Diversity quotas suck. Here’s why
blog.alinelerner.com
blog.alinelerner.com
For every fun example of "Oh, only a tattoo wearer would recognize that this product wouldn't work for people with tattoos! Diversity would have halped!" there are the Silicon Valley giants, achieving ridiculous financial results with highly non-diverse teams.
The advocates need to further explain why it is good for the team and such. Since many companies, including my own, implementing the diversity initiative by lowering the bar for the diversity candidates, even those that are being turned down at earliest stage, still interestingly, being rescued later. However, even people take issues about the candidate in question, they are afraid to voice any opinions, because the fear of being labeled as bigots. As ironic as it gets, during the group meeting, that the said candidate apparently doesn't have any deliverable for 6 months, still being praised with great efforts and hard works, while others in the team are pressured to take over what is supposedly to be done by the candidate.
I don't think this put the diversity initiative into good lights. It only further convinces people that it is more to the vanity of higher management other than productivity/benefits of the team/company itself.
If this does happen at your company, there's no need to use a throwaway account.
The company is a big one, a named one, but with some age on it. As absurd as it sounds, the candidate is hired for a hotly sought after position, that supposedly requires production ready coding skills AND battle tested machine learning skills. Yet the candidate comes with no knowledge of Python or Linux.
And, if your claim is that this doesn't happen: I have to say, I've seen people hired for many reasons. Because they're related to X (perhaps including that they can't find a job otherwise). Because they or someone in their immediate environment is disabled (which gets your boss a tax advantage). Because ...
So I would argue in Europe, it's almost the opposite. People are really treated as interchangeable, and hiring for ability is really more the exception in many companies. Hiring to make the boss look good, hiring fox tax reasons, ...
And before you say "that doesn't happen at fancy FANG company", I would like to remind you that all of Google, Microsoft, Amazon, Apple and Facebook have LARGE Irish offices, and I've visited the Facebook one: not many Irish (like 30% of the office at most, and they apologize during the interview for trying to get you to Ireland, so it's not like they like it. I'd bet good money the other companies are similar).
How much do you want to bet that the "dutch Irish sandwich" tax avoidance scheme requires a minimum headcount in Ireland ?
So let's just state this as a fact: ALL the FANG companies hire less competent people because of location (this must be the tradeoff) in trade for tax advantages in Europe.
Every last one of them.
Every last one.
And if they do it, which are these mythical companies that hire because of the bar, wherever ? We complain a lot, but when it comes to employees, nobody really doubts that FB, GOOG, MSFT, AMZN are the top, la creme de la creme. Who is this mythical company that does not lower the bar ? And if they do lower the bar, why would they not lower it below whatever you consider a reasonable level, especially for positions where they can't really hire well anyway ?
Frankly, while it has its flaws, affirmative action is the best we have. A truly blind hiring process is very uncommon (I have never heard of it before this article), and everyone knows that the best way to get a job is through a friend, reinforcing the culture which already exists at a company.
One of the biggest problem with affirmative action is that it is always used in a discriminating way. If one were to sort the worst gender or race segregated profession guess how proportional affirmative action is being applied? A while back I took a list of the top 20 gender segregated profession here in Sweden and most of them were government employed, and there is no policy in regard to hiring procedures to address that. No one is even suggesting affirmative action to address the worst professions with 99%/1% and higher gender segregation.
To give some context for Sweden, 90% of the employed people work in a profession that is considered gender segregated. That is to say if you take a random employed woman or man there is a 90% chance that they work somewhere where the ratio is worse than 60/40. If affirmative action was only used to address gender segregation then 9 out of 10 new hiring should be using it.
I think the problem with affirmative action in the gender space is that there is some evidence to suggest that a purely unbiased world would not be one where every industry had a perfect 50/50 split.
So when you approach the problem from the 'outcome' end, it's hard to know what outcome you should be aiming for.
in a competitive market, the company that performs well would be the one that's diverse, if the theory is to be believed. I have not yet seen much evidence for or against though.
This argument goes both ways. We also don't have concrete evidence proving that "diversifying the engineering team" will necessarily result in success.
You say Apple has almost none, yet is doing great. So what? Lots of things that were or are shit have done well. Even things that are now illegal. Do you like discrimination?
Every company the size of Apple should have people of all backgrounds, ages, sex, race. For every room full of twenty somethings designing a "great" product achieving ridiculous financial results there will be a dozen features or possibilities missed to invisibly cater better for more groups (potential customers).
Perhaps they need to use it in subtly different ways, or have different goals that needs an option or overlay not seen by the room full of bros. Perhaps when you hit 50 and your eyesight isn't quite so sharp so you need reading specs before you reach for the phone you realise Apple's much vaunted, industry leading, accessibility is shit. A few 40 year olds on all the teams might have noticed that and come up with better ideas and solutions than the current dross.
As to diversity being a drawback in execution, what utter twaddle.
I'd argue "diversity quotas" are the very archetype of discrimination.
The best plan is to make the process as fair as possible so that everyone has an opportunity if they want it, then let people do what they desire. It may not "look" like what some people want it to look like, but the whole point is that doesn't matter anyway.
What's missing in this is the acknowledgement that "diversity quotas" while, of course, imperfect and clumsy, reflect an effort to counterbalance institutional discrimination and measure our success. Of course arbitrary targets sometimes lead to suboptimal outcomes in specific cases. But, in aggregate, they are better than no measurement at all and I believe they do improve things.
There are many forms of institutional discrimination that alter the candidate pool before people even apply. Your blind process can be perfect but that's the last link in the chain. Quotas are a blunt tool to force us to compensate for unfairness that occurred before our process for which we are not responsible in order to build stronger teams.
As such I believe quotas are a good thing, because institutional bias and discrimination are bullshit and need to be eradicated. When they are, we can drop the quotas. If we come up with anything better than quotas, we can drop the quotas.
