing bada boom, the institution is getting the cream of the crop for whatever quantitative definition of "cream" they want to set. People will be naturally disincentivized from applying to overly competitive programs in which they're likely to fail. There are probably many effective variations of this idea, too, like using a geographically distributed population-weighted lottery for applicants meeting some threshold score instead. What are the failure modes of such quantitative approaches? Could the outcomes be worse than those of our current system? What are some better alternatives?