Hacker News Reader

Comment by Bostonian | Hacker News Reader

"While this is a substantial change in the demographic composition of the Class of 2028 compared with recent years, I want to be clear that it does not bring any aggregate change in the quantifiable characteristics we use to predict academic success at MIT, such as performance in high school or scores on standardized tests. By these measures, this cohort is no more or less prepared to excel in our curriculum than other recent classes that were more broadly diverse."

I doubt this and would like to see the numbers. To simplify a bit, racial preferences existed because relatively few blacks and Hispanics had the test scores needed to get into MIT. If MIT has stopped using racial preferences, the average test scores of admitted students should have risen a little.

View on HN

Not necessarily - if you assume there are more qualified entrants than there are seats available (as is true for MIT), then scores could be unchanged - e.g. there could be, say 200 students in line to get in, all have 1600 SAT scores, 100 are black/Hispanic; 100 are Asian/white. Now the question is what is your tie-breaker - the old tie-breaker was Affirmative action, so more of the former got in, the new tie-breaker is, say, volunteer work, and more of the latter qualify for that and get in

In your example, the average SAT score will be unchanged at 1600, but the new student body will still be better from the point of view of volunteer work experience

And extracurriculars, which are expensive.

Or just have harder tests.

I imagine that more people applying to MIT than they can accept have maxed out all of the opportunities they've been given.

"Relatively few blacks and Hispanics had the test scores needed to get into MIT" because:

- These racial groups have far fewer parents who can afford the coaching and external resources required to have kids achieve those test scores

- These racial groups were subject to a long history of systemic segregation that locked them out of opportunities to build generational wealth during one of the most opportune times in the US's history to do so (post-World War II)

That is part of the reason why schools like MIT chose to use race as a factor.

Tons of literature has been (and continues to be) written on this topic.

MIT admits presumably ~1000 students as freshmen each year (divide ~4k by 4 [1])

A baseline of 25% means 250 students. Are you going to make the claim that there aren't 250 students who identified as Black, Hispanic, and/or Native American and Pacific Islander that can't get either a perfect score or a score higher than somebody outside their group but admitted to MIT?

[1]: https://registrar.mit.edu/stats-reports/enrollment-statistic...

My assumption is that they exist, but are now competing in a larger pool of asian and white applicants - who may have better access to extracurricular opportunities that are still considered in admissions decisions - and therefore are less likely to be admitted.

Haha it is true that one solution to this problem is to truncate test scores on the right so your pool of applicants all look the same academically--then you can use swarthiness as a tiebreaker!

>Are you going to make the claim that there aren't 250 students who identified as Black, Hispanic, and/or Native American and Pacific Islander that can't get either a perfect score...

Sometimes that is what happens with some standardized tests. For the LSAT for example, there are years where no black or latino test takers score above 174 on the LSAT. See e.g., figure 14 here: https://www.lsac.org/sites/default/files/research/tr-22-01_j...

Interesting thing about the Gaussian curves on pages 34 & 39 giving the performance by gender and race: the smartest group, by quite a bit, is the group that put their race as "No Response".

Personally, my race is "human" (I swear!) and I identify as not identifying much, so go go team Non Responder!

>Are you going to make the claim that there aren't 250 students who identified as Black, Hispanic, and/or Native American and Pacific Islander that can't get either a perfect score or a score higher than somebody outside their group but admitted to MIT?

As AlanYx said, yes, that's quite possible.

Another example: Americans do well on PISA compared to their ethnic relatives. <https://www.unz.com/isteve/the-new-2018-pisa-school-test-sco...> Asian Americans do better than Asians; whites do better than Europeans; Latinos do better than Latin Americans; and blacks do better than Africans.

Hispanics and especially blacks' scores drag the US average down. Both white and Asian Americans score higher than Canada (and white+Asian is essentially Canada's racial makeup), and higher than New Zealand, Australia, the Netherlands, Belgium, Germany, France, Norway, and the UK; Estonia is below US Asians but above white Americans. Norway is by far the wealthiest Nordic state but its average is only two points higher than the US national average, despite not having a demographic that is 13% of the population and scores 85 points below the white American average.

(The above discusses 2018 results. 2022 results: <https://x.com/cremieuxrecueil/status/1732087511327908128>)

Bar for admission and bar for success are very different things. MIT admits ~1100 students a year, but far more than 1100 globally have the drive and ability to succeed in the program. So they have to set some criteria. If the filter is arbitrarily "must have a 1580 SAT score to get in" and the applicant has 1560 instead, that doesn't mean they are doomed to fail.

> To simplify a bit, racial preferences existed because relatively few blacks and Hispanics had the test scores needed to get into MIT.

I think you’re just claiming the opposite of the quote here. I feel that doesn’t really add anything to the conversation, your claim also lacks any numbers. I do agree numbers would be helpful though — I’ll request numbers from both sides please :)

I thought this point was addressed in the interview.

Reply on news.ycombinator.com