> As such I believe quotas are a good thing
You don't address the article's claim that quotas make stereotype threat worse.
This is a real, lived experience. In high school, I (an overrepresented minority) failed to get in to MIT, and my best friend (an underrepresented minority) got in, despite me looking better 'on paper'[0]. He chose not to go even though I pled with him that it would be a better life choice. Not going to MIT was disastrous because his college experience was bad, including an unjustified arrest, and dropping out. Don't worry, it all worked out - he's the CEO of a startup that's doing well now. I can't claim to know what was in his mind in high school but a part of me thinks there was real stereotype threat in play.
Second issue with quotas: In the case of Harvard with their overrepresented minority, they usually reject based on soft qualities like "no leadership potential". This really irks me, as my father in his government job was denied promotion several times because of that exact phrase, "no leadership potential". This is considerably ironic since for a brief stint he took a sabbatical and went full-time with his job in the navy, as an O-6 (captain, as in Picard, not America). During this year, the team he commanded completed the US Navy's first ever fully databased inventory system, on time and under budget, saving the navy billions of dollars. "no leadership potential" and phrases of its ilk are basically cover for discrimination, as they can be so easily issued even in the face of blatant evidence to the contrary.
[0] for example I had great scores in the AP sciences and a high-performing science fair project that placed at the international level - which are relevant skills for MIT.
But that's why we should do the best we can, to make it as fair as possible, and while there is much progress to be made around the world, there are also some clearly wrong ways to go about it and diversity quotas are about as worse as it gets.
Give people freedom and the opportunity to do what they want, and then the result will naturally be the fairest possible.
If we all started with the same options, this would be true. Our limits would only be reflections of our preferences and natural abilities. Like in an RPG where we pick wether to be a warrior or mage or whatever - you go down a skills path by choice based on what you want to achieve and what resources (xp etc) you have.
Unfortunately people are born into an active system wherein millions of events from before they were born are exerting pressures on their lives, pressures that we are mostly unaware of until it’s too late.
There’s no neutral “fair” option. This sucks. It’s bad. It’s stupid. People’s location or race or sex at birth should not bless them or condemn them. It should just be an interesting detail. But those things have real effects.
Ignoring this means accepting the current biases as somehow natural, and continues rewarding those already born with advantages. Fighting it means adding NEW biases, and reducing the advantages inherited by people with privilege. Which naturally seems unreasonable if you currently enjoy the benefits of those advantages.
Solutions like yours, which require no work be done to change things by the privileged, and ask the less privileged to reinterpret the status quo so that they realize things are fine as they are, should be suspect. They are not all definitely wrong, but they are especially attractive because of their convenience. And that pretty exterior can mask their lack of actual substance.
Don't focus on the outcome and try to engineer what you think is fair. That only leads to chaos because you can't possibly understand every individual's life experience.
Life will never be fair. That is a fundamental aspect of reality, so trying to arrange the future without perfect omniscience means you are doing nothing more than adding another roll of the dice. It may seem like you're equalizing things, but you're not.
Like I said, the focus should be on opportunity. That is vastly different than outcome, and it does mean that we should strive to make it so that anyone from any background has a chance to do what they want to do. That doesn't mean that some won't have to worker harder than others, again life isn't fair and that'll always be a part of it, but the chance is what matters. As long as people have a chance, then they have a chance to achieve their goals and dreams and actually find happiness. That's the only way it works.
Diversity quotas are by their very nature institutional discrimination. You can’t jusify it by saying the people they discriminate against deserve it.
Sure, they are institutional discrimination in the sense that institutions recognize the differences between gender/races and the differences in the treatment they receive, which is often based on negative prejudice (the other meaning of "discrimination" [1]). Discrimination is not negative per se, as long as it's the "neutral" meaning of the word (e.g., having different toilets is also discrimination).
[1] https://en.oxforddictionaries.com/definition/discrimination
You don't have to agree, of course. I wonder if you think that there is prejudice/discrimination that quotas are trying to fix (in a deeply flawed way, according to your view; in an imperfect but still better-than-nothing way according to mine), or if the whole starting point is wrong.
Sad truth is, it does not work this way atm, and fixing it is painfully hard for multiple reasons. Quotas try to do that in an imperfect way, sure. But I think it's good that someone is at least 1. acknowledging the problem 2. doing something practical to fix it.
So, concretely, how do you propose we fix prejudice and discrimination in this context? Because saying and knowing how it "should" be, does not magically make it happen...
[edit]: ..and unfortunately I don't think anonymous interview can be used everywhere.
Discussion about equality usually don't get to this point, but in hope that it might someday get started I will try to give an proposed answer.
First we need to expand the context, and I will use Swedish numbers for that. 90% of the employed population work in a gender segregated profession, same for both men and women. It is a known phenomena (sometimes called a paradox) that the more a society works towards equality the more segregated society gets. The best theory I have heard for it is that people will self segregate when given multiple similar valued choices, as going with the group one identify with carries with it a higher self confidence level in that choice on average.
So one obvious fix is to address that very aspect. Understanding that people will have on average a lower self confidence when they decide to go into a group where they are a minority. Make the interview process or trial period better suited for low confidence participants and it will benefit minority groups more, while at the same time not being discriminate to the majority. A mentor program is a prime example which has shown good results in equality programs, and I would bet that is because it addresses this.
Less practical on a company level but possible on a national one would be to address how and when individuals make career choices, and what the influences are. If we expect that a 15 year old to make a critical decision to decide which path to study and later find a job in, it should surprise no one that friends make for a major influence. Look at the school yard and its clear how self segregated kids become based on gender, race, sexual orientation, economic class and so on. The school system could be designed to improve self confidence and self-esteem, and in times where suicide has risen to the most common cause of death for men between the age of 15 to 60 it seems it would also be a major health benefit.
