7 ms·

Perhaps there are disparaging comments because people love to quote misleading statistics... kind of just like you did. From the glassdoor wage gap study you m

Perhaps there are disparaging comments because people love to quote misleading statistics... kind of just like you did. From the glassdoor wage gap study you mentioned in [2]: > After taking into account differences in education and experience, men working at tech companies overall make 5.9 percent more than women do From a brief scan of the study they didn't control for hours worked and they are already down to 5.9% Since full-time men work 5% longer than full-time women[3] just adjusting for hours worked almost entirely eliminates the gap. [3] https://www.forbes.com/sites/karinagness/2016/06/30/new-report-men-work-longer-hours-than-women/#2f35760118b4 https://www.forbes.com/sites/karinagness/2016/06/30/new-repo... Edit: Sorry I put the overall adjusted US pay gap of 5.4% when I meant to put the adjusted tech pay gap of 5.9% from the glassdoor study.

ajross 10y ago
So... your response to finding an uncorrected variable in the study (which I don't see the value of, honestly, given that tech workers are near-uniformly salaried in this country) is to go out and cherry pick a "correction" from an entirely unrelated study that confirms your worldview? Who's quoting misleading statistics now? Yeah, I've gone through and scanned that BLS report you linked to (actually you didn't, you linked to Forbes, but this is what they were talking about): https://www.bls.gov/news.release/pdf/atus.pdf https://www.bls.gov/news.release/pdf/atus.pdf It doesn't say what you think it does at all. This isn't a study of men vs. women working identical jobs and doing less work, this is a finding that across the entire workforce, a men's "work day" is somewhat longer than a woman's. Which is to say: Women are more likely to work part time jobs.
- sheepmullet 10y ago
  > Which is to say: Women are more likely to work part time jobs. Please stop trying to mislead people. From your link on the very first page: > However, even among full-time workers (those usually working 35 hours or more per week), men worked longer than women—8.2 hours compared with 7.8 hours. (See table 4.) Like I said this is an ~5% difference. If you don't like this particular study the results have been replicated dozens of times across a wide range of industries. If you have some evidence that full time male tech workers work fewer hours than full time female tech workers I'd love to see it.
  - flukus 10y ago
    Can someone start a "no overtime for equality" campaign?
- sheepmullet 10y ago
  > which I don't see the value of, honestly, given that tech workers are near-uniformly salaried in this country You can't see how working extra hours can increase productivity enough to warrant a small increase in pay?
  - ajross 10y ago
    I don't see how a number cribbed from a comparison of "work day length" (NOT total hours worked!) across the entire US workforce (and thus not correcting for the fact that women and men have different career statistics) says anything about whether women in the tech industry covered by this glassdoor study work more or less hours than their coworkers. What you did is go out through the whole world of statistics to find one value that kinda/sorta "fixes" this inconvenient finding in the direction you want. That's just baldly ridiculous cherry picking, and you should be ashamed.
    - sheepmullet 10y ago
      I'm not saying the difference is explained by difference in hours worked. I'm saying it's a possible explanation. You on the other hand are trying to draw conclusions that aren't backed by the data.
      - ajross 10y ago
        What? That's 100% up-is-downism. What you say is simply incorrect. The conclusions is backed by data. The data is in the link. Go look at it. See? Glass Door says women are underpaid. That's data! What you are doing is saying "No, I don't think the conclusion should be what the data says", and then you're going out and finding ("cherry picking") different data about related but not identical subjects that appears to contradict it. Sorry, but the burden of proof is on you if you want to make a numeric argument here. And you're doing it with extraordinarily bad analysis.
        flukus 10y ago
        Not all data is equal. The collection methodology for glassdoor isn't much better than an online poll.
T-R 10y ago
> they didn't control for hours worked Are the programmers in question being paid by the hour? Does programmer pay within each gender correlate closely to hours worked?
- gozur88 10y ago
  You don't think people who are willing to work longer hours end up getting paid more? Are people who work more than forty hours each week doing it for nothing?
- sheepmullet 10y ago
  I wouldn't know. It seems reasonable. All the academic studies I have seen either adjust for it explicitly or put forward a strong case on why it isn't being included.
  - T-R 10y ago
    It's certainly reasonable for a studies across fields with hourly pay or, e.g., including both part-time and full-time (such as the BLS survey cited by the Forbes article), but I'm not sure we can make that assumption for software development - programmer productivity in a given time period is, on its own, somewhat mythically considered to vary by orders of magnitude.
    - sheepmullet 10y ago
      Fair point. I'd argue it's a more logical and likely explanation than sexism though.
    - wanderer2323 10y ago
      Anecdotal evidence matches this up pretty closely. Better programmers tend to work more, it's just a fact of life.
owenversteeg 10y ago
Funny enough, I agree with both you and the person you're arguing against. I agree that the wage gap (as defined by the difference in how much men and women earn when working in the same position with the same skills) is probably pretty close to 6%. [0] I also agree that there is a big problem currently. In my opinion, the problem is twofold. First of all, this gap - whatever you think it is, 6% or 9% or 21% - it is a problem. Most Americans live pretty close to financial ruin, like it or not. 76% of Americans live paycheck to paycheck [1] and 63% of Americans are one missed paycheck away from the streets [2]. A 6% pay gap is about two paychecks. Even though it may not seem like much to a HN reader with significant savings or a safety net, this small gap spread across the whole population actually causes problems. The second part of the problem is societal: women are discouraged from pursuing higher-paying jobs by society. This isn't the fault of the women themselves, or the people doing the hiring, but sexism and biases hidden everywhere. In my opinion, this is the best support for affirmative action: more women doing a certain job translates to more women interested in that career. These two things combined are a double whammy that currently mean that less women work and women who do work are not always financially secure - and that's the big problem here. If you're a woman from an upper class background and have been encouraged by those around you, you may never experience any of these problems, and that's great. But if you're an average American woman, these things have a very real impact on your life. This also explains the giant gulfs of opinion I see here on HN. Most people here are relatively wealthy - remember, if you earn more than 36k a year, you earn more than most working Americans. $70k/yr means you earn more than 75% of American workers. I'd bet the vast majority of US programmers on here earn above 70k/yr. It's easy to get caught in a bubble and say 'well the women around me seem to have no problems!' and forget about the majority of the country. At the same time, if you live in a low-income area, those two "missed" paychecks might be absolutely killer. [0] I have a personal, unfounded feeling that it's closer to 10% - maybe 8-9% if I had to guess. [1] http://money.cnn.com/2013/06/24/pf/emergency-savings/ http://money.cnn.com/2013/06/24/pf/emergency-savings/ [2] http://www.marketwatch.com/story/most-americans-are-one-paycheck-away-from-the-street-2016-01-06 http://www.marketwatch.com/story/most-americans-are-one-payc...
