Remix.run Logo
AnthonyMouse a day ago

This is presented as an argument against objective metrics, but a) the alternative is subjective metrics, which is even worse, b) "disparate impact" is just another metric subject to Goodhart's law, and c) resources aren't the only thing that determines test scores.

If you're poor but determined, you can't afford a high cost test prep course, but you can go to the library. The rich kid has their private tutor come to their house and then saves time that allowed them to be chauffeured to tennis lessons. The poor kid has to take the bus to the library and spend twice as long with the study books and then doesn't get any tennis lessons, but it's possible for someone to do that if they actually care about it. Whereas, how is a low-income white kid supposed to overcome a race quota where every slot for their race was already filled by nepotism?

Timon3 a day ago | parent | next [-]

> This is presented as an argument against objective metrics

No, it's not. It's presented as an argument that measuring objective metrics doesn't mean you're measuring merit.

> If you're poor but determined, you can't afford a high cost test prep course, but you can go to the library.

The library most likely doesn't have the same specialized learning materials that rich kids can afford, so this doesn't mean the poor kid has equal opportunity.

> The rich kid has their private tutor come to their house and then saves time that allowed them to be chauffeured to tennis lessons. The poor kid has to take the bus to the library and spend twice as long with the study books and then doesn't get any tennis lessons, but it's possible for someone to do that if they actually care about it.

You're presenting this like it's purely an issue of free time, which is obviously not the case. The poor kid possibly can't reach the library without their guardian, who may or may not have time to drive them there. The poor kid possibly doesn't have anyone to teach them using the learning material there, while the rich kid likely has either their guardians or even specialized tutors for this purpose.

Your comment is a wonderful example of how people arguing for meritocracy can ignore reality - the bare minimum is supposed to be enough for the disadvantaged, even though there's a massive difference in effectiveness.

AnthonyMouse a day ago | parent [-]

> No, it's not. It's presented as an argument that measuring objective metrics doesn't mean you're measuring merit.

The goal is to measure merit. Objective metrics are the nearest thing we have to achieving that goal. If there is a better metric, you use that instead. But if the best metric we have isn't perfect, that's no argument for doing something even worse.

> The library most likely doesn't have the same specialized learning materials that rich kids can afford, so this doesn't mean the poor kid has equal opportunity.

And yet this is still more of an opportunity than being locked out by race quotas.

> The poor kid possibly can't reach the library without their guardian, who may or may not have time to drive them there.

Libraries are generally in higher density areas with mass transit, and in the worst case you can walk there. Moreover, primary schools generally have libraries and then the kid is already there for school.

> The poor kid possibly doesn't have anyone to teach them using the learning material there, while the rich kid likely has either their guardians or even specialized tutors for this purpose.

Which is why it takes longer. But the point is that determination has an effect. It's something you can choose rather than something you can't control. Whereas telling people that it's not a level playing field so therefore they shouldn't even try is how you perpetuate the problem forever, if not actively make it worse.

Timon3 a day ago | parent [-]

> The goal is to measure merit. Objective metrics are the nearest thing we have to achieving that goal. If there is a better metric, you use that instead. But if the best metric we have isn't perfect, that's no argument for doing something even worse.

You may notice that I didn't argue against objective metrics at all? All I've said is that objective metrics don't mean you're directly measuring merit. It's important to keep this in mind, for example by ensuring equal access to specialized training materials.

> And yet this is still more of an opportunity than being locked out by race quotas.

DEI doesn't necessarily mean race quotas - it's telling that you think it does.

> Libraries are generally in higher density areas with mass transit, and in the worst case you can walk there. Moreover, primary schools generally have libraries and then the kid is already there for school.

But the kids don't necessarily live in higher density areas! So not only do the kids get worse material and less help, they also have a much harder time accessing those worse materials. And again, a library usually doesn't have the same specialized materials that rich kids can afford. I've been trying to show that this should be kept in mind and remedied, but you're arguing against me by arguing against things I haven't said. This is usually what happens.

> Which is why it takes longer. But the point is that determination has an effect.

But kids don't have infinite time! So the rich kids still have unfair advantages, so the tests aren't directly measuring merit. And again, you're arguing that the bare minimum should be enough for the disadvantaged. You seem to effectively be arguing that meritocracy is either impossible, or should not be the goal.

AnthonyMouse a day ago | parent [-]

> You may notice that I didn't argue against objective metrics at all? All I've said is that objective metrics don't mean you're directly measuring merit. It's important to keep this in mind, for example by ensuring equal access to specialized training materials.

But then who are you arguing against? Is there someone strongly opposed to providing equal access to training materials, e.g. by making them available in school libraries?

> DEI doesn't necessarily mean race quotas - it's telling that you think it does.

That's how it's most commonly implemented in practice whether de jure or de facto and that's its opponents' primary objection to it.

> But the kids don't necessarily live in higher density areas!

In general the poor kids live in higher density areas and the affluent kids live in the suburbs.

> You seem to effectively be arguing that meritocracy is either impossible, or should not be the goal.

Actual perfect meritocracy is impossible because actual perfect anything is impossible. But meritocracy is the goal and what you want is to get closer to it. Which providing better study materials in school libraries can do, but that isn't what anybody is complaining about when they're complaining about DEI.

mystraline a day ago | parent | next [-]

> In general the poor kids live in higher density areas and the affluent kids live in the suburbs.

The poor kids also live rural.

Remind me again, where there are lower costs, but also lower income, less opportunity, harder to get anywhere, less education? And also, who did most of rural vote for?

In most situations, rural = poverty = trap. Our society is nowhere near prepared in addressing the rurality and poverty trap.

But really, this whole dei being a proxy for this gender or that race issue is looking around the real problem. In the end, its all about access to 2 resources: money and time. The bourgeoisie have it, the proletariat do not. As long as there is a massive gulf between the 2, we'll argue this in different names and forms (civil rights, affirmative action, political correctness, DEI)

Timon3 a day ago | parent | prev [-]

> But then who are you arguing against? Is there someone strongly opposed to providing equal access to training materials, e.g. by making them available in school libraries?

The materials aren't available. Why do you think that is?

> In general the poor kids live in higher density areas and the affluent kids live in the suburbs.

Ah, and that means we can ignore poor kids who don't fall into this pattern, as well as poor kids who live too far away from libraries?

> Actual perfect meritocracy is impossible because actual perfect anything is impossible. But meritocracy is the goal and what you want is to get closer to it. Which providing better study materials in school libraries can do, but that isn't what anybody is complaining about when they're complaining about DEI.

You're ignoring that this is only part of the equation, as the tutoring etc. is also missing. There must be special programs for the disadvantaged to level the playing field here, but that's what anti-DEI advocates also complain about!

Actual meritocracy isn't the goal when you argue that the disadvantaged should be fine with far worse resources and opportunities, as you've done in this thread. You've repeatedly argued that it's fine if they have far higher time investments and far worse materials, as long as they theoretically could achieve similar things as the rich. That's simply not meritocracy.

AnthonyMouse a day ago | parent [-]

> The materials aren't available. Why do you think that is?

To begin with, they often are. A lot of school libraries actually have test prep materials available. They don't all have them because libraries are locally administered and each locality gets to make its own choices, but if that's the case in your locality then you can direct your complaints to the town council rather than the federal government.

> Ah, and that means we can ignore poor kids who don't fall into this pattern, as well as poor kids who live too far away from libraries?

This is the thing where perfect is impossible. If you live in an urban area, having a library within walking distance is feasible because there are enough people there to justify it. If you live in a rural area, it isn't. What do you propose to do about it?

> You're ignoring that this is only part of the equation, as the tutoring etc. is also missing. There must be special programs for the disadvantaged to level the playing field here, but that's what anti-DEI advocates also complain about!

Rich people will pay for things that aren't scalable. If your parents make $20M/year, they can spend $1M/year on their kid. If you spent $1M/year on each of the 74M kids in the US, the cost would be $74 Trillion, which exceeds the US GDP. And there is a threshold past which additional spending has diminishing returns. Again, the goal is to get as close to measuring merit as feasible; "closer than now" is possible but perfection isn't.

Timon3 a day ago | parent [-]

> To begin with, they often are. A lot of school libraries actually have test prep materials available.

They have some materials available, but often older or less specialized ones. That's my whole point: rich people have access to better materials. This is simply a fact.

> This is the thing where perfect is impossible. If you live in an urban area, having a library within walking distance is feasible because there are enough people there to justify it. If you live in a rural area, it isn't. What do you propose to do about it?

How about introducing DEI programs that help these disadvantaged people access the same materials? Again, you're basically saying that they have to suck it up and accept their position. That's not meritocracy.

> Rich people will pay for things that aren't scalable. If your parents make $20M/year, they can spend $1M/year on their kid. If you spent $1M/year on each of the 74M kids in the US, the cost would be $74 Trillion, which exceeds the US GDP. And there is a threshold past which additional spending has diminishing returns.

There's obviously an incredibly large gap between "spend $1M/year on each of the 74M kids in the US" and "poor kids should either have no access at all, or have to walk large distances to public libraries, only have access to worse materials and have no tutoring available". The latter simply isn't meritocracy, yet you keep arguing that it is, and keep arguing against DEI programs.

AnthonyMouse a day ago | parent [-]

> That's my whole point: rich people have access to better materials. This is simply a fact.

"Rich people have more money" isn't an interesting fact, it's just the definition of rich people.

> How about introducing DEI programs that help these disadvantaged people access the same materials?

The term "DEI" has been applied to disparate impact rules and other policies that amount to race quotas and correspondingly garner strong opposition. If you want to advance good policies, you should stop using the same term to apply to them as is used to apply to bad policies with strong opposition.

> There's obviously an incredibly large gap between "spend $1M/year on each of the 74M kids in the US" and "poor kids should walk large distances to public libraries, have access to worse materials and have no tutoring available".

There is equally obviously a point at which the threshold of diminishing returns is met, and high-quality individualized private tutoring is plausibly beyond that threshold because it is very expensive. It's also still not clear how you expect to feasibly provide a high density of libraries in an area with a low density of people.

Timon3 a day ago | parent [-]

> "Rich people have more money" isn't an interesting fact, it's just the definition of rich people.

That's not what I said. This is bordering on bad faith, please don't do that.

> The term "DEI" has been applied to disparate impact rules and other policies that amount to race quotas and correspondingly garner strong opposition. If you want to advance good policies, you should stop using the same term to apply to them as is used to apply to bad policies with strong opposition.

First, what term would you have me use instead? Second, I don't believe it matters what term I choose, because it will get demonized just like DEI did.

> There is equally obviously a point at which the threshold of diminishing returns is met, and high-quality individualized private tutoring is plausibly beyond that threshold because it is very expensive.

There is still a large gap between "high-quality individualized private tutoring" and "poor kids should walk large distances to public libraries, have access to worse materials and have no tutoring available".

But that's besides the point, which was: objective metrics don't mean you're measuring merit. You've shown wonderfully how those advocating for "meritocracy" often don't care about actual merit. Thank you for the discussion, but I don't think it makes sense to continue, as you seem to simply not care about the issues with your position.

wredcoll a day ago | parent | next [-]

Personally I appreciate your merit. Starting the race 100 meters ahead of the other runners probably doesn't get you a very accurate measure of who is the fastest.

AnthonyMouse a day ago | parent [-]

The issue is that it's not an athletic competition. If someone is better at heart surgery, it doesn't matter if it's because their parents could afford books and someone else's couldn't, that's still the person you want doing heart surgery.

If you then want to buy books for people who can't afford them, that's an entirely different proposal than giving the job to someone who isn't as qualified.

wredcoll 21 hours ago | parent [-]

> The issue is that it's not an athletic competition. If someone is better at heart surgery, it doesn't matter if it's because their parents could afford books and someone else's couldn't, that's still the person you want doing heart surgery.

Well, an athletic competition would make more sense because that actually determines whose the best. We don't test heart surgeons to see who the best is, we test to see if they can do the job.

That's what people tend to... conveniently overlook... in these conversations. No one is hiring "the best" or only accepting "the best" into their college or whatever else. They pick a good one from the pool of candidates they have available.

Trying to pretend that "using race to pick between two equally qualified candidates" is the same thing as "picking unqualified candidates" is, well, damn close to a lie.

AnthonyMouse 21 hours ago | parent [-]

> Trying to pretend that "using race to pick between two equally qualified candidates" is the same thing as "picking unqualified candidates" is, well, damn close to a lie.

When you have a competitive major university that gets thousands of applicants and you base admission strictly on test scores, you'll end up accepting only 1% black applicants because their test scores are lower for various reasons. If you wanted to accept 14% black applicants as reflects their proportion of the US population, you would have to be turning down other applicants with significantly higher test scores. It's not just about accepting someone who got a 1520 instead of a 1530, the difference is hundreds of points.

wredcoll 20 hours ago | parent [-]

That seems rather unlikely, you have any data to back that up?

Someone else linked this article https://www.aei.org/carpe-diem/new-chart-illustrates-graphic...

> For the 2015-2016 academic year, the average GPA of all students applying to medical schools was 3.55 and the average MCAT score was 28.3 according to data from the Association of American Medical Colleges (AAMC).

> The middle set of bars in the top chart above show that for applicants to US medical schools between 2013-2016 with average GPAs (3.40 to 3.59) and average MCAT scores (27 to 29), black applicants were almost 4 times more likely to be accepted to US medical schools than Asians in that applicant pool (81.2% vs. 20.6%), and 2.8 times more likely than white applicants (81.2% vs. 29.0%).

Seems like they're in the same applicant pool.

AnthonyMouse 18 hours ago | parent [-]

That link shows that (presumably in order to meet diversity targets) black applicants with a GPA of 3.2-3.39 had nearly the same acceptance rate as Asian applicants with a 3.6-3.79 GPA. 0.4 points is an entire standard deviation for GPA.

AnthonyMouse a day ago | parent | prev [-]

> That's not what I said. This is bordering on bad faith, please don't do that.

The premise of a meritocracy isn't that everyone is the same, it's that everyone is subject to the same standard. The alternatives are things like racism or nepotism where someone gets the position even if they're not expected to do a better job, because of their race or because their father owns the company.

But merit isn't a fixed property. If you spend your time studying physics, you'll make yourself qualified to do certain types of engineering when spending that time playing football wouldn't.

Money, then, can be used to improve merit. You can e.g. pay for tuition at a better school that someone else couldn't afford. If that school actually imparts higher quality skills than a less expensive school (or no school), a meritocratic hiring practice will favor the graduates of that school, because they're actually better at doing the job.

You can then argue that this isn't fair because rich people can afford better schools etc., but a) that will always be the case because the ability to use money to improve yourself will always exist, and b) if you would like to lessen its effect, the correct solution is not to abandon meritocracy in hiring decisions, it's to increase opportunities for the poor to achieve school admissions consistent with their innate ability etc.

> First, what term would you have me use instead? Second, I don't believe it matters what term I choose, because it will get demonized just like DEI did.

The demonization comes from rooting the concern in race rather than economic opportunity, because the people obsessed with race are interested in dividing the poor and pitting them against each other in tribal warfare, and then any term you use for that will be demonized because it will become infected with tribal signaling associations.

> There is still a large gap between "high-quality individualized private tutoring" and "poor kids should walk large distances to public libraries, have access to worse materials and have no tutoring available".

And then we're back to, what is even the dispute? You can't close the entire gap because part of the gap is a result of things that are infeasibly expensive at scale and no one disputes that. There are cost effective and reasonable policies that could close some of the gap, but many of those have already been implemented or could be adopted with minimal opposition if they were simply proposed in the places not already doing them, because they're cost effective and reasonable. It's literally only a matter of going to your town council meeting and convincing them that it's a good idea.

People don't strongly oppose libraries that stock study books. They oppose race quotas.

wredcoll a day ago | parent [-]

> People don't strongly oppose libraries that stock study books. They oppose race quotas.

People absolutely do oppose libraries. They also oppose programs that pay for tutors for poor kids, programs that allocate more money to schools in poorer neighborhoods and basically anything else you can think of.

But I do admit it must make your life incredibly simple to just pretend racism doesn't exist and everyone ends up in the exact position they deserve.

AnthonyMouse a day ago | parent [-]

> People absolutely do oppose libraries. They also oppose programs that pay for tutors for poor kids, programs that allocate more money to schools in poorer neighborhoods and basically anything else you can think of.

Opposition to spending in general is distinct from opposition to a specific policy because the policy has a deleterious effect, and is much easier to overcome if you would e.g. source the money from a constituency that supports the policy, or offer to cut something else to make room in the budget.

> But I do admit it must make your life incredibly simple to just pretend racism doesn't exist and everyone ends up in the exact position they deserve.

Straw man.

croes a day ago | parent | prev [-]

That would need some inclusion laws that enforce preferred hires from lower financial backgrounds instead of better prepared candidates with more resources.

AnthonyMouse a day ago | parent [-]

So why is that not the proposal instead of DEI?

croes a day ago | parent | next [-]

That’s still DEI.

Poor have less chances? Enforce hiring of the poor.

Non-white habe less chances? Enforce hiring of non-whites.

Non-male have less chances? Enforce hiring of non-males.

Disabled have less chances? Enforce hiring of disabled.

If you have less chances because of a attribute you aren’t responsible for, enforce hiring of people with such an attribute to normalize the attribute in the workspace is DEI.

AnthonyMouse 21 hours ago | parent [-]

The premise of that proposal is that the test scores are inaccurate as a result of the economic disparity, because people with less income have less resources to prepare for the test. That would apply in the case of an economic disparity because having more resources allows you to artificially receive a higher score. It's not about accepting someone with less merit out of charity but rather about adjusting for a measurement error.

But the economic disparity is the reason for the racial disparity, because otherwise we expect people of different races are equally intelligent, right? So the economic disparity is the real one and accounting for that inherently accounts for the racial disparity as well, and you don't need both.

Which is the reason doing the latter is controversial.

Timon3 17 hours ago | parent [-]

> But the economic disparity is the reason for the racial disparity, because otherwise we expect people of different races are equally intelligent, right? So the economic disparity is the real one and accounting for that inherently accounts for the racial disparity as well, and you don't need both.

That's only true if you assume prejudices like racism and sexism don't exist anymore, but they do. Even today, these are the lived experiences of many people in society. As examples, there are black people who don't get jobs because they are just assumed to be worse candidates, even when they are more qualified and put in more work. There are women who don't get jobs because they are just assumed to be worse candidates, and so on.

These are real implicit biases, and they don't go away by just ignoring them.

Whoppertime 6 hours ago | parent [-]

Yes but what prejudices we accept and which we do not is arbitrary. You have no choice over your height. Taller people can have more promotions and more dating opportunities, but we don't have affirmative action for short people, and we don't treat women who say they don't date short guys like we treat women who say they won't date black guys. There's always going to be discrimination, what form of discrimination is acceptable or unacceptable is still arbitrary.

Timon3 5 hours ago | parent [-]

First, let's not talk about dating and work in the same argument - they are fundamentally different in many important ways, and it's not conductive to the conversation.

Second, at least call a spade a spade - according to you, when people say meritocracy they actually mean "meritocracy with handicaps for non-whites and non-males". Let's not call that "meritocracy", okay?

wredcoll a day ago | parent | prev [-]

Because it's incredibly complicated and proxying it via race is easier and humans like easy?

AnthonyMouse 20 hours ago | parent [-]

Using race as a proxy for other things is what racism is all about. We're trying to get rid of that rather than perpetuate it.