81 comments

[ 0.22 ms ] story [ 169 ms ] thread
No. This is mathematical nonsense.

Let's say there's a fitness function for each candidate in a candidate pool.

If you filter candidates by any criteria, the expectation of fitness function will be less or equal to expectation before filtering.

Humans are not numbers. If your workforce has a craptastic gender ratio you will hammer that in because potential new members of the minority will be less interested in joining the company.
There are plenty of numbers associated with humans in workforce. For example, their salary.

> If your workforce has a craptastic gender ratio you will hammer that in because potential new members of the minority will be less interested in joining the company.

I don't see any complaints about "diversity" in female-dominated or minority-dominated professions.

I don't see any complaints about "diversity" in female-dominated or minority-dominated professions.

Then you haven't looked very hard. There are even programs with financial incentives, e.g. http://www.independentnurse.co.uk/news/new-bursary-introduce...

That story seems to come up a lot in these discussions. IIRC it appeared in a prior HN discussion of the same thing.

The bursary in question is only £3,000 which is nothing. And the total fund is only £30,000 - so enough for 10 nurses.

This doesn't even begin to compare with organised political lobbying on a continental scale to force women into jobs where people earn £3,000 for simply sitting in a single board meeting.

This doesn't even begin to compare with organised political lobbying on a continental scale to force women into jobs where people earn £3,000 for simply sitting in a single board meeting.

Of course, because no such jobs exist from which men are mostly absent. You can't expect a response of the same size when the problem is nowhere near the same size.

You're changing the subject. The original poster's point was that there isn't any real advocacy towards the benefit of men - in other words, that "diversity" doesn't mean diversity, it means benefits for women.

You responded with an article about a sum of money so trivial it would cover 10 men for perhaps a couple of months depending on where they live.

If that's the best you can find, this doesn't really disprove his point.

No, parent said "any", not real; you introduced that term. Which is the kind of ambiguous language that impedes debate.

If you're saying the efforts are not of the same absolute magnitude, then I would agree; my point is that they are similar relatively to the size of each problem.

If you insist that diversity efforts are only "real" if the same effort is applied to solving two problems which are of clearly different orders of magnitude, then I can only say that I find your definition of "real" perplexing.

I wouldn't define gender disparities as a problem to begin with, so arguments of the form "but the problem is big so it's OK if the actions are big" have no weight with me.

What remains is that the word "diversity" doesn't mean diversity. It means focusing on getting women into positions of wealth and power - which you appear to be agreeing with me about, but for you it's an "ends justify the means" situation whereas for me and other posters here it isn't.

If diversity advocates cared about diversity they'd attack all gender disparities with equal vigour because it'd be the disparity itself they care about. But obviously they don't: they only go after the ones where there are big benefits of money and power to be had. Gender disparities that favour women or that don't yield large streams of money and power are ignored.

However they don't admit that. This is what rankles. Instead of talking about money and power they claim they merely want "gender diversity", which is false.

If diversity advocates cared about diversity they'd attack all gender disparities with equal vigour because it'd be the disparity itself they care about

Yeah, to me this makes no sense. It's like saying that if you care about infant hungriness, you have to help with equal vigor both those starving in war-ridden areas and those in wealthy neighborhoods whose parents occasionally forget to give lunch money. Or that if you care about forest fires, you give the same number of fire engines to California and to Delaware.

In my opinion, if you care about a problem, you dedicate more resources to where its effects are more pronounced.

Please give a suggested way to measure the effect in a objective way.

With hunger this is easy. Measure the number of children sick of malnutrition or that dies of hunger per population and square miles. With forest fires you measure the number of forest fire per square miles and affected people per square miles.

So what should we use in order to measure gender disparities. What about gender segregation, ie the professions with the highest rate of gender segregation get also the highest number of resources dedicated to it. We could measure by effectiveness, looking at the historical rate that gender segregation goes down per number of resources in gender equality. We could look at gender disparities in the form of Maslow's Hierarchy of Needs, addressing each layers gender disparities.

The parent comments implies that money and positional power is the metric being exclusively used here to measure anything.

Picking a natural effect that is very widespread, defining the "badness" of it as being measured by money and power, and then only caring about the instances where there is money and power to be gained quite obviously opens up people to charges of hypocrisy - there is no way to tell the difference between such diversity advocates and people who don't care about diversity at all.

