Join 3,433 readers in helping fund MetaFilter (Hide)


you got dick'd
May 8, 2010 1:23 PM   Subscribe

Is a Woman's MBA Worth Less? $4,600. That's how much less women made than men in their first post-MBA jobs, according to research by Nancy Carter and Christine Silva of Catalyst. And it's not because women tend to start at lower positions than men — though they do start at lower positions than men, on average, that's a separate problem. The research controls for job level and industry. What's more, the salary lines aren't parallel; men's salaries start higher, then rise faster. The gap widens over time, even after controlling for factors like having children or differing aspiration levels. The pay just isn't equal.
posted by infini (96 comments total) 17 users marked this as a favorite

 
Anyone who does equal the amount and quality of work as anyone else, regardless of sex, age or skin color, should get the same amount of pay. Anything less is unacceptable. I think that should be the arguing point, if the point is ever argued in any judging public venue, rather than 'women make less than men', because that sounds too victimizing. I don't think companies should go out of there way to hire women over men, just like I don't think they should go out of there way to make sure they have enough of any certain skin color of people, but I think they should be fair on who does best at the job. Certainly, there are women who perform better than men, and visa versa.
posted by Malice at 1:49 PM on May 8, 2010


. . . rather than 'women make less than men', because that sounds too victimizing.

Huh? We should avoid mentioning something that's true, and a problem that needs fixing, because it would be "victimizing?" The fact that racism still exists in America -- we shouldn't mention that, because it's "victimizing?" The fact that homosexuals are discriminated against -- we shouldn't mention that either, because it's "victimizing?"

Anyhow, this is a good article, even if what it says is, sadly, not at all surprising, given that we still live in a deeply patriarchal day and age.
posted by Frobenius Twist at 1:56 PM on May 8, 2010 [3 favorites]


What really bothers me is that this article isn't saying what it should say.

$4600 over a year seems not that big of a deal...the REAL big deal is the fact that women on average (regardless of education or...ahem...that GLORIOUS MBA) are paid about a quarter less than men are. Thats about 25%, not $25...or even that shitty $4600.

So yeah...that $4600 doesn't surprise me...for jobs in which men get paid about $20-25K/year.

I know you guys love being elitist, harvard...but if you're going to talk about a problem, you have to talk about how it affects all women, not just the ahem...MBA holders.
posted by hal_c_on at 2:08 PM on May 8, 2010 [2 favorites]


$4,600 less sounds pretty good actually. According to this, the average MBA graduate makes $80,000 to start. So $4,600 would be roughly 6% less*. This is much better than in the rest of the economy where the gap is much higher, at 22%.

*= The $80,000 figure presumably includes men and women making the wrong number to use here.
posted by euphorb at 2:11 PM on May 8, 2010


the REAL big deal is the fact that women on average (regardless of education or...ahem...that GLORIOUS MBA) are paid about a quarter less than men are. Thats about 25%, not $25...or even that shitty $4600.

That's a good point, although my understanding of that statistic (which could be flawed) is that there isn't a disparity of 25% within the same jobs, but rather that women make 25% less than men considered across all jobs; i.e., the lucrative jobs go disproportionally to men. So this article can be taken as additional evidence: not only do men get higher-paying jobs than women in general, but even when women do get those jobs, they are still paid less.
posted by Frobenius Twist at 2:15 PM on May 8, 2010 [1 favorite]


I have never gotten the hang of negotiating a salary. Often, I've been in the position of being slightly desperate to get a job because whenever I've been unemployed, I've felt stir-crazy and scared to death and embarrassed. In my previous job, I worked very, very hard and did really excellent work that was praised, but my colleague, who was also a woman and white, who did not produce the same level of results, but had some characteristics that probably made her seem less "foreign," received a raise and a promotion. It was extremely hurtful, so I got a new job and didn't communicate the slight. My colleague was my friend, and we kept in touch afterwards. She said she received the new title and responsibillities, but they delayed her actual raise for months. She didn't want to keep bothering HR about it, she said, when I told her she was getting shafted, that they owed her from the time they gave her new responsibilities.
posted by anniecat at 2:17 PM on May 8, 2010


Please read the following as a meta comment on studies and human sociological biases and not as saying that pay differences do not exist. This research was done for Catalyst which is an advocacy group for women in the workplace. It would be utterly shocking if they came to any other conclusion. If this were, say, Pfizer or Glaxo putting out research that a lot more people should be taking prescription meds everyone would be falling all over themselves to ridicule it as self-serving. But there's a well-known human cognitive bias where we accept uncritically things that we want to believe or align with our existing politics.

In other words, I'm sure women are discriminated against in the workplace. But I'm pretty sure that was the conclusion that Catalyst was going to come to no matter what. Because Catalyst is about discrimination against women in the workplace. It's like the Family Research Council discovering that the gay agenda harms children. Except, you know, not evil. I'm not sure that because we agree with something we should ignore the facts, though, which are that studies by advocacy groups are virtually meaningless because the result is pre-determined.
posted by Justinian at 2:21 PM on May 8, 2010 [17 favorites]


While the pay gap is real, how and where it comes from is something of a mystery.


It would have been nice if they had bothered to formulate a follow-up survey for the respondents that addressed this. Others have made attempts:


About 10 years ago, a group of graduate students lodged a complaint with Linda C. Babcock, a professor of economics at Carnegie Mellon University: All their male counterparts in the university's PhD program were teaching courses on their own, whereas the women were working only as teaching assistants.

That mattered, because doctoral students who teach their own classes get more experience and look better prepared when it comes time to go on the job market.

When Babcock took the complaint to her boss, she learned there was a very simple explanation: "The dean said each of the guys had come to him and said, 'I want to teach a course,' and none of the women had done that," she said. "The female students had expected someone to send around an e-mail saying, 'Who wants to teach?' " The incident prompted Babcock to start systematically studying gender differences when it comes to asking for pay raises, resources or promotions. And what she found was that men and women are indeed often different when it comes to opening negotiations.

