Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

But this isn't just a typical year. This immediately after racial discrimination was banned. Yale had previously insisted that absent race-based affirmative action there'd be an even larger overrepresentation of Asians and reduction in diverse student enrollment. This is what was observed at other universities, like MIT [1].

Instead the group that the Supreme Court had determined was being discriminated against in SFFA vs. Harvard saw a decline when this discrimination was (supposedly) removed.

Imagine a company is taken to court and found to have been discriminating against women. They insist that they've resolved the discrimination, but next year their number of women hired is even lower. That doesn't look suspicious at all?

1. Before racial discrimination was prohibited: https://mitadmissions.org/apply/process/composite-profile/

2. After: https://mitadmissions.org/apply/process/profile/



It’s literally not statistically significant at all. Numbers of students ebb and flow, as does their makeup. Asians are no less represented than they were a few years ago, if you believe that non-affirmative action is “racist” then how do you explain the previous dips when AA was still around?


For the third time, you're ignoring the fact that this the the first year of admissions after racial discrimination was banned. Many other elite institutions saw rises in admissions of Asian applicants. The courts found that race based affirmative action suppressed Asian representation. Attributing the decline to noise and ignoring the fact that this is the first year that anti-asian discrimination was supposedly banned is a very naive analysis.

Again: Imagine a company is taken to court and found to have been discriminating against women. They insist that they've resolved the discrimination, but next year their number of women hired is even lower. That doesn't look suspicious at all?


No, it’s not suspicious. And for the second time, the reason is because it’s NOT statistically significant. I don’t care whether it was the first year, a singular data point is not proof of anything suspicious, especially when it fits within typical statistical data. Please, for the love of God, take a stats class before trying to read a graph and draw conclusions. Your supposed gotcha is “hmmm isnt is suspicious if X is less than I think it should be?” When X is altered by so many variables that it is LITERALLY IMPOSSIBLE to draw your conclusions from it.

Maybe less asians applied? Maybe less asians qualified? Maybe more asians bowed out than normal and runnerups took their slots, maybe, due to negative publicity, the asians went to different colleges? That seems far more likely than your conspiracy theory.


The fact that Asian enrollment is about the same as it was during the years that racial discrimination was legal is exactly why it's suspected that Yale is still engaging in discrimination. I'm seriously confused as to why you think you're helping your argument by pointing out how similar Yale's enrollment is post-AA ban as it was pre-AA ban.

Imagine University A stops discriminating against Asians and University B decides to continue affirmative action secretly. Which one would have admissions rates in line with years when affirmative action was legal? Which one would see a rise in Asian enrollment?

> Maybe less asians applied? Maybe less asians qualified?

Publishing the stats on how many Asians applied and the average SAT scores of Asian admits and diverse admits would shed light on this. Notable, Yale has not released this data.


> It’s literally not statistically significant at all

You have no basis on which to make that claim at all. We cannot infer the variance from this chart.


> We cannot infer the variance from this chart

We can observe similar drops in Asians' share of admission, as well as similar levels, before the race-neutral treatment began. That's enough to, at a glance, dismiss this as evidence of anything per se.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: