From the article: ""Americans have come to view diversity as integral to learning and to trust that the path to leadership is open to all," lawyers representing Harvard University wrote in a court filing."
Do you see the "open to all" statement? Is this true?
What about the Asians who were denied entry based on their race? Is this "open to all"?
Which precedent? The one where the 14th Amendment forbids racial discrimination, or the narrow carve out 20 years ago that has allowed it to continue in higher education?
It is when "a court decides what the outcome of the case should be and then works backward to determine the reasoning that will reach the desired conclusion"
If you think conservatives have a monopoly on this, then take a look at some of the most important Supreme Court cases from FDR through the 1970s or so.
The Supreme Court is a political organization that likes to pretend it's something else. I think that's always been the case and I agree it isn't inherently one party or the other.
That said, with the notable exception of the Warren court, SCOTUS has been conservative for its whole history.
>That said, with the notable exception of the Warren court, SCOTUS has been conservative for its whole history.
I think that depends a lot on one's perspective. If you're very far-left and support all the equity you can dream up, then yes I see how one might think that. I disagree, but still...
If that were objectively true, I don't think it's the damning claim you may have meant it to be if you consider the overall history of the USA and contributions to the world.
The US constitution and the bill of rights are based on classical liberal ideas, like inalienable rights, the consent of the governed, secularism, etc.
Not at all. Those ideas were formed in a time when they were not the norm. State religion is about as conservative as you can get, it has existed for all of human history, and was the default until very recently. When the US constitution was adopted it was a progressive change. However, yesterday's progressivism is always today's status quo.
SCOTUS has frequently made decisions that overturned a conservative prescedent in favor of a liberal one. Many 'landmark' SCOTUS cases are exactly this.
It strikes me as naive to think it would ever work otherwise. Justices are political appointees and political actors. I see no reason to expect anything different. Believing they are somehow "above politics" strikes me as similarly fanciful to believing in the divine right of kings.
Judges are NOT political appointees. Within the civil service a political appointee is one that can be and usually is dismissed in a change of government. Heads of agencies and such. Supreme Court justices serve lifetime appointments specifically so that they are not subject to political interference.
Words have meaning, even if that meaning is not clear from the direct composite words used, as in this case. The word “political appointee” has the technical meaning I stated.
Appointed by politicians = political appointee in my book. But yes, they are lifetime political appointees.
"Supreme Court justices serve lifetime appointments specifically so that they are not subject to political interference."
I also think this is naive. In my view, they serve lifetime appointments because they don't want to give up power, and no one's been able to force them to so far.
(1) it is not the case that they serve lifetime appointments "because they don't want to give up power." They serve lifetime appointments because the framers of the constitution decided it was better for the court to be structured that way hundreds of years before they were born.
(2) Forcing them to give up power, as you put it, would require a constitutional amendment, which is not something any one person can do.
What I'm questioning is that the motivation for lifetime appointments ever had anything to do with avoiding political interference. Regardless of the claimed justification, I think lifetime appointments were just a naked, cynical power grab by factions that favored oligarchy over democracy, and have remained so to this day.
On the contrary, any officer appointed by the President, Vice President, or an agency head, with or without Senate confirmation, is a “political appointee”.
> Within the civil service a political appointee is one that can be and usually is dismissed in a change of government.
It is true that within the executive branch, political appointees often serve at the pleasure of the President (though even there are a number of exceptions where political appointees have fixed terms and cannot be arbitrarily dismissed.)
There are plenty of liberal reasons to dislike race-based admission processes as well. I agree with the spirit of what these programs are trying to accomplish, but punishing Asian applicants for their race is a perfect example of what is wrong with this solution from a fundamental level.
This is the distinction between old-school liberal (judge by content of character not colour of skin; what you do behind closed doors is your business as much as possible; free speech) and more recent liberal, which seems far more likely to keep the middle of those three, but jettison the others. Bill Maher's journey over the last 2 years has been interesting to watch as he tries to berate the modern left into returning to liberal roots.
I don't think so. Old school liberal and modern libertarian are fairly closely related. The term libertarian has appeared I think due to the changing of the word liberal.
> MLK, of course, coined the phrase "content of character not color of skin" and he was very much in favor of affirmative action.
Sure, at least sort of. But this conflates a principle (the thing I quoted) with one man's politics.
This court has been shredding the rule of law in this country for the last year, and is systematically replacing our democracy with fascism.
Their ruling is definitely going to be more damaging to minorities than anything I can come up with.
Hell, they banned abortion by eliminating the right to privacy in the US.
They were even complete assholes about it. Despite being minority (and arguably illegally, in some cases) appointed, they blame the (underage) women their rulings victimize for not voting enough.
The text of the ruling explicitly states that no ruling the court makes could be sexist or victimize women, since women are allowed to vote.
I fully expect this ruling to use similar logic to blow an equally large hole in our constitutional rights.
I’m certainly not defending this court, their motivations, or any tangential effects their opinions might contain.
Purely on the merits of the topic: taking an admissions slot from an Asian student and giving it to someone of a different racial background is, plainly, racial discrimination.
If we want to help people of a disadvantaged background, we need to find a better way to measure that directly rather than using race as a proxy. Using race as a proxy in prejudging someone isn’t okay even if you think you’re right. Prejudiced people don’t think that they’re wrong.
Yes, that is what's being claimed by the plaintiff and if you accept their characterization it does seem like an easy case. That's probably true for most cases.
> This court has been shredding the rule of law in this country for the last year, and is systematically replacing our democracy with fascism.
If you think that, it's probably a sign you need to come up for air. What you write is simply not true unless you're examining things with a partisan funhouse mirror that shows myside as righteous and true and the otherside as cartoon villians. Admittedly, those mirrors are getting handed out right and left, because they're extremely useful for partisan GOTV efforts.
What do you think is the conservative outcome? It is not so obvious. The allegation is that Harvard is discriminating against Asians (and thus favoring Whites). California colleges are not allowed to consider race. Check Asian vs. White student percentages at UCSD. Is that the outcome conservatives want?
Republicans have been generally against affirmative action since its inception. Trump was explicitly against it. The particulars of this case and this argument don't really matter.
I find this view to be pretty naive. The vast majority of judges in the US legal system are naturally going to be conservative in that the job is to follow the letter of the law. To the extent that progressive justices move up through the legal system is only through political appointments.
It seems easier to me to say the vast majority of progressive opinions from the Supreme court have been outcomes driven and the current court is reverting to process-driven (you have to actually pass legislation for legislative changes).
> To the extent that progressive justices move up through the legal system is only through political appointments.
All federal and most state justices only move up through political appointments. The rest of state justices are directly elected, which is even more political.
> All federal and most state justices only move up through political appointments.
Federal justices have either 0 or 1 steps they can move up, and of the ones that can move up, there is only 1 position for them to move into, that of Chief Justice.
> The vast majority of judges in the US legal system are naturally going to be conservative in that the job is to follow the letter of the law.
C'mon, who's being naive now?
The law was already settled on this issue. This case is being heard with the goal of overturning precedent. It's not because we learned new things about the Constitution in the last 20 years.
Nobody--yourself included--wants us to honor every bad precedent. Should we be stuck with Citizens United for all of eternity? Of course not--everybody wants that sumbitch overturned!. Pretending that you believe precedent to be sacred is both silly and transparently self-serving.
> The law was already settled on this issue.
The idea of "settled law" is a rhetorical device and nothing more. SCOTUS has been overturning precedent on an almost annual basis for centuries[0].
> It's not because we learned new things about the Constitution in the last 20 years.
Funnily enough, from the majority opinion of the "precedent" they may overturn: "The Court expects that 25 years from now, the use of racial preferences will no longer be necessary to further the interest approved today."
""Americans have come to view diversity as integral to learning and to trust that the path to leadership is open to all," lawyers representing Harvard University wrote in a court filing."
This is not true, and the Asians who were denied entry based on their race would agree.
I mean with reference to a sophistry technique to shut down discussion. Were the individual example to be computer scientists instead of surgeons, then the definition of a successful computer scientist can be redefined to a suitable outcome as proof that racial discrimination does not harm consumers of surgeons or computer scientists. Its avoiding the core discussion via wordplay. Furthermore if no individual example were provided then the sophistry method would be to complain there is no specific example of an individual being harmed therefore there is no harm and racial discrimination is an inherently good process.
I don't know why you believe the purpose of medical school is to predict how good a doctor is at surgery. Most doctors aren't going to be surgeons.
I've never been to med school but my understanding is you learn how to be a doctor in residency, where there are no grades. Medical school gives you theoretical knowledge but not the practical knowledge you need to actually do the work.
Lots of people don't care about what doctor they get. I have heard endless stories from doctors that see patients all day where so many patients don't care about their own health to the point of having limbs amputated and shrugging it off like its no big deal. Or having complications from diabetes and declining free-to-them care. Not taking medication. There are so many of these people. They seem to be mostly on Medicare from the stories. If you care about your health there are options out there unless you have to deal with the VA because of your coverage. Then you're genuinely screwed. For those that don't care about their care there are reasonable defaults available.
Studies have shown that surgery outcomes are highly correlated with experience in that particular procedure. So if I needed a heart bypass operation, I would want the surgeon who does 5 of them every week and nothing else. Where they attended medical school is unimportant.
You might have a stronger argument in other medical specialties where theory is more important. But heart surgeons are more like athletes in a way.
We should do what pilots have to do: start with simulators, then move to copilot, then primary pilot with an experienced copilot, then finally primary with inexperienced copilot. The medical system gets the later parts right, but there are no simulation components at the beginning.
IMO it would even be helpful for surgeons to do some time as carpenter, mechanic, fitter/turner etc for some time to really drill the hand/eye before they start on humans.
> We should do what pilots have to do: start with simulators, then move to copilot, then primary pilot with an experienced copilot, then finally primary with inexperienced copilot. T
And this is EXACLY what they do, using cadavers, schooling, and internships. Their work on cadavers is graded, much like most of what they do. This goes back full circle, grades do have an input into the quality of a surgeon. It may not be the "best" or "only" indicator but it shouldnt be outright discounted.
People with high GPA's work hard to get them and have a high level of understanding of the topics at hand.
My ex is excellent in math and always scored very high on tests, but she is not very practical.
If i a math problem i'd have full confidence her answer was correct. If i asked her some non-math questions my trust would be adjusted.
Me too. Unfortunately new doctors have to start practicing every now and then, or we'll run out of experienced ones. Which new surgeon would you prefer.
Contrary to the a priori hypothesis that care would be compromised by so many of the top cardiologists being gone, they find that adjusted mortality for high-risk heart failure and cardiac arrest is lower at teaching hospitals during the cardiology meetings than during the rest of the year. And although patients were less likely to get a percutaneous coronary intervention for a myocardial infarction during the convention times, there was no associated decrement on myocardial infarction–associated mortality.
Two students are under consideration to get into University to become surgeons.
One student went to the most expensive private school and had private tuition. They recieved AAABB. Anothers student went to a poor school and never had private tuition and recieved ABBBB. Which should be accepted for a single University place?
The one with more As. Standardized tests are actually one of the most fair ways to evaluate two candidates. y = mx + b regardless of your skin color or family wealth. Is it perfect? no; is it more fair than the alternatives? I think so
The best engineers I've worked with were ones that performed terribly at college, so I'm not sure about using GPA as a proxy for how well they perform in the workforce.
This is a funny example because I'm probably the only person in this thread who went to medical school (MD programme in Canada, but did not complete or obtain my degree), and I think this is a bad example.
The meritocracy trap obviously does not imply that competence/merit does not matter: it implies only that we should be much more careful and critical about whether we measure competence/merit fairly and properly, especially a posteriori.
I would say that your example illustrates a very shortsighted conception of competence/merit.
A true meritocracy wouldn't be a friendly world to live in.
Think about the current state of education. Parents with resources over-optimize their kids' education because they understand the inefficiencies in the system and/or understand what it takes to become successful in our society.
some may argue that affirmative action, as it currently is implemented, does not directly address disparities in resources, but rather uses race as a proxy for resources (which would be controversial)
> Parents with resources over-optimize their kids' education because they understand the inefficiencies in the system and/or understand what it takes to become successful in our society
This is a perfectly fine response to meritocratic pressure. With tuning to ensure the response is productive learning, versus e.g. rote memorisation, the outcome should be a more-productive citizenry.
In any event, the case is about race-based affirmative action. Nobody is going after means-based anything.
I'm not sure about how I feel about Affirmative Action and I do appreciate another member's suggestion to look at socio-economic background. But is this really our goal?
> should be a more-productive citizenry
I'm not sure I care that much about that. Especially w.r.t college admissions. If you wanted to be more productive you'd not even send many people to college in the first place and just have them go straight into the labor force.
> wanted to be more productive you'd not even send many people to college in the first place and just have them go straight into the labor force
This is incorrect. The question of education's impact on GDP per capita is thoroughly researched [1][2][3]. When one broadens to measuring living standards, the results get starker.
Yeah this is the problem with meritocracy, it doesn't exist in a vacuum. The reason someone has merit is arguably _only_ due to external factors. You are the way you are because of your DNA and environment. You are who you are because of who your parents are and how they chose to raise you, but they are only who they are because of their parents. And that recurses all the way back to the first organisms.
Even a staunch defender of meritocracy will admit that life isn't fair if they are honest.
The important thing to consider is the benefits afforded by meritocracy. When we have adequate filtering mechanisms to help put competent trustworthy people in positions of power, we get less bridges that collapse, less space shuttles that explode, less government officials demanding bribes, and less subtle failures at many levels of the economy. These failures are years downstream of the decisions that set people on their path in life. Most people have not yet mastered their profession; helping this process along is good for everyone.
Those who criticize meritocracy should be honest with themselves as well that these failures can happen.
That's only part of the equation. The fact is, generally, anyone in the US who has a capable brain and really tries can excel. There's no shortage of examples of people growing up poor who did excellent in school and/or careers. Do they have a disadvantage in not having the same resources as a wealthy family? Definitely. But it's only a disadvantage, not a roadblock.
I feel like a lot of the argument is little more than regret filled excuses and sour grapes.
> The fact is, generally, anyone in the US who has a capable brain and really tries can excel.
This is a great populist message but not sure it can be backed up with enough data at any meaningful enough scale to enact policies. Just like if I posited, "Generally, there are a lot of smart, capable and hardworking people who are just surviving check to check." It feels good and there are plenty of anecdotes to back it up, but it's too vague to make policy decisions upon. Raj Chetty's research [0] on social mobility dives into the complexity of it all.
You really nailed the crux of it here in terms of equality of opportunity vs. equity in opportunity. The unhappy compromise to avoid irreparable divisions in U.S. society lies somewhere between removing roadblocks vs. addressing disadvantages.
Well it's easier for me to say, having grown up relatively poor in a relatively poor area. The kids who listened, took notes, participated, for the most part all did well in school and went off to college. Most kids did not, they were disruptive, asleep, rude, violent, etc. Public school was very frustrating to me for that reason, so I guess it struck a nerve.
The equity vs equality is hard to square. Imagine someone who came from real poverty who worked 2 jobs while studying to give his family a better life. He or she makes it, finally, only for them and their kids to be punished for it come admissions time. Is that fair? I don't have great answers, I just feel like everything they try today isn't right, either.
I empathize with your experience having also grown up in relative poverty. But now that I have kids of my own and see other kids being disruptive, asleep, rude, violent, etc. I'm looking at those kids' parents and guardians like WTF?
So just like it's not fair for the person working hard and studying to get their family a better life (equality of opportunity), it's also not fair to penalize minors who don't have those types of parents or can't outlast their broken environment to reach a better life (equity in opportunity).
I don't have the answers either but think we should keep trying rather than accept some large number of kids can work hard and still fail due to circumstance beyond their reach. Especially when at the same time, some large number of kids can just be kids and eventually figure it out due to circumstance beyond their reach.
If it were a basketball training program you were applying to, that would make sense. I fail to see your point regarding universities using academic merit as admission criteria.
I think the correct move here is not to go full-on "merit-based access", but change the affirmative action to be based on socio-economic status. Boom, you just solved the issue the race-based AA was claiming to address, but by going directly to the source of the problem (socio-economic inequality).
I am not saying this out of some abstract concern about asian kids with perfect scores failing to get admitted to Harvard. While that's definitely a problem, imagine how much more difficult it is for asian kids from poor socio-economic backgrounds to get literally into any decent college. They dont have the socio-economic advantage of upper-middle class asian kids, but have all the disadvantages of coming from a poor family AND then get shafted by race-based affirmative action on top of that. This is just plain disgusting in my eyes.
I realize that this approach might fail to address everything perfectly, but it is better than race-based AA on literally every single measurable metric. So we might as well switch to that, while in search of a "perfect" solution.
That already exists. Universities have quotas for a certain number of poor kids and/or first generation college applicants. Race-based affirmative action is a separate category of discrimination.
Maybe some tiny percentage of wealthy can game it, i honestly dont know much about MediCAL to addres it directly.
However, I would much rather have a tiny percentage of wealthy game it, than a much larger number of poor people of the "wrong race" getting shafted and another number of wealthy people of the "right race" "gaming" it. Because race is pretty deterministic, you don't need to game anything. Poor people of "wrong races" get hurt by the current AA system heavily. While under AA based on socio-economic status, at least the poor won't get screwed over.
> I think the correct move here is not to go full-on "merit-based access", but change the affirmative action to be based on socio-economic status. Boom, you just solved the issue the race-based AA was claiming to address, but by going directly to the source of the problem (socio-economic inequality).
Also move the efforts down the prestige ladder. Affirmative action at Harvard does diddly-squat do address anything except allow Harvard to signal its own virtue.
Society would be much better served to take a large fraction of Harvard's endowment, and use it to send disadvantaged kids to state land-grant colleges. You could help a lot more of them, and they'd be less likely to flame out to do culture clash and/or relative lack of preparation.
"socio-economic status" means nothing. This is so often used after race (discrimination) apparently is no longer a good argument "yeah but it's socio-economic". It's not falsifiable and therefore it should be dismissed as an argument.
I literally said "poor" in my example in the second paragraph.
Also, "socio-economic status" is a spectrum between poor and wealthy, including all of them. Just like "race" includes all races. We don't call it "non-white-and-asian-based affirmative action", so i dont see why you have an issue with me not calling it "poor-based affirmative action". It is a sliding scale, with all groups within it having different weights attached. So calling it "poor-based affirmative action" makes zero sense to me.
The weight is decided based on the socio-economic status, hence why we refer to it this way.
> Also, "socio-economic status" is a spectrum between poor and wealthy, including all of them. Just like "race" includes all races. We don't call it "non-white-and-asian-based affirmative action", so i dont see why you have an issue with me not calling it "poor-based affirmative action". It is a sliding scale, with all groups within it having different weights attached. So calling it "poor-based affirmative action" makes zero sense to me.
In total: One can make up the groups ... and the weights. It's completely arbitrary. That's why socio-economic is closer to astrology than a clearly defined term. Sure, you can make up cool calculations (with statistics even!), but it does not mean anything.
Would that make my definition clearer and less arbitrary, if i said that by "socioeconomic status", in this specific scenario, I am referring to where you fall on the spectrum of "your family's income + savings"?
Yes, in that case you're actually calling for/proposing something specific. Socioeconomic is a placeholder term, where everyone can project something into it (and this is also regularly done).
I appreciate your response. I genuinely was assuming it was well and clearly understood what I meant by "socio-economic status", but looks like I was wrong. In light of that, I totally understand your initial pushback to it. I hate vague and non-explicitly-defined terms, when it comes to policymaking, as well.
Merit is derided as a concept by people who wish to discriminate to meet a goal.
I view this as misguided because while people who wish to discriminate often have good intentions(theoretically discrimination simply means "recognition and understanding of the difference between one thing and another" after all) it's emblematic of the difficulties of discrimination that the most prevalent description of discrimination is "the unjust or prejudicial treatment of different categories of people or things, especially on the grounds of race, age, or sex."
I would say merit is a goal - it's never achieved as an end state and requires constant vigilance to stop entrenched powers from corrupting it.
It’s a complex problem. If the process of evaluating someone’s merits is biased or restricted in some way, then the merits are an inaccurate way of evaluating them.
We’ve seen this exaggerated throughout history. If you deny groups the opportunity to learn how to read then they appear less competent based on their “merits”. When you then take those groups and give them the opportunity to learn how to read, then their “merits” improve over time.
But it’s complex because no two students are the same, nor are any two reading teachers, nor any two reading tests.
So, rather than making it unconstitutional to enfranchise groups, I hope the court comes up with some sort of test to see if a group remains disenfranchised or not. But that’s also an incredibly complex problem.
Right, it can be tricky and no method would be perfect.
Sometimes courts when faced with an inherently ambiguous problem use a “reasonableness test”, which is basically taking it on a case by case basis and deferring to the current cultural climate to call BS on suspect claims.
Including race (among other factors) is more of a meritocracy then not, in my opinion. To fully calculate merit, schools should consider challenges and starting points for students. Students who had to overcome more difficulty to get where they are have more merit then students who are at the same place, but had an easier path to get there.
I'm not sure how much race plays an impact on difficulty of academic achievement, but I'm sure there's data on it that Universities are using. Black students, on average, probably have to overcome more adversity, since racism does exist. Being able to include that in admissions should mean more of a meritocracy, not less.
For the context of this case, don't forget we're talking about Harvard.
This school, if they wanted, could limit themselves to selecting ONLY from applicants with PERFECT test scores, PEREFECT GPA's, ideal extracurriculars and still have far too many applicants than they could admit. They would then have to resort to either a lottery system to select randomly from this pool of perfect applicants or maybe come up with ever more extreme and rarefied criteria to select the "best" from "perfect". Is that the way to go? Many don't think so.
Here's a spicy take: Merit mostly doesn't exist. In order to invent the concept of merit (for college admissions anyways), you need to weigh colleges, universities, high schools etc with varying degrees of power which inherently is extremely subjective.
Is someone earning say, a 3.0 at Harvard better than someone that earned a 4.0 at state? What about the highschool student that graduated with perfect As in a lax highschool? Do you consider AP classes more valuable over the student that took college classes instead? How do you address the fact that these sort of things disproportionately affect different minority groups?
If you start taking a lot of these factors into mind, you end up with something very similar to affirmative action. There is no one-size-fits-all solution.
Families with more resources apply those resources to the creation of apparent “merit.” Violin lessons, Kumon, French tutors, international travel. The mythology that what we call merit arises de novo with each subsequent generation is pervasive indeed. This investment process is intergenerational wealth transfer by another name; it’s just paid on the front end instead of the back end.
Whether university admission is the right locus of intervention or not is an open question; but kidding ourselves that merit exists solely as a stand-alone property separate from the developmental opportunities purchased by parents is a commonly-held belief that undermines advancement; but I suppose it makes us feel better about the structure of society. But if we were serious about allowing equal opportunity for merit to emerge we’d intervene much earlier than university admission. But then again, I doubt you’d find a consensus about what a just society would look like.
The problem with "meritocracy" has always been that so many of its proponents never define "merit". Often they don't even have a specific definition in mind; they just know it as a term that's aligned with other parts of their favorite ideology. Sometimes they do have a definition, but also know that it's too obviously tied to legacies of inequity and discrimination. Nobody likes to admit that they're a neo-feudalist.
True merit is hard to define and even harder to measure. As another commenter has already pointed out, that measurement has to include challenges and circumstances overcome. Unfortunately, those don't easily reduce to numbers. Sure, we can measure income or wealth, but how do we measure the impact of being denied opportunities to prove oneself? Of being dragged down by family and friends who need help because of the discrimination they also face? Of living in a place with poor schools, poor food, poor health care, for the same reasons? Note that some of these can apply to (especially rural) white people as well as others, but they remain hard to measure.
All we're left with is statistical measures and proportions, which will inevitably result in some errors. No matter what policy is enacted, those errors make it easy to complain about quotas and tokenism. Responses are possible, but less compelling. So we end up back to the fake meritocracy of grades and test scores. "What can we do" say the people who have already benefited and will continue to benefit from a system that ignores systemic inequality.
I have a feeling that in hindsight, 30-40 years from now, this will seem like no brainer and that the period prior was heavily anti-meritocratic and counterproductive.
This sort of thing is endemic to most parts of the world, and I doubt the rest of the world will come around in that period. The political forces that drive this will likely be as strong in fifty years, just as they were as strong fifty years ago.
People just leave this critical detail to our intuitions for some reason.
I think we want something good and fair... What's the measure of "merit" that's good and fair?
People who are good at the SATs usually want SAT scores to matter. People who get good grades in high-school want high-school grades to matter. Not unreasonable... but that's still just people defining "merit" in a self-serving way.
I think we want to focus on outcomes...
Let's suppose the college experience and fact of an elite university degree confers significant advantages of reach and power to those who get it... and that there are 10x more people able to take advantage of those advantages than there are spots available.
I think the question is: of those 10x people, who are the 10% we want to give a spot to?
I think we want to give them to the people who are most likely to benefit all the rest of us (completely fine by me if they benefit themselves as well -- In fact, I think it will work best that way, by a large degree.)
So that's the "merit" I'm looking for.
(I think when left to intuition, people tend to image "merit" as something that would favor people like themselves and their families... usually, in effect, something quite narrow with a heavy self-serving bias.)
People want “merit” to be the determining factor… but they don’t want to express or discuss what merit actually is.
As someone has plenty of the characteristics commonly associated with merit, but who also has a strong sense of fairness, I’m awfully skeptical of the faith (and I used that word deliberately) people put in the concept. Why the hell do the talents I was born with or advantages I was given make me any more deserving of material wealth than anyone else?
That’s all I see anyone doing with “merit”… it’s used as a “drop the mic” concept that to justify self-serving criteria for conferring material advantages.
Personally, I think ability should confer obligations more than advantages. (A wildly radical concept, I know, in the morally retarded era we live in.)
Oral arguments are today. The justices vote afterwards but we won't know the result until the opinions are issued, which will be several months from now.
> Nine US states - including California, Florida, Georgia and Michigan - currently prohibit the use of race as a factor in admissions for public universities
Sounds like instead of doing polls on opinions and "gut think" that there are already large real-world examples that can be examined precisely to see what that policy does or doesn't accomplish.
It was 1996, pretty long ago. Looking at a UC vs a private uni, the most obvious difference is how UCs tend to have majority Asian students.
UC Berkeley chancellor publicly denounced the ban when it was upheld a couple of years ago. Then Cal removed SAT/ACT as an admission requirement. Next admission cycle, she sent out an email to all students and alumni (me) cheering that the school is now more diverse. The percentages of White, Black, Hispanic, etc students in the incoming class all increased. Percentage of Asian students sharply decreased. If you ranked the deltas, it was similar to the ranks of average SAT scores for each group (Asian students have the highest SAT scores, etc). I guess the most diverse class would be achieved using a pure lottery system.
The whole idea behind it is stupid and racist... "if you're a certain skin color, you're too stupid, to compete on merit"... then when people complain, they add the "socioeconomic" factors: "if you're a certain color, you're too poor to compete on merit, but instead of looking at your family income, we'll look at your skin color".
I have no idea how something like this gets support in developed western countries.
That is because the gaps persist when controlling for income. The gaps actually increase the higher up the income scale you go!
This happens in a democracy because whilst a tiny minority of people care about principles, most people care about power. AA empowered the right people so happened. It's a mistake to try to understand its justification from a first principles approach.
Specifically, it was invented because Jewish people performed too well on objective metrics and of course we cannot have too many Jews. So they did away with those pesky objective metrics and came up with an opaque system that no one could easily audit but at the same time they could easily apply their racial biases
On a completely separate and unrelated note Harvard is also getting rid of objective metrics now.
It gets support because it gets votes. If Democratic voters weren’t asking for affirmative action, the Democratic Party would not pursue it, regardless if it’s bad policy or not, it’s pandering to the base.
We should stop differentiating people by race, but whichever side loses this particular case can take solace in the fact that it won't make a huge difference either way. Studies have shown very little difference in average long-term career outcomes between highly selective and less selective colleges, after controlling for SAT scores and high school class ranking. Let's stop fetishizing schools like Harvard.
That's what always gets my goat. People seem to be less concerned with actual equality and improving things and more interested in claiming symbolic prizes.
Why bother trying to make to her schools and opportunities better when what we really want is just to cash in a piece of the pie for ourselves. Another example is the push to STEM for women and minorities, it's not about actually learning the skill or contributing it's that being a coder is seem as an easy path to the good life and so we want to get in on that. If it were really about equality or income there would be more people pushing for representation in being an oil worker as those guys make great money too, but it just isn't seen as high status.
The average person is not someone who moves the needle in society and it's not what you would target if you were a parent pushing to give you child the best achieving something remarkable. You want to give your child an outsized chance to end up the very top top of the band: successful founder a large business, national level politician, C-level executive, high successful doctor, well compensated finance professional, managing partner at a law firm. Or more crucially to be know those people and get opportunities the average person does not through their network.
If both places produce the same average outcome, then you would be crazy to not pick the place that is several times more likely to have an amazing outcome.
Are news outlets not allowed to include links to actual source material or something? Why is it that I almost always find these links in HN comments rather than the article itself? It's really annoying and I genuinely don't understand why this is so common, just like not linking to the actual studies for "science" news. To me it just seems sneaky, like if the article is correct, why not link the source so I can verify that myself? I know most people won't do that, but it feels like including the source when once exists and is already publicly available should be a hard requirement for any piece of journalism.
I agree, and it irks me. When there is clear online source material, like in an article about a court case or viral video, the newspaper should link that material.
I think many of these outlets must have a no-external-linking policy. But it really hurts their credibility and even their readability IMO.
> Institutions like Harvard say they consider an applicant's race as one of several factors - including economic status and religious belief - in order to build classes that accurately represent the racial and ethnic diversity of the country.
I mean, this is all kind of window dressing for elite institutions to be elite, right? There are millions of people qualified to attend these university, Harvard gets to hand select from all of them who gets to join their walled garden based on whatever Harvard Admissions is into right now. Notice that they don't care about diversity of viewpoints or politics.
If these universities actually cared about minority enrollment, they could always expand the size of their enrollment. But none of them want to admit that the only real cache of their education is the exclusivity.
> they could always expand the size of their enrollment
There are reasons beyond manufactured exclusivity for why this doesn’t work. College education isn’t just about students soaking up material. We need more online vocational schools where that’s the sole purpose. But let’s not throw out the baby with the bath water.
Berkeley is highly exclusive [1], albeit 3x less so than Stanford [2]. The class-size difference, about 12,000, is noise against the millions of degrees conferred to Americans every year.
Meanwhile, Stanford produces different graduate as a result of its process. Forcing homogenisation to one model for ideological purity is misled.
tl;dr Berkeley is still exclusive even after you're admitted
Berkeley has its own way of filtering students once they're admitted. This is most obvious with the computer science majors, of which there are two that involve almost the same coursework. EECS is super hard to get into to begin with. Whereas anyone in the school can intend L&S CS, but there's a rather high GPA cutoff (at like 30th %tile but don't quote me) for declaring, and you're effectively barred from most upper-div CS classes until you've declared, usually second year IIRC.
If you miss the cutoff for L&S or one of the other exclusive majors, you end up in an alternative. CS usually becomes the other CS: cognitive science. Yes it's a real field, but at Berkeley that just means you couldn't make comp sci. Cog sci has very few classes within its own major; instead everything is cross-listed with other majors, and you have a very hard time getting a spot in the class. The other common duo is Haas vs economics.
And many of the required lower divs are "weeders," huge classes with low average grades that students perceive mainly as obstacles to the upper divs.
I also want to highlight that this dynamic messes up those adjacent programs.
I was a cognitive science student because I thought it was fascinating, and it's unfortunate that the program gets a bit diluted by being branded "the other CS" and some share of students that don't actually care about the other components.
Yeah, it kinda stinks. I want to edit my comment saying that cog sci is just mostly the alt bucket for comp sci, but it's too late. My roommate was cog sci because he actually cared about it, then he switched majors.
I largely agree, except: When I look back on my own college experience, the actual academic content might only account for, like, a quarter of the value I got out of it.
And data seems to back this out - entirely online degrees don't nearly drive as much success as the in-person experience.
So universities are certainly not going to admit that the content and learning portion of their experience are not important. But moreover I suspect what Harvard and others don't want to admit is that they don't fully understand what makes alumni successful, and are terrified of changing the recipe too much.
> Notice that they don't care about diversity of viewpoints or politics
Ever talked to someone from Harvard? (they'll usually tell you in the first 4 minutes you meet them telling you they went to school in Boston). They admit quite a number of conservative people if that's what you're worried about, and even their more left people are pretty center-right, as in they really don't want to pay more taxes.
> they could always expand the size of their enrollment
Really the feds could (and imo should) force them to do so, given a ton of their money come from federal grants. It's always been insane to me that private schools can receive 50k+ per grad student from my taxes, knowing that most people will never get in. Also insane how the elite UCs for example educates more people than the entire ivy league combined.
> Really the feds could (and imo should) force them to do so, given a ton of their money come from federal grants.
People frequently tell me I don't understand what endowments are for. Harvard has $51 billion. If we take an average of 5% per year and this assuming they make a $2.5bn return on that investment each year (and raise zero money in addition to it), that would allow them to pay for 25k students (at $100k each. I'm rounding up by like $20k btw). That's more than Harvard has at any given time, by a few thousand.
I get that endowments are supposed to be safety mechanisms and allow funds for things like new buildings, tuition assistance, and gaining more investments. But at what point do we recognize that these are just money making schemes. Many public universities have multi-billion dollar endowments. Money held away and not being used is not adding value to our economy. I'd argue that education adds a lot of value. But that's an argument for cheaper education, not allowing it to become a highly profitable system. Clearly what is happening right now isn't sustainable.
Harvard already effectively pays for the entire tuition if you can't afford it. They have decided that preserving the endowment for future crises as well as funding other aspects of a college (running a research university actually costs a ton of money, around if not more than $100k per student) is a more important goal than ensuring that students who have families making $300k+ a year go free.
I went to Harvard and your comment about the leaning of the school is incorrect. Certainly, "even their more left people are pretty center-right" is just... wrong.
Expanding the size of enrollment isn't just something that can flipped like a switch. One cannot simply build a high-rise dormitory without increasing the capacity of all other facilities in lockstep - and a university designed to be fully walkable can only accommodate so much simultaneous construction.
And more generally, you have the use of applicants' race backwards - the entire point of affirmative action is to take into account an applicant's merit-based performance relative to the opportunities they received growing up - and it is difficult to evaluate this fairly without taking into account the opportunities that might have been denied to the applicant as a result of their race.
To put it another way, two hypothetical applicants from two families with different races but equivalent wealth etc. in the same community may have been treated incredibly differently based on their race. An admissions office forced to be "blind" to an applicant describing this discrepancy in opportunity would be less effective at evaluating applicants based on merit.
Ivy schools have some of the largest endowments, rivaling the GDP of some small countries. If Harvard, and Yale were serious about diversity, they would let people attend for free.
If they admit you in the first place, they pretty much do, which was indeed a pretty salient perk of the top-tier colleges among the Asian immigrant parent crowd, at least a decade or two back.
For a particular cohort of second-gens, you were going to an "elite" institution or you were going to state—the family finances weren't going to support sticker price without a near-full scholarship.
> A university designed to be fully walkable can only accommodate so much simultaneous construction.
Well, your implication here is that accessible, affordable colleges are clearly a second-rate experience. I'm not necessarily disagreeing here, but it's something academia themselves have not admitted.
Your example forgets that they are still allowed to look at financial/hardship data to determine merit. So presuming that someone with identical upbringing must have had a harder life because of the color of their skin only exists to perpetuate bias.
I'd certainly like to live in a world where students of the same age, with the same family income/savings etc., with the same teachers and classrooms, with equally supportive parents, would have equivalent opportunities throughout their K-12 careers. In such a world, I would indeed be the one "perpetuating bias" with my comment. But we live in a world where the race of those students, controlling for everything else, is statistically shown to have an impact on things we'd generally agree are an important part of a child's learning experience.
Now, we might reasonably disagree on whether admissions departments are overreaching in trying to ensure that their merit-based decisions take these factors into account. And we might reasonably disagree as to whether the government should have the ability to intervene if such overreach were found to exist. But the idea that any of this is only being done to perpetuate bias, ignores the fact that bias already exists.
> relative to the opportunities they received growing up
There are so many better ways to do this than race, and they have zero interest in looking at any them. Measures that are both more accurate and more objective like.
Do your parents have a college degree?
What's you family income?
What opportunities that affect GPA/SAT are going to be denied to anyone because of their race?
Everything makes sense when you think of Harvard and similar universities as selfish private institutions that want to maximize their reputation. They want the children of world leaders, billionaires, and actual super-geniuses. They don't want to dilute their brand with thousands of extra students with mundane backgrounds that end up being moderately successful professionals who make mid-six-figures. Affirmative action students give the school more credibility among URMs, so they offer more to Harvard than a somewhat more qualified Asian who isn't quite IMO medalist level.
It's more than this. It's way better to think of Harvard et al as large hedge funds abusing the school tax loophole. They do want the children who will be world leaders, billionaires etc. but it is because those are the most likely to donate to the hedge fund. They don't want to grow admission because it might hurt hedge fund donations but also because educating people is secondary to their actual purpose and should only be engaged in to the point that their primary purpose is maximized (middle management doesn't make the kind of donations worth worrying about for a hedge fund, they want a higher minimum efficient scale).
“...but we also recognize principles such as inclusiveness and equality, which many members of the Harvard community consider of paramount importance to our mission..."
How well you do academically is influenced predominantly by genetics. It’s unfair to idolize these institution as something to aspire to and worship. Doing so is a gross misallocation of precious resources as a society. Instead they should be viewed as trade school for academically gifted. We should find different ways to harness the remaining 95 percentile. Striking it down would be the step in the right direction
I’m of two minds here. Obviously race based admissions is unfair. But Harvard isn’t trying to admit the next generation of scholars, it’s trying to admit the next generation of power. Will that generation be 75% Asian, the way a class would be if formed with standardized testing? Doesn’t seem likely.
Yeah, but the Supreme Court isn't trying to help Harvard accomplish its goals. It's trying to determine the legal principles to resolve an issue. Whether that helps or hurts Harvard shouldn't be a concern.
> The fact that we get any Black students in under-resourced neighborhoods testing within 100 points of White students, who have been prepping for the test since the eighth grade, that is a miracle of over-performance
Not the poster, but the implication is that all white students are highly privileged in well resourced school districts. It's not technically what was said, but I agree the implication is pretty strong and unreasonable. You can have other readings ofc.
Honestly the boundary between race/money/class/privilege is a pretty rough space to hang around.
Standardized tests are an opportunity for students to distinguish themselves no matter what school they're stuck in. Yes the wealthier students on average study more for the SAT, but anyone can study for it with widely available materials. UC Berkeley got rid of the SAT/ACT requirement and is making their own test instead; if other schools did this, having so many different tests would ironically favor those with the most resources.
And what's the alternative? There's not a lot else to go off of since GPAs are whatever.
I'm curious as to why affirmative action in the US has always been race-based and not economic-status or household income based?
Asking as someone with no skin in the game; I'm merely curious if anyone has any constructive opinions on why they think economic status based affirmative action may/may not work?
I'm a Harvard alumnus but obviously don't speak on behalf of Harvard. The Harvard admissions office does consider folks from disadvantaged socioeconomic backgrounds, and does not only look at race. There is a dirty history to the origins of schools steering away from entrance exams towards more holistic admissions, but it really is the case that they consider what opportunities were available to students based on the high schools they attended, whether applicants had time for extracurricular activities or were busy working a job or caring for a family member, and plenty of other circumstances that are not just limited to pure meritocracy and race.
The term "socioeconomic" is really malleable, when in actuality you probably meant "low income" or "poor". This might be a better measure, but can be gamed also as you point out. This however doesn't necessarily mean the measure is bad.
"What are socioeconomic examples?
Social and economic factors, such as income, education, employment, community safety, and social supports can significantly affect how well and how long we live. These factors affect our ability to make healthy choices, afford medical care and housing, manage stress, and more."
Notice it includes "Income"? it includes other factors as well such as poor safety nets, etc.
> "What are socioeconomic examples? Social and economic factors, such as income, education, employment, community safety, and social supports can significantly affect how well and how long we live. These factors affect our ability to make healthy choices, afford medical care and housing, manage stress, and more."
This is a bit non-sensical. If you take education as an example for college admission, you need to support the ones that are the least suitable for exactly that.
I know you certainly don't mean that, but you said it. Make it concrete and you notice the error. Socio-economic is a handwavy term.
Because no one has ever worked for cash and lied about it?
Do you suspect that someone who launders dirty money for a living did their taxes and listed "money launderer" as occupation and reported their income?
Assets like this are often hidden behind shell companies to prevent law enforcement from taking them under "proceeds of crime" charges.
There are so few people this would apply to (wealthy individuals with large amounts of unreported cash income who are applying to college) that its not even worth considering as an edge case for college admissions.
The point isn't to eliminate every possible way college admissions can be gamed but to get the best method that works for the majority of applicants.
> There are so few people this would apply to (wealthy individuals with large amounts of unreported cash income who are applying to college) that its not even worth considering as an edge case for college admissions.
You would be mistaken.. This is actually a rather serious problem.
This seems like multiple reports on a single batch of incidents connected to sketchy accounting of foreign capital (and in Canada). What about this doesn't scream "edge case" and needs to be handled as a general case of college admissions and not by anti-money-laundering?
Race is easy to game too, in fact far easier. I know people who were around 5% Hispanic who would then write
Hispanic on college applications. They're technically not lying.
There's also just plenty of people that just straight up lie about their race on college applications. It's not like its being verified at all.
I went to high school with a lot of White - Asian mixed kids. None of them wrote Asian on their college applications. I also knew a lot of Indians (like me) who'd select Other in order to avoid being denied opportunities due to being Asian.
There was even a United States senator who claimed to be Native American, was awarded academic positions based on that, but a DNA test proved her native ancestry to be so minuscule as to not even be measurable.
The part about 'was awarded academic positions based on that' is debatable, e.g. this claim was investigated by a newspaper (article below is titled 'Ethnicity not a factor in Elizabeth Warren's rise in law'):
Realistically, we should eliminate the category of hispanic entirely. It is orthogonal to race and the easiest one to game.
My college counselor told me to mark Hispanic because my parents are portuguese. I did not out of principle because I consider myself pretty standard white.
That's true, and those laws are supposed to right decades of misappropriation of education, finance, basic rights back to those that were wronged. The intent is sound, but the execution of it? I'm not sure it's been done fairly. Consider this:
Asians have been discriminated in many cases in the past as well. Goes way back to the days of slavery:
> After slavery was abolished in the United States, Chinese laborers were imported to the South as cheap labor to replace freed Blacks on the plantations [...] Starting as early as 1865, Southern newspapers began printing editorials and letters calling for Chinese labor to be the new labor supply.[2] This interest was sparked in part by accounts boasting that the Chinese contract labor attributed to the increase in Cuban agricultural imports. The Chinese effectively became the new labor supply but were positioned societally at the same level as African Americans.
Then you have the railroads era, WWII internments camp era, the Chinese Exclusion Act, 2021 Atlanta Spa shooting, and more. Then factor in all the other racist policies against Asians: minority roles in film, in white collar job sector, bamboo ceilings, violence/assaults against asians vs police response, US government witchhunt on Asian researchers during Covid, dealing with "Model Minority" myth...
Yet, Asians are constantly excluded when it comes to Affirmative action in these modern times, why is that? We can clearly see they've had an unpleasant history too, so why are institutions increasingly hostile against them? This isn't even restricted to Colleges, even though the lawsuit stemmed from it, we can see it affecting the workplace as well:
> "Asian Americans Are the Least Likely Group in the U.S. to Be Promoted to Management"
We should embrace Asians to be part of the diversity and Affirmative Action policies, not reject them, so I feel the current implementation of AA needs to be addressed, which I hope the ongoing college lawsuit will do.
