Snowflake CEO Says Worker Merit Should Outweigh Diversity Goals
bloomberg.com
bloomberg.com
Do people go around office and then think "Oh we don't have anyone blonde in the team, we should hire a blonde person"? Of course not, nobody cares.
There should be no "diversity goals". Roles should be filled on merit (that means whatever objective criteria were set when defining the role). Race, gender, etc. should be totally irrelevant.
That's the ideal. "Diversity goals" are a symptom of a problem but they also perpetuate the problem by labelling people. The goal should be colour-blindness, gender-blindness, etc. and we won't get there by continuously caring a lot about these attributes and keeping tabs on them.
On a smaller team, hiring high performers is obviously important. But for large organizations, there is a WIDE tolerance for incompetence. Hiring staff consistently underestimate how much can be learned on the job. Mid-managers are much, much more easily replaced than anyone wants to admit - and if you are already hiring lifestyle employees, you may as well be hiring a diverse set of project jockeys.
I won't go so far as to say merit is a myth, but adequately assessing merit at scale during a hiring process does feel pretty fraught.
I think managers-- bad ones-- pigeonhole people and make decisions on personal preference.
I once worked for a Fortune small cap-- and there was this incompetent manager (IM), nice guy-- assigned to the hardware portion of the project. There was this very competent manager (CM), assigned to the software portion.
It was recognize by management that IM was running his half of the project into the ground-- but the management "liked" him, so they assigned all his work to the CM and then promoted IM so he was now CM's boss and didn't have to do the work he couldn't do.
CM took on twice the workload and lead the project to success.
The perverseness of this is staggering. IM is given more money to do nothing. CM has to do twice the work for the same amount of money.
Later, the company tried to fire him for an injury he sustained on the job, claiming he was "no longer able to perform his job function."
Or, less charitably, the Gervais Principle strikes again.
The candidate could have no video and obfuscated audio, and I'd still enthusiastically hire them if they fell into the 10%. Most people I've spoken to have the same experience. Programming is an employee's market and stands to be into the future. As such, I don't think anyone who can code is at a significant disadvantage.
For example, 90+% of programmers are male, so not all firms can meet a 50-50 gender balance. Top firms try, but that will pull from somewhere. The same goes for race/ethnicity.
Diversity begins in schools.
It's not like random people disconnected from the startup cared enough to write or edit an article about it.
Is it just a massive ego stroke for those in charge?
And it feels unfair, because bias is unconscious and we usually aren't aware of it. Unless you fully accept that you are a person with biases that are unconscious and invisible to yourself, it will feel like you're compromising on merit.
Embracing diversity isn't something I have much of a problem with, it's the way diversity it's defined which suggests attributes like race and gender are the defining characteristics of an individual that I have the problem with.
I agree that bias is a problem when hiring, but I also think we go too far when we as a society suggest the primary reason for Asian over-represented in tech is racism, and then actively discriminate against Asians in hiring. I'd argue this line of reasoning is deeply flawed and immoral for numerous reasons.
I'm not going to downvote you, but I really don't like this semantic maneuvering and I think it's disingenuous. By definition, merit is supposed to be free of bias and be the perfect correlate between qualification and result.
If the argument is that humans are flawed and they sometimes don't reach this ideal because of prejudice or what-have-you, that's another discussion entirely. But no, merit does not have a high degree of subjectivity and bias associated with it.
I'm open to introspection, but at some point "bias" has to become something falsifiable. If it remains something unconscious and invisible, then how does one actually bear responsibility for it?
In many ways your quote reminds me of Christianity. Simply replace the word 'bias' with 'sin'.
Hiding $protected_variable can be difficult in in-person interviews, but should at least be possible for a CV, and when that has been tested, it has been observed. (Names imply race and gender, hide the names and get different aggregate results).
Am I a diverse hire?
I feel like the problem started when diversity switched from a group measure to an individual label.
Individuals aren't diverse. A 100% Black team isn't racially diverse. A 100% gay team isn't sexual orientation or gender diverse. Asking whether a bisexual Latina is "diverse" or a "diverse hire" should be meaningless.
Not every commitment to diversity has this conflict. For instance, committing to diversity by anonymizing applications is not. But pursuit of representation targets largely is.
Many of the best programmers in the world move to SF specifically because of the opportunity.
Logic/math does not bear out this well-intended belief.
If you select N employees for the company based entirely on merit, you get some set U.
If you select N employees entirely based on diversity, you get a set V.
