It's unfortunate that Alphabet declined to present their data, that way we'd likely be able to correct for things like location and seniority along with having solid deviation rates.
It wasn't accurate at all, because you don't know who flagged the story. Moreover you're breaking the HN guidelines not only by calling names but by insinuating astroturfing/shillage without evidence. Plus you've done this before. Please stop.
Yes, it would be nice to see the raw data, but note that they've bucketed Level 1 through Level 6 employees, and have still found gender discrepancies.
Are you suggesting that different jobs might occupy the same level but have widely differing compensation? That might be true - I don't know Google well enough - but that's not how salary bands usually work.
It's exactly how salary bands work - they're designed to be opaque.
If you have women who largely choose to work in Foo with a compensations which puts them at the top of the band 1 and at the bottom of band 2, and men who work in Bar that starts with compensation at the bottom of band 1, but a promotion gets you to the middle of band 2, then you'll get that in band 1, women earn more, but in band 2, men earn more. This is why you must know the actual distributions (which includes weighting and deviation/variance), and the job role.
Indeed, in level 2, seems like women earn more - maybe? might not even be significant though.
In any case, I'll bet job role correlates far more strongly with compensation than gender.
We would need a lot more data, I work for a much smaller company vs Google yet we can easily have people who while formally at a lower level can make beyond VP level comp. because of their unique tech. skills and experience. I imagine Google has a fairly large group of famous eng. etc. who could scew results greatly should they get to be part of the sample.
Why is it that stupid memos are leaked while the raw data to this prefabricated conclusion isn't? If properly anonymized I don't see harm done to anybody.
One of the things this misses is that even if these numbers were even, systemic gender bias means that things would likely be unfair.
Even if Google were perfectly fair internally, women coming in from academia are likely to have faced gender bias [1], meaning less impressive resumes and therefore lower starting positions. Similar logic applies to women coming in from industry: one component to hiring pay and title is previous pay and title; external bias can easily be imported.
But it would be a miracle if Google were perfectly fair internally given our society's long history of sexism, documented gender bias in academia (from which Google draws some of its culture and a lot of its people), and the ongoing issues in the tech industry. I'd expect Google's promotion ladder to reflect that to one extent or another, meaning that the levels themselves have some level of bias to them.
> Even if Google were perfectly fair internally, women coming in from academia are likely to have faced gender bias [1], meaning less impressive resumes and therefore lower starting positions.
To play devil's advocate, if the performance review process was good/perfect and skill = compensation, factors like "less impressive resumes" should even out pretty quickly, right?
I doubt it. I once had an employee who had accepted an entry-level engineering position when they really were qualified for two levels above that. I was explicitly told they couldn't skip levels, and that I could only put them up for promotion so often.
But even at a place that didn't have slow-down-there bias built in to the promotion process, the way you prove you're qualified for the next level is by doing work that's above your current level. Nobody will give staff engineer work to somebody in an entry level position; there's too much jockeying for that work by people one level down. So there's a practical limit to the rate at which somebody can advance even with no bias.
Another thing to bear in mind: this is self-reported data. Even if the population numbers were perfectly fair, there are known differences in the way men and women report things in surveys (example[1]). Personally, I conjecture that this bias and biases in response rates that affect men and women differently account for some of the reported differences in salary.
I'm more concerned about stuff that wouldn't show up in this data. For example, assuming a woman who is equally qualified as a man, I expect that the woman would be at a lower level on average, because e.g. of studies like those cited in [2]. I suspect that Google is better about this than most places, but I have heard too many stories from my woman coworkers to believe that things are perfectly fair. But I suspect this article makes things out to be perhaps slightly worse than they are, at least on the compensation front.
You over generalize you citation [1]. it's not reasonable to assume the way we report our sexual history is the same as the way we report anything else.
I am not "disproving" the scientific study. One study isn't proof of anything; it's just one entry in decades of academic dialog. I am pointing people to more of that dialog.
And no, I don't have to do better than that. It's not my job to argue some random anonymous dude into understanding sexism. The people who want to understand the topic will do the work of contextualizing that one study.
Those who never really wanted to understand, like you and the anonymous dude who tried using that study as a trump card, will go on with your opinions no matter what anybody tells you.
You, sir/madam, seem to enjoy making illogical claims. First you link worthless bragging in attempt to prove that a peer-reviewed scientific study is "deeply flawed". Now you proceed to claim that I "never really wanted to understand sexism" because I question these sources and that I "will go with my opinions no matter what".
Ironically enough, based on the above evidence, the last statement likely applies to you.
Ok, Mr. Evidence. Prove me wrong. List the books on sexism that you've read, with brief summaries. Show me your college transcript with the courses you took.
I'll add that I have evidence for my assumption that dudes are biased when evaluating material about sexism. Aside from my 20 years of experience talking with dudes in tech about tech's gender bias, there's also a study showing this effect: https://www.wired.com/story/why-men-dont-believe-the-data-on...
I don't need to be an expert on sexism to make judgments on it. This is one benefit we get from science: it allows one to make accurate judgments without becoming an expert. If one had to be an expert to make any judgment at all, one would never be able to leave their house.
Of course, some disclaimers apply. Any conclusions you draw based on a study have to stay faithful to its constraints. And you have to keep in mind that some fields are more trustworthy than others. E.g., so-called 'social sciences' seem to be as much overrun by extreme-left fanatics as the 'progressive' mass media, which is evident in ignoring obvious control factors (e.g. taking into account marital status and number of children reverses results on sex bias [1]) and in publication bias [2].
Also, it's you who needs to prove something wrong, since it's you who doubts a scientific study. And please stop linking news articles, I'm no longer entertained to learn more ways in which the mass media spins the Damore story.
Ah, right. You are so brilliant that you don't need to study anything. You can just know that most of a field is bunk because you perceive the practitioners as "extreme-left fanatics". Fake science!
You also get to ignore any article about scientific research (even if it is written by a scientist) because of those darned politically biased journalists. Fake news!
You have constructed a nice, comfortable political bubble for yourself. One you're only willing to talk about under cover of anonymity. This could be because you are the true lone genius you see yourself as. Or it could be because you are picking and choosing anything that you find politically inconvenient. Perhaps, as a genius, you can guess which one I'm thinking here.
It's funny how you try to spin this around and claim that it's me who lives in an echo chamber, not the extreme-leftist mass media and SJW 'science'.
So it's okay from your point of view that major news outlets, such as Bloomberg TV, switch into propaganda mode and start making anti-factual claims, such as 'there is no research to support the idea that there are biological causes for the lack of women in tech' [1]. It might be too much to expect commercial media to be objective, but outright fabricating facts is nothing less than betrayal of everything journalism stands for.
Is that a singular case of leftist propaganda? Far from it. If you took a look at Google News the week this story was hot, you'd see your run-of-the-mill articles that open by twisting facts in the title. E.g., 'Anti-diversity manifesto' was apparently considered a mild way to describe it [2]. Some of them proceeded to make false claims, e.g. that Damore was "criticizing the company’s efforts to increase diversity" (no he was not, he considers diversity a good thing and suggests better ways to increase it) or arguing that Google "should stop offering programs to minorities and women" (he was only discussing racist and sexist programs, not _all_ programs) [3].
As to 'fake news' - I take it you're trying to paint me as a Trump supporter. So let me tell you I'm not one, and I didn't even know what that was all about (as I never bothered with US mainstream media before) before it hit me where it hurts with the Damore story. Now I don't know how much pro-Trump media twisted facts, maybe it was the same. All I know is I was shocked to find the mainstream US media to be so ideologically charged as to attack a guy for his [very mild and well-substantiated] opinions with forged facts and labels. Because I know that guy could have been me.
In summary, you, together with the mainstream US culture, need to pop your own obvious, well-documented and plainly visible ideological bubble before you start making claims that people with true liberal, pro-science, pluralistic views like myself are in one.
It feels like these things become self-fulfilling prophecies, because of the way this issue is reported on. If "seven of nine technology companies, including Apple, Amazon.com and Microsoft, [disclosed] that data", would it be too much for an article to compare the unofficial Google numbers to that data?
I'm not a big fan of large companies, I don't think they can win. Even if you have good numbers, that doesn't fit the narrative of "Silicon Valley has established itself as the boys’ club". Say for example, we applauded companies who are doing better, maybe even close to statistical significance, then you could say, "Hey, it's still not quite there yet, but it's close and we're doing better than others. Come work for us and help us make it right." And then that would be actionable advice, and companies would have a great reason to strive for this. But we'd also have to be honest about statistical error rates, and maybe some other stuff.
I'll close on this, I've heard people say stupid and offensive shit when they're drunk, but never "Hey, let's pay women less", although I'm not an executive. So I feel like it isn't a common thought... unless you keep reading articles like this one.
I've hired plenty of people, and been in three times as many discussions of pay with everyone from the janitor to the CEO and I've never heard someone insinuate that we could pay less because the person is female, but I've heard people say "...and we get the bonus of having her be female to make our numbers look better..." before.
Here's what I think a big driver is (you're going to have to walk down premise lane with me as I don't want to spend the time on citations ATM): Presuming that women are in general more agreeable than man in the workplace and that they don't fight as hard and as often for raises, I would expect an otherwise equal male counterpart that is more aggressive to have higher pay. Why? Because the squeaky wheel gets the grease.
Every year when we did promotions we'd start out with a zero-sum raise pool. Let's say if everyone got exactly the same, they'd get 2%. But if I give Bob a raise of 4% I have to go 'take' that extra 2% from people or everyone. If I'm worried about Bob leaving, or he's asked for a raise and plead a good case, he's going to get it. If Alice or Charlie have been silent all year but have been doing an otherwise good job, they're going to get what's leftover- in general.
There have been a number of cases where we idtenified someone silently kick major ass and threw money at them, but that I think it is because we had very conscientious managers that worked closely with their subordinates, unlike BigCo.
I'm not even going to touch on all the other reasons that people have laid out as possible reasons why women tend to earn less. I think if you've got two people working well, doing the exact same job and producing the exact same outputs, they should be paid nearly equally, but we're humans and life is messy. Some people haggle better than others, some had a very low salary before that the new job used to baseline, etc.
If you really want to maximize your salary at an given job, argue for every last penny coming in. All future raises (for the most part) will be based off that number and it will haunt you for your entire tenure at the company- most likely.
I think greasing the squeaky wheel is a bad executive decision. It sets precedent, and rewards political behavior above actual work output. If you have to ask for a raise, you either don't deserve it or management is broken.
You can ask for a raise in two ways (in my book) that are indicitave of someone who is in fact a high performer, or at least savvy.
1. You approach a manager with expanded responsibilities that would justify more pay. Or a detailed account of your undocumented responsibilities that justify more pay. Don't come at me asking for more money for the same job if you're not already (or plan on) doing more work than we agreed to.
I went from Architect to Director this way.
2. You have another offer from another company and give your old company a chance to match. This could have ramifications that everyone goes around and around about. The general rule of thumb is not to do this, but if you believe in your current company, what you're doing them is informing them you're below market for YOU, and they've got an opportunity to fix that.
I did this at one company and it worked fine, but I left a year later anyway. There's some stats that say a large portion of people who do this are gone within a year.
> If Alice or Charlie have been silent all year but have been doing an otherwise good job, they're going to get what's leftover- in general.
...and Alice and Charlie just find another job in a few years because their raises haven't been keeping up with market rates. It's not necessarily the direct managers' fault unless corporate gives:
* estimates of market rates
* recommendations that an average employee gets an X% raise to track with market rate (or even stay ahead of it)
* a big enough pile of money to compensate everyone in line with market rates
* even more money to reward high achievers (and probably squeaky wheels, too)
This seems like a tall order to most budget-makers, so instead wise employees change jobs every so often to make sure their salaries keep up with market rates.
I would expect someone who is less predisposed to negotiate for a higher salary to also switch jobs less frequently, even if they could get higher pay elsewhere.
If you like your colleagues and the work you do, leaving the company for a raise seems more aggressive than first telling your manager that you are unhappy with your pay.
In addition to that, the person being less aggressive about asking for raises in their current job is going to start salary negotiations at their next workplace from a lower number. In aggregate, this is a major source of income inequality.
Every year when we did promotions we'd start out with a
zero-sum raise pool.
Well, there's your problem.
If I'm an employee and I do some great work that brings the company an extra $X per annum, and I ask for a raise of $0.1X the company is welcome to take that out of the $X I made them.
If the company decides to instead take the $0.1X from someplace else, that's their decision not mine - claiming raises are zero-sum is just bullshit to make people feel guilty about asking for a raise.
If my efforts have made my employer an extra $X per annum, the financial reality is the company has an extra $X per annum.
We can debate how that $X should be divided between workers, executives, creditors, shareholders and reinvestment. But there's no "financial reality" that means workers' share has to be zero.
Every organisation has rules and budgets and procedures, but they're made by people and if they don't work right they can be fixed by people. It's not like the CEO received the budget from God on stone tablets on a mountain 2000 years ago.
If I wanted to work in some calcified bureaucracy where rules were more important than outcomes and worker performance was irrelevant, I wouldn't be working in the private sector!
Ultimately money is finite - no matter how much you have or how it's allocated. What if there literally isn't any more money left to be given by the time they're done giving raises?
You're missing their point entirely. If my efforts made the company an additional $10,000 per year above and beyond what they would have made, they can offer me a raise of up to $10,000 without affecting any one else's chances at bonuses or raises.
Not if your efforts were wiped out by a project that went sideways and lost 20K on another side of the company.
It's also highly unlikely that your singular effort created 10K more profit. That 10K was likely the culmination of a number of people who had varying levels of participation in the effort.
Unless you're working at a company that tracks your hours to every single possible project you're working on, there's often a huge blind spot as to where exactly your effort is going as it pertains to revenue. Unless your in professional services or some other type of billable position, then I actually agree with your notion to a degree...
With regards to people in billable positions, I once workthink d for a company that did something I really thought was neat. At the end of every quarter they'd announce a profit sharing rate based off of whatever it was that defined the rate. Since we were all billable this rate was at $X per billable hour. So if you billed more, you participated in a greater share of the revenue. I frequently saw $10-20K quarterly bonuses based entirely off my effort. It was quite motivating.
Because the profit sharing was the same rate for everyone and took in to account the entire company's financials, you were assured that your billable efforts were going to be rewarded on the same scale as the next person.
And if you make a mistake that costs the company $10k is it also okay to withhold $10,000 from your pay that period too? You seem to be tying reward _directly_ to profit, while ignoring the risk involved. Usually the higher the risk, the higher the opportunity for reward - but if it fails, the business needs to absorb the risky loss and insulate the workers as much as possible. This means that when there IS a big profit, part of that profit needs to be kept for future losses too in order to smooth things out so they don't have to do layoffs if they have a bad month.
If you fly too close to the sun and your weekly pay directly relates to the value you put in that week you'd might have job security but you'd have absolutely no income security!
Just from my limited experience, it's possible for people to be working on multi-year projects that may not bring in much or any money but which have the strong potential to grow. On the flip side, it's possible for an employee to claim he's 'bringing in' $X while he's simply maintaining work that was developed by someone else. Trying to to draw a line from revenue to your work is something you like to do if it is favorable and not if it's not (I've been in both positions). OTOH raises are about retaining people and the relative differences between who gets a raise is usually about who's more important to the future of the group and / or who is harder to retain / replace.
> Presuming that women are in general more agreeable than man in the workplace and that they don't fight as hard and as often for raises, I would expect an otherwise equal male counterpart that is more aggressive to have higher pay. Why? Because the squeaky wheel gets the grease.
If that were true, wouldn't the salaries start off the same and then diverge? Doesn't seem to quite match up with these unofficial numbers.
Besides, at most $MegaCorp, isn't compensation reviewed periodically, and some kind of peer review is included? Which means that agreeableness is less of a factor, which means the process is working.
I recall this programming teacher I knew with a civil engineer education. While looking for his dream job, he sent in applications for other work but demanded way above average pay while knowing that most would reject it. In the end one place did accept it which was a teacher position with the same pay as coworkers who had been in the place for 10 years.
If there existed statistics for such strategy, I wonder if it would have a gender attribute to it.
> If that were true, wouldn't the salaries start off the same and then diverge?
No, because starting salary is something that is negotiable, too. And you need to push back on the offer and ask for more if you want to make more. If you're not the type to push back, you start lower.
I think it's also quite likely that women are discouraged from making cases for a raise because they have a harder time convincing male managers of their competence. I've seen many men who simply ignore a woman's plea about a problem, but then a man will reiterate it and then it's suddenly an issue. Imagine trying to get a raise from someone who ignores you and only listens to the men on your team.
I'd ask what constitutes "same level", and what's the overtime gap between genders, if there is one, but why do that when we can have a good ol' laugh at how Google made their own bed, and now we can watch them use the same "it's more complicated than that" arguments used by people who Google think are "propagating sexist stereotypes".
I keep wondering what is the economical explanation for the existence of the gender pay gap?
Assuming women and men employees are (generally speaking) equally skilled yet differently compensated (for whatever reason), rational employers would actually prefer to hire women over men. Just like with any other purchase, all things being equal, cheaper is better.
Perhaps "old" companies with a "tradition" might have a status-quo bias, but new startups/smart CEOs that really try to minimize their burn rate would obviously see this.
Over time more and more women employees would be sought after more then men, which would eventually lead to an equilibrium and remove the pay gap.
Because it's more complicated that. Hiring decisions, career progression decisions, salary decisions are not made by an algorithm trying to balance the sheets. They're made by humans with biases.