These differences, Babcock and other researchers have concluded, may partially explain the persistent gender gap in salaries, as well as other disparities in how people rise to the top of organizations. Women working full time earn about 77 percent of the salaries of men working full time, Babcock said. That figure does not take differing professions and educational levels into account, but when those and other factors are controlled for, women who work full time and have never taken time off to have children earn about 11 percent less than men with equivalent education and experience.

[...]



Similar ground was trodden and kicked around the web earlier this year.
posted by ferdinand.bardamu at 2:22 PM on May 8, 2010 [9 favorites]


I'm not seeing a discussion of methods in the research report, but the way it's presented makes me wonder if these findings make interpretable sense.
This research is essentially asking, "How would men and women with MBAs differ on wages in first year post-MBA if they did not differ on job level, industry, having children, or aspiration levels?" The thing is, they do differ on these factors. You can't just remove them and pretend you're talking about something that happens in the real world. You can parcel out the effect of out these kinds of factors if you're confident that they're "noise," like if you're doing an fMRI study looking at some brain region and want to take out the effect of head movement, but to my understanding, you just can't do these types of analyses on preexisting groups (e.g., men and women). Some things are so intimately intertwined that you just can't separate their influence.
I'm not an expert statistician by any means, and I happen to be a feminist, but I think that our arguments, as long as they're research-based, should be based in research done with sound methods.
See Miller & Chapman (2001)warning: pdf for more information.
posted by emilyd22222 at 2:23 PM on May 8, 2010 [4 favorites]


I'm afraid the whole 'patriarchy' bit does not yield the clearest picture. We're talking about fairly innate perceptions of both men and women.

In particular, there have been experiments that demonstrated : If you create artificial academic hiring committees consisting entirely of women, given the curriculum vitae to evaluate, but randomly assign each fake applicant's gender via name choice, the hiring committee will discriminate against the female sounding names.

I'm sure the study controls for issues like aspiration levels as well as possible, but they simply cannot control for people's expectations, especially if those expectations are partially biological. We know for example that gender is influenced by birth order.

African Americans succeeded early in sports largely because sports uses almost perfectly objective criteria for measuring performance. We should therefore expect that women will be treated fairly wherever your evaluation criteria are extremely objective.
posted by jeffburdges at 2:25 PM on May 8, 2010


I guess I'm trying to say what emilyd2222 just said; we need to hold research we agree with to exactly the same high standards as that which bothers us (such as the anti-climate change stuff) both for reasons of integrity and more practically because if we do not, we will be rightly dismissed as partisan.
posted by Justinian at 2:26 PM on May 8, 2010 [2 favorites]


While the pay gap is real, how and where it comes from is something of a mystery.

Passing on establishing cause also leads to an unstated argument that the result is automatically unjust, a weasely appeal to the perception that inequal outcomes are sufficient evidence of discrimination and injustice. But establishing a convincing and irrefutable cause takes more work and studied investigation than survey research and isn't a good return on limited issue-advocacy organization dollars.
posted by ferdinand.bardamu at 2:30 PM on May 8, 2010 [2 favorites]


ferdinand.bardamu brings up an excellent point.

Consider also that it seems many women place an extremely high value on men's income and earning potential when considering them as mates. And as far as I know, the opposite generalization does not hold true; that is, I know of no research which suggests that men concern themselves very much with how much their potential girlfriends or wives might earn relative to other qualities they might possess.

As long as we're looking at this ugly business of income disparity between the genders, it merits notice that higher salaries are incentivized for men in a way that they are not for women.
posted by millions at 2:31 PM on May 8, 2010 [6 favorites]


Consider also that it seems many women place an extremely high value on men's income and earning potential when considering them as mates.

Could you please provide a citation for this, other than "the women I know"? If we're going to play Anecdote Wars, none of the women I know place particularly much value on a potential partner's income.
posted by Mrs. Pterodactyl at 2:36 PM on May 8, 2010 [8 favorites]


So, the Equal Pay Act of 1963 is not really working, and we will never pass the Equal Rights Amendment. Neat!
posted by Houstonian at 2:38 PM on May 8, 2010


many women place an extremely high value on men's income and earning potential when considering them as mates...it merits notice that higher salaries are incentivized for men in a way that they are not for women.

The study controlled for aspiration levels. I suppose you could argue that having high aspirations because you want the money and prestige is qualitatively different from having high aspirations so you can attract women, who are a bunch of golddiggers.

I'm still waiting for any company, or anyone in the "equal opportunities doesn't guarantee equal outcomes" crowd, to advocate a hiring process in which job applicants' names are replaced by numbers on resumes, and maybe applicants to technical positions can take anonymous tests as part of an initial screening. I'm not optimistic.
posted by transona5 at 2:47 PM on May 8, 2010 [2 favorites]


Could you please provide a citation for this, other than "the women I know"? If we're going to play Anecdote Wars, none of the women I know place particularly much value on a potential partner's income.

Yes, data directly relating to the issue presented would be good.

Relatedly, there are two strong trends in place: 1) More women than men are entering and completing college. 2) Marriages between those of different educations levels are decreasing.

These do indicate that less educated men, who can generally also be described as less wealthy, will be at a disadvantage in the marriage market versus their more educated (and, generally, more wealthy) peers.
posted by ferdinand.bardamu at 2:55 PM on May 8, 2010


educations
posted by ferdinand.bardamu at 2:56 PM on May 8, 2010


Huh? We should avoid mentioning something that's true, and a problem that needs fixing, because it would be "victimizing?" The fact that racism still exists in America -- we shouldn't mention that, because it's "victimizing?" The fact that homosexuals are discriminated against -- we shouldn't mention that either, because it's "victimizing?"

It's not about not mentioning it. As a woman, if I say "I work as hard as any man here. Clearly, (x, x, x) shows so. I think in all fairness my salary should reflect my hard work.", then I am sounding respectful and it is, of course, an empowering statement. If I say, "Men get paid more than me and it's because I'm a woman" then I sound like a victim.

This is just my experience in the work force.
posted by Malice at 3:02 PM on May 8, 2010 [1 favorite]


It's not about not mentioning it. As a woman, if I say "I work as hard as any man here. Clearly, (x, x, x) shows so. I think in all fairness my salary should reflect my hard work.", then I am sounding respectful and it is, of course, an empowering statement. If I say, "Men get paid more than me and it's because I'm a woman" then I sound like a victim.