But you won’t convince me that institutional racism and sexism are justifiable, no matter how much you think some group deserves it.
I also find it hilarious that you have come to the defence of institutional racism and sexism on the basis of equal opportunity...
Quotas are inherently unfair and take opportunity away from others.
Isn't it better to treat people as individuals?
James Damore suggested something like that, it didn't go down well. (Though I completely agree).
> Isn't it better to treat people as individuals?
This is an equally impossible goal. The system is already unfair. You want to preserve the current unfairness. I think we should instead calibrate the unfairness to optimize for undoing the weirdness caused by humanity’s awful history of subjugating, robbing, and denying the rights of various kinds of people over many centuries. People born today experience various benefits and drawbacks related to conditions completely beyond their control.
If we have privilege this is the god damn least we can do.
> This is an equally impossible goal.
For being such impossible goal it seem strange then that 21th century society is basically built around that exact concept. Banks lend out money based on it. Courts make judgement based on it. Government gives out social support based on it. Insurance rates and medical care is all about individual assessment. Even education here in Sweden is by law now individualized and each teacher must teach to the ability of each individual student (a new law since a decade ago).
A few hundred years ago that was not the case. If a person were of the lower class then they were treated as such. Government was not expected to operate on anything except category of people, and the idea of treat people as individuals was seen as indeed impossible. Education in particular was extremely bound by this idea, and a person worth was categorized and put in predefined box from a very young age.
However impossible it is to treat people as individuals, it is better to do so and fail rather then to return to the old ways. If you get a wave of low economic immigrants should we throw up our hands and say “we’ve done our best” with individual assessment in the legal system and just prosecute based on class, or should we double up on it and try to do better? The political winds here is swinging, and I honestly hope that treating people as individuals wins over any attempt of short-term gain from treating people based on class. A person born under low economic class should not have their life hindered because others before them of same low economical class had a higher rate of crime that higher economical classes. That would be a failure of society to treat people as individuals. History does not dictate the future and it doesn't create any benefits and drawbacks unless we treat people based on category and class rather then individuals.
... these are all examples of places where bias is rampant. Banks, courts, government assistance, insurance premiums, and even medical care are not delivered to all individuals equally. Not at all. Not remotely.
Through activism and lawsuits that these areas have made strides towards fairness... but you seem to be holding them up as an example of treating people as individuals “done right” or “done successfully”.
To briefly take these one by one:
> Banks lend out money based on it.
- banks had to be forced to do this.
> Courts make judgement based on it.
- this is laughable. The legal system in the US punishes African Americans more strongly that Caucasians for the same crimes.
> Government gives out social support based on it.
- maybe?
> Insurance rates and medical care is all about individual assesment
- doctors provide more medication for pain at the same level to white patients vs black patients. Proxies for race are all over insurance. Women had to pay more for healthcare until very recent.
Without active efforts that were made over many years to tip the scales and correct these bad outcomes, all of this would be much much worse.
> History does not dictate the future and it doesn't create any benefits and drawbacks unless we treat people based on category and class rather then individuals.
I respectfully suggest that the above statement is completely incoherent with reality, and represents weird wishful thinking.
Where courts make predefined judgment and don't even bother to giver a person their day in court?
Where government make one size fit all decision based on gender, race, religion, or sexual orientation?
Where insurance companies refuse insurance based on race?
Where doctors use force sterilization, or outright refuse service based on race or economical class?
Where kids of low economical class is not allowed to go to school.
> Through activism and lawsuits that these areas have made strides towards fairness... but you seem to be holding them up as an example of treating people as individuals “done right” or “done successfully”.
Through equal respectfully suggestion, the above statement is completely incoherent with reality of today, and represents weird negative thinking. The improvement on those area are real, and suggesting that the old ways are better is sad.
> I respectfully suggest ...
If you are trying to insult someone don't bother to wrap it up like that. If you want a thread full of insults then just insult, and if you do not want that then rethink what you wrote and why you wrote it.
However your questions in this reply, and the idea that I think the "old ways" are better ... I mean that's not in my argument. That's creating a false binary and trying to contort my statements into something that you can readily dismiss. You are insisting that somehow continuing to improve things means returning to the old ways.
It also seems to be a common tactic used by people with power to justify the status quo. I'm ok with dropping it here since we're apparently talking right past each other.
Lets apply a quota system to those system above that today operate on individual assessment, using current imbalance that statistical evidence can detect.
Banks have data on who defaults on loans based on any number of conceivable categories, and likely incorporate some of that during an assessment. A quota system where they would grant or deny loans based on a single identity would not only be less fair because individuals can be part of multiple groups, but it would also institutionalize discrimination based on historical data. For the benefit of society it would be better if the system was more blind and relied less on the fact that people from low economic class defaults more on their loan that those from the higher economic classes.
Insurance companies is in the same boat. Having people of different categories pay higher rates make sense from a historical perspective, but it make for a worse society when women automatically pay higher medical insurance and men pay higher car insurance. Quotas where people are sorted into specific rates create a worse society.
Thankfully for both banking and insurance, individual assessment is more economical from a corporate perspective. A strict quota system which simply represented a single category of assumed risk is less effective at predicting a individual person. Individual employees of companies might still use discriminative quotas as shortcuts, but thankfully in most places that is illegal under anti-discrimination laws. There is no quota on how many African Americans can get a loan regardless of the rate of defaulting that that group historically has.
Some people have jested and suggested to have quotas in the legal system to correct existing bias. They say: Simply put more white people (and women) in jail! Obviously a quota system here would be terrible and the issue of racism and sexism in the legal system should be fixed through better individual assessments rather than quotas. Race and gender blindness would be a major improvement, while quotas would make society a much darker place.
See the theme here where blindness is the general cure to the ailment of bias and where quotas based on historical data would be terrible? Let see some examples where quotas are used.