Because the "pay gap" comes from the fact that men and women in aggregate make different choices with respect to how they develop their human capital. When you compare groups of men and women who make the same choices with regard to their careers and what to study, the gap disappears.
In the case about 20% of college graduates who studied Computer Science are female, and about the same proportion are seen employed as Software Engineers in the industry. Those women are not any cheaper than their male counterparts, because they're just as qualified.
Why women and men make different choices in aggregate, as well as if any of that is problematic, are topics of further debate that I don't think we have concrete answers to yet.
A possible explanation is that men, in the aggregate, are more motivated by compensation and willing to put up with more hardship or danger in exchange for more money.
There are a lot of jobs that pay a lot of money because they're dangerous, like working on an oil platform. Suppose those people get paid $90,000 whereas below average tech workers get paid $70,000 and above average tech workers get paid $100,000+. If you're a below average male tech worker who is in it for the money, maybe you take the oil job instead for the extra $20,000/year. The equivalent women take the $70,000/year non-dangerous job, causing the male average for "tech jobs" to be higher than the female average.
It's also possible that more men than women are willing to work longer hours or jobs with less flexible hours in exchange for a higher salary. Employers rationally pay people willing to do that more than people who aren't.
> A possible explanation is that men, in the aggregate, are...
Ironically, thats the type of talk that can get you fired from Google. We have to be able to address the reasons if we want a solution, even if it hurts your sensibilities.
That to me is the answer to this question: So why hasn't this happened (yet)?
Companies actually do irrational things all the time.
Have you seen how many people on here report they have to change jobs to get a raise? Or how often companies will spend $200 worth of employees' time on bureaucracy before replacing a $100 broken chair?
Humans are mostly irrational. Massively so. Rationality isn't the default state, it's not to be expected, it's a rare and precious jewel.
So many nerds dismiss the idea of systematic bias because "rational people wouldn't do that." This is true. It's also true that leprechauns wouldn't do that. Both statements are equally tied to reality.
Except that the "systematic" bias only seems to be happening in some fields, and ironically in a field that kind of relies on rationality, or at least logic.
Obviously, it's because everybody in that field is either a frat boy or evil, and there can be no other explanation /s
> We are all biased. The only differences are a matter of degree, and whether or not we acknowledge it.
That should even out then, right? Or is bias somehow not normally distributed?
I'm happy to call BS on "bias" and "irrational" explanations - sounds like something you make up when you want to dismiss something and can't be arsed to come up with a proper explanation.
There are preferences for prettiness/beauty and especially tallness. They are biologically routed, well documented, and have good predictive power. I guess you could call that a bias. From looking at high competency fields such as doctors or lawyers, there doesn't seem to be a gender bias, neither biologically, nor culturally.
So which systematic bias are you proposing that would employers to act so irrationally, but yet somehow opt for the more expensive labour cost. Even if e.g. a CEO is racist, he's not going to lose sleep over offshoring jobs. Is gender different, if so why/why not?
On the flip side, if it is a systematic bias like tallness, do we have an obligation to combat it? Height doesn't get nearly the same treatment. Either way, the "bias"/"irrationality" argument is lazy and incomplete.
Remember when black people were terribly oppressed in the South, so big companies rationally took advantage of this to hire labor for cheap and out-compete others, thereby ending racial discrimination through the power of economic rationality?
Do you seriously suggest that bias should be evenly distributed? For every person biased against women in some field, there's another person equally biased against men?
After all of these discussions about this topic on HN, I'm starting to wonder if HN actually has users posting from a parallel universe. Because some of the things being described don't match the one I live in at all.
> Remember when black peoples were terribly oppressed in the South
Actually I don't remember, I wasn't alive or in the US for that. But from my understanding, calling that "bias" is a bit of an understatement.
> For every person biased against women in some field
What bias and which people? A general anti-women bias in society? In all men? White men? Only in tech? Only men in tech against women in tech? Any specifics at all for your hypothesis?
Fact is, there are fewer women than men doing CS degrees, about 20% AFAIK. That seems broadly to carry over to the ratio of genders in tech roles. Women seem to earn less than men. We don't really know why either is the case, but people are working hard to "correct" that, even though we have no idea what correct looks like. Some people think 50% would be good, because that's roughly the ratio in the general population. Looking at other majors and fields, 50% sometimes happens, sometimes doesn't.
Maybe there are better ways to approach this than to claim everybody is prejudice - maybe that's true, but is it constructive? Why is the ratio in tech disproportionate? Why is the ratio in veterinarians disproportionate? Are these equally important issues? Are they getting the same amount of coverage?
At the moment, it seems that all the nuance we've got is a hammer, and publications like the New York Times are sure reporting on a lot of nails.
Bonus question: Is Big Bang Theory good or bad for the tech industry, and do you think it will increase, decrease, or have no effect on gender ratios in tech?
You ought to remember, because it still happens today. Take a resume and put a typically black name at the top, and your chances of getting an interview plummet compared to a typically white name.
There's a lot of reporting on nails because there are a ton of nails. Gender bias has been a thing literally since before recorded history. There are blatant examples quite recently, too. Until the 1970s, it was routine for banks to refuse to issue credit cards to women without their husbands' permission. Am I to believe that his problem, which is literally thousands of years old (probably more) has been solved in the past 40 years?
I have no idea about Big Bang Theory, never saw it.
It seems that the most likely explanation for the pay gap is the gap in dedication and energy spent at work. Until recently, many studies had found a bias against hiring women women in education and tech. But once corrected for lifestyle/marital status (i.e. single men without children compared against single women without children, etc), we observe the opposite picture. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4418903/pdf/pna...
I don't understand why Google gets so much heat when they seem to be one of the better companies in the world in regards to diversity.
I've been at several companies where literally all managers are men and women no more than objects that are hired based on their looks. Executive meetings are spent discussing female coworkers attributes and who has managed to get laid with who.
That doesn't sound like the kind of behavior anyone would want to encourage.
I will say (anecdotally) that the women that I hired as individual technical contributors and promoted to managers were (on average -- with a small sample size) better at adapting to their new responsibilities than the men.
I didn't hear much of the kind of talk you are talking about -- but I discouraged employee fraternization of that type -- and maybe I was wrong to do so. (This was a while back - maybe norms have changed?)
how does that apply for salaried positions google doesn't pay by the hour outwith janitorial and similar blue collar staff - who are probably out sourced
What Google pays men versus women is at the heart of a dispute with the Department of Labor. The company is fighting over how much data it needs to hand over
What are they hiding? They must know what that data will reveal is the only explanation.
For a good (if scary) insight into the "hidden bias" and "forced diversity" culture inside Google, have a listen to the recent James Damore interview at Joe Rogan's podcast:
They recount the ways people have been misrepresenting James' (rather meek) position, right down to accusations of sexism, nazism, malicious removal of academic references and outright lies. A sad testimony of this day and age and the current media climate in the US.
It's like people think covering their ears shouting "LALALA NOT TRUE, WILL NOT DISCUSS" has ever been a constructive way of addressing social issues.
I wish Google would come clean and release the raw stats. I understand that's one politically loaded and sensitive dataset, but nothing beats drawing your own conclusions. Right now, Google is getting flak from all sides, lose-lose.
Heck, if I recall, Google hasn't even disclosed what exact part of their own code of conduct Mr. Damore was in violation of. Kind of important, don't we think? I suspect these things are purposefully written to be so vague that virtually anything can be interpreted as being in violation. Their reluctance to disclose what specifically he violated (they wouldn't even him when they fired him) suggests they know they would be laughed at.
He was fired for insinuating women are genetically inferior to men.
The fact he couched it in pseudo-science, doesn't make that any less true. That's harmful to Google's image, and no woman would want to work with him after that, so he's toxic. They have every right to fire such a person.
I wonder whether the effort in question was scientific. In particular, did it control for life-style (e.g., single without children, married with children)? Sociological results on sex-related biases have been reversed when properly controlled for this parameter [1], which warrants skepticism towards any study that doesn't give it a serious consideration.
It doesn't just matter that men are paid more than women, it also matters why. The article is more nuanced than the title on this point, but it still doesn't come off as very nuanced. But then none of the discussion around this topic ever does.
> The self-reported Google salary spreadsheet [...]
That's all I needed to see. Nothing worthwhile here. The people who'll tend to report will be the people who are worried about how much they earn. I'd say all this shows is that men are less worried about how much they earn than women.
Did they also look at the reasons why men are paid more?
Because "men are paid more than women" alone simply says NOTHING.
In the company I used to work there were HUGE salary differences between men with the same job title, and the difference sometimes could run up hundreds of euros. The main reason was that the better paid men negotiated their terms of employment WAY better than those that got paid less.
I suspect to see something very much alike here. Also, I would like to see the salary distribution within men and women, because I can almost certainly guarantee that some women are paid more than some men, but I don't think we will ever see that because that would destroy the entire "all men are paid more than women for no reason other than they are men" narrative some people are still trying to push, while it is related to a LOT of other things other than just pure gender.
BTW, if a salary difference is SUCH a big problem why stick with the companies who do this? Why not find employment somewhere else, or start your own company with better working conditions? It isn't like you are FORCED to work for Google or any other of the big corps in Silicon Valley if you don't like how they treat people there.
89 comments
[ 2.7 ms ] story [ 133 ms ] threadhttps://news.ycombinator.com/newsguidelines.html
> At Google, Employee-Led Effort Finds Data Scientists Are Paid More Than Web Designers
> At Google, Employee-Led Effort Finds AI Researchers Are Paid More Than Project Managers
I could do this all day...
Any chance they'll release the data that backs up this statistical claim?
It was probably a pretty half baked, self-reported Google form so the data may not be worth much.
Are you suggesting that different jobs might occupy the same level but have widely differing compensation? That might be true - I don't know Google well enough - but that's not how salary bands usually work.
If you have women who largely choose to work in Foo with a compensations which puts them at the top of the band 1 and at the bottom of band 2, and men who work in Bar that starts with compensation at the bottom of band 1, but a promotion gets you to the middle of band 2, then you'll get that in band 1, women earn more, but in band 2, men earn more. This is why you must know the actual distributions (which includes weighting and deviation/variance), and the job role.
Indeed, in level 2, seems like women earn more - maybe? might not even be significant though.
In any case, I'll bet job role correlates far more strongly with compensation than gender.
I highly doubt it.
Even if Google were perfectly fair internally, women coming in from academia are likely to have faced gender bias [1], meaning less impressive resumes and therefore lower starting positions. Similar logic applies to women coming in from industry: one component to hiring pay and title is previous pay and title; external bias can easily be imported.
But it would be a miracle if Google were perfectly fair internally given our society's long history of sexism, documented gender bias in academia (from which Google draws some of its culture and a lot of its people), and the ongoing issues in the tech industry. I'd expect Google's promotion ladder to reflect that to one extent or another, meaning that the levels themselves have some level of bias to them.
[1] e.g., http://www.pnas.org/content/109/41/16474.abstract and http://www.apa.org/pubs/journals/releases/apl-0000022.pdf
To play devil's advocate, if the performance review process was good/perfect and skill = compensation, factors like "less impressive resumes" should even out pretty quickly, right?
But even at a place that didn't have slow-down-there bias built in to the promotion process, the way you prove you're qualified for the next level is by doing work that's above your current level. Nobody will give staff engineer work to somebody in an entry level position; there's too much jockeying for that work by people one level down. So there's a practical limit to the rate at which somebody can advance even with no bias.
I'm more concerned about stuff that wouldn't show up in this data. For example, assuming a woman who is equally qualified as a man, I expect that the woman would be at a lower level on average, because e.g. of studies like those cited in [2]. I suspect that Google is better about this than most places, but I have heard too many stories from my woman coworkers to believe that things are perfectly fair. But I suspect this article makes things out to be perhaps slightly worse than they are, at least on the compensation front.
[1]: https://www.pri.org/stories/2013-05-30/women-underestimate-s... [2]: https://hbr.org/2013/08/why-do-so-many-incompetent-men
> women coming in from academia are likely to have faced gender bias
Did you mean bias in their favor ?
"Women have substantial advantage in STEM faculty hiring, except when competing against more-accomplished men" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4611984/
"Women preferred 2:1 over men for STEM faculty positions" http://news.cornell.edu/stories/2015/04/women-preferred-21-o...
"Study of accreditation exams reveals biases actually favor women in STEM positions" https://phys.org/news/2016-07-accreditation-exams-reveals-bi...
https://othersociologist.com/2015/04/16/myth-about-women-in-...
http://www.chronicle.com/article/The-Myth-That-Academic-Scie...
https://feministphilosophers.wordpress.com/2015/04/15/michae...
The study about French would-be teachers seems of very low relevance to me to software engineers.
And no, I don't have to do better than that. It's not my job to argue some random anonymous dude into understanding sexism. The people who want to understand the topic will do the work of contextualizing that one study.
Those who never really wanted to understand, like you and the anonymous dude who tried using that study as a trump card, will go on with your opinions no matter what anybody tells you.
Ironically enough, based on the above evidence, the last statement likely applies to you.
I'll add that I have evidence for my assumption that dudes are biased when evaluating material about sexism. Aside from my 20 years of experience talking with dudes in tech about tech's gender bias, there's also a study showing this effect: https://www.wired.com/story/why-men-dont-believe-the-data-on...
So let's see your data.
Of course, some disclaimers apply. Any conclusions you draw based on a study have to stay faithful to its constraints. And you have to keep in mind that some fields are more trustworthy than others. E.g., so-called 'social sciences' seem to be as much overrun by extreme-left fanatics as the 'progressive' mass media, which is evident in ignoring obvious control factors (e.g. taking into account marital status and number of children reverses results on sex bias [1]) and in publication bias [2].
Also, it's you who needs to prove something wrong, since it's you who doubts a scientific study. And please stop linking news articles, I'm no longer entertained to learn more ways in which the mass media spins the Damore story.
[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4418903/pdf/pna...
[2] http://www.sciencedirect.com.sci-hub.cc/science/article/pii/...
You also get to ignore any article about scientific research (even if it is written by a scientist) because of those darned politically biased journalists. Fake news!
You have constructed a nice, comfortable political bubble for yourself. One you're only willing to talk about under cover of anonymity. This could be because you are the true lone genius you see yourself as. Or it could be because you are picking and choosing anything that you find politically inconvenient. Perhaps, as a genius, you can guess which one I'm thinking here.
So it's okay from your point of view that major news outlets, such as Bloomberg TV, switch into propaganda mode and start making anti-factual claims, such as 'there is no research to support the idea that there are biological causes for the lack of women in tech' [1]. It might be too much to expect commercial media to be objective, but outright fabricating facts is nothing less than betrayal of everything journalism stands for.
Is that a singular case of leftist propaganda? Far from it. If you took a look at Google News the week this story was hot, you'd see your run-of-the-mill articles that open by twisting facts in the title. E.g., 'Anti-diversity manifesto' was apparently considered a mild way to describe it [2]. Some of them proceeded to make false claims, e.g. that Damore was "criticizing the company’s efforts to increase diversity" (no he was not, he considers diversity a good thing and suggests better ways to increase it) or arguing that Google "should stop offering programs to minorities and women" (he was only discussing racist and sexist programs, not _all_ programs) [3].
As to 'fake news' - I take it you're trying to paint me as a Trump supporter. So let me tell you I'm not one, and I didn't even know what that was all about (as I never bothered with US mainstream media before) before it hit me where it hurts with the Damore story. Now I don't know how much pro-Trump media twisted facts, maybe it was the same. All I know is I was shocked to find the mainstream US media to be so ideologically charged as to attack a guy for his [very mild and well-substantiated] opinions with forged facts and labels. Because I know that guy could have been me.
In summary, you, together with the mainstream US culture, need to pop your own obvious, well-documented and plainly visible ideological bubble before you start making claims that people with true liberal, pro-science, pluralistic views like myself are in one.
[1] https://youtu.be/s4WoeOkj2Ng?t=2m1s
[2] https://www.theverge.com/2017/8/5/16101978/google-employee-w...
[3] https://www.vanityfair.com/news/2017/08/google-fires-employe...
I'm not a big fan of large companies, I don't think they can win. Even if you have good numbers, that doesn't fit the narrative of "Silicon Valley has established itself as the boys’ club". Say for example, we applauded companies who are doing better, maybe even close to statistical significance, then you could say, "Hey, it's still not quite there yet, but it's close and we're doing better than others. Come work for us and help us make it right." And then that would be actionable advice, and companies would have a great reason to strive for this. But we'd also have to be honest about statistical error rates, and maybe some other stuff.
I'll close on this, I've heard people say stupid and offensive shit when they're drunk, but never "Hey, let's pay women less", although I'm not an executive. So I feel like it isn't a common thought... unless you keep reading articles like this one.
Here's what I think a big driver is (you're going to have to walk down premise lane with me as I don't want to spend the time on citations ATM): Presuming that women are in general more agreeable than man in the workplace and that they don't fight as hard and as often for raises, I would expect an otherwise equal male counterpart that is more aggressive to have higher pay. Why? Because the squeaky wheel gets the grease.
Every year when we did promotions we'd start out with a zero-sum raise pool. Let's say if everyone got exactly the same, they'd get 2%. But if I give Bob a raise of 4% I have to go 'take' that extra 2% from people or everyone. If I'm worried about Bob leaving, or he's asked for a raise and plead a good case, he's going to get it. If Alice or Charlie have been silent all year but have been doing an otherwise good job, they're going to get what's leftover- in general.
There have been a number of cases where we idtenified someone silently kick major ass and threw money at them, but that I think it is because we had very conscientious managers that worked closely with their subordinates, unlike BigCo.
I'm not even going to touch on all the other reasons that people have laid out as possible reasons why women tend to earn less. I think if you've got two people working well, doing the exact same job and producing the exact same outputs, they should be paid nearly equally, but we're humans and life is messy. Some people haggle better than others, some had a very low salary before that the new job used to baseline, etc.
If you really want to maximize your salary at an given job, argue for every last penny coming in. All future raises (for the most part) will be based off that number and it will haunt you for your entire tenure at the company- most likely.
1. You approach a manager with expanded responsibilities that would justify more pay. Or a detailed account of your undocumented responsibilities that justify more pay. Don't come at me asking for more money for the same job if you're not already (or plan on) doing more work than we agreed to.
I went from Architect to Director this way.
2. You have another offer from another company and give your old company a chance to match. This could have ramifications that everyone goes around and around about. The general rule of thumb is not to do this, but if you believe in your current company, what you're doing them is informing them you're below market for YOU, and they've got an opportunity to fix that.
I did this at one company and it worked fine, but I left a year later anyway. There's some stats that say a large portion of people who do this are gone within a year.
...and Alice and Charlie just find another job in a few years because their raises haven't been keeping up with market rates. It's not necessarily the direct managers' fault unless corporate gives:
* estimates of market rates
* recommendations that an average employee gets an X% raise to track with market rate (or even stay ahead of it)
* a big enough pile of money to compensate everyone in line with market rates
* even more money to reward high achievers (and probably squeaky wheels, too)
This seems like a tall order to most budget-makers, so instead wise employees change jobs every so often to make sure their salaries keep up with market rates.
If you like your colleagues and the work you do, leaving the company for a raise seems more aggressive than first telling your manager that you are unhappy with your pay.
If I'm an employee and I do some great work that brings the company an extra $X per annum, and I ask for a raise of $0.1X the company is welcome to take that out of the $X I made them.
If the company decides to instead take the $0.1X from someplace else, that's their decision not mine - claiming raises are zero-sum is just bullshit to make people feel guilty about asking for a raise.
We can debate how that $X should be divided between workers, executives, creditors, shareholders and reinvestment. But there's no "financial reality" that means workers' share has to be zero.
Every organisation has rules and budgets and procedures, but they're made by people and if they don't work right they can be fixed by people. It's not like the CEO received the budget from God on stone tablets on a mountain 2000 years ago.
If I wanted to work in some calcified bureaucracy where rules were more important than outcomes and worker performance was irrelevant, I wouldn't be working in the private sector!
It's also highly unlikely that your singular effort created 10K more profit. That 10K was likely the culmination of a number of people who had varying levels of participation in the effort.
Unless you're working at a company that tracks your hours to every single possible project you're working on, there's often a huge blind spot as to where exactly your effort is going as it pertains to revenue. Unless your in professional services or some other type of billable position, then I actually agree with your notion to a degree...
With regards to people in billable positions, I once workthink d for a company that did something I really thought was neat. At the end of every quarter they'd announce a profit sharing rate based off of whatever it was that defined the rate. Since we were all billable this rate was at $X per billable hour. So if you billed more, you participated in a greater share of the revenue. I frequently saw $10-20K quarterly bonuses based entirely off my effort. It was quite motivating.
Because the profit sharing was the same rate for everyone and took in to account the entire company's financials, you were assured that your billable efforts were going to be rewarded on the same scale as the next person.
If you fly too close to the sun and your weekly pay directly relates to the value you put in that week you'd might have job security but you'd have absolutely no income security!
If that were true, wouldn't the salaries start off the same and then diverge? Doesn't seem to quite match up with these unofficial numbers.
Besides, at most $MegaCorp, isn't compensation reviewed periodically, and some kind of peer review is included? Which means that agreeableness is less of a factor, which means the process is working.
If there existed statistics for such strategy, I wonder if it would have a gender attribute to it.
No, because starting salary is something that is negotiable, too. And you need to push back on the offer and ask for more if you want to make more. If you're not the type to push back, you start lower.
Assuming women and men employees are (generally speaking) equally skilled yet differently compensated (for whatever reason), rational employers would actually prefer to hire women over men. Just like with any other purchase, all things being equal, cheaper is better.
Perhaps "old" companies with a "tradition" might have a status-quo bias, but new startups/smart CEOs that really try to minimize their burn rate would obviously see this.
Over time more and more women employees would be sought after more then men, which would eventually lead to an equilibrium and remove the pay gap.
So why hasn't this happened (yet)?
In the case about 20% of college graduates who studied Computer Science are female, and about the same proportion are seen employed as Software Engineers in the industry. Those women are not any cheaper than their male counterparts, because they're just as qualified.
Why women and men make different choices in aggregate, as well as if any of that is problematic, are topics of further debate that I don't think we have concrete answers to yet.
A possible explanation is that men, in the aggregate, are more motivated by compensation and willing to put up with more hardship or danger in exchange for more money.
There are a lot of jobs that pay a lot of money because they're dangerous, like working on an oil platform. Suppose those people get paid $90,000 whereas below average tech workers get paid $70,000 and above average tech workers get paid $100,000+. If you're a below average male tech worker who is in it for the money, maybe you take the oil job instead for the extra $20,000/year. The equivalent women take the $70,000/year non-dangerous job, causing the male average for "tech jobs" to be higher than the female average.
It's also possible that more men than women are willing to work longer hours or jobs with less flexible hours in exchange for a higher salary. Employers rationally pay people willing to do that more than people who aren't.
https://www.bls.gov/news.release/pdf/atus.pdf
Ironically, thats the type of talk that can get you fired from Google. We have to be able to address the reasons if we want a solution, even if it hurts your sensibilities.
That to me is the answer to this question: So why hasn't this happened (yet)?
Have you seen how many people on here report they have to change jobs to get a raise? Or how often companies will spend $200 worth of employees' time on bureaucracy before replacing a $100 broken chair?
Well there's your problem.
Humans are mostly irrational. Massively so. Rationality isn't the default state, it's not to be expected, it's a rare and precious jewel.
So many nerds dismiss the idea of systematic bias because "rational people wouldn't do that." This is true. It's also true that leprechauns wouldn't do that. Both statements are equally tied to reality.
Obviously, it's because everybody in that field is either a frat boy or evil, and there can be no other explanation /s
The alternative is not that everybody in that field is a frat boy or evil. It's that they're... human.
We are all biased. The only differences are a matter of degree, and whether or not we acknowledge it.
That should even out then, right? Or is bias somehow not normally distributed?
I'm happy to call BS on "bias" and "irrational" explanations - sounds like something you make up when you want to dismiss something and can't be arsed to come up with a proper explanation.
There are preferences for prettiness/beauty and especially tallness. They are biologically routed, well documented, and have good predictive power. I guess you could call that a bias. From looking at high competency fields such as doctors or lawyers, there doesn't seem to be a gender bias, neither biologically, nor culturally.
So which systematic bias are you proposing that would employers to act so irrationally, but yet somehow opt for the more expensive labour cost. Even if e.g. a CEO is racist, he's not going to lose sleep over offshoring jobs. Is gender different, if so why/why not?
On the flip side, if it is a systematic bias like tallness, do we have an obligation to combat it? Height doesn't get nearly the same treatment. Either way, the "bias"/"irrationality" argument is lazy and incomplete.
Do you seriously suggest that bias should be evenly distributed? For every person biased against women in some field, there's another person equally biased against men?
After all of these discussions about this topic on HN, I'm starting to wonder if HN actually has users posting from a parallel universe. Because some of the things being described don't match the one I live in at all.
Actually I don't remember, I wasn't alive or in the US for that. But from my understanding, calling that "bias" is a bit of an understatement.
> For every person biased against women in some field
What bias and which people? A general anti-women bias in society? In all men? White men? Only in tech? Only men in tech against women in tech? Any specifics at all for your hypothesis?
Fact is, there are fewer women than men doing CS degrees, about 20% AFAIK. That seems broadly to carry over to the ratio of genders in tech roles. Women seem to earn less than men. We don't really know why either is the case, but people are working hard to "correct" that, even though we have no idea what correct looks like. Some people think 50% would be good, because that's roughly the ratio in the general population. Looking at other majors and fields, 50% sometimes happens, sometimes doesn't.
Maybe there are better ways to approach this than to claim everybody is prejudice - maybe that's true, but is it constructive? Why is the ratio in tech disproportionate? Why is the ratio in veterinarians disproportionate? Are these equally important issues? Are they getting the same amount of coverage?
At the moment, it seems that all the nuance we've got is a hammer, and publications like the New York Times are sure reporting on a lot of nails.
Bonus question: Is Big Bang Theory good or bad for the tech industry, and do you think it will increase, decrease, or have no effect on gender ratios in tech?
There's a lot of reporting on nails because there are a ton of nails. Gender bias has been a thing literally since before recorded history. There are blatant examples quite recently, too. Until the 1970s, it was routine for banks to refuse to issue credit cards to women without their husbands' permission. Am I to believe that his problem, which is literally thousands of years old (probably more) has been solved in the past 40 years?
I have no idea about Big Bang Theory, never saw it.
I've been at several companies where literally all managers are men and women no more than objects that are hired based on their looks. Executive meetings are spent discussing female coworkers attributes and who has managed to get laid with who.
I have no idea how google rates on any employee scale but why "surely"?
I will say (anecdotally) that the women that I hired as individual technical contributors and promoted to managers were (on average -- with a small sample size) better at adapting to their new responsibilities than the men.
I didn't hear much of the kind of talk you are talking about -- but I discouraged employee fraternization of that type -- and maybe I was wrong to do so. (This was a while back - maybe norms have changed?)
Education and extra-curricular activities don't seem to be taken into account, which can easily explain the gap in the lower tiers.
What are they hiding? They must know what that data will reveal is the only explanation.
https://www.youtube.com/watch?v=uQ1JeII0eGo
They recount the ways people have been misrepresenting James' (rather meek) position, right down to accusations of sexism, nazism, malicious removal of academic references and outright lies. A sad testimony of this day and age and the current media climate in the US.
It's like people think covering their ears shouting "LALALA NOT TRUE, WILL NOT DISCUSS" has ever been a constructive way of addressing social issues.
I wish Google would come clean and release the raw stats. I understand that's one politically loaded and sensitive dataset, but nothing beats drawing your own conclusions. Right now, Google is getting flak from all sides, lose-lose.
The fact he couched it in pseudo-science, doesn't make that any less true. That's harmful to Google's image, and no woman would want to work with him after that, so he's toxic. They have every right to fire such a person.
https://news.ycombinator.com/newsguidelines.html
[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4418903/pdf/pna...
That's all I needed to see. Nothing worthwhile here. The people who'll tend to report will be the people who are worried about how much they earn. I'd say all this shows is that men are less worried about how much they earn than women.
Because "men are paid more than women" alone simply says NOTHING.
In the company I used to work there were HUGE salary differences between men with the same job title, and the difference sometimes could run up hundreds of euros. The main reason was that the better paid men negotiated their terms of employment WAY better than those that got paid less.
I suspect to see something very much alike here. Also, I would like to see the salary distribution within men and women, because I can almost certainly guarantee that some women are paid more than some men, but I don't think we will ever see that because that would destroy the entire "all men are paid more than women for no reason other than they are men" narrative some people are still trying to push, while it is related to a LOT of other things other than just pure gender.
BTW, if a salary difference is SUCH a big problem why stick with the companies who do this? Why not find employment somewhere else, or start your own company with better working conditions? It isn't like you are FORCED to work for Google or any other of the big corps in Silicon Valley if you don't like how they treat people there.