"A whining crying race may be pitied but seldom respected." - Booker T. Washington
posted by ferdinand.bardamu at 3:07 PM on May 8, 2010


I don't think you too terribly many people valuing some quality of potential mates highly before dramatic percentages of the other gender adopt the quality, meaning the effect millions describes almost surely exists. Yes, this study's control for aspiration would hopefully cover the incentivizes millions describes. Also, I'm sure society can handle a few more uneducated men delaying marriage because finding an uneducated women takes longer. ;)
posted by jeffburdges at 3:07 PM on May 8, 2010


"A whining crying race may be pitied but seldom respected." - Booker T. Washington

Precisely.
posted by Malice at 3:09 PM on May 8, 2010


I was going to say what transona5 said above but I would have added that anonymizing information would only work for new applicants. I don't think it would be practical in most companies as it would apply to promotions. It's worth a try though I think.

And here is a weird thing to further complicate (and that I hope doesn't apply to my marriage):
Wealthy men give women more orgasms
posted by vapidave at 3:15 PM on May 8, 2010 [1 favorite]


I'm afraid the whole 'patriarchy' bit does not yield the clearest picture. We're talking about fairly innate perceptions of both men and women.

In particular, there have been experiments that demonstrated : If you create artificial academic hiring committees consisting entirely of women, given the curriculum vitae to evaluate, but randomly assign each fake applicant's gender via name choice, the hiring committee will discriminate against the female sounding names.


So the fact that the victims of oppression internalize that oppression- that women who are raised in a misogynist and patriarchal society adopt misogynist and patriarchial attitudes- somehow means that patriarchal attitudes are "innate"? That's completely inane.
posted by Pope Guilty at 3:22 PM on May 8, 2010 [12 favorites]


You're over thinking things again, Pope Guilty. According to some of my learned colleagues at work, women just need to work smarter, not harder. And if a woman is willing to be paid less, why, that's just the market being efficient. And if she can't get over it, maybe she's not cut out for the workplace, in which case they'd be happy to impregnate her and turn out the next generation of Little Johnny Howards.
posted by obiwanwasabi at 3:33 PM on May 8, 2010 [3 favorites]


How do the prevalence of misogynist patriarchal attitudes account for the established and increasing rate at which females outnumber males on college campuses? Isn't this likely, over time, to erode and perhaps even invert the income gap?
posted by ferdinand.bardamu at 3:40 PM on May 8, 2010


I knew my penis would come in handy eventually.
posted by doublehappy at 3:40 PM on May 8, 2010 [5 favorites]


Yes, ferdinand, the existence of one field in which women are (at least at the present time) doing better than men disproves the existence of systemic misogyny. Similarly, the fact that black men comprise a larger percentage of professional athletes than of the national population proves that our culture isn't racist.
posted by Pope Guilty at 3:47 PM on May 8, 2010 [2 favorites]


If female employees are as productive as male employees, wouldn't a savvy company be able to save money by replacing its male staff with female staff? It seems like there would be a strong incentive to profit from work that was erroneously undervalued in the labor market.
posted by Human Flesh at 3:49 PM on May 8, 2010


Is A going to change B at some point?

Yes, ferdinand, A is proof that there is no B.
posted by found missing at 3:50 PM on May 8, 2010


forgot to close my tag
/Pope Guilty
posted by found missing at 3:52 PM on May 8, 2010


If female employees are as productive as male employees, wouldn't a savvy company be able to save money by replacing its male staff with female staff? It seems like there would be a strong incentive to profit from work that was erroneously undervalued in the labor market.

One of the funnier things about libertarians is the claim that the free market will, in the manner you've identified here, eliminate sexism/racism/etc. That this never fucking happened is yet another data point in the mountain of evidence that libertarians live in a fantasy world.
posted by Pope Guilty at 3:53 PM on May 8, 2010 [6 favorites]


"Scientists find gender inequality continues to exists, film at eleven"
posted by rebent at 3:54 PM on May 8, 2010


Where is the claim that the market will eliminate sexism? I only see the question as to whether the market would exploit sexism.
posted by found missing at 3:56 PM on May 8, 2010 [1 favorite]


Is A going to change B at some point?

Yes, ferdinand, A is proof that there is no B.


You skipped the first part of the question. How is it that a pervasively misogynistic and patriarchal society has more females successfully attaining more (and increasingly more) education than males?
posted by ferdinand.bardamu at 3:58 PM on May 8, 2010


I know you guys love being elitist, harvard...but if you're going to talk about a problem, you have to talk about how it affects all women, not just the ahem...MBA holders.
posted by hal_c_on at 2:08 PM on May 8


that's the main reason this caught my eye. Unlike other situations, the Bschools never mention this aspect, emphasize the high salaries and fancy jobs their graduates get and encourage more women to apply in order to meet their targets.

This then implies that simply an MBA alone isn't enough - as was pointed out upthread, very well I believe, women also tend to be more passive in salary negotiations but I also include the fact that there's a bias still towards "men as heads of households" - for eg, my former business partner was from South Africa and his CV clearly stated "married with two children" as that would change the salary level he could command/demand almost by default than "divorced/no kids" ... there's more here but my main issue is with the positioning and marketing of the MBA program as this automatic pass to high salaries/fancy jobs regardless of background, gender, race etc

I interviewed at Kellogg and at MIT but ended up choosing a second tier program that offered me a 70% scholarship - a decision that sometimes I regretted in the early days of the job boom but serves me well today as I have no heavy weight of a debt burden on me to carry nor the painful sense of disappointment and disillusionment I've seen in female friends with similar backgrounds face after they graduated from top notch schools in Chicago with debt the size of a small mortgage and no dramatic change in their employment situation. Its a false promise in a way. Its also a given that any woman who puts herself through the whole MBA admissions through to graduation process is ambitious and career minded, not to mention a professional.


otoh, I recall a couple who were just husband hunting on a very premium scale
posted by infini at 3:59 PM on May 8, 2010


Did anybody else look at the data? I know its getting into the weeds a bit but it would certainly seem relevant to judging the validity of their conclusions. Take the regression analysis for first post-MBA job level, full sample, standardized coefficients. The R2 value for gender alone is only 0.02 but thats to be expected since we know there are a lot more factors at play. But when they include age, region, and industry in order to conclude that gender is a strong factor even after controlling for other factors the β for gender goes from -0.13 to -0.11. Which is okay since thats still p<0>2 has only risen from 0.02 to 0.10 after including age, region, and industry. That's a really, really low R2 to decide you've factored in all the variables, particularly when β has moved towards 0 by almost 20% at that value. It would seem to me that it's not unreasonable to think that all of the dependent variables have not been accounted for and that β for gender might lose significance once they are.

Women in their sample had an average first post-MBA β of 7080 (or $7080) less than men, but that's with an R2 of 0.01! Once they include age, job level, region, and industry β drops to $4600, which is the number they conclude is the "gender penalty" women receive. But R2 is only up to 0.30 at that point. Now, 0.30 is a whole lot better than 0.10 but considering the gender gap has fallen by almost exactly a third (from 7000 to 4600) at that point, it is again not totally unreasonable to think that there are additional variables which would further decrease the gap.

So my impression is that the data really does say what they claim so far as that goes. Women appear to make less than men after controlling for age, region, and industry. By about $4600. But, and this seems important, there are also decent indications that there are other unrecognized variables here and, furthermore, each variable introduced so far has decreased the apparent pay gap by a noticeable amount.

Like I said, this in no way is evidence against the hypothesis that women with the same background and experience are paid less for the same job. But neither does it strike me as hugely convincing evidence for that hypothesis. I know that drawing conclusions based on R2 values is chancy but, really, 0.10 is goddamn low and 0.30 isn't going to inspire any kind of confidence.
posted by Justinian at 4:00 PM on May 8, 2010 [5 favorites]


You skipped the first part of the question. How is it that a pervasively misogynistic and patriarchal society has more females successfully attaining more (and increasingly more) education than males?

This is a very recent development, and is the product of many people working to counteract and eliminate the pervasive misogyny and patriarchy. Pretending that there is only one force driving events is dumb and useless.
posted by Pope Guilty at 4:01 PM on May 8, 2010 [1 favorite]


Pope Guilty, I'm not a libertarian, and I didn't claim that free markets would eliminate sexism. I'm just wondering if there are known employers who benefit from other companies' ignorant hiring practices.
posted by Human Flesh at 4:04 PM on May 8, 2010


Oops, metafilter ate part of my comment because it included less than and greater than signs which it interpreted as a tag. The problematic section follows:

"Which is okay since thats still p<0.01, but R2 has only risen from 0.02 to 0.10 after including..."
posted by Justinian at 4:04 PM on May 8, 2010


As I've noted around here before, pay is strongly correlated with the perceived ethics of the employer. From Disciplined Minds:
None of this is to imply that new professionals are left without goals. Ironically, however, the primary goal for many becomes, in essence, getting compensated sufficiently for sidelining their original goals. Robert H. Frank, a Cornell University economics professor, tried to find out exactly how much compensation people deem sufficient for making this sacrifice. He surveyed graduating seniors at his university and found, for example, that the typical student would rather work as an advertising copywriter for the American Cancer Society than as an advertising copywriter for Camel cigarettes, and would want a salary 50% higher to do it for the cigarette company. The typical student would want conscience money amounting to a 17% salary boost to work as an accountant for a large petrochemical company instead of doing the same job for a large art museum. Indeed, employers that are seen as less socially responsible do have to pay a "moral reservation premium" to get the workers they want. Frank found that men are more likely than women to sell out, and this accounts for at least part of the gap in average salaries between equal men and women.[1]

1. Robert H. Frank, "Can Socially Responsible Firms Survive in a Competitive Environment?" in David M. Messick, Ann E. Tenbrunsel, editors, Codes of Conduct: Behavioral Research into Business Ethics, Russell Sage Foundation, New York (1996), ch. 4 (pp. 86-103). Chronicle of Higher Education, 21 February 1997, p. A37.
posted by Chuckles at 4:05 PM on May 8, 2010 [7 favorites]


The size of R-sq is important if your study is attempting to explain as much variance as possible in the outcome variable. Of more important here to the hypothesis test is the size and significance of the coefficient for gender, after controlling for other factors. (That said, there is much unexplained variance, but that's a different question.)
posted by found missing at 4:06 PM on May 8, 2010


Could be a generational artifact but I don't count 1991 as being particularly recent. Further, if there was a great push to enroll females in post-secondary education in order to improve their status in re males, why is there no concern, if some notion of parity is a goal, at how male enrollment is not only lagging but falling further and further behind?
posted by ferdinand.bardamu at 4:08 PM on May 8, 2010


Sorta sorry to do this but:

Metafilter: ...but that's with an R2 of 0.01!

Seriously, I'm really happy to have people that understand these things here.
posted by vapidave at 4:15 PM on May 8, 2010 [1 favorite]


found missing: What you are saying is true, but doesn't a very low R2 suggest that they haven't necessarily sufficiently controlled for other factors?
posted by Justinian at 4:17 PM on May 8, 2010


Similarly, the fact that black men comprise a larger percentage of professional athletes than of the national population proves that our culture isn't racist.

The pool of college graduates dwarfs that of professional athletes. The success or failure of a particular group vis-a-vis the latter is statistically meaningless in drawing wider conclusions about society. The former is not.
posted by ferdinand.bardamu at 4:18 PM on May 8, 2010


Justinian: Nope. So long as they've controlled for factors that are systematically related to both the IV of interest, and the DV, there is no need to explain all the variance in the DV.
posted by found missing at 4:22 PM on May 8, 2010


Malice: Anyone who does equal the amount and quality of work as anyone else, regardless of sex, age or skin color, should get the same amount of pay.

There's an old saying in business: you don't get what you're worth, you get what you negotiate. I wonder if some of the issue might not be the socialization most women get to 'get along' with groups? Good salary negotiation requires a touch of arrogance and self-centeredness, which are typically taught to women as being character flaws. And that woman upthread, who didn't get her raise when she got the new job... had that been a guy, his (professional-level) friends would have been telling him to be an asshole about it if he had to. I wonder if her close friends would have said the same thing?

Getting a good salary is a separate skill from just doing the work, and that's something, sadly, you learn mostly from being taken advantage of when you're young.
posted by Malor at 4:37 PM on May 8, 2010 [5 favorites]


found missing: Whether they've controlled for all the factors that are systematically related to both is exactly the issue, though. They've controlled for some factors but all of them? I'm positing that they have not. For example; what if one of the explanations is that starting salary after an MBA is strongly correlated with aggressiveness of negotiation and men tend to be more aggressive negotiators? That would be systematically related to both the IV of interest and DV and is not addressed in this research, and would lead to a lot of unexplained variance like what we see in the data.

I dunno, I suppose this is a roundabout way of saying that their conclusions look valid from the data but that those conclusions are being stated in a misleading way consistent with the slant of the advocacy group.

The data supports "It appears likely that women receive a lower salary than men in their first job after obtaining an MBA while controlling for age, region, and occupation. Other factors which may explain the difference have not been ruled out." What gets reported: "Women are paid less than men even after accounting for other factors which indicates gender discrimination."

That conclusion is only valid if age, region, and occupation are (as you rightly point out) the only factors systematically related to both the variables of interest, and the high variance may indicate that they are not. As, it seems to me, does common sense.
posted by Justinian at 4:43 PM on May 8, 2010 [1 favorite]


The gap widens over time, even after controlling for factors like having children or differing aspiration levels.

How exactly do you control for those things?

I understand that a lot of the quantitative difference in statistical salaries comes from qualitative choices made more often by women, especially those with families - eg for more flexible working hours, shorter working hours, offices closer to home, fewer demands for paid or unpaid overtime, less stressful jobs or environments.

I'm not sure how you can simply control for those kinds of "aspirational" choices, as in "let's just pretend that you were working 70 hours a week instead of 30 for the past five years; surely you'd be VP of Marketing by now...?"

It can't be a simple matter of extrapolating earnings out on a pro-rata basis, because career advancement depends so much on success in positions or projects, and it's a qualitative matter whether or not one is even in a position to deliver those kinds of results in the first place, depending on the work/life balance that one seeks.
posted by UbuRoivas at 4:44 PM on May 8, 2010 [3 favorites]


Justinian: Sure, then we agree about one thing we should be scrutinizing the study for: important missing control variables. My point is that R-sq is *no* indication of whether there are any. This is a matter of theory not numbers, and your arguments about negotiation aggressiveness exemplify how such variables are identified.
posted by found missing at 4:50 PM on May 8, 2010


It's easy to control for the children thing, Ubu, because you can easily look at the data set for all employees and then only those employees without children and see how much it influenced salary and advancement. Controlling for aspiration level is trickier. It appears to me that they simply asked people what level they aspired to. That strikes me as a rather flawed methodology for a lot of reasons. Personal motivations would be a very hard thing to account for reliably. They may have been better off just leaving it out altogether.
posted by Justinian at 4:53 PM on May 8, 2010


My point is that R-sq is *no* indication of whether there are any.

My understanding was that in practice you tended to find lower R2 in cases where important control variables were missing. But if I'm wrong about that it's no skin off my bask as it's been rather a number of years since I did a lot of statistics work and the important point is, like you said, whether they've accounted for all the variables.

Of course now you've got me interested so I am forced to look up whether I am wrong or not. I will undoubtedly be back to post if I am correct and will disappear from the thread if I am wrong. This is the internet.
posted by Justinian at 4:56 PM on May 8, 2010


I'm not sure how you can simply control for those kinds of "aspirational" choices, as in "let's just pretend that you were working 70 hours a week instead of 30 for the past five years; surely you'd be VP of Marketing by now...?"

That's what's insidious about these sorts of generalizations. Money isn't the only compensation out there, and everyone takes a job (or employs someone) on an individual basis. And now everyone feels bad.
posted by gjc at 4:57 PM on May 8, 2010


Okay, so let me clarify that you would be correct--that we could rely on R-sq--in the (impossible) case where we could explain 100% of the variance. In the real world, in studies such as this, the R-sq will always be fairly low, simply because the DV is so damn complicated to explain in its entirety. So, any incremental increase in R-sq tells you that you have explained more of the variance, but it should not give you any confidence that you have found more (or all) of the confounds, just as a seemingly small R-sq does not on its face indicate that you have missed confounds.
posted by found missing at 5:04 PM on May 8, 2010


Grgrgrgr.
(Pitfalls in) Interpreting R2 and Adjusted R2
...
3. A high R2 or R2 does not mean there is no omitted variable bias. (Nor does low R2 or R2 mean that there is omitted variable bias.)
I have two comments. First, I cling to the technicality that it is different to say that a low R2 means there is omitted variable bias and to say that a low R2 value may indicate there are omitted variables and that's my story and I'm sticking to it. And second, I'm going to punch found missing in the face.
posted by Justinian at 5:06 PM on May 8, 2010


Of course elsewhere we have:
Thus, the first diagnostic for omitted variables is a low R2 statistic. However, R2 less than 1 is a necessary but not a sufficient condition for omitted variable bias.
So I claim that my weak statement that the low R2 values here may indicate omitted variables cannot be proven wrong. Har.
posted by Justinian at 5:11 PM on May 8, 2010


However, R2 less than 1 is a necessary but not a sufficient condition for omitted variable bias.

As I said above in so many words.
posted by found missing at 5:14 PM on May 8, 2010


This is the part when I punch you.
posted by Justinian at 5:25 PM on May 8, 2010


:)
posted by found missing at 5:28 PM on May 8, 2010


How do the prevalence of misogynist patriarchal attitudes account for the established and increasing rate at which females outnumber males on college campuses?