There is an infamous school "lottery" in some places where kids from rich families have multiple choices (since they can sigh up to any number of good schools and move to places where they "win"), while kids from low economical class do not have that option and simply has to hope that they get one of the token slots from the good schools that their class get a quota from. Quotas has made matters worse, as been argued by teacher, parents and politicians alike. You don't fix the problem of good and bad schools through quotas, but rather by fixing the schools. The school should be blind to how much money a child parents has and simply provide the teaching that each individual child need.
Which lead us to the last area, medicine. Here quotas is currently used as a cost saving mechanism, where instead of individual assessment you have screenings and vaccines based on race, gender and age. It is a definitively worse system but since individual assessments are costly and doctors are limited in numbers it is still being used. The idea to give people medical care without medical assessment is at odds with modern practice of not doing harm, and the optimal solution is always to do individual assessments based on personal needs. Quotas is used as a last ditch effort when the benefits clearly outweighs the cost through careful and very conservative considerations. Quotas are dangerous and medical history is full of the harm they have done.
This is why I brought those examples. Each of them operate today on treating people as individuals and while neither system is perfect, adding quotas to them makes things worse rather then better. At best it can be used as a last ditch effort, but only if all other methods have been explored and one recognize the massive potential for harm that it introduce. I understand why it quotas feel attractive but to claim that it is impossible to treat people as individuals is practically giving up on the foundation of modern society.
“Blindness” is ideal once people get into the stages of the process where they are being evaluated on an individual level. It should be the goal.
Quotas are not targeted at those problems.
The important piece that a lot of us people seem to miss is that the ongoing effect of historical racism and sexism messes things up and creates unfairness before people even get in the room. Massive unfairness. Compounded in a thousand different ways over somebody’s life. Creating pressure away from one kind of life and towards another simply through historical inertia, because for most of history people like them did not have full rights as human beings.
Quotas seem bad at solving the problems you point out because they are not for that. They are an attempt to measure, and correct for, much bigger trends.
Well-designed quotas at the right places in processes could let the pool of people (who are then evaluated blindly) be a much more representative group. Quotas (or some other measurement) would let you know if some proxies for discrimination were still getting through the process.
Insurance companies my not use race, but a zip code can be a proxy for race, and the kind of neighborhood a person lives in can make them more likely to suffer a break-in or make a claim for some other reason.
I really do believe that it’s impossible to treat people as individuals with 100% certainty that that’s what you are doing. No matter how hard you try. Even in all the examples you gave, there is progress but things are not done.
Hopefully this makes my point more clear.
It is not clear the system will automatically balance itself, and if it does it will probably take multtiple generations.
That's not really fair is it? The company is overly diverse if anything. In a way they're stealing women from the workforce of other companies.
Diversity needs to happen from the ground up. Affirmative action is more effective and more appropriate the earlier in life that it is. Special coding classes and programs for underserved groups (e.g. Rails Girls) are the best, because a lot of girls don't get the same opportunities. In New Zealand, a lot of single-sex girls' schools don't teach programming at all. It's the same with schools in lower income areas, they often have very underdeveloped STEM programs.
Having lower entry requirements to university for less privileged populations also makes sense, although using ethnicity is a very blunt stick, as there are often external factors that mean that equally talented students have worse educational outcomes in school, such as poor nutrition, working a job after school to help pay the bills, and low quality teachers (most good teachers want to teach at good schools). There external factors tend to be ironed out at university.
Obviously not every type of affirmative action is appropriate for every situation. Requiring lower grades for women to get into engineering programs at university probably isn't appropriate, since it's not like women grow up in less privileged households than men.
Affirmative action in hiring makes a lot less sense than at earlier stages. There's only a certain percentage of women in the workforce, hiring over that percentage isn't going to help with diversity or inclusion. If only 20% of programmers are women, then your company's programming team should be somewhere around that number. If it's a lot higher or a lot lower than this, there's something wrong.
> That's not really fair is it? The company is overly diverse if anything. In a way they're stealing women from the workforce of other companies.
This was always a mystery to me as well. Unless you are expanding the workforce, it's a zero-sum game, so exhibiting a higher ratio merely means that someone else must have a lower ratio. Aiming for a higher ratio isn't empowering at all.
If a 50/50 ratio is to be meaningful, you have to fix the workforce ratio, first.
If there was a company that was made up of 90% men when the workforce was 50%, would you say that "plenty of men independently decided that that company was a great place to work in?"
I know its challenging to rank people in real world jobs compared to say scores on a test but humor me here. Say you score people on an objective metric from 100 to 0 and you have 5 times more applicants than jobs. One logical way to hire than is to hire the top 20% of the population.
If the overall population is 20% female than all things being equal 20% of the top 20% will also be female.
How then do you arrive at 50% female while still trying to maintain quality. You have to remove the lowest ranking male and replace him with a female from the applicant pool. All the woman above this man are logically all in the to hire pool already. The only applicants available are those ranked below him so you must be logic replace a more qualified man with a less qualified woman.
As you travel from 20% to 50% you must pick increasingly less qualified individuals to hire in order to reach your goals. Presumably your last hire would be from the 35% percentile instead of the 20th?
The real world is multidimensional, complicated, and much harder to rank but bias logically has to reduce overall fitness on the larger scale.
Increasing your percentage of women increases your candidate quality, not decreases it, because it forces you to address all the issues that make it challenging for you to attract/retain women in the first place.
Hint hiring women in marketing to balance the male software developers wouldn't prove much.
I was more commenting that in general trying to get 50% women on your tech team isn't going to help the industry as a whole.
The argument being that quotas actually don't help with diversity because, in the example of high end colleges like MIT, a huge chunk of quota kids end up dropping out, or changing to non-technical majors, or stay in the bottom 10th percentile of their class. They end up feeling terrible, their classmates (who on average will be 300 SAT points above them) end up not taking them seriously, so instead of this being a positive experience for everybody, it just creates more resentment on both sides.