Look at it this way. I make the charge that so-called "diversity advocates" don't care about diversity, that it's not a matter of prioritisation - gender is just a smokescreen. I posit they could achieve 100% female boards but then they'd stop. They wouldn't go on to try and tackle the dominance of women in teaching because they never cared about diversity to begin with: just money and power.

What evidence or arguments do you have that could convince bystanders I'm wrong? There are none. The behaviour we observe can be explained in both a complex way (yours) and a simple way (mine), so Occam's Razor applies.

I live in Austria and I know two fields were men voice their concerns: nursing and kindergarden teachers / daycare.
Female-dominated professions generally doesn't suffer from lack of diversity at the top. There are less male nurses, but plenty of male doctors. There are less male teachers, but plenty of male professors. Etc. That is specifically what quotas for company boards deal with long term. Because a de facto career filter for gender is equal to the filer example you mentioned.
A professor is not above a teacher. A professor doesn't tell a second grade teacher what to do. They have almost nothing in common.
(comment deleted)
But it help reduce needless lawsuit, which cost lots of time money and mindpower
What is the fitness function currently in use to select board members?
With respect, I think what you just wrote is statistical nonsense. If your ultimate selection from a pool is heavily biased towards an irrelevant fact (gender) then countering that bias can lead to a better selection.
A better solution would be to do gender-blind hiring, not gender quotas.
I was only responding to the suggestion that quotas are "mathematical nonsence". I don't know if they are the best solution or not.
This would also account for transgender and intersex individuals too.

Why shouldn't the right person be hired just because they have both sets of genitalia? How does that fit in to the quota system?

That's true. But consider that in some majors such as software engineering there are 5 times as many men as women. If you try to hire women, your pool will be so much smaller you'll almost certainly choose worse candidates.
One problem with this reasoning is that it assumes one can assign a fitness function to each single employee in isolation, independent of their different roles and how they relate to each other.
How are you supposed to choose a candidate without filtering by any criteria?
The statement is true for all possible criteria.
The counter argument is that in practice, the true fitness function is only very roughly approximated by the selection heuristics used. And if - consciously or otherwise - $ismale happens to be weighted positively in the selection heuristic (more so than any positive effect maleness might actually have on their performance in the job), a quota which artificially inflates the number of females might actually go some way towards redressing that mismatch.

Obviously this is considerably more likely to actually be the case in some businesses and industries than others.

What very confident nonsense!

Individual fitness is not independent, let alone group fitness. And without independence you are just plain wrong.

Sociology dressed up as a priori mathematics is rarely credible. Even worse if the math simply isn't correct.

> Individual fitness is not independent, let alone group fitness. And without independence you are just plain wrong.

Who said that fitness in my description cannot include group fitness? The same argument goes for adding a worker to the workforce. Does the worker fit the existing workforce?

> Who said that fitness in my description cannot include group fitness?

I did. Because doing so contradicts your conclusion. Though I also said it is not necessary. Lack of independence alone makes you wrong.

> The same argument goes for adding a worker to the workforce. Does the worker fit the existing workforce?

Which same argument? The argument you made was about filtering lowering fitness.

> I did. Because doing so contradicts your conclusion.

Going for the conclusion instead of the argument. You can't demonstrate where my reasoning is wrong, but saying that the conclusion is wrong. Congratulations, you're an ideologue.

> You can't demonstrate where my reasoning is wrong

I specifically said why your argument was wrong. Without independence it is simply untrue.

> Congratulations, you're an ideologue.

I am. Mathematical flummery triggers me.

That's meaningless.

If it were true then it would apply to selection by any attribute e.g. experience or qualification. The only way you'd get around that would be to include the attribute in the fitness function. You're then back to deciding what effect gender has on the fitness function i.e is gender selection good for the business?

Of course that assumes you can select the best candidate from a pool. Feminists argue that there is bias that prevents this. However studies on the gender pay gap reveal that there's very little bias and most differences can be explained by education, experience, job choice, hours worked, etc. More recently psychological differences between men and women have been shown to contribute to the pay gap.[1]

My opinion is that affirmative action is naive and that society is better off to strengthen meritocratic hiring. Removing names from CVs may help women slightly but there's no reason to believe that men and women would achieve the same outcomes given equal opportunity.

[1] https://academic.oup.com/oep/advance-article/doi/10.1093/oep...

Why assume technical skill in one area is the only variable in determining merit?
I didn't do that. Merit is meant broadly to encompass many things.
Humans aren’t cogs in a machine. Our value to an organization is not absolute and based solely on our technical proficiency in a single area. An individuals value is relative to the team.