Just because there is a gender gap in college enrollment (which I assume is what you mean by "on college campuses") does not mean we do not live in a misogynistic and patriarchal society. Even if we accept your argument that the actions of college admissions committees accurately reflect the cultural attitudes of our entire country (which I find extremely unlikely), the "gender gap" you point to still does not prove we're now egalitarian (or even matriarchal). The gap between men and women is far, far larger in college applications--which means a smaller percentage of female applicants are accepted and enroll relative to male applicants. I'm not saying this itself is obviously misogynistic, but it does mean that the on-campus "gender gap" is a misnomer, because a larger percentage of male applicants are able to attain education than are female applicants.

why is there no concern, if some notion of parity is a goal, at how male enrollment is not only lagging but falling further and further behind?

Because, according to the link above, male enrollment is actually increasing relative to applications and people are concerned about gender discrimination against female applicants.

If you want to shift the discussion to the gender gap in college applications, then I'm surprised to hear you think no one is concerned, because it seems to be bellowed from the rooftops fairly often.

Isn't this likely, over time, to erode and perhaps even invert the income gap?

If this "gender gap" has been in place since 1991, and the income gap has resisted erosion and inversion, perhaps that's a sign that some other pervasive force is keeping it in place--say, our misogynistic and patriarchal society?
posted by sallybrown at 5:35 PM on May 8, 2010 [5 favorites]


Malor, being less assertive about salary negotiations is certainly part of the problem, but there's a complicating factor that you've not elucidated. Women who are assertive about their compensation are less likely to be perceived as aspirational and ambitious and more likely to be perceived as pushy and selfish. (I actually had a manager tell me "Well, that's not very ladylike" when I straight-out asked him for a raise.) Those perceptions can negatively affect their ability to advance in their careers. You end up in a damned-if-you-do, damned-if-you don't situation, where you can be either thought of as a cutthroat bitch who cares about nothing but lining her own pockets, or you can seethingly settle for $5K a year less than your colleagues are making.
posted by KathrynT at 5:45 PM on May 8, 2010 [10 favorites]


The gap between men and women is far, far larger in college applications--which means a smaller percentage of female applicants are accepted and enroll relative to male applicants. I'm not saying this itself is obviously misogynistic, but it does mean that the on-campus "gender gap" is a misnomer, because a larger percentage of male applicants are able to attain education than are female applicants.
Are you sure that that is not due to very poor statistical analysis? A similar effect was shown to be an example of Simpson's paradox (not accounting for confounders) in this paper:

Examination of aggregate data on graduate admissions to the University of California, Berkeley, for fall 1973 shows a clear but misleading pattern of bias against female applicants.... [and] stems not from any pattern of discrimination on the part of admissions committees, which seem quite fair on the whole, but apparently from prior screening at earlier levels of the educational system. Women are shunted by their socialization and education toward fields of graduate study that are generally more crowded, less productive of completed degrees, and less well funded, and that frequently offer poorer professional employment prospects.

posted by esprit de l'escalier at 5:58 PM on May 8, 2010 [1 favorite]


Oh, and ferdinand.bardamu, if you're right that this "gender gap" in college enrollment has been in place for close to two decades, I would consider workplace inequality an even more disturbing sign of a misogynistic and patriarchal society, because your argument would mean that fewer and fewer men still manage to dominate the ever-growing number of women entering the workforce, at least when it comes to positions of workplace power:

A recent report by the White House Project shows that U.S. women are still just 3 percent of Fortune 500 CEOs, and less than a third of politicians and law-firm partners—despite being the majority of college graduates and a near majority of law-school grads. A woman may have won the Oscar for best director this past weekend—for the first time in the history of the Academy Awards—but 83 percent of the writers, producers, and directors of 2007's 100 highest-grossing films were men, a recent U.S.C. Annenburg study found; in those films, fewer than 30 percent of speaking roles were played by women. Women have made up the majority of college journalism majors since 1977 and two of the three network TV anchor chairs are now occupied by women, but female bylines at the 11 top political and intellectual magazines are still outnumbered by a rate of 7 to 1, according to the White House Project.

Apologies for sounding "too victimizing."
posted by sallybrown at 5:59 PM on May 8, 2010 [3 favorites]


Thus, the first diagnostic for omitted variables is a low R2 statistic. However, R2 less than 1 is a necessary but not a sufficient condition for omitted variable bias.

You guys are just making stuff up now, right? These are like statistician in-jokes and you guys are laughing at us. I'm sure of it.
posted by marxchivist at 6:10 PM on May 8, 2010 [1 favorite]


My understanding was that in practice you tended to find lower R2 in cases where important control variables were missing.

Only in highly deterministic processes with very low randomness and very low measurement error.

Dependent variables about human beings tend to be rather loosely deterministic, have a strong apparently random component, and are ridden with measurement error. Usually a low R2 just means that the thing you're predicting is more random.

More broadly, you're concerned about omitted variable bias.

But omitted variables affect the regression through their correlations with variables that are in the regression. If you have an omitted variable that is uncorrelated with any explanatory variable, omitting has no effect on the variables in the model.

So either being female directly causes lower pay, or being female is strongly correlated with other things that cause lower pay. But in that second case, being a women may itself be causing those other things, not merely correlated with them, and there would still be a strong policy interest in the pay gap.
posted by ROU_Xenophobe at 6:21 PM on May 8, 2010 [2 favorites]


Are you sure that that is not due to very poor statistical analysis?

I'm no statistician, so I'd be lying if I claimed to be sure. I'm also not clear on what that paper is saying, and how it applies to general college application and admission rather than specific graduate programs. In addition, it dates from 1975! That's only a few years after many major universities went coed. When it comes to college and graduate education, 1975 is like a different world.

That being said, did you read the articles I linked? The head of one admissions committee flat-out admitted (no pun intended) in the NYT that she accepted a smaller percentage of female applicants with equal numbers, and (though I know better than to assume that the L.A. Times would spot "very poor statistical analysis") their support seems to suggest this is not statistical error:

A 2007 analysis by U.S. News & World Report, based on the data sent by colleges for the magazine's annual rankings, found that the admissions rate for women averaged 13 percentage points lower than that for men. But percentages don't tell the whole story. It could be that the men were stronger candidates, or they might have applied in areas of engineering and science where women's numbers are still lower. But such justifications, even if true, are unlikely to fully explain these numbers. At schools such as the University of California, where admissions rely overwhelmingly on statistical measures of academic achievement such as grades and test scores, the disparities don't appear. Far more women than men applied to UCLA -- the UC's most selective campus -- last year. The university accepted about the same percentage of each, with a slight edge to the women. As a result, the freshman class has close to 800 more women than men.
posted by sallybrown at 6:23 PM on May 8, 2010



Are you sure that that is not due to very poor statistical analysis?