I suppose many people believe that throwing unqualified kids into the meatgrinder is worth it just for the diversity exposure alone, and their sacrifice is for a greater cause.
This is not to say that exposing different clusters of society to each other isn't beneficial to overall social cohesion and acceptance. Doing it through quotas, however, doesn't seem to have been proven to be an effective way of accomplishing that goal.
Race is an awfully blunt way of looking at these societal clusters. Mashing all the white people together makes no sense. The Italians, the Polish, the Irish were discriminated against heavily early in the century. Russian-Americans make 30% more on average than French-Americans. West Indian blacks make 50% more on average than American blacks. Africans who move to the states for work are about as successful as Asians or Whites, despite supposed heavy racial discrimination. Race includes none of that nuance in it.
I agree quotas suck. A lot. When people explain why we need them, they fall back to "it's morally good" because the other choice is "monoculture" or "racism" which is "bad" or "evil" depending on who you ask. My problem with that is that I don't accept purely moral arguments, these are usually subjective and depend on someone's culture. Additionally they easily lead to tribalization where people form two camps over a problem and neither side is willing to compromise because their tribe is good and the other is evil/bad.
A diversity of thought seems intuitively to be more effective at solving "programmer didn't think of X because they're not in group Y" problems. Diversity by skin color or gender... not so much. Atleast, it doesn't seem to be a good solution for the above problem and when it is, only for a very very narrow range of problems. It obviously doesn't mean you shouldn't hire women or people with other skin colors or ethnicity. But also that you shouldn't hire them primarly.
Hiring 50/50 male/female workers doesn't solve your ignorance of deaf or blind users of your software. Or people who lost limbs or are otherwise inhibited in movement (just look at Microsoft, their new controller is an amazing statement for diversity).
A purely blind hiring process would solve some of this. Additionally the hiring process would also have to find people that challenge the status quo at the work place, atleast in the relevant matters.
Writing an app in React? Hire someone who is good at Angular and React but advocates Angular. You'll be hearing about problems in your App you haven't thought about. Make this person satisfied with your App.
Creating "Uber but for X"? Hire people who do X and who use X and don't want X disrupted by "Uber but for X". They'll tell you all about what's wrong with the implementation. Make them satisfied.
A diverse group of people is a group that when asked about a problem will find at least as many solutions as people in the group.
Ask your team "what would the ideal calculator look like" and see what they come up with. They can make it phyiscal or software, any arrangement or features they like. But each should have their own unique calculator.
I would say this is a good start point for further thought and discussion on this, though obviously I'm reducing the complexity of such a wide, subjective and deep rooted problem.
> I think it’s important to ask ourselves what we want to accomplish with diversity quotas in the first place.
Like the article says, maybe the company has found that hiring diverse candidates leads to more profits and they are seeking them out and letting the rest fall into place, i.e. they get higher salaries, better compensation packages... It would be no different than figuring out graduates from Stanford's such and such department end up doing particularly well in the company.
Or maybe the company does want to right a wrong and want to not just improve its bottom line but also help marginalized groups who have been neglected, abused and pushed out in the past. Companies the size of Google, Amazon, Facebook and others can certainly, and maybe should take that position. They can influence the technology field and their policies can affect a huge number of people. But I think they shouldn't be shy about declaring their mission, and should be more bold, as opposed to trying to hide behind the former reason. It is this hiding that leads to controversy and conflict.
Also in case of this article, the title could have been a bit less controversial too. It should have been something like "Diversity Quota Needs To be Explained".
Honestly, I believe it's the last reason. Levelling the playing field would be relatively easy: blind auditions (easy in some fields, like music, more difficult in ours — but doable). I'm really not convinced that diversity of thought has all the benefits we attribute to it. At the end of the day, I think it's fashion: we want to be seen as fashionable, and diversity is in fashion.
Regarding diversity of thought, her example is a great one: 'And look, if you put a gun to my head and asked me, given absolutely identical abilities to do the job, whether I should hire a woman who came from an affluent background, aced her SATs because of access to a stellar prep program and supportive parents, went to a top school and interned at a top tech company over a man who dropped out of high school and worked a bunch of odd-jobs and taught himself to code and had the grit to end up with the requisite skills… I’ll take the man.' I'd take the woman, not because she's a woman, but because she almost certainly has more of the requisite skills than the man in the situation. If they actually have identical relevant skills, then maybe he'd be more interesting, just because he's so unusual, but judging by appearances the woman is more likely to have the skills I'd need.
This is the problem we have today.
However, I do not understand your rationale for hiring the woman. By assumption, both the woman and the man have the same skills. It is the histories of how they got those skills that differ.
If I actually did know that they were exactly identical skill-wise, it'd be a different story.
Correcting prejudice by introducing more prejudice doesn't see like the best way to go about it.
> So, what about diversity of thought? If you’re really going after candidates who can bring fresh perspectives to the table, their lived experience should trump their gender and ethnicity (though of course, those can correlate heavily)
Diversity of thought is a dangerous term in this current culture of witch hunts and personal destruction: https://techcrunch.com/2017/10/13/apple-diversity-head-denis...
(As an aside, this article linked from yours https://www.forbes.com/sites/ruchikatulshyan/2015/01/30/raci... states a correlation between racial, ethnic and gender diversity and company performance. I wonder to what degree this is infuenced by the fact that many high-performing companies are tech companies which now fervently pursue diversity as a goal in itself?)
Jonathon Haidt's research shows there are trade-offs between mono-cultures and diverse cultures. Neither is "better" than the other. Each have different traits. As a quick example, diverse groups (on average) are more creative at problem solving than mono-cultures, while mono-cultures (on average) have higher "Moral Cultural Capital" and are thus better at group allegiance and out competing other groups when the tasks are known.