The problem is there but the current AA is not helping.
Here is what might work:
- Harvard is required to do only merit base admission
- Harvard is required to increase diversity. They will be required to achieve that by investing in high schools in poor areas. Basically university will recruit poor students by helping them during middle and high schools. Something like opening Harvard middle or high school.
(This is one of things what communists did right after WW2)
If race is barred then the next method could be increased emphasis on perfect price discrimination to distinguish applicants socioeconomically. Outrageous tuition prices coupled with generous "scholarships" can also create a diverse student body.
Presumably if you cannot use race to decide who gets a place, you cannot use race to decide who get's a scholarship. At least as Harvard University, different rules might apply for the NAACP etc.
Who said anything about race? If you can afford Choate, you can afford $250k/year at Harvard. But the kid from Biloxi, MS gets a "scholarship" so she only pays $25k/yr.
Of course, for 250k*4y I could probably just buy a house in Biloxi... And many of the people the current system removes from the pool come from areas that would be just as economically under-performing as other groups.
Less clear than you might think. Choate isn't as rich as Harvard, but it's still on a tier with, say, Wesleyan (for context, not as a fair comparison), with the finaid apparatus to match. They are absolutely capable of letting you "afford" to attend if they want you there, not quite orthogonal to whether you can afford $250k/year (more than Harvard sticker price anyway) but maybe at a 60-degree angle or so.
I look at this through the lens of the San Francisco public schools. When I went to the magnet high school there, Lowell, they had explicit race based standards. We all got scores out of 70 based on our standardized tests and middle school grades. The differences were pretty stark.
Chinese kids needed a 68, which could mean they got just one B. White kids needed a 62, black and latino even lower. So I totally get why Chinese families were screaming. The standards for them were way higher than for other races.
But also just as jarring, the middle schools that fed this system were incredibly biased. My middle school, Hoover, had a segregation system that split the school in half. The honors half had at most one black kid per grade. We even had a different lunch period.
The honors half was being groomed for schools like Lowell or at least to look at High School as an on ramp to college.
It's not like we didn't have a lot of black students. We actually bussed in two, maybe four busloads of black kids from another neighborhood. But only one of those was ever admitted to honors. I just can't believe that there were no kids on those busses that were smart. I knew a lot of them through sports and they were plenty smart, together, hard working.
If this level of segregation can happen in San Francisco, I can't imagine how bad it is in the rest of the country.
It's complicated, and you can have your own takeaway, but mine is that merit often defined wrong. I'm on the lookout for people who did more with less, or even did equal or near to it with less. To me, that is more a measure of merit than top line measures like test scores.
> It's complicated, and you can have your own takeaway, but mine is that merit often defined wrong. I'm on the lookout for people who did more with less, or even did equal or near to it with less. To me, that is more a measure of merit than top line measures like test scores.
I agree with this, and the problem is how many schools look directly at race for that. This reminds me of predictive policing ML models that take race as a factor. Yes it gets more accurate results than if it didn't use that information, but that doesn't make it fair.
So giving Black students a slightly higher chance to get into Harvard, all other things being equal, reminds you of sending the cops to overpolice neighborhoods Black people live in? I guess they're the same in a way. Doing bad things reminds me of doing good things because things are being done in either case.
> This reminds me of predictive policing ML models that take race as a factor.
In statistical systems you often have to take in race as a factor to remove it as one. You can't debias a system if you can't determine the bias. There are many things that highly correlate with race: a notorious one being zip code (byproduct of red lining). But there's also a lot of non-obvious ones and considering historical data is used, the machine will find those correlations. So only with that data can you then analyze and tweak your model to remove the racial biases.
The system developer has to look at race to de-bias a system, but the system shouldn't take race as input, unless it's somehow able to de-bias itself. In this example, I'd wonder why ZIP code is even an input.
So I'd consider that part of the system. You could also train your model to debias by using a cost function that incorporates race and minimizes the KL divergence between different ethnicities. (Obviously it would be best to use interpretable models and/or causal ones) A Mahalanobis distance might also be good here. But when training your model you might want to PCA it and if race isn't an included factor then you can't see if it is a main contributing component or not. You should be training your model such that the racial variable contributes near nothing to the model's outcome (easier said than done).
> In this example, I'd wonder why ZIP code is even an input.
The school you go to is strongly correlated with zip code. So... if any of your schools are considered on your college app (hint: they are) then you have extracted the zip code (which again, points back to red lining[0]). In the crime example, the zip code is going to also correlate with the frequency and type of crime as crime (especially when broken down into types of crime) is not homogeneously distributed across a country/state/city (to be honest, not even a zip code! But the more you zoom in the more accurate you get. Aggregation bias is a pain).
This stuff is a pet peeve of mine. Our tiny human brains weren't designed to figure out how all these things are connected. That's why we use data science and statistics. It is why data science and statistics is so fucking hard too! Even more so when you get into causal statistics. Contributions may frequently be obvious post hoc, but they rarely are a priori (which is why "novelty" is so absurd). The world is extremely complex and we have long chains of causality that we have to work with. The problem here is our brains are designed to simplify things. To take first or second order approximations. The problem with this is that these approximations are not sufficient for navigating the major issues we face in the modern world. These low order approximations often lead us in the wrong direction rather than being a small step in the right direction. If you're taking a gradient descent in a complex feature space (thousands of features/dimensions) then approximating it in 2D will get you nowhere near an equilibria. There's probably dozens, or hundreds, of features that are highly significant. All before we get into the mess of coupling and confounders.
This is a problem because: 1) we fight over things we're not even qualified to have an opinion on; 2) the actions we take are frequently determined by popularity with a hint of analysis and the popularity is determined by low order modeling; 3) because we have a limited view of our actions we can easily convince ourselves that low order approximations are good because they work locally (spacial and temporal) but not over the global solution space (humans are notoriously terrible at long term thinking/planning and this too is highly correlated with income levels/life stability).
Our modern world is highly dependent upon specialization but we have a culture that tries to convince us that expertise in an extremely narrow domain generalizes really well. It doesn't and that is destroying us. This isn't just a "you" problem, so you're not being targeted. These issues are highly common on HN and anywhere nerds congregate because that cultural pressure convinced us that our self-worth is dependent upon the generalization of our knowledge. Einstein nor Feynman said you don't understand something unless you can explain it to a simpleton/your grandma/a child. Most stuff you can't because domain expertise is required for even basic understanding. So I'm trying to put social pressure to solve this problem and I hope others will join.
tldr: The world is fucking complex, treat it that way
[0] One of the major issues with red lining (there are a lot) is that where you live determines the quality of education you get. School funding is often related to housing taxes. Cheaper houses? Less funding. Less funding? Lower education. Lower education? Increased difficulty in climbing the socioeconomic ladder. Which correlates with crime rates. Which correlates with a lot of things. These downstream effects aren't actually dependent upon race but if you correlate an upstream effect with race then the downstream ones do too.
I don't assume this is easy or even possible to solve. If the model is trying to determine something like the applicant's level of adversity based on whatever info college admissions has, of course the result will have a divergence between races, and historical racism will have shaped some of your input variables (like high school) to some degree that we'll never fully understand. Using some statistical difference between races as a cost function wouldn't be satisfactory either.