Adding another programmer to my company would be less valuable then adding a salesperson. Similarly, adding people with diverse perspectives is more valuable then hiring only people that grew up in the same city, at our income level, with same gender, ethnicity, age, etc.

An assumption I suspect you’re making is that we already have THE correct fitness function.

They aren’t just trying to filter the fitness function, they’re trying to improve it.

> Humans aren’t cogs in a machine.

Humans aren't gender either. Doing gender quotas reduces candidates to their gender.

> An assumption I suspect you’re making is that we already have THE correct fitness function.

> They aren’t just trying to filter the fitness function, they’re trying to improve it.

Ohh, no. My assumption is that current hiring is not perfect, but gender quotas are waaay worse. Gender-blind hiring would be better than both.

"This is mathematical nonsense."

Board members might be the least likely of all people related to the company to have the right credentials. It's networks, power, relationships. Not usually talent.

Though I'm totally against board quotas, it's very easy to see how women are locked out of this game. Far, far more so than for normal roles within an organization.

Not even close. Networks and relationships can be important.

It's the friends of low level managers who get hired who sometimes have the least credentials.

Paywall? I see no paywall.
Some people do, evidently.

There's a limit on articles per month, at least in some regions, and maybe other restrictions in different regions.

Huh. I'm using a VPN exit in the US, to avoid GDPR bullshit.
I’ve always said to hire the people who can do what we need as a business, regardless of their gender
Which is a noble sentiment but simply doesn't happen in the business world (or in many other places). So the hard question is, how do you effectively encourage business to choose the best people regardless of gender?

Some have tried blind interviews but these are very hard to do properly and effectively. A 40% quota is another way to push business into looking beyond gender.

Then a minority-only company should be both cheaper and more productive, by supply and demand.
I think there are three potential issues with this, firstly, minorities are being discriminated against and paid less in the scenario so it wouldn't be desirable on that metric. Secondly competition in many markets isn't robust enough to have an even playing field if you wanted to start a new company with all "minorities". Third, if the company was all "minorities" you would be drastically reducing the hiring pool so even with good competition it would be hard to hire well (which could wipe out any minor efficiency gains).
> Which is a noble sentiment but simply doesn't happen in the business world (or in many other places).

On the contrary, all the evidence I've seen points to discrimination being a very small part of the gender pay gap.

The problem is that men and women are different.[1] So it becomes very difficult to even measure if women are being discriminated against much or are just different to men.

[1] https://academic.oup.com/oep/advance-article/doi/10.1093/oep...

(comment deleted)
(comment deleted)
That's not the goal of gender quotas. (With this I don't mean that I support them.)
I remember 2nd term math lecture. There were students from different engineering disciplines. There were 600 students with ~15 females among them.

Some years later situation is no different: ~100 people on the floor, 5 females here: 2 secretaries, 2 export specialists and senior researcher with PhD degree.

Where find suitable candidates for these quotes when women avoid technical stuff at university level?

Seems the answer to this more and more is to discourage men and encourage women. I wonder personally if men and women just have different interests. I worry that we're perhaps pushing men and women into jobs they're unhappy with than if just left to their own devices?
I read a study last year, that at least in the US actual STEM graduates are more and more balanced between male and female with females having better degrees. When they looked at the jobs the graduates got it was not at all representing of the quota that graduated. E.g. 30% of graduates were female but only 3% of accepted job applicants (nmumbers are made up, just to demonstrate the point of the study).
There was a similar government study done on teacher graduates here in Sweden and it got the same results, through genders were reversed.

The conclusion they had is that by drop rates and gender segregation get amplified in every step between first year student to long term employee.

Before the comments go too far down a rabbit-hole, I encourage everyone to respond to the article instead of answering the question in the headline. The article is talking about boards, not employees, and states that there is no causation proven, despite some correlations found in Europe after they required quotas on company boards.
I was surprised and encouraged to see that the Economist actually cited the papers they were discussing - this seems new, and progress. Way too often, news outlets refer to "studies" or "a study" without making it easy to find out what studies were actually cited.

I'll queue up the studies to read later, but I am skeptical. I've looked at studies that claimed quotas for women in boards = more profit before and without fail, they weren't controlling for confounding variables. Usually it was as simple as "companies in Norway do better than companies in Saudi Arabia, look at the gender differences!" which is of course totally invalid. Unfortunately the way governments have forced women into boards via legislative fiat means it's very hard to control for the pre-existing economic state of countries: you'd need to compare two economies that were basically the same except for gender differences, or compare the same country before and after and then try to control for all other changes in the world economy, which is very hard.