- brighteyes 10y ago
  > The second part of the problem is societal: women are discouraged from pursuing higher-paying jobs by society. This isn't the fault of the women themselves I really think that point of view is both wrong and harmful. It's true that society influences people in various ways. But you're assuming that what men do is the "default", and so if women do it less, they are discouraged. Instead, looking at it from the women's side, perhaps it is men that are overly encouraged to do certain things. For example, men are highly overrepresented in prisons compared to women, and that's not good. That might be an extreme example, but the fact is that many studies show that men care more about careers that make money, while women care more about making a positive difference in the world (obviously, both only on average). Given that, perhaps society pushes men too much towards tech and finance and so forth. It would be better if we nudged men instead towards nurturing professions, where you help people every day, like nursing or teaching. In other words, society influences both genders in various ways. Any difference could be either women pushed one way or men pulled the other.
throwaway91111 10y ago
Ok, I can appreciate skepticism. But do you have any arguments to advance yourself? Wage gaps are hugely complex, and the stats put out by the labor department (fuck Glassdoor, whatever) are both controlled, detailed, and fairly damning. Women make less on average in almost every field, women are employed less in high paying jobs, and the jobs that are heavily weighted towards women do not pay nearly as much. I don't understand how people can brush off single statistics while ignoring the absolutely massive cache of freely available evidence. Also, starving people of hours is just as effective at reducing costs as lowering the hourly. I don't think it's in your favor to argue that's something that should be controlled.
- SomeStupidPoint 10y ago
  The claim is that women (im aggregate) make different life choices than men, and once we account for those, the "wage gap" disappears. I honestly don't see anything wrong with different life choices leading to different outcomes, since the same is true within genders! That is, there's nothing sexist about that fact. The truth is the people advocating for equalizing pay when the sexes make different choices are arguing for sexism, not against sexism.
  - throwaway91111 10y ago
    The only sexist thing here is your support of a baseless claim. If it's not baseless, presumably you have something to offer. However, without evidence, you are most likely wrong. Does that wrongness amount to sexism? Hard to say; my inclination says yes since you're referring to shuffling hundreds of dimensions defining employment into "life choices". Yes, that is probably a factor; culture is a powerful force. But it's also probably not the only factor; there are likely myriad explanations. Are you blind to that, or just dismissive?
    - SomeStupidPoint 10y ago
      The only contention I have is one you agreed to elsewhere: that the actual aggregate wage-gap once you've controlled for common variables (such as life choices) is 3-6% (with women earning more in some fields), which likely isn't significant or correctable. A difference on that magnitude can easily be something about aggregate behavior correlated to sex, an uncontrolled for variable, or just random chance. So let me ask you: do you have any evidence that 3-6% gap is the result of sexism instead of merely being a fact correlated to sex? (Particularly if we were to confine the data to people under 40, to control for historical biases that take time to age out.)
KirinDave 10y ago
> From a brief scan of the study they didn't control for hours worked and they are already down to 5.9% Why is it a foregone conclusion that the majority of women in this population worked fewer hours? A forbes study which averages across a large canvas of american workers is not a very good thing to compare to glassdoor's targetted subsection.
- sheepmullet 10y ago
  The women surveyed could have worked more hours than the men. But by not controlling for it you reduce the studies explanative powers. You don't get to just ignore it because it is convenient. It doesn't even pass the "more likely than not" standard of evidence.
  - KirinDave 10y ago
    Right, but uh... The vast majority of positions Glassdoor cares about are salaried positions which don't track hours. They're exempt from overtime, as well. It also seems to me that that signal would show up weakly in the data because it wouldn't vary the immediate effect on apples-to-apples salary, only on promotions and possibly referrals. So I think your complaint is specious and the inclusion of a totally unrelated data point from a radically different dataset is either very misguided or profoundly disingenuous.
    - sheepmullet 10y ago
      > The vast majority of positions Glassdoor cares about are salaried positions which don't track hours Yes. And this makes any kind of data analysis a lot harder. On paper I work the same number of hours as my Japanese coworkers. In reality they work more than double. It's the same way on-call for many companies is just rolled into salary. I've managed projects as a senior dev while the senior dev sitting next to me was purely technical. Etc etc. The truth is two senior devs sitting right next to each other can have wildly different actual job duties. I suspect glassdoor has taken at least some of those differences in work duties vs stated title into account. > because it wouldn't vary the immediate effect on apples-to-apples salary, Why wouldn't it? Its not uncommon to stay at a certain level/title for 3-5 years. You can't just ignore hours worked because "salary". I come from a strong union background where you would get overtime or be able to take the time off later and where there were fairly rigid job duties. Where interviews and scoring candidates is all done in a very standardised fashion. And oddly enough after adjusting for these explicit factors we had a very small unexplained gender pay gap. Much smaller than average for the tech industry. Perhaps it's in our interest to make these factors explicit rather than implicit?