I'm no statistician, so I'd be lying if I claimed to be sure. I'm also not clear on what that paper is saying, and how it applies to general college application and admission rather than specific graduate programs. In addition, it dates from 1975! That's only a few years after many major universities went coed. When it comes to college and graduate education, 1975 is like a different world.


Their point still applies because the statistics you quoted are for overall admissions, and don't take into account the confounders mentioned later in the same article: that the applicants should be divided by department and applicant strength. The statistic over the whole presented with the intro "Consider some of the numbers at leading schools"(!) is not meaningful. ("Consider" what? The writer's statistical inaptitude?)

Also, this statement is begging the question: "But such justifications, even if true, are unlikely to fully explain these numbers." If you assume that taking into account applicant strength won't explain the discrepancy, then you are assuming gender bias, which is the thing you're trying to prove. You are supposed to try to take these things into account, even though it's very hard to do.

This kind of poorly-written article only gives fire to the convinced using illusory evidence.
posted by esprit de l'escalier at 7:13 PM on May 8, 2010


Regarding what the paper is saying: It is describing an example of Simpson's Paradox that appears in college admissions. Simpson's paradox is: an apparent paradox in which a correlation present in a group is reversed when the groups is split into components. The wikipedia article goes on to describe the paradox as introductory statistics.
posted by esprit de l'escalier at 7:29 PM on May 8, 2010


So either being female directly causes lower pay, or being female is strongly correlated with other things that cause lower pay. But in that second case, being a women may itself be causing those other things, not merely correlated with them, and there would still be a strong policy interest in the pay gap.

Yes, I agree with that. The example with regard to the possibility that aggressive negotiation is strongly correlated with being male would be an example. If the pay gap were due to women tending not to push as hard as men during negotiations, that wouldn't make it an unimportant thing to identify and address. And, of course, as someone above said there's the possibility that bosses interpret aggressive negotiation differently depending on the sex of the negotiator.

My experience with statistics has, indeed, been in large part with more deterministic processes with low measurement error. Which would explain my impression that low R2 values were more strongly associated with missing variables than appears to be the case.
posted by Justinian at 7:33 PM on May 8, 2010


Well, basic statistics about average salaries based on education levels suggests that women go on to higher education more than men because men have a significantly higher average pay for high school education only than women do, so it's more worthwhile for women to forego a few years of earning money than it is for men.
posted by jeather at 7:35 PM on May 8, 2010 [1 favorite]


The gap between men and women is far, far larger in college applications--which means a smaller percentage of female applicants are accepted and enroll relative to male applicants. I'm not saying this itself is obviously misogynistic, but it does mean that the on-campus "gender gap" is a misnomer, because a larger percentage of male applicants are able to attain education than are female applicants.

Yes, schools have implemented policies to mitigate the rapid pace at which college campuses are becoming predominantly female. One of the results is that a smaller percentage of female applicants are admitted and a larger percentage of male applicants are admitted. Duh. Why, however, are there so few equally qualified male applicants and so few graduates versus the apparent abundance of qualified female applicants and graduates?


If this "gender gap" has been in place since 1991, and the income gap has resisted erosion and inversion, perhaps that's a sign that some other pervasive force is keeping it in place--say, our misogynistic and patriarchal society?

Or generational change requires generations. On the anecdotal level, those in my personal cohort (mid-late 20s) who are female are certainly doing far better materially than the males. I fully expect most of my female friends and acquaintances to earn lifetime wages far in excess of mine and most of my male friends and acquaintances. This pervasive income inequality that's apparently driven by a deeply patriarchal and misogynistic culture is simply not something that is visible among those I associate with.
posted by ferdinand.bardamu at 7:35 PM on May 8, 2010


If this "gender gap" has been in place since 1991, and the income gap has resisted erosion and inversion, perhaps that's a sign that some other pervasive force is keeping it in place--say, our misogynistic and patriarchal society?

Furthermore, it is flat out wrong to say that the income gap has resisted erosion. Trends in male-female income disparity have definitely shifted toward parity over the past thirty years.
posted by ferdinand.bardamu at 7:42 PM on May 8, 2010


The gap between men and women is far, far larger in college applications--which means a smaller percentage of female applicants are accepted and enroll relative to male applicants. I'm not saying this itself is obviously misogynistic, but it does mean that the on-campus "gender gap" is a misnomer, because a larger percentage of male applicants are able to attain education than are female applicants.

To reiterate here since I wasn't particularly forceful earlier: What you are complaining about are the results of specific admissions policies implemented to stem the growing sex ratio disparity.
posted by ferdinand.bardamu at 7:45 PM on May 8, 2010


Furthermore, it is flat out wrong to say that the income gap has resisted erosion. Trends in male-female income disparity have definitely shifted toward parity over the past thirty years.

I didn't say nothing has changed. Considering we're talking about the last twenty (nineteen, actually) years (1991-2010), I disagree it's "flat-out wrong" to say the pay gap has resisted erosion, when, with all the societal changes we've had and how far women have supposedly come, we've only (at least according to your Wiki link) moved from 71.6% to 76.5%. Clearly there's some resistance going on, whether conscious or unconscious.

This pervasive income inequality...is simply not something that is visible among those I associate with.

That's all your anecdote illustrates.
posted by sallybrown at 7:58 PM on May 8, 2010


That's all your anecdote illustrates.

Yes, undoubtedly. However, my anecdote was corroborated in the link:

"though young women have started to outearn young men in some large urban centers with young women earning up to 20% more than their male counterparts [3]."

This reinforces my supposition that there is not just erosion but a growing trend toward an inversion among the youngest cohorts. Whether this continues and broadens to other cohorts and areas remains to be seen but it is difficult to square this with notion that systemic injustice continues to be a formidable impediment in the way of female income equality.