The article is quite balanced about the limits of these studies though. Which is good.

Shouldn't we ask why this is only about boards and not employees, and why state and governments that issue laws like this don't lead by example and mandate gender diversity on themselves.

Why is the scope so narrow defined as to exclude any criticism on how state and government operate as the largest group of organizations with paid employees. Seems very convenient from the position of power that government and states enjoy.

I've never worked in a company that hires based on quotas, but I have found organically diverse teams seem to make better products. Having people who are able to see that a product doesn't work for them means features don't get added/dropped as easily. I've never made software as accessible as the code I wrote with a disabled person on the team, or as friendly as when 50% of the team were women.

I suspect the reason is that building software that works for users is about asking the right questions, and more diverse groups ask more diverse questions. That doesn't mean an all-male or all-white team can't ask the right questions but in my limited experience it seems they don't.

If you only discover that a product doesn't work for some of your users because one of those users is on the development team, then you probably need a better way to get user feedback.

You can hire developers of diverse gender, ethnicity and age, but those are most likely still very similar along other dimensions like income, location and ability to deal with complex systems.

Rather than trying to hire more diverse developers, you might be better off paying someone to find out what your actual users think about your product.

(comment deleted)
This is an implementation of pure form of gender type discrimination.

Employee should be of a neutral color, creed or gender.

I love the idea that there is a scarcity of "golden skirts" that make this too hard. It presupposes that the current male members are all "golden trousers", the cream of the crop, the best of the best.

I've met many many board members over the years and they are often very unimpressive. I'm sure we can find some equally unimpressive women to replace them! Better still anyone with even a modicum of talent would do!

I have a different take;

If we suppose a meritocratic hiring process is the best way forward, and that girls are equal to boys (and only considering those genders (if we surmise that there are more)):

Then, if we take the stance that there has been historic bias against girls then it leads to them not having the experience to be in the same category as boys.

It's not about "all guys being good", it's that historically girls have not necessarily had the same opportunity to gain experience and learn. (Except for the most dedicated or those who went to shops that did not practice any kind of discrimination). This leads to a small pot of experienced girls but a large pot of experienced guys.

If you hire "the most qualified candidate" you'll end up with mostly guys because guys have had the opportunity to work unfettered, and if you have a quota to hire more women and want someone experienced, then you're competing against many in a small pool of experienced women.

--

EDIT: I should add that I haven't read the article; but the above is a topic I think about often in the context of discrimination and its role in the future. I don't believe in hiring quotas, I believe the problem should be fixed at its root- But this is not the current topic- and I would kindly ask those who downvote me to argue their points if they wish me to be more understanding.

There are much better options than quotas, but that doesn't make them inherently bad.

The question is are they an improvement over the current system or not?

However they should be at most a stop gap solution, a truly diverse system would make it impossible to know the race and gender of your workers, so implicit basis would be impossible. Or maybe we could train a neural net to have no implicit bias.

"I've met many many board members over the years and they are often very unimpressive."

This is an extremely difficult thing to assess frankly.

There are so many reasons for board members.

Often they could be from a family fund. Maybe they are from an institutional investor who own a lot of share. They are not there to provide strategic guidance, rather just to sit there and keep an eye on things. They may be simply there as an 'extra head' for another party, like the CEO who wants to keep the power. And sometimes they're there because they have deep networks that you might not know about.

Instead of forcing women on boards, they should try to convince companies that having female board members is a good thing, maybe provide some data, and then create some networks for hiring.

"maybe we should just tell the people that benefit from the sexism that the sexism is bad, surely they will stop"
Well, your project, PouchDB has 20 'contributors' listed on Github, it would seem all but one are male. Give or take maybe one.

So until your own project isn't blatantly gender biased, maybe you should refrain from waiving the 'sexism' card to broadly. Because I'll gather your probably not trying to 'oppress women' over there, rather, there are any number of other more nuanced and complex reasons for the absence of female contributors.

There are many reasons boards are mostly male, and most of it is systematic. They're no more trying to 'keep women off boards' than you're trying to stop women from making contributions to open source. Approaching boards with information, and especially access to women's business networks where boards can find qualified women has actually worked in other areas (Finland) and it's a considerably more responsible approach. (Although in Finland there is mandating as well).