ajross 10y ago

So... your response to finding an uncorrected variable in the study (which I don't see the value of, honestly, given that tech workers are near-uniformly salaried in this country) is to go out and cherry pick a "correction" from an entirely unrelated study that confirms your worldview? Who's quoting misleading statistics now? Yeah, I've gone through and scanned that BLS report you linked to (actually you didn't, you linked to Forbes, but this is what they were talking about): https://www.bls.gov/news.release/pdf/atus.pdf https://www.bls.gov/news.release/pdf/atus.pdf It doesn't say what you think it does at all. This isn't a study of men vs. women working identical jobs and doing less work, this is a finding that across the entire workforce, a men's "work day" is somewhat longer than a woman's. Which is to say: Women are more likely to work part time jobs.

sheepmullet 10y ago
> Which is to say: Women are more likely to work part time jobs. Please stop trying to mislead people. From your link on the very first page: > However, even among full-time workers (those usually working 35 hours or more per week), men worked longer than women—8.2 hours compared with 7.8 hours. (See table 4.) Like I said this is an ~5% difference. If you don't like this particular study the results have been replicated dozens of times across a wide range of industries. If you have some evidence that full time male tech workers work fewer hours than full time female tech workers I'd love to see it.
- flukus 10y ago
  Can someone start a "no overtime for equality" campaign?
sheepmullet 10y ago
> which I don't see the value of, honestly, given that tech workers are near-uniformly salaried in this country You can't see how working extra hours can increase productivity enough to warrant a small increase in pay?
- ajross 10y ago
  I don't see how a number cribbed from a comparison of "work day length" (NOT total hours worked!) across the entire US workforce (and thus not correcting for the fact that women and men have different career statistics) says anything about whether women in the tech industry covered by this glassdoor study work more or less hours than their coworkers. What you did is go out through the whole world of statistics to find one value that kinda/sorta "fixes" this inconvenient finding in the direction you want. That's just baldly ridiculous cherry picking, and you should be ashamed.
  - sheepmullet 10y ago
    I'm not saying the difference is explained by difference in hours worked. I'm saying it's a possible explanation. You on the other hand are trying to draw conclusions that aren't backed by the data.
    - ajross 10y ago
      What? That's 100% up-is-downism. What you say is simply incorrect. The conclusions is backed by data. The data is in the link. Go look at it. See? Glass Door says women are underpaid. That's data! What you are doing is saying "No, I don't think the conclusion should be what the data says", and then you're going out and finding ("cherry picking") different data about related but not identical subjects that appears to contradict it. Sorry, but the burden of proof is on you if you want to make a numeric argument here. And you're doing it with extraordinarily bad analysis.
      - flukus 10y ago
        Not all data is equal. The collection methodology for glassdoor isn't much better than an online poll.

