Right, because the policy at highly scientifically productive countries is to hire scientists (or student scientists), not hire an ethnic group due to their ethnic group’s proclivity for science, regardless of their actual capability.
-
This Tweet is unavailable.
-
-
Replying to @elliot_leonie @negatingspirit and
If you considered both kinds of info, you'd do better.
2 replies 0 retweets 11 likes -
Replying to @gcochran99 @elliot_leonie and
Right. Considering nationality among immigration applicants is like how Harvard treats a 3.9 GPA from Loyola HS as a better predictor of success at Harvard than a 3.9 GPA at Michelle Obama Global Success Academy.
1 reply 1 retweet 12 likes -
Replying to @Steve_Sailer @elliot_leonie and
Worse than that: you need to adjust standardized test scores in a ways that allow for regression to different means.
2 replies 1 retweet 8 likes -
Replying to @gcochran99 @elliot_leonie and
That's probably a bridge too far even for me: if the black kid gets a 1500/1600 on the SAT, I'm fine with colleges treating that as legit, even though it's more likely to have been a lucky fluke if the applicant is black than if he is Asian.
2 replies 0 retweets 10 likes -
Replying to @Steve_Sailer @elliot_leonie and
If we were recruiting for some job that actually mattered - something our fate depended on - naturally, you'd want to do that suboptimally.
0 replies 0 retweets 10 likes -
This Tweet is unavailable.
-
Replying to @negatingspirit @Steve_Sailer and
Yep. Loosely related: what fraction of the people running the show understand regression to the mean? Surely not 1 in 20.
2 replies 0 retweets 11 likes -
Replying to @gcochran99 @negatingspirit and
For clarity there are two different relevant effects in the area that can be called "regression to the mean" Meaning 1 - High IQ person from a low IQ group, likely to have lots of IQ boosting random alleles - children will be lower IQ because won't get as many at random.
2 replies 0 retweets 3 likes
Meaning 2 - Hi *measured* IQ test result of a person from a lo IQ group - the test is likely to have erred in measuring IQ too high because of variance + Bayes theorem. A single IQ test is evidence of individual IQ but group membership is also evidence and both have to be weighed
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.