FYI PouchDB is amazing, by the way.

The reasons that boards (and OSS projects) systematically exclude women ... is what sexism is.
It's important to differentiate between overt sexism (i.e. 'women can't or shouldn't be in business') to the kinds of negative externalities that result from a gendered world.

We have gender (despite what a fringe group of academics might believe) and 99.9% of the world is fine that that.

This creates gendered outcomes in various fields which in some cases might be, or perceived to be exclusionary. The difference is more than nuanced to the pointe wherein I think it's wrong to merely point at any gendered situation and say 'sexism!'.

I'll bet $100 that OSS projects are not actually very exclusionary at all, and more actually representative of the fact that women for whatever reason, don't want to do that kind of stuff to the extent men do. At least today. Of course, one could argue than the ranks of female engineers are thin due to lack of role models, sexism in early years and that would be true, but even without this ... there would still be a mismatch.

Screaming sexism at OSS teams or boards is counterproductive.

Imagine if the government mandated that 25% of all OSS contributions of 'major projects' had to be made by women? I don't think that would make sense, nor would it be well received.

But - I'll bet that the managers of major open source initiatives might be open to some facts and figures, right? Or open to networks that have more women? If the 'women's coding club of London' indicated they had a bunch of young ladies who wanted to learn the ropes of OSS would you help out? Probably, right?

And shareholders are not some pasty group of old white men either. They'll listen and change will happen, though fairly slowly admittedly.

    I'll bet $100 that OSS projects are not actually 
    very exclusionary at all, and more actually 
    representative of the fact that women for whatever 
    reason, don't want to do that kind of stuff 
    to the extent men do.
* http://theeverygirl.com/why-women-are-leaving-stem-and-how-w... * https://thenewstack.io/women-open-source-still-fighting-good...

If you could donate that $100 to a relevant charity (maybe https://www.womenwhocode.com/)

And read up why the "negative externalities that result from a gendered world" is called "sexism" and why its not counterproductive to able to identify sexism in order to work towards solving it, it is actually pretty critical.

Is there an idea where to go with these quotas in the long run? I mean as I understand it, the numbers are just pulled out of thin air.

So we have a company, following the however implemented rules and has, let's say has 60% of group A and 40% of B on their board.

But maybe the ideal quota is 70% / 30%. Or 10% / 90%. Or the standard deviation is so high that there is a high probability that everything from 20 % / 80% to 70% / 30% was achieved without any bias but just because of differntly distributed skillsets. Point is, by setting a quota like that without any research, you're gonna end up with one group having it statistically harder to get their well deserved board place as the other, which is the problem that we're trying to fight in the first place, right? And now it is locked in stone, because abandoning the quota would have big political implications.

I understand that it is an attractive short term fix and that we might not have good other solutions, but where are we going with it?

No, the "problem" they're trying to solve is that women want more power and money. Have you noticed how pushes for "diversity" always ignore jobs that are dangerous and unskilled? Brickies are 95%+ male, you don't read about the importance of setting quotas for female bricklayers do you? Primary age teaching is essentially a female only job at this point, we don't see governments imposing quotas for male teachers.

Board seats and other senior management roles however get enormous pressure from feminist campaign groups - because that's where the money and power is.

This is truly unfortunate, I can understand giving incentives to bring more women into upper management, but a quota is somewhat punitive in nature.

The main objective of the article it seems is to allay fears that more women in boardrooms would not be a bad thing. I honestly don't care for such studies. What I do care about is the fact that more and more businesses are going to be told _how_ to hire. This is a terrible idea.

Furthermore, the article goes on to say that there have been no changes in the profits reported and that the companies across Europe have seen no change whatsoever. This might say much more about boards and their lack of effect more than anything else.

What I see happening is that due to the increased regulation on how businesses operate, more of them will not go public and thus prevent a lot of the population will miss out on good investment opportunities.

Last time I looked, “gender” had been redefined to include many more than just two genders according to various universities. Multiple countries are likely to legislate in extra genders. So mandating one logically requires mandating all of them eventually. Just mandating sex and racial attributes alone will cause significant board changes.

Interesting times ahead. The market will have fun sorting out the profitability of these changes in due time.

No it’s not good for business. And it’s not bad, either.

I don’t think anything measurable will happen.

Men and women are different in many qualities, but doing business in modern relatively civilized society is not one of them. Both genders do just fine.