3 ms·

Before we get our pitchforks out here, let's remember that Yale's percentage of Black students is 7.4%[1]. That's not exactly a gigantic portion. In fact Yale i

Before we get our pitchforks out here, let's remember that Yale's percentage of Black students is 7.4%[1]. That's not exactly a gigantic portion. In fact Yale is on the lower end compared to Harvard's 14.3%[2] and MIT's 12% [3]. If Yale is truly discriminating in favor of Black and Hispanic students, well, they're not doing an excellent job of it. [1]: https://oir.yale.edu/sites/default/files/factsheet_2018-19_031020.pdf https://oir.yale.edu/sites/default/files/factsheet_2018-19_0... [2]: https://college.harvard.edu/admissions/admissions-statistics https://college.harvard.edu/admissions/admissions-statistics [3]: https://web.mit.edu/facts/enrollment.html https://web.mit.edu/facts/enrollment.html

dba7dba 6y ago
Meanwhile, I read on this thread that UCSF admitted less Asians. It was 40% last year. It's been now cut to 22%. That is a gigantic shift of a demographic of a new medical school class. Within one year.
virtuous_signal 6y ago
It's actually not very hard to see how discimination-in-favor can be compatible with overall low numbers. The simplest way is if they comprise less than 7.4% of the applicant pool. Another way is if they comprise less than 7.4% of "qualified" applicants, in some definition of qualified. But those breakdowns of the applicant pool are typically not shared except from lawsuits.
swimfar 6y ago
But what percentage of college students are black? (sincere question, couldn't find the answer from a quick search). Without knowing that I don't think it's fair to say they aren't doing a great job. Just because other colleges are doing better, doesn't mean Yale is doing bad. edit: This report says 14% of college enrolled U.S. residents were black in 2017. But that makes it sound like it doesn't include international students. https://nces.ed.gov/fastfacts/display.asp?id=98 https://nces.ed.gov/fastfacts/display.asp?id=98
- hardwaregeek 6y ago
  According to this [1], Black students make up 15% of the enrolled student population. Therefore Harvard, MIT and Yale are all under par. [1]: https://nces.ed.gov/fastfacts/display.asp?id=98 https://nces.ed.gov/fastfacts/display.asp?id=98
Petrova 6y ago
Most of the black students are probably African immigrants. I see it all the time. If you just browse through LinkedIn and look at black Harvard/Ivy League alumni you'll immediately notice all the African immigrants
- xiaolingxiao 6y ago
  This is very true, and speaking as a non-black/african person, I noticed they're very different culturally. They carry a sense of national pride related to their home country, and draw from it constantly. I'm thinking about Nigerian immigrants specifically.
- hardwaregeek 6y ago
  Yes, that's a very interesting phenomenon that should be taken into account. Likewise I've wondered what the sub-demographics are of Asian ethicity students. It's a little ridiculous to lump children of Vietnamese boat people in with nouveau riche direct from China.
- supernova87a 6y ago
  I think you will find this article very relevant. Yet it's 15 years old. This issue has not gone away. https://www.nytimes.com/2004/06/24/us/top-colleges-take-more-blacks-but-which-ones.html https://www.nytimes.com/2004/06/24/us/top-colleges-take-more...
supernova87a 6y ago
But here is the fallacy you have to be very careful about. You cannot say that just 7% is low or 14% is high. Even looking at other universities. You have to know what source population that is being drawn from, and what is the criterion for drawing. It could be that 4% of people applying to Yale are black, 3% are qualified, and 7% outcome is already overrepresented. Or it could be "terrible" if 10% are applying, and 7% are qualified. You have to look at what % pass the filter you are applying, to those who apply. With that further criterion applied, any of these figures could be "reasonable". It is not true that a selective process should simply match the proportions of <x> group that you see in the general population. That is the wrong and very misguided / misrepresented way of reporting the problem. But too often it's the politically simple (or intellectually lazy) approach to framing the issue. It sounds good to say "we should look like our customers". But that is often not true for a selective process. It is true, if and only if, the qualities you're selecting for are represented in an unbiased way, among the people who are your customers, and among those who apply.

dba7dba 6y ago

Meanwhile, I read on this thread that UCSF admitted less Asians. It was 40% last year. It's been now cut to 22%. That is a gigantic shift of a demographic of a new medical school class. Within one year.

virtuous_signal 6y ago

It's actually not very hard to see how discimination-in-favor can be compatible with overall low numbers. The simplest way is if they comprise less than 7.4% of the applicant pool. Another way is if they comprise less than 7.4% of "qualified" applicants, in some definition of qualified. But those breakdowns of the applicant pool are typically not shared except from lawsuits.

swimfar 6y ago

But what percentage of college students are black? (sincere question, couldn't find the answer from a quick search). Without knowing that I don't think it's fair to say they aren't doing a great job. Just because other colleges are doing better, doesn't mean Yale is doing bad. edit: This report says 14% of college enrolled U.S. residents were black in 2017. But that makes it sound like it doesn't include international students. https://nces.ed.gov/fastfacts/display.asp?id=98 https://nces.ed.gov/fastfacts/display.asp?id=98

hardwaregeek 6y ago
According to this [1], Black students make up 15% of the enrolled student population. Therefore Harvard, MIT and Yale are all under par. [1]: https://nces.ed.gov/fastfacts/display.asp?id=98 https://nces.ed.gov/fastfacts/display.asp?id=98

Petrova 6y ago

Most of the black students are probably African immigrants. I see it all the time. If you just browse through LinkedIn and look at black Harvard/Ivy League alumni you'll immediately notice all the African immigrants

xiaolingxiao 6y ago
This is very true, and speaking as a non-black/african person, I noticed they're very different culturally. They carry a sense of national pride related to their home country, and draw from it constantly. I'm thinking about Nigerian immigrants specifically.
hardwaregeek 6y ago
Yes, that's a very interesting phenomenon that should be taken into account. Likewise I've wondered what the sub-demographics are of Asian ethicity students. It's a little ridiculous to lump children of Vietnamese boat people in with nouveau riche direct from China.
supernova87a 6y ago
I think you will find this article very relevant. Yet it's 15 years old. This issue has not gone away. https://www.nytimes.com/2004/06/24/us/top-colleges-take-more-blacks-but-which-ones.html https://www.nytimes.com/2004/06/24/us/top-colleges-take-more...

supernova87a 6y ago

But here is the fallacy you have to be very careful about. You cannot say that just 7% is low or 14% is high. Even looking at other universities. You have to know what source population that is being drawn from, and what is the criterion for drawing. It could be that 4% of people applying to Yale are black, 3% are qualified, and 7% outcome is already overrepresented. Or it could be "terrible" if 10% are applying, and 7% are qualified. You have to look at what % pass the filter you are applying, to those who apply. With that further criterion applied, any of these figures could be "reasonable". It is not true that a selective process should simply match the proportions of <x> group that you see in the general population. That is the wrong and very misguided / misrepresented way of reporting the problem. But too often it's the politically simple (or intellectually lazy) approach to framing the issue. It sounds good to say "we should look like our customers". But that is often not true for a selective process. It is true, if and only if, the qualities you're selecting for are represented in an unbiased way, among the people who are your customers, and among those who apply.