T-R 10y ago

> they didn't control for hours worked Are the programmers in question being paid by the hour? Does programmer pay within each gender correlate closely to hours worked?

gozur88 10y ago
You don't think people who are willing to work longer hours end up getting paid more? Are people who work more than forty hours each week doing it for nothing?
sheepmullet 10y ago
I wouldn't know. It seems reasonable. All the academic studies I have seen either adjust for it explicitly or put forward a strong case on why it isn't being included.
- T-R 10y ago
  It's certainly reasonable for a studies across fields with hourly pay or, e.g., including both part-time and full-time (such as the BLS survey cited by the Forbes article), but I'm not sure we can make that assumption for software development - programmer productivity in a given time period is, on its own, somewhat mythically considered to vary by orders of magnitude.
  - sheepmullet 10y ago
    Fair point. I'd argue it's a more logical and likely explanation than sexism though.
  - wanderer2323 10y ago
    Anecdotal evidence matches this up pretty closely. Better programmers tend to work more, it's just a fact of life.

owenversteeg 10y ago

Funny enough, I agree with both you and the person you're arguing against. I agree that the wage gap (as defined by the difference in how much men and women earn when working in the same position with the same skills) is probably pretty close to 6%. [0] I also agree that there is a big problem currently. In my opinion, the problem is twofold. First of all, this gap - whatever you think it is, 6% or 9% or 21% - it is a problem. Most Americans live pretty close to financial ruin, like it or not. 76% of Americans live paycheck to paycheck [1] and 63% of Americans are one missed paycheck away from the streets [2]. A 6% pay gap is about two paychecks. Even though it may not seem like much to a HN reader with significant savings or a safety net, this small gap spread across the whole population actually causes problems. The second part of the problem is societal: women are discouraged from pursuing higher-paying jobs by society. This isn't the fault of the women themselves, or the people doing the hiring, but sexism and biases hidden everywhere. In my opinion, this is the best support for affirmative action: more women doing a certain job translates to more women interested in that career. These two things combined are a double whammy that currently mean that less women work and women who do work are not always financially secure - and that's the big problem here. If you're a woman from an upper class background and have been encouraged by those around you, you may never experience any of these problems, and that's great. But if you're an average American woman, these things have a very real impact on your life. This also explains the giant gulfs of opinion I see here on HN. Most people here are relatively wealthy - remember, if you earn more than 36k a year, you earn more than most working Americans. $70k/yr means you earn more than 75% of American workers. I'd bet the vast majority of US programmers on here earn above 70k/yr. It's easy to get caught in a bubble and say 'well the women around me seem to have no problems!' and forget about the majority of the country. At the same time, if you live in a low-income area, those two "missed" paychecks might be absolutely killer. [0] I have a personal, unfounded feeling that it's closer to 10% - maybe 8-9% if I had to guess. [1] http://money.cnn.com/2013/06/24/pf/emergency-savings/ http://money.cnn.com/2013/06/24/pf/emergency-savings/ [2] http://www.marketwatch.com/story/most-americans-are-one-paycheck-away-from-the-street-2016-01-06 http://www.marketwatch.com/story/most-americans-are-one-payc...