An example from the real world might be that the United States has a more diverse culture and has a lot of creative output in its economy, while China or Japan are mono-cultures and have less creativity but are out competing others when the task undertaken is well known (ie not requiring creative problem solving).
I can also understand the reasons for having civil rights laws to protect members of those groups, and I fully acknowledge the long history of discrimination in the US (I'm from the US) against Native Americans, blacks, women, non-heterosexuals, non-Protestants, and more.
But I can't wrap my head around what "diversity of thought" means, in any meaningful way.
Look at all the places with 'No-Assholes' hiring policy. Aren't "assholes" part of the diversity of thought?
I think it's pretty obvious that there may be good reasons to not hire people based on which thoughts they decide to express and how they express them. (I assume 'diversity of thought' refers to expressed thoughts - our mind reading devices aren't that good.)
The author doesn't address this topic, and instead proposes an evaluation based "entirely on performance in anonymous technical interviews."
How does that evaluation have anything to do with "diversity of thought"? If people from school X are 80% likely to pass the interview and people from elsewhere are 20% likely to pass the interview, then woudn't that process result in a monoculture of thought?
freedomben? Can you tell me how one might meaningfully characterize "diversity of thought"? For lack of a better definition, by "meaningful" I mean "incorporate it into hiring decisions such that the result is to the overall benefit of the company and the society which allows the company to exist".
For that to be true, you'd have to be unaware of the fact that people can process data and reason through complex problems differently.
> I assume 'diversity of thought' refers to expressed thoughts - our mind reading devices aren't that good.
That is a poor assumption that requires the redefinition of the word "thought". Our mind reading devices are pretty good: "Here is a complex problem that can be solved in more than one way - show your work."
> How does that evaluation have anything to do with "diversity of thought"?
By clearly demonstrating an applicant's competing priorities by way of solution selection? It sounds like you've never worked on a team that had that one guy who would regularly shoot down seemingly sensible textbook solutions, and instead offer a solution that either sounded like more work or just plain insanity. Have you never heard the story of Mel [0]? I can tell you from my own experience, as a former full stack developer who had to do a lot of interdepartmental work at a non-software company, diversity of thought is very useful.
[0] https://groups.google.com/forum/#!topic/net.jokes/k2JVKQzJSp...
This was in the context of hiring, and I gave my operational definition - how does one apply 'diversity of thought' to the hiring process?
> "Here is a complex problem that can be solved in more than one way - show your work."
Presenting a solution, or an attempt at a solution, is an expression of one's thoughts.
> "diversity of thought is very useful"
How do you apply it to the hiring process? How does the proposed method (skills testing) result in improved diversity of thought?
I'll be more concrete with two examples. Example 1: Suppose there are only three schools which teach COBOL programming, and you are a bank which is looking to hire junior COBOL programmers. Do you design your test for COBOL competency, which will result in preferential selection from those small number of schools and hence reduce the diversity of thought? Or do you hire for more general diversity of thought and expect more overhead to train people in COBOL?
Example 2. Suppose you put the applicant on an Ubuntu box and ask them to write a sort program in C++ which takes two command-line arguments. The first is an input file name containing set of lines (ending with a '\n'), the second is the output file name containing the input lines but sorted by byte value.
Of those that pass the test, 50 do the 'standard' vector<string> with a sort from the standard library, but only 10 do I/O failure checking, and 1 of the 50 decides to make the parent directories if the filename's directory doesn't already exist.
But wait, another 40 treat C++ as C, and use stdio along with a resizing array of char * . Of those, 30 use qsort, 8 implement quicksort, and 2 use bubble sort.
(Oh, and one person who didn't finish was 90% of the way through implementing a Timsort.)
That's a diversity of thought. Do you give preference to them for thinking about the problem in a different way?
But wait, another 7 think about it a bit, realize that 'sort' is on the Ubuntu box, and construct system("sort filename1 -o filename2").
That's surely thinking outside the box, so should be encouraged in the name of diversity of thought, yes?
Except another 2 think about it further, realize that the filenames may contain spaces or shell metacharacters, so construct the exec call directly. Security thinking is hard to train, so they should have preference, yes?
And finally, the last one writes a C++ program which uses the Python C API to pass in a Python program as a string.
Surely that's even more diverse thinking - and completely within the test protocol as given.
Now to reverse the direction. I know all of these solutions. If I were given the task and I know that the employee wants to hire for "diversity of thought". Which implementation should I do to improve my chances of being hired?
That is, should I do something which I know is less maintainable simply because I know it's more obscure and thus shows my diversity of thought?
All that matters is that at the end the calculator is the most ideal one for you as the candidate. All the features you wish for and none of the downsides.
So you get a bit of paper to draw on and maybe write some short text.
Compare the results to what your team has drawn when they were hired (or you asked them separately).
Of course the comparison is still subjectively but ideally you pick the candidate with the most different calculator from the others.
That's better than coding tests since you can look at the result and it's independent of coding skill.
Similar tests are possible where you don't test coding skill but instead test if and how much a person thinks outside the box and what it looks like. This would better help to assemble diverse teams than a simple coding tests.
Another one would be to look for people that are contra to the current team. If you develop COBOL look for a developer that can do COBOL but hates it and would love to switch elsewhere. These kinds of people will make your team diverse since they can find problems more efficiently.
I think the problem there is that your diversity tests still only test for coding skills, which aren't as important when you consider that someone with a fresh and contra perspective to your current team could be much more helpful.
The real goal - meaning, the one which decides if I will be hired - is to design the model which is most different from others.
Since I want the job, shouldn't I disregard the instructions and instead create crazy looking calculators which aren't ideal for me but which are different enough to stand out, and thus get me hired?
As a test designer, aren't you making a bias toward hiring people who are good at gaming the system? Is that really good for the company?
Going back to my examples, which you did not address, would your team hire the "C++ programmer" who implemented the sorting challenge by using the Python/C API to write a Python program? Because that's certainly thinking outside the box. But it doesn't tell you if the programmer would be a good C++ hire.
> If you develop COBOL look for a developer that can do COBOL but hates it and would love to switch elsewhere.
What? If you want to hire a COBOL developer with the expectation of working for 20+ years at a bank, why would you want that developer to hate that job? Do you have any evidence that they will be "more efficient" at it? Why wouldn't they do the bare minimum need to do a satisfactory job, while looking for a new job on the side. And when that happens you'll need to retrain your new employee.
It seems like money down the drain.
That is actually a helpful reminder - the context of the conversation is a blog entry that struggles over the objectives of quotas and the utility of using race and sex as a proxy for diversity quality beyond... race and sex.
> Presenting a solution, or an attempt at a solution, is an expression of one's thoughts.
Yes... are you trying to defend your prior conflation of behavior and thought?
> Example 1 ... do you hire for more general diversity of thought and expect more overhead to train people in COBOL?
That depends entirely upon your organization's capabilities and priorities. Hypothetically lets say that the bank's long term objectives don't include a migration from COBOL, the IT department doesn't have a long history of successful inhouse training, and being a bank - is generally risk averse. First, filter for candidates that demonstrate an acceptable level of competency in COBOL. Second, filter for candidates that have skills and interests that are not organic to your team - but could feasibly be useful (in your mind, that is all we've got). Third, ask the candidates for examples of times that they've come up with novel solutions to difficult problems. Here is a personal example: I once accidentally landed a contract when I was having lunch with a friend and his boss, I was later told the clincher was my long exposition on fault tree analysis in ballistic missiles. The contract involved the integration of time management and security systems.
You seem to be struggling with the prioritization of diversity of thought over skills that are actually need to perform a job. Here is a hint: the first order of business is getting somebody who you imagine can do the job (skills, job history, etc). If you have more options that positions, of those people, select the one who demonstrates the ability to reason in a way unique to your team.
> Example 2.
You hire the guy who points out that you did a poor job of framing the problem. Not only did you describe it in a way that could be interpreted to demand an implementation that spits in the face of POSIX utility conventions (newlines as argument delimiters), but you also failed to establish a success metric (time, maintainability, performance, etc).
> Surely that's even more diverse thinking - and completely within the test protocol as given.
You described a bunch of potential implementations, not different ways of thinking. The protocol you gave wouldn't be useful to measuring diversity of thought, unless you modified it to include the possibility for interviewer-interviewee interaction, where you might get some clues about their thought process. "What is the success metric?", "Are the sorted values bounded?", "Is the source untrusted?", "How does this fit into the larger process flow?"
> That is, should I do something which I know is less maintainable simply because I know it's more obscure and thus shows my diversity of thought?
No, even in the cartoon funhouse of an example you provided - they might already have a Python weirdo running amuck, you'd add nothing. You don't know that ahead of time.
Where do I do that?
You wrote: "You seem to be struggling with the prioritization of diversity of thought over skills that are actually need to perform a job."
First, I am struggling over how to identify 'diversity of thought' during the interview process. Is it something different than "can come up with innovative solutions" or "out of the box thinking" or "creative problem solver"?
Eg, freedomben writes that 'diversity of thought' can include "different socio-economic status ... different political views, different learning styles"
While your focus is only on "the ability to reason in a way unique to your team".
These seem like two different interpretations of that phrase, and I lean towards ubernostrum's down-voted comment that "diversity of thought" seems often used as a euphemism for "put up with assholes" (my interpretation).
You wrote: "You hire the guy who points out that you did a poor job of framing the problem."
Which means you end up biased towards rules lawyers. Which may be what you want, but bear in mind that you are presenting one performance goal while you have withheld a secret goal that you are actually looking for.
Here's a less secret goal: the test is meant to see if you know what modern C++ is like (and not simply the C-ish subset), and if you have a good idea of what the POSIX mindset is like (so you don't end up asking pointless rules-lawyer questions).
Which of these possible secret goals should the interviewee try to optimize?
You commented "unless you modified it to include the possibility for interviewer-interviewee interaction".
Well, yes. But no matter what scenario you come up with, the essential problem remains - does "diversity of thought" differ from "highly competent and creative problem solver"?
You wrote "newlines as argument delimiters". In my description, the first argument is a filename. The contents of the file are a set of lines, terminated by a \n. The goal is to sort those lines by byte value. Otherwise the desire to sort the contents makes no sense.
"I assume 'diversity of thought' refers to expressed thoughts..."
> First, I am struggling over how to identify 'diversity of thought' during the interview process.
I think we've covered this pretty well, so your struggle isn't over the "how" - but the "why".
> Is it something different than "can come up with innovative solutions" or "out of the box thinking" or "creative problem solver"?
It is no different, with one exception: it is measured in relation to your existing organization. If all your programmers are proponents of the functional programming paradigm, hiring another Haskell programmer, while relatively novel to the rest of the industry - likely does little to increase your organizations diversity in thinking (without additional screening parameters).
> These seem like two different interpretations of that phrase...
They are: one is selecting for proxies, presumably as a shortcut. The other is directly addressing what is desired. I'm always amazed at how proponents for such selection mechanisms are totally oblivious to how ridiculously prejudice it is.
> ...I lean towards ubernostrum's down-voted comment that "diversity of thought" seems often used as a euphemism for "put up with assholes" (my interpretation).
Clearly.
> Which means you end up biased towards rules lawyers. Which may be what you want, but bear in mind that you are presenting one performance goal while you have withheld a secret goal that you are actually looking for.
Boom, point proven. You have at that point learned something about that candidate's way of thinking, select on it or don't. If that was a hidden goal then it worked, if it wasn't then disregard.
> Here's a less secret goal: the test is meant to see if you know what modern C++ is like...
Not a diversity of thought test.