Whenever U /= V, the criteria are necessarily in conflict. The conflicting choices are represented in the symmetric difference of U and V: elements in U that are not in V and vice versa.
The merit criteria selects some people U\V (those in U who are not in V); the diversity criteria disagrees with that and insists they be replaced with people in the V\U difference.
You can mix the criteria: somehow hedging between diversity and merit. There is more than one way of doing that and those ways can be in conflict with each other as well as with pure merit or pure diversity.
Ideally you want merit only because only that is relevant to the organization. What may happen instead is that selection takes place based on a mixture of merit plus some undesirable bias. Diversity advocates rightfully point out that this is in conflict with pure merit.
What diversity should ideally mean is the elimination of such a bias, rather than introduction of compensating biases.
However, I’m much less certain of the value or sanguine about the prospects of diversity-focused hiring in Silicon Valley, which is very check-box based, is reliant on a definition of diversity that is very US-focused, and has malleable standards. The last point is a contentious issue, but I’ve encountered instances in high-profile companies (Facebook, Netflix, Google) - it’ll be interesting to see how this works out, and I for one will closely look at all the data that I have handy.
Downvoters: What is so problematic about this sentence verses the parent's sentence?
Care to explain?
Which is why companies in SV sponsor visas, spin up satellite offices, and acqui-hire.
Do you think society as a whole (and by extension, said company) has a responsibility to create a system where there is an equal opportunity to develop merit?
E.g., if society defines merit as the ability to play tennis, it would probably be best served (ha) to ensure everyone has access to a tennis court to develop those valued skills. Otherwise, you won't really get the "best" players you could, you'll essentially commit the equivalent of sampling error.
We know training works to improve the measures we use to estimate merit. It's not a coincidence, for example, that those who have resources to dedicate to SAT prep tend to have better SAT scores. I think the argument is this tends to concentrate merit, perhaps under a erroneous notion that merit. E.g., those who have the resources to develop merit tend to get paid more which allows them to invest more in developing merit in themselves and their offspring...rinse and repeat. I would argue merit can be used as a bad-faith argument if there isn't equal opportunity to develop those attributes.
Taking your example of SAT score, to which the main contributing factor is clearly time spent studying. If we investigate further, we find that the main factor in determining whether young adults spend time studying isn't economic, but rather cultural.
Students who grow up in households that highly prize grades, study and education unsurprisingly spend more time studying, regardless of household wealth. These students also unsurprisingly have much more well-developed skills and habits around study and educational attainment.
The question is what would you do about this? Should you do anything? I personally don't think it's any of our collective business how a household chooses to raise their children beyond the extremes of physical abuse and neglect.
Only speaking for myself, I think society has a role in helping overcoming those disadvantages that aren’t the fault of the child. Otherwise, you are asking them to pay for the sins of their fathers. I also think society as a whole would be better off for it and the overall aptitude would increase by truly seeing who has the most merit, rather than who won the cultural lottery.
Are these children "disadvantaged" from your perspective? Perhaps, but how would you respond to a parent who suggests publicly-schooled children are far more disadvantaged in metrics important to them? From where do you derive the authority to say that "economic achievement" is the supreme value we should optimize for?
I would say from the social contract of acknowledging that, as a part of a society, I have at least some duty to that society in exchange for what that society provides (whether that is just a functional judiciary, a stable economy, rule of law, infrastructure, collective defense etc). Its not me who has authority in a representative democracy but the people from whom the government derives power. So it’s not really if I think that’s the right metric but rather if society does. To a certain extent, they have already voted on this with the creation of a public school system. And it’s not to say there is a single “supreme value” to optimize for because society is too complex to have a single cost function and dangerous to pretend there is (which I worry your ‘paper clip’ comment alludes to), but we can still acknowledge there is an acceptable base level for specific metrics.
I think the libertarian ideal is seductive because if there’s never an authority, you never have to be wrong. In the words of a philosopher, “ Perhaps this is why believing in absolute freedom is so tempting — if we follow that path, we are always “right,” which is easy.”[1] It’s an easy, but overly simplistic, framework. I tend to find the focus is disproportionately on what the government owes us without acknowledging that in exchange we have an obligation to society.
I can understand why someone advocates for ultimate freedom because of that seductiveness but I can’t really understand how they can also claim it will create a meritocracy unless you reduce yourself to a myopic view. I can reduce my worldview all the way down to the single room from which I’m typing and proclaim I’m the most meritorious person, but that position gets harder to defend as I acknowledge I am part of a larger society.
[1] https://www.nytimes.com/2021/05/24/special-series/mike-schur...
If you have a master who can tell you what to do, compel your cooperation with the threat of violence, and from whom you have no legitimate recourse, are you free? Does it matter if that master is elected or not?
I'm not saying that I'm "right". I'm simply saying no one is, and that words like "right" and "good" are human constructs, not factual observations. They vary from person to person, community to community and culture to culture. Given that we are all equal as people, it seems to follow that no one has the right to impose their ideas of "right" or "good" onto someone else, regardless of whether the imposition is democratic or authoritarian.
> There is only Steve, Jamel, Selena and 332 million other siloed ego-states; a confederacy of individuals for whom freedom, safety and the ability to flourish are zero-sum competitions.
This seems to be an accurate description of the real situation though, doesn't it? We're not ants. People fundamentally disagree on basic things. We lack an oracle to mediate these disagreements. One side forcing the other's cooperation is morally untenable. And a confederacy of individuals does not equate to freedom, safety and flourishing being zero-sum. Voluntary cooperation has and continues to achieve great things, why not let it be?
Regarding meritocracy, what does one look like in your mind? What happens to individuals who lack merit in a meritocracy?
>How does the imposition of obligations to "society" through an involuntary social contract differ from slavery itself?
There are a few distinctions that I don't think your analogy captures. It would be an apt analogy if 1) slaves got a chance to vote out their master at reasonable intervals and 2) you have the ability to go to different plantations.
You could go to another society if they are willing to have you. But then you'll be governed by their rules. The only way out of this is to find a patch of land outside every other society's jurisdiction. So yes, we have to balance our own interests with that of the group we associate with. You've probably heard this before, but I'll take a chance at outlining a hypothetical: Assume you're on your condo balcony on the 20th floor but trip and fall. Luckily, the person directly below you on the 19th floor has installed a flag pole that you grab ahold of and you can slowly start walking yourself back hand-over-hand. But hearing the commotion, the neighbor comes out and proclaims you must get your hands off their flagpole. There's really only two possibilities if you believe in the sanctity of personal property rights: 1) you ignore your neighbors commands and walk yourself back to safety, meaning you are immoral or 2) You let go and plunge to your moral death.
I think almost everyone picks scenario #1 because we implicitly understand that certain ideas like personal property rights are guiding principles and not dogmas. Meaning we understand they are not absolutes and can be modified when there is enough at stake. We may debate about what constitutes "enough at stake", but they are not meant to be a suicide pact.
>This seems to be an accurate description of the real situation though, doesn't it?
This may be getting to the heart of where we disagree. I don't think it's a particularly good representation of the real situation, mostly because society is too complex.
I think there's a reason that most society's evolved toward nation-States. There are certain economies of scale that need to occur as society gets more complex. (This isn't to say it's the end-state of the way societies will be governed, either.) At a certain point, we no longer have the bandwidth to make all the decisions necessary for a completely individual decisions to truly be cooperative. We just didn't evolve to operate in systems that include millions of direct and indirect interactions. So we develop systems to mediate these interactions, like representative democracies and home owners associations and all kinds of other groups. And with this comes balancing some of those individual principles with the needs of the groups. I also don't think I'd agree on your assumption that safety and flourishing are zero sum; can you elaborate on what you mean? I think the libertarian principles are romantic notions that fits well under the Jeffersonian ideal of a yeoman farmer, but largely incompatible with the complexity of modernity. To that extent, I'm not 100% convinced that going to the extreme would increase flourishing writ large.
>What happens to individuals who lack merit in a meritocracy?
I don't think they flourish as well as meritorious individuals but I think a moral society has an obligation to provide some baseline level of quality of life. But this question is a bit of a digression from the original topic regarding whether or not society is structured to actually identify those with the most merit.
Likewise =)
> There are a few distinctions that I don't think your analogy captures. It would be an apt analogy if 1) slaves got a chance to vote out their master at reasonable intervals and 2) you have the ability to go to different plantations.
I don't think it would matter to a slave being whipped on a plantation that the person doing the whipping was elected by a narrow majority of his fellow slaves. Nor would it matter that he could leave to another plantation with a different master whose whip stings just the same. Slaves who elect their masters are still slaves, are they not? Only people without masters are free.
If you would allow me to proselytize a little, how would you know you are free? Is there a test? I think there is. It's the same test you'd use to tell if you are being raped or having sex. Does your consent matter? Will revoking your consent be respected by other parties involved? If you're having sex with someone whom you yourself chose, and you suddenly and vocally revoke your consent (ie. say "stop"), will they stop? or will they use violence to force you to continue? If they choose the latter option, in that moment they have, in a very literal sense, become your master and you their slave. I think the same test can be applied to any human relationship, interaction or form of cooperation, would you agree? If not, which relationships does this test not apply to? Why not?
> You could go to another society if they are willing to have you. But then you'll be governed by their rules.
Yes indeed, you could move to a society which practices child cannibalism if you are willing to be governed by their rules. I'm asking whether you think there should be limits on the kinds of rules that people can enforce on each other with an involuntary agreement. If there aren't, then is all that's stopping us from reinstating slavery or organizing a holocaust a democratic majority? I also think telling someone whose rights your violating "well, you can always move somewhere else" doesn't hold any water. People have rights where they are.
> We hold these truths to be self-evident, that all men are created equal, that they are endowed by their Creator with certain unalienable Rights, that among these are Life, Liberty and the pursuit of Happiness.
These are the axioms I'm starting from. If we're all created equal, and we are all entitled to life, liberty and the pursuit of happiness (in that order), does everything else not follow? You'll also note that "a baseline quality of life" is not present in this list.
> Assume you're on your condo balcony on the 20th floor but trip and fall ... There's really only two possibilities if you believe in the sanctity of personal property rights: 1) you ignore your neighbors commands and walk yourself back to safety, meaning you are immoral or 2) You let go and plunge to your moral death.
I agree wholeheartedly. I also think each individual has a right to make this choice for themselves. Many people would prefer to starve than beg. Many others have chosen to commit suicide rather than burden their families with their care. Whichever moral code you live by, and whichever ways you are willing to bend or break it in dire circumstances; these are deeply personal choices. In regards to the legal aspects of your example, there are several layers of due process that are in place exactly to prevent punitive action for reasonable violations of the law as-written. Namely policing discretion, prosecutorial discretion and trial by jury. These are clearly imperfect, but I don't see them convicting the trespasser in your example.
> At a certain point, we no longer have the bandwidth to make all the decisions necessary for a completely individual decisions to truly be cooperative.
I'm not sure that's true - some of the most complex things we do and build are overwhelmingly voluntary and cooperative in nature. The tens of thousands of people collaborating at Airbus to build an A380 are cooperating splendidly without being forced to, as are the employees of every major company which has contributed so much to the advancements in our quality of life.
> I also don't think I'd agree on your assumption that safety and flourishing are zero sum;
I think you misread what I wrote:
> And a confederacy of individuals does not equate to freedom, safety and flourishing being zero-sum.
However I might say that flourishing, based on the hierarchical nature of our social psychology, might be zero-sum. Only so many people can be at the top of the social pyramid, regardless of wealth. Most of everyone else seems to be wired to derive misery from the fact they do not occupy such a place. It does seem possible to overcome these instincts with practice however.
> I think the libertarian principles are romantic notions that fits well under the Jeffersonian ideal of a yeoman farmer, but largely incompatible with the complexity of modernity. To that extent, I'm not 100% convinced that going to the extreme would increase flourishing writ large.
I agree! The Jeffersonian homestead is the perfect example of functional liberty. I was going to cite it in my last post. Don't you think it could have been maintained with the political will and/or stronger constitutional protection? The US may not have become the global superpower it is today, but I don't see that as all together terrible, given all the awful things we have done to people in order to achieve and maintain this status.
>elected by a narrow majority of his fellow slaves.
In the US at least, we already have checks and balances that avoid the tyranny of the majority.
Nor would it matter that he could leave to another plantation with a different master whose whip stings just the same.
In this case, it only holds if the assumption is that any social arrangement devolves into a master/slave relationship. Why hold onto this assumption if we've already said the "slave" is free to go as well as determine who runs the plantation? To me, that's sounds more like an egalitarian arrangement than a master/slave one. Maybe this is where the slave/master analogy has its limits.
>Does your consent matter? Will revoking your consent be respected by other parties involved?
Yes, consent matters. But I think this may be one of the fundamental areas where we disagree. I think the complexity of society requires numerous implied consents, I get the impression you only favor explicit consent. My larger point is that explicit consent is impossible in a complex society. When I get onto a public road, I do not get explicit consent from the 300+ million citizens who helped pay for that road. Likewise with most of the other social arrangements. The Founding Fathers understood this as well; Washington was a notorious disciplinarian who did not look for consent from individual soldiers regarding their willingness to continue serving because there is a pragmatic limit to instances of individual consent. Soldiers in today's army literally sign a contract and they don't get to revoke that consent merely because they've changed their mind about their willingness to uphold that agreement when things get hard. This is very similar to the idea of the social contract except one is explicit because it's relatively narrow in scope while one is implicit because of the complexity of the context in which it operates.
>These are the axioms I'm starting from. If we're all created equal, and we are all entitled to life, liberty and the pursuit of happiness (in that order), does everything else not follow? You'll also note that "a baseline quality of life" is not present in this list.
I mean, Jefferson wrote beautifully in the Declaration of Independence but the actual social contact is the Constitution. And within that, the federal government has specific enumerated powers, including using those specific powers for the "general welfare" of the citizens (Article I, Section 8, Clause 1). This general welfare is exactly the quality of life I'm talking about. Also note that a right to property is not in your axioms. Now I'm assuming, from your perspective, this is implied by the right to liberty. My other main point is that rights and responsibilities are two sides of the same coin. If we have implied rights protected by society, we also have implied responsibilities to said society.
>there are several layers of due process that are in place exactly to prevent punitive action for reasonable violations of the law
This is actually speaking to my point that they are general principles and not hard-lined ideologies. You've indicated you have a hierarchy of rights, starting with the right to life and that lesser rights are subjugate to that primacy (like the right to property, like the flag pole). I agree. So it sounds like we agree that liberty isn't absolute in a society where interactions between rights have to be balanced and that some rights are subjugate to others.
>collaborating at Airbus to build an A380 are cooperating splendidly without being forced to
I agree. But the complexity of the operation means they aren't knowledgeable regarding every operation. The complexity actually makes certain conventions necessary. A software engineer may have little or no clue what a structural engineer is doing. They coordinate by creation of social orders and delegation of duties. It's the same thing we do in order to make a complex society work; we sacrifice explicit consent by delegating authority to people like political representatives. The SWE doesn't get to explicitly approve the decision of Airbus to sell military equipment. We build a system of implied consent to manage the complexity of interactions.
Sorry, I did misinterpret what you had previously wrote. Thank you for correcting me.
>Most of everyone else seems to be wired to derive misery from the fact they do not occupy such a place.
Roosevelt said "Comparison is the thief of joy" and I tend to agree with both of you. I do think this is where those balances to counteract the tyranny of the majority come in.
>The Jeffersonian homestead is the perfect example of functional liberty.
I think we agree here. But one of my main points is that this isn't practical in today's society. As romantic of a notion that it is, I think it's really just a hypothetical indulgence rather than a pragmatic approach to policy. I doubt many people in US would like to return to an agrarian economy because, even though it might have more principled liberty, it would lower their quality of life. That ship sailed over a hundred years ago when America decided to become an global superpower and integrated into a worldwide economy. You can get awfully close to a libertarian ideal by joining a hunter-gatherer society, so why don't people do so? My premise is that nation-states evolved as a social response to the risks of a increasingly complex society. I think the libertarian thought experiments are fun but not particularly useful.
I think so too, and I'm grateful for your replies. Hard to find a good conversation these days.
> In the US at least, we already have checks and balances that avoid the tyranny of the majority.
Indeed, but my thesis is that these have been insufficient to protect liberty, as it was initially practiced and understood, from erosion so significant as to change its very meaning.
> Maybe this is where the slave/master analogy has its limits.
> Soldiers in today's army literally sign a contract...
> ...one is explicit ... while one is implicit
It does seem as though this is the key area where we disagree. To me there is no difference between a soldier who wants to leave combat (but can't because he'll be shot/imprisoned by his unit) and a slave who wants to leave his plantation. These people are both decidedly not free, and must take orders from others. Indeed the soldier may have signed a contract, but we've long established that you can't sign yourself into bondage, and that the penalty for breach of contract should be financial only, backed by bankruptcy protection. For a conscript, this is doubly true, as he didn't even have the opportunity sign his freedom over to begin with, making him indistinguishable from a slave. At least the volunteer soldier had the opportunity to consider the terms of his contract before signing, and has the opportunity to renew or reject those terms every contract period. Neither the slave, nor the conscript, nor any of the rest of us, have that opportunity.
Fundamentally, implicit, lifelong obligation to collaborate, with no opportunity to opt-out, is indistinguishable from slavery.
Say instead of emancipating slaves completely, they obligated them to work one day a week on their masters' plantations. They would be free to do what they wanted with the rest of their time. If they moved, they would be obligated to find new masters to serve, but still only one day per week. Would these people be free?
Returning to your point about defense, this is the strongest moral case for enslaving people: the imminent threat of total destruction. However under these circumstances almost no one objects to cooperating anyways due to overwhelming social pressure. Indeed, this case falls into the broader category where forced cooperation is justified: maintaining life and liberty themselves. Thus the draft to serve in WWII and the draft to serve in Vietnam are clearly in separate categories, with more than public support and opinion differentiating them.
> This general welfare is exactly the quality of life I'm talking about.
Regarding US policy, we've pretty much narrowed it down to Hamilton and Madison regarding the general welfare clause, ultimately decided in United States v. Butler. My objection to this decision is based in the overall vagueness of the interpretation of "general welfare". It's like saying the government is empowered to do "good". Much like religion, it can be used to justify _anything at all_ to suit any particular agenda. It also seems to be a backdoor to the otherwise carefully and narrowly limited powers set out for the government in the constitution. How do you think the founding fathers would react to the modern US? How do you think the United States would change if this decisions was reversed?
> But the complexity of the operation means they aren't knowledgeable regarding every operation
You don't need to know what's going in every corner to meaningfully consent. Much like you don't need to understand where the tea you are drinking was grown, and by whom it was harvested in order to consent to an offered cup (you can of course inquire); you only need to consent to direct interactions between yourself and others. If you don't like your manager's tie at Airbus, you can just get up and leave. If you have a contract, then its terms will be used to decide who owes what.
> So it sounds like we agree that liberty isn't absolute in a society where interactions between rights have to be balanced and that some rights are subjugate to others.
Yes! In my eyes liberty is a graph, where each possible action in a given moment represents a node. The more actions are permissible (note - not possible, permissible - nature has a say too), the larger the graph, the more free you are. Our society, whose stated goal is to optimize for liberty second only to life itself, ought to design policy that will keep every individual's graph as large as possible. There is no requirement to return to an agrarian economy to put this into practice. In this light, GoFundMe is a much more ethical platform to fund personal medical expenses than taxpayer-funded public health insurance.
If we keep optimizing for the collective good, we end up in China, who has unquestionably been overwhelmingly successful in lifting its (surviving) population out of poverty, and keeping a large and diverse population (relatively) peaceful and productive. The collective good has been served, but at what cost?
> My premise is that nation-states evolved as a social response to the risks of a increasingly complex society. I think the libertarian thought experiments are fun but not particularly useful.
I'm not disparaging the nation-state, just proposing limits to its power. I don't think these are thought experiments at all, but rather important boundaries, without which policy decisions are just tools politicians use to advance their careers.
So no surprise some are not onboard with "diversity goals"
Now watch people misinterpret this comment and call me names...
There are instances where there is bona fide qualification for being a specific gender.
It's not common, but does exist.
Megan Fox will not likely win a lawsuit alleging discrimination based on EEO Law if I did not cast her in the role of Barack Obama in my HBO Miniseries on the Death of Osama bin Laden.
Often, they justify this by claiming that Asian and white people have to score much higher on the SAT to gain admissions. But that's probably the wrong metric. The purpose of college is to grow students, which means a better metric of merit is not a snapshot of how proficient a student is at a particular moment in time, but by how much they have improved from some baseline.
We also know that society puts BIPOC at a disadvantage because of their race. For example, historical and present day discrimination means that they have less economic power, which directly translates to educational disadvantage (worse schools, worse nutrition, less time interacting with parents who may work longer hours). BIPOC are also less likely to be recommended for gifted programs at the same level of skill. Therefore (because of society) their baseline is lower.
So given the above two points, I don't really see admission of BIPOC as discrimination at if you shift the metric to "ability to grow." IMO this is also reflected in the admissions criteria; the ability to overcome adversity is sometimes cited, and adverse life events usually accompany some amount of growth. And even though the average SAT score is lower for BIPOC applicants, that average is not low enough for those people not to be able to handle the courseworks.
That's not to say that all BIPOC applicants come from a lower baseline than all Asian/white applicants; of course there will be some distribution overlap. But using growth as a metric (instead of absolute score) could probably explain most of the test score/GPA discrepancy.
Your comment describes the meaning encapsulated in the term "positive discrimination". Groups have been disadvantaged throughout history, so the institution is deliberately assigning advantages for members of the group to offset that.
> And even though the average SAT score is lower for BIPOC applicants, that average is not low enough for those people not to be able to handle the courseworks.
This is not entirely true. Lower SAT scores are correlated with dropping out of certain majors at much higher rates. This is a significant factor in lower URM representation in STEM graduates: https://lawreviewblog.uchicago.edu/2020/10/30/aa-arcidiacono...
This statement needs more careful examination for two reasons. First, students drop out for various reasons that are not academic, including personal reasons. It's possible that if you come from a lower-income family, your family circumstances change such that you can no longer attend school. Even those who dropout for academic reasons may have their academics affected by their personal life, and without personal difficulty they may have been able to handle the coursework. So such a finding would not contradict my claim that most people have the ability to handle the coursework.
Second, because of Simpson's Paradox (https://en.wikipedia.org/wiki/Simpson%27s_paradox). When you aggregate all students, you may find that lower SAT scores correlate with high dropout rates, but that would also be consistent with a school environment that discriminates against BIPOC. (Just because the admissions environment attempts to compensate for discrimination by measuring "growth" doesn't mean that the schools themselves cannot exhibit elitism and/or racism.) Your article doesn't provide any data (so I have no evidence to support this), but it's possible that if you separated SAT scores out by race, the score has little to no effect on dropout rates.
Is it possible that it's actually a predictor of who will predict later extenuating circumstances inhibiting their studies? And one that works with remarkable consistency across all demographics? Maybe but I feel confident enough in the former that I'm going to expect to see data on those circumstances to which you allude before shifting conclusions.
https://www.google.com/search?q=SAT+score+STEM+attrition+cor...
Also, such a paper should analyze dropout rates at elite colleges, as it's possible that the correlation degrades after a certain threshold.
And yes, it is limited to an elite college (Duke University).
> However, even though we have shown substantial evidence that Duke does indeed possess private information about the student’s future performance, we cannot conclude that there is mismatch.
It (claims to) disprove the hypothesis that there is no mismatch because Duke possess no private information, as private information is a necessary (but not sufficient) condition for mismatch.
> In sum, Duke’s private information on student performance is important beyond the student’s grade point average, affecting both the student’s subjective well-being as well as the likelihood of persisting in a difficult major. It is important to note, however, that we have still only established that some of the necessary conditions for mismatch are met. For mismatch to exist, information must not only be important but the revelation of the information must change the institution the individual would have chosen to attend. Hence, even if individuals find out they are not as academically strong as they thought, the information still may not have changed their decisions.
Let's not lose track of what I initially claimed: that lower levels of academic preparedness lead to students changing out of science/engineering majors, and that this was consistent across race. The study did find that this was true. And this was limited to students at an elite university. Mismatch is about what students would have done when given this information, which is not directly related to my claim
Another point is that when a group of people call for more diversity, which would be attained by employing more people from this group, I tend to see tribalism more than diversity. Of course in a way it's fine, it's the basis of human behavior, but it still feels weird to me.
In reality, diversity goals aim to compensate for otherwise discriminatory hiring practices. That's a good thing.
What if I had 2 head count, and I first posted to an internal job board before posting to our public careers page. Would I still be in violation of EEO Law?
The point I'm making is while diversity goals are dumb sounding on the surface--in theory they force companies to examine innocuous policies and processes that skew hiring towards a certain group of people.
To give a toy example: If half the population has green hair, and the other half blue, and there is no difference between them other than hair colour, you should expect half the candidates to be of each hair colour, and half of the successful applicants of each hair colour. If you don’t see that, that should be a strong signal that your hiring process unreasonably discriminates.
The issue we see is that the blue haired people simply aren't applying for our positions, despite the best efforts of the recruitment department who have been told to hire more blue haired people because it makes the green haired management look bad.
The people doing the hiring are then told they are discriminatory and are sent on courses.
The real issue of course is much more complex and has everything to do with culture, and genetics (women generally have differing interests to men for example - for evolutionary reasons).
Why do you think I listed two fictional hair colours for my hypothetical?
Also, why I chose to split into two clauses the fragment you quoted?
Diversity goals are not created in order to encourage hiring less capable people. They are challenging companies to work harder on finding the capable "diverse" candidates, reevaluating what it means to be qualified, and whether those qualification criteria actually indicate a capability to perform the job.
Nope, "Snowflake Inc. Chief Executive Officer Frank Slootman said the cloud software maker..." he's not A snowflake, he's CEO OF Snowflake. That headlines reads different knowing the difference there.
They're not calling the CEO a Snowflake. This is how news organisations commonly refer to CEOs of companies. Here are a few examples.
- Facebook CEO reveals three new features WhatsApp is working (https://timesofindia.indiatimes.com/gadgets-news/facebook-ce...)
- GM CEO Barra’s pay rose to $23.7 million in 2020 (https://www.reuters.com/business/autos-transportation/gm-ceo...)
- Starbucks CEO Kevin Johnson explains why ... (https://www.businessinsider.com/starbucks-ceo-kevin-johnson-...)
I’ve had at least one person on here (years ago now) assume I must be a woman after I wrote something feminist.
Nothing. It purely depends on how you define "merit"? For example a lot of tech companies only hire entry level people from elite colleges. That leads to a workforce that is largely white and largely middle to upper class. Sure, the average MIT grad is better than the average boot camp grad, but there are also plenty of boot camp grads that are better than plenty of MIT grads.
If '...to "make the world a better place"' means putting in diversity quotas, then it is completely pointless to this goal.
Hiring the best and the brightest regardless of their race gets to this goal faster. As long as they are developing the tools that make the company money much faster and on time, I don't care about this 'diversity' nonsense.
So to get the change at the end of the hiring pipeline you want you need to make sure the top of the pipeline are skewed strongly in the direction you want to move the company. Maybe I'm off base but this takes two things:
1. Stop advertising in forums / regions / services that don't match the target demographics you need at the top of the pipeline. 2. Start advertising in ones that do.
(1) Don't get me wrong, I would support changing laws to make it easier to explicitly correct historical iniquities.
I’d assume that any CS grad would apply to all these firms. I certainly did when I was in university.
We end up with bizzare scenarios, where someone form Nepal, or Palestine, is considered as 'privileged', as they are asian, and weirdly arabs are counted as 'white'. While someone from Argentina or Chile, is considered as unprivileged as they are Latinos.
Which is ridiculous, as Nepal's gpd is just 1k USD per capita, while Argentina and Chile are 10x and 14x that.
Also, some South American/Latino countries are actually richer than many Eastern European countries, (countries like Moldova and Bosnia are poorer than Mexico).
Separating people by race/ethnicity feels deeply wrong.
> We can’t get distracted in that mission.
he sees diversity as a distraction.
I read somewhere an interview where he says some time ago he was rejected for a job because they gave it to a diverse candidate. this is his revenge I guess.
edit: it was here https://archive.is/dMTEM
> In 2019, during an all-hands meeting at Snowflake, Slootman told the story of how early in his career he was rejected for a job he was well qualified for because the company needed to fill a quota of hiring more minorities, a former senior Snowflake employee said.
it also implies that people get hired based on professional skills alone. so if a minority person gets lower score it must mean that they don't have enough skill. but in fact, the whole discussion is about (unconscious) bias against minorities. so you can't really assume the interview is fair.
When we have a candidate who is from a "minority" group and we treat them differently because we assume they suffered more to get where they are so they paid their due and now are entitled to the job position, isn't that an unconscious bias too?
When we look at white men and assume they had an easy ride in life, isn't that an unconscious bias too?
Look, everyone belongs to different groups at the same time so they have more privileges and advantages in some areas, while in a minority group in another. How far should we go? should we also check who has the least privilege and representation within a minority group before we hire them? Do people in minority groups ever run out of privilege once they acquire it?
This whole idea of treating people differently based on their race or other irrelevant attributes is discriminatory and racist. How the hell did we get here? I don't know.
Is there, as in actually?
Of the top of my head, the NBA, the entire of India, and all of China are crushing the competition without barely any diversity.
You should investigate the need to consider 'comprehensive evidence'. Even if it doesn't suite your conclusions.
Definitely has worked for the FAAMNG companies.
Downvoters: How exactly is having some form of diversity quotas system more fairer than merit? I would rather hire the best regardless of their background than to have a fixed set amount of candidates because of the colour of their skin. The latter is extremely discriminatory.
Get with the times man and quit being medieval. Relevant quote of the day. [0]
[0] https://twitter.com/ThomasSowell/status/1197869084479766529
But yes, let's just keep on doing things the way we've always done them. I'm sure it's working great.