All I'm saying is at the bare minimum, don't feed race as an input. It's a vague factor that the applicant has zero control over, and a model (or human reviewer) making any assumptions based on race would be inherently racist, which is unjustifiable. Another bad example is how UCSD asks how feminine (rate 0-5) the applicant considers themselves. Only use factors that have some legitimate purpose; at least the high school matters because of grade inflation, but standardized tests are actually fairer. US hiring is this way, where many companies consider it an unwanted risk to even know about protected categories (age, race, religion, marital status, etc). It's far from an airtight solution.
This isn't really about science, and the article is off-topic for HN.
I meant a cost function that the model trains on. The expert can look at the cost function to double-check whether the inputs are appropriate. Maybe the applicants' names got in there by mistake, and the names serve no purpose other than giving away their ethnicities.
If the output seems to strongly favor some races over others, the expert debugs the inputs by temporarily swapping some out and rerunning. This is only a second line of defense beyond checking the inputs manually. Maybe the expert doesn't realize that some factor has a huge correlation with race without saying much about the candidate's merit.
Beyond that, if there's a divergence between races, it is what it is. The expert shares what factors are being considered, and it's politics at that point.
> If the output seems to strongly favor some races over others,
You'd typically do this with a cost function.
> the expert debugs the inputs by temporarily swapping some out and rerunning.
The best way to do this is to use that cost function to optimize your parameters.
> Of course, could do this without race data just by proofreading.
I thought I've adequately demonstrated why this is much easier said than done.
> Beyond that, if there's a divergence between races, it is what it is.
Except it isn't. It depends if the divergence is because you've trained a model on poisoned data (i.e. the historical racism gives momentum to your model). These things are rather tricky.
It really feels like we're going in circles here and I'm trying to get you to understand why we do what we do, but you're avoiding getting specific to support your assumptions. I work with data all day every day. I'm sure you've done some data analysis too, but to do high quality work you have to consider a lot of factors. It's okay to not have this domain expertise. You're a general software engineer and probably far better than I am at that. But it is a bit naive to argue with domain experts about something that isn't in your domain. It would be naive for me to argue with you about your expertise. That's okay though. Our world is highly specialized. No one can know everything. But that's why I wrote so much before, because this is a toxic pattern in our culture that we need to curb.
Is your domain expertise in college admissions, data science, or a combination? You assuming this model should try and undo some effects of historical racism, but that isn't the engineers' job to decide. If that's the intended behavior, I understand that the difficulties involved are beyond my experience, and I would leave it to the data scientists. That's not what I think the goal of college admissions should be, and our government is making a decision on that now.
> The best way to do this is to use that cost function to optimize your parameters.
> I thought I've adequately demonstrated why this is much easier said than done.
I understand how this works, but that's specifically what I don't want, for race (or other data that serves no purpose but to indicate the race, like name) to be part of the training data. Those inputs are manually selected at some level.
My domain expertise is in data science. We're not talking about specifically school admissions but something bigger: handling race in data. That's decently general. Were I to actually create a model for predicting admissions I'd definitely be working with admissions to see those specific nuances. But you're attempting to pigeonhole the conversation.
> I understand how this works, but that's specifically what I don't want, for race to be part of the training data.
Tough. The problem, is that it is really difficult to move. That's what you aren't understanding. Making a model that doesn't explicitly incorporate race doesn't mean you've removed the race variable. It is just hidden now. There's plenty of examples of this happening in real life and they're fairly easy to Google. The historical data you train your model on doesn't suddenly lose it's correlation to race (there's major confounding variables) just because you didn't explicitly incorporate it. Dealing with hidden variables is hard and that's all you're doing.
At least since 2007, that sort of explicit race based discrimination undertaken voluntarily is not legal in public school districts, see PICS v. Seattle.
What about this, which has been on SFUSD's web site for the past 12 months?
Tutoring for African-American SFUSD Students Citywide
Posted on September 23, 2021
The Citywide Tutorial Program is partnering with after school programs
across San Francisco to offer homework help and tutoring
for African-American students enrolled in SFUSD schools,
kindergarten through 12th grade.
SFUSD Access and Equity Department and SFUSD State and
Federal Programs Department provide a SFUSD credentialed
teacher to tutor African-American students at each location,
with a virtual option:
> Off-Topic: Most stories about politics, or crime, or sports, unless they're evidence of some interesting new phenomenon. [...] If they'd cover it on TV news, it's probably off-topic.
I think this is being framed poorly. The "-based" in "race-based" suggests that race is a dominant factor. But it sounds like the real question is, can admission decisions consult race even as a minor consideration.
> The court is hearing challenges to their admissions policies, which consider race among many factors when evaluating applications.
> Institutions like Harvard say they consider an applicant's race as one of several factors - including economic status and religious belief - in order to build classes that accurately represent the racial and ethnic diversity of the country.
In which case -- sure, perhaps this is a valuable conversation to have. But if one considers the efficient frontier of tradeoffs between the goals of having the most "meritocratic" class of students with the best grades and test scores or most AP credits etc, or having a "representative" class which reflects the diversity of the nation, I'm guessing there are bigger opportunities for improvement. Legacies and athletic recruitment, perhaps?
If the explicit consideration of race is removed from admissions, are there not plenty of ways for admissions offices to try to consider similar factors other than personal academic performance, which would leave the plaintiffs still unsatisfied? Non-race factors for which admissions offices could try reflect national statistics in incoming classes:
- percent public school vs private school
- percent households which rent vs own their primary residence
- distribution over home values in admits' zipcodes of residence
- share of admits who had a parent incarcerated for some portion of their childhood
- personality big 5 scores
We get indignant in public about different treatment on the grounds of race. But I think students and families get frustrated because of the (inevitable) feeling of a loss of control when being judged based on criteria out of their control. If instead of race, one had the sneaking suspicion that one was not admitted because there were too many other private school, owned-home, 800k - 1.2M home, non-incarcerated-parent, high-conscientiousness applicants ... would you feel better?
> A 2021 Gallup poll found that 62% of Americans were in favour of affirmative action programmes. A separate poll released by the Pew Research Center earlier this year, however, found that 74% of Americans - including a majority of black or Latino respondents - believed that race should not be factored into college admissions processes.
I think my takeaway is that even if you give a horrendously unpopular program a catchy enough name, people will think they like it.
Affirmative action isn’t just for college admissions, and outside of college admissions, affirmative action programs generally do not involve decision preferences, but are limited largely to training programs, outreach efforts, and other funnel-shaping efforts.
So, the poll results are perfectly consistent with people understanding the question and either supporting AAP in general but not in college admissioms, or supporting it in general but not when it goes beyond funnel-shaping to include decision preferences. They are also consistent with them supporting it in college, and involving decision preferences, but only on ethnicity, gender, and other non-race factors, though that seems less likely.
I think that's a fair point. Then it's worth noting the Supreme court is not considering ending affirmative action at all, just the race-based selection.
> Students for Fair Admissions, accuse Harvard of discriminating against Asian American applicants in order to boost representation from other groups.
Why does this argument need to be framed entirely as discrimination against Asian Americans? Affirmative action clearly harms White people too. Are White people so incapable of advocating for themselves that they can only point out a policy is bad when it harms another minority group?
Well when the narrative is that all white people are at least slightly racist and that only white people can be racist to begin with, white people aren't really allowed to advocate for themselves, and any instance of it would immediately be designated as a hate group.
I'm waiting for the day that "caring about discrimination against asian americans" is listed by the ADL as a dog whistle for white supremacy.
reply