> ...and if you have a good idea of what the POSIX mindset is like (so you don't end up asking pointless rules-lawyer questions).
That is a diversity of thought test. Is the candidate willing to, in the face of ambiguity, insert his own opinion instead of speaking up?
> Which of these possible secret goals should the interviewee try to optimize?
As I said earlier: the one that, in your experience, indicates an ability to do the task that the candidate is hired for. Should there be a tie, the one that arrived at a solution that you did not anticipate. And no, that doesn't mean you should hire a guy who insists on sorting files using Node.js, just because he was the only one... you can't determine rationale based only on language selection. If he does it in shell script, ask why. You may learn that binary compatibility concerns are higher on his priority list due to some past experience that you wouldn't have considered.
> ...the essential problem remains - does "diversity of thought" differ from "highly competent and creative problem solver"?
Yes, but first I'll point out that "highly competent" is an unrelated concept. Because effectively nobody has unbounded useful creativity over an entire problem domain, you want to select individuals who's constraints overlap as little as possible - thereby covering more of the problem domain. Personal example: I took over a database from the engineering department because the guy maintaining it retired. I was horrified when I looked inside, the architect was obviously a PLC programmer - using triggers and views to form a hellish logic ladder. I was trying to figure out how I'd be able to untangle everything into a more tradition normalized database when I got a request from one of the engineers to insert a new trigger to account for some upcoming process change, I told him how much I didn't want to do that and asked if I could spend a day with him in order to get a better grasp on their problem domain - he was annoyed, but agreed. Well it turns out the old engineer wasn't totally insane, the problem domain was pretty much unbounded and constantly changing: new metrics, new datatypes, new requirements - totally normal for their department. Whereas I would have used a Domain-key normal form in that situation, he just created a new table and added a trigger. Both styles work, each has different weaknesses and strengths. Neither of us would have arrived at the other's solution. Diversity of thought.
> In my description, the first argument is a filename. The contents of the file are a set of lines, terminated by a \n.
Lol, even your attempt at clarification adds confusion. The ambiguity this time: is the set of lines terminated by \n, or is each line in the set terminated by \n, or is the file terminated by \n? Yes, I'm pretty confident that I know what you mean - but I have seen 0x1f used for stuff like this in production, bad things would have happened had I just made an assumption.
> Otherwise the desire to sort the contents makes no sense.
Sure it does: if instead of interpreting the explicit mention of '\n' to mean line termination within the file, one interpreted it as the termination of the first parameter. Again, one can guess the intent due to convention - but it is ambiguously phrased and immediately led to two competing possibilities of intent in my mind. Maybe that is because I think differently from you...
The only time I screamed at my management was the day he was laid off. The organization never recovered.
When interviewing candidates, I am constantly afraid that I'm selecting for clones of myself.
You make some good philosophical points (that "assholes" are part of the diversity of thought), but practically I don't think "assholism" is what you want to look for.
When I say "diversity of thought" I think of things much like the author of the piece mentions: different socio-economic status (if you've never been dirt poor, you think differently than someone who has been well off. both viewpoints can be good), different political views, different learning styles, etc.
Political diversity is also good I think. I think progressives bring good stuff to the table, as do conservatives, libertarians, and socialists. When one view predominates, the organization tends to suffer (not a universal truth, just an observation I've made (so clearly not scientific)).
Your desire for a "meaningful" definition however that one could "incorporate it into hiring decisions such that the result is to the overall benefit of the company and the society which allows the company to exist" is really hard however (at least for me). I'm not currently part of hiring, but I have been in the past (tho my involvement is typically limited to technical questions, and I rarely delve into anything other than technical stuff). I like to look for areas where we are strong and weak. For example, if we have mostly people that love UI design and front end work, a candidate that feels strong on the backend may be more desirable. I think it's good to remember that diversity of thought is insanely broad, and encompasses tons of areas (which seem to me to be the point you are trying to make). As such, coming up with an example that generally demonstrates it is likely not possible.
I feel like I just rambled and didn't answer very well. Sorry about that. I think it's time for bed ;-)
I can understand the issues regarding so-called 'diversity hiring' - it's due to the generations of systemic racism, sexism, homophobia, and other bigoted practices.
I think some turn it into a "diversity for the sake of diversity" argument, but I think that's because we (meaning Americans) often do that to avoid facing up to our history, and the problems of how to right those imbalances.
That's why it's so easy for me to reject a "diversity of thought" argument - it's not quantifiable or actionable.
I also think it's a misnomer. No one cares about "thoughts", only the expression of those thoughts.
That word meaning "appearances" suddenly appeared and became all the rage a year ago. Writers and commentators think they're hip and cool using it, rather than what it actually is - camp-following and boring.
Blech. Just use "appearances".
Some dumb-ass dudes may tell you: "oh you're here because of diversity quotas, not for your inherent quality", and it will be annoying for you. But the purpose of the quotas is that the next generation of students can look at the distribution and tell themselves: "oh, actually this field is not only populated by bros, there is a lot of women in it, maybe I'll give that a shot".
Ok, it's not meritocracy, but come on, meritocracy is only unbiased if the culture is unbiased...
That cuts both ways. Many people will not hire someone who "doesn't look the part" because they think that means they can't do the job.
It's not about this generation; it's the next.
If this was really a problem, then every field that was slightly skewed would only trend towards the extremes but that hasn't happened. This constant need to balance groups without any understanding that people are different just comes across as nonsense.
[1]] https://medium.com/@CodeKingdoms/why-did-women-stop-coding-1...
It looks like advertisers in the 80's had a demographic they were marketing too, they didn't create one. That article start with the assumption that everyone studying a field is interested in it as opposed to wanting a career and goes downhill from there.
Did you read the article? Aline claims that a completely blind application process lands more diversity in tech.
Fair enough. Besides the obvious "diversification" of the company's populace, what other success factors can be directly attributed to quotas?