Is this normalized by number of each demographic that sits for the test? (Sorry if that's answered in the PDF but maybe someone with more time or statistical understanding than me can check it.)
If the mainstream discourse was "yes, whites are underrepresented, but they deserve it because they're dumb", that would be progress compared to the current "there are too many whites in the Ivies due to white supremacy (and the SAT is racist)" narrative. Sources:
If the mainstream discourse was "yes, whites are underrepresented, but they deserve it because they're dumb",
This is the mainstream discourse imo. The SFFA case didn't argue against affirmative action for non-white people, they argued against it for white people. They tried to make the case that whites were disproportionately legacy and athlete admits so therefore were dumb. (In my other comments I show that legacy students at Harvard actually had higher SAT scores). The SC of course had an ax to grind against Affirmative Action but the actual evidence, especially in light of the SAT distribution from GP, that there was AA for whites specifically I think is extremely lacking.
It's quite the opposite. Actually, using the same arguments SFFA made, it looks like Ivies discriminated against whites more than Asians. Remember that almost 1 million white students took the SAT in 2021. That leaves a huge number of high scoring students but somehow between only about 25-40% representation at Ivy League and Ivy League adjacent universities.
I think the myth of the privileged, entitled dumb white student with rich parents is pretty commonly used now. The excuse that high scores are a result of having rich parents who bought tutoring only seems to matter when the race of the student is white. It's ironically also the case that whites are the least likely group to take test prep and that test prep only seems effective in the style that East Asians tend to use which is the cram school variety or hagwon.
The SFFA case used some research from Princeton(I was told this so I never verified it was from Princeton) which showed that among Asians with a 1550+ on the SAT, the top one percentile of Asian students taking the test exceeded the number of seats at Harvard, etc. for freshman. The SAT has some stats here https://satsuite.collegeboard.org/media/pdf/sat-percentile-r...
So the SFFA case argued that the Ivies must be suppressing high scoring Asians. Although ultimately the court case came down to whether or not it was legal for schools to use race for admissions. But, for example, nearly 1 million white students took the SAT in 2020. The top 1 percentile of 1 million is ~10000. Yet the white enrollment at Harvard was only about ~37% that year but discrimination against whites didn't seem to be a factor because SFFA argued that legacy and athletic admissions were secret backdoors allowing AA for white students, without any evidence for this claim.
There is a fair bit of research done about the East-Asian style "shadow education" as they call it. It's worth looking into. As well, the College Board has historically not done a good job of securing the test especially on the international level where a lot of cheating scandals have arisen. The same test prep companies that operate in China and Korea also have operations in the US. This is worth looking at https://theconversation.com/test-prep-is-a-rite-of-passage-f... because it presents evidence that test prep is only really effective in the way East Asians use it which is for long periods of time.
Furthermore, why use 1500 SAT score? Because it gives the distribution you like? Harvard's cutoffs are 1100 for Blacks, Native Americans and Hispanics, 1350 for Asian women and 1380 for Asian men, and 1310 for "white wannabe students in states that don’t see a lot of Harvard attendees":
Comments
Distribution of 1500+ SAT Scores
White 0.381106
Black/African American 0.0544724
Hispanic/Latino 0.111558
Asian/Asian American 0.414271
Native Hawaiian/Pacific Islander 0.00100503
American Indian/Alaska Native 0.00261307
Two or More Races 0.0349749
Sources:
https://nces.ed.gov/programs/digest/d19/tables/dt19_226.10.a...
https://satsuite.collegeboard.org/media/pdf/sat-percentile-r...
Is this normalized by number of each demographic that sits for the test? (Sorry if that's answered in the PDF but maybe someone with more time or statistical understanding than me can check it.)
Out of all the people that scored 1500+ on the SATS, this is the percentage breakdown by race.
If the mainstream discourse was "yes, whites are underrepresented, but they deserve it because they're dumb", that would be progress compared to the current "there are too many whites in the Ivies due to white supremacy (and the SAT is racist)" narrative. Sources:
https://news.ycombinator.com/item?id=36992516
https://news.ycombinator.com/item?id=36992739
This is the mainstream discourse imo. The SFFA case didn't argue against affirmative action for non-white people, they argued against it for white people. They tried to make the case that whites were disproportionately legacy and athlete admits so therefore were dumb. (In my other comments I show that legacy students at Harvard actually had higher SAT scores). The SC of course had an ax to grind against Affirmative Action but the actual evidence, especially in light of the SAT distribution from GP, that there was AA for whites specifically I think is extremely lacking.
It's quite the opposite. Actually, using the same arguments SFFA made, it looks like Ivies discriminated against whites more than Asians. Remember that almost 1 million white students took the SAT in 2021. That leaves a huge number of high scoring students but somehow between only about 25-40% representation at Ivy League and Ivy League adjacent universities.
I think the myth of the privileged, entitled dumb white student with rich parents is pretty commonly used now. The excuse that high scores are a result of having rich parents who bought tutoring only seems to matter when the race of the student is white. It's ironically also the case that whites are the least likely group to take test prep and that test prep only seems effective in the style that East Asians tend to use which is the cram school variety or hagwon.
Could you elaborate on this?
I'd love a source for this, if you still have it.
The SFFA case used some research from Princeton(I was told this so I never verified it was from Princeton) which showed that among Asians with a 1550+ on the SAT, the top one percentile of Asian students taking the test exceeded the number of seats at Harvard, etc. for freshman. The SAT has some stats here https://satsuite.collegeboard.org/media/pdf/sat-percentile-r...
So the SFFA case argued that the Ivies must be suppressing high scoring Asians. Although ultimately the court case came down to whether or not it was legal for schools to use race for admissions. But, for example, nearly 1 million white students took the SAT in 2020. The top 1 percentile of 1 million is ~10000. Yet the white enrollment at Harvard was only about ~37% that year but discrimination against whites didn't seem to be a factor because SFFA argued that legacy and athletic admissions were secret backdoors allowing AA for white students, without any evidence for this claim.
https://academic.oup.com/sf/article-abstract/89/2/435/223527... <- if you can access this article. But here is a quote from a different researcher citing this article (https://www.jstor.org/stable/40984541):
Blacks and Hispanics are more likely than whites from comparable backgrounds to utilize test preparation
This paper also shows a similar outcome and includes Asians. Asians are the most likely to take test prep. https://journals.sagepub.com/doi/abs/10.3102/000283121142560...
There is a fair bit of research done about the East-Asian style "shadow education" as they call it. It's worth looking into. As well, the College Board has historically not done a good job of securing the test especially on the international level where a lot of cheating scandals have arisen. The same test prep companies that operate in China and Korea also have operations in the US. This is worth looking at https://theconversation.com/test-prep-is-a-rite-of-passage-f... because it presents evidence that test prep is only really effective in the way East Asians use it which is for long periods of time.
Furthermore, why use 1500 SAT score? Because it gives the distribution you like? Harvard's cutoffs are 1100 for Blacks, Native Americans and Hispanics, 1350 for Asian women and 1380 for Asian men, and 1310 for "white wannabe students in states that don’t see a lot of Harvard attendees":
https://nypost.com/2018/10/17/harvards-gatekeeper-reveals-sa...