brighteyes 10y ago
> The second part of the problem is societal: women are discouraged from pursuing higher-paying jobs by society. This isn't the fault of the women themselves I really think that point of view is both wrong and harmful. It's true that society influences people in various ways. But you're assuming that what men do is the "default", and so if women do it less, they are discouraged. Instead, looking at it from the women's side, perhaps it is men that are overly encouraged to do certain things. For example, men are highly overrepresented in prisons compared to women, and that's not good. That might be an extreme example, but the fact is that many studies show that men care more about careers that make money, while women care more about making a positive difference in the world (obviously, both only on average). Given that, perhaps society pushes men too much towards tech and finance and so forth. It would be better if we nudged men instead towards nurturing professions, where you help people every day, like nursing or teaching. In other words, society influences both genders in various ways. Any difference could be either women pushed one way or men pulled the other.

throwaway91111 10y ago

Ok, I can appreciate skepticism. But do you have any arguments to advance yourself? Wage gaps are hugely complex, and the stats put out by the labor department (fuck Glassdoor, whatever) are both controlled, detailed, and fairly damning. Women make less on average in almost every field, women are employed less in high paying jobs, and the jobs that are heavily weighted towards women do not pay nearly as much. I don't understand how people can brush off single statistics while ignoring the absolutely massive cache of freely available evidence. Also, starving people of hours is just as effective at reducing costs as lowering the hourly. I don't think it's in your favor to argue that's something that should be controlled.

SomeStupidPoint 10y ago
The claim is that women (im aggregate) make different life choices than men, and once we account for those, the "wage gap" disappears. I honestly don't see anything wrong with different life choices leading to different outcomes, since the same is true within genders! That is, there's nothing sexist about that fact. The truth is the people advocating for equalizing pay when the sexes make different choices are arguing for sexism, not against sexism.
- throwaway91111 10y ago
  The only sexist thing here is your support of a baseless claim. If it's not baseless, presumably you have something to offer. However, without evidence, you are most likely wrong. Does that wrongness amount to sexism? Hard to say; my inclination says yes since you're referring to shuffling hundreds of dimensions defining employment into "life choices". Yes, that is probably a factor; culture is a powerful force. But it's also probably not the only factor; there are likely myriad explanations. Are you blind to that, or just dismissive?
  - SomeStupidPoint 10y ago
    The only contention I have is one you agreed to elsewhere: that the actual aggregate wage-gap once you've controlled for common variables (such as life choices) is 3-6% (with women earning more in some fields), which likely isn't significant or correctable. A difference on that magnitude can easily be something about aggregate behavior correlated to sex, an uncontrolled for variable, or just random chance. So let me ask you: do you have any evidence that 3-6% gap is the result of sexism instead of merely being a fact correlated to sex? (Particularly if we were to confine the data to people under 40, to control for historical biases that take time to age out.)

KirinDave 10y ago

> From a brief scan of the study they didn't control for hours worked and they are already down to 5.9% Why is it a foregone conclusion that the majority of women in this population worked fewer hours? A forbes study which averages across a large canvas of american workers is not a very good thing to compare to glassdoor's targetted subsection.

sheepmullet 10y ago
The women surveyed could have worked more hours than the men. But by not controlling for it you reduce the studies explanative powers. You don't get to just ignore it because it is convenient. It doesn't even pass the "more likely than not" standard of evidence.
- KirinDave 10y ago
  Right, but uh... The vast majority of positions Glassdoor cares about are salaried positions which don't track hours. They're exempt from overtime, as well. It also seems to me that that signal would show up weakly in the data because it wouldn't vary the immediate effect on apples-to-apples salary, only on promotions and possibly referrals. So I think your complaint is specious and the inclusion of a totally unrelated data point from a radically different dataset is either very misguided or profoundly disingenuous.
  - sheepmullet 10y ago
    > The vast majority of positions Glassdoor cares about are salaried positions which don't track hours Yes. And this makes any kind of data analysis a lot harder. On paper I work the same number of hours as my Japanese coworkers. In reality they work more than double. It's the same way on-call for many companies is just rolled into salary. I've managed projects as a senior dev while the senior dev sitting next to me was purely technical. Etc etc. The truth is two senior devs sitting right next to each other can have wildly different actual job duties. I suspect glassdoor has taken at least some of those differences in work duties vs stated title into account. > because it wouldn't vary the immediate effect on apples-to-apples salary, Why wouldn't it? Its not uncommon to stay at a certain level/title for 3-5 years. You can't just ignore hours worked because "salary". I come from a strong union background where you would get overtime or be able to take the time off later and where there were fairly rigid job duties. Where interviews and scoring candidates is all done in a very standardised fashion. And oddly enough after adjusting for these explicit factors we had a very small unexplained gender pay gap. Much smaller than average for the tech industry. Perhaps it's in our interest to make these factors explicit rather than implicit?