Unfortunately, times change but orthodoxy never yields an inch.
posted by ferdinand.bardamu at 9:10 PM on May 8, 2010 [2 favorites]


Considering we're talking about the last twenty (nineteen, actually) years (1991-2010), I disagree it's "flat-out wrong" to say the pay gap has resisted erosion, when, with all the societal changes we've had and how far women have supposedly come,we've only (at least according to your Wiki link) moved from 71.6% to 76.5%. Clearly there's some resistance going on, whether conscious or unconscious.

1991 Was simply the year females eclipsed males in college attendance and graduation. The gap began to close beginning in 1970. It's therefore disingenuous to ignore the closing of the income gap during the entirety of the time. The link gave numbers from 1980 to 2004 (from 1970 wasn't available to me) showing the total gap closing from 60.2% to 76.5%. And these are overall numbers, not adjusted for industry, position, experience, education, etc. This cannot be termed anything other than erosion of the income inequality gap.

I am also curious about your proposals regarding the elimination of these unconscious factors you posit are interfering with women closing the income inequality gap. How are they to be identified and what measures do you suggest be implemented to cure them?
posted by ferdinand.bardamu at 9:21 PM on May 8, 2010


these unconscious factors . . . How are they to be identified and what measures do you suggest be implemented to cure them?

The Washington Post article you linked, early in the thread, cites other studies aimed at identifying them:

. . . women's reluctance was based on an entirely reasonable and accurate view of how they were likely to be treated if they did. Both men and women were more likely to subtly penalize women who asked for more -- the perception was that women who asked for more were "less nice".

Advice to individual women from one of the authors of that study: [Women face less backlash if they use] “relational accounts” that justify the salary request while communicating concern for organizational relationships are effective for women in achieving both social and economic objectives in compensation negotiations. Bowles emphasized that the goal of this research was not to solve gender inequality in organizations by “fixing the women,” but rather to suggest that women can play a role in “fixing the gender inequalities” through their own interpersonal interactions.

Human Resource Executive Online suggests that HR departments 'educate those involved in the interviewing process [and] "Of course, the organization then needs to foster a culture which is clear on, and consistently rewards, desired behaviors, regardless of gender . . . Otherwise, an individual may be hired but will end up not being 'successful' there, which may result in turnover, lost productivity and a lack of faith in management's ability to bring in the right leaders."'
[I'd have liked for more detail on this bit: "when [male applicants] affected a more cooperative demeanor. . . . they "were viewed as less socially skilled than female" applicants with the same demeanor, "suggesting that atypical men may also risk social backlash" and they received the lowest ratings of all." Same coin, different side. Googling suggests there are studies in the works, but nothing ready for publication yet.]

Babcock (quoted in the Washington Post link) observes in one of her books that the status quo (women's disinclination, generally speaking, to ask, and the backlash that many face when they do) unfairly penalizes women for a social problem; in a blog post, she goes into what is perhaps a more broadly persuasive angle, the considerable costs to a company's productivity of passively accepting this status quo. What Managers Can Do about it.
posted by cybercoitus interruptus at 12:44 AM on May 9, 2010 [3 favorites]


Mrs. Pterodactyl:Could you please provide a citation for this, other than "the women I know"? If we're going to play Anecdote Wars, none of the women I know place particularly much value on a potential partner's income.

Of course. In this video, a Stanford professor lecturing on human behavioral biology references a study by David Buss which canvassed around 50,000 people across a variety of cultures and revealed (among other things) that women generally expressed a greater attraction for men with higher earning potential. @1:39:15

And for good measure, here are a couple more supporting quotes:


From an APS article: "A forthcoming study conducted by Günter Hitsch, Ali Hortaçsu (both at University of Chicago), and Dan Ariely (Duke University) confirmed existing evolutional theory, finding that in a sample of 22,000 online daters women weigh income more than physical attributes, including facial attractiveness, height and body mass index, when deciding who to contact (Hitsch et al., 2009)."

Also: Dr. Michael Dunn discussing his study published in the British Journal of Psychology: “There’s a wide variety of evidence that does suggest that females are more influenced by wealth and status. It’s not a recent phenomenon. It is very ingrained and the evidence is not just anecdotal."

So there you have it. I didn't originally include a citation because I felt that it was a generally accepted, uncontroversial concept. And just to be clear, I didn't make that first comment as some sort of chauvanist statement against women, I just think that if we're going to discuss the income gap we shouldn't ignore that men and women, often and generally, may have different motivations for pursuing a higher salary. It stands to reason that if one group's attractiveness as potential mates is entangled in their income, that group has at least one reason to pursue a higher salary more intensely than a group whose attractiveness as mates doesn't depend on their income.
posted by millions at 12:55 AM on May 9, 2010 [2 favorites]


To my knowledge, women have never been paid as much as men and still aren't. I remember 2 years ago in my freshman year when taking Sociology 100, we saw figures that had shown women are only paid 76% of what men are in most jobs. I'm not saying it's right, which it's not, but it is a well known fact.
posted by Chocomog at 4:28 AM on May 9, 2010


with all the societal changes we've had [since 1991] and how far women have supposedly come,

Funny thing is, speaking from my local perspective in Canada, in Ontario, and in Toronto, we've gone backwards since then.. Believe it or not, the Ontario Government of the time was bringing in legislation to ban scantily clad women from advertising.
posted by Chuckles at 12:46 PM on May 9, 2010


I feel a need to preface my remarks. I am completely in favor of equal work for equal pay. I think there have been good comments in this thread about how prevalent this is, and I believe--having seen it-that it is very real.

I do not see any way in our crypto-capitalist economy that this can be legislated or enforced. As has been noted upthread, it is in a business's best interest to hire the best employees for the least pay. There certainly are cultural factors that have influenced the apparent facts that men are offered more and that the gap increases over time. Missing above is any real process for righting this wrong. Teaching women to expect more, to ask for more? No. Because it is also ingrained in business to not talk about how much one is compensated. Will business open up their hiring to say, we have this job and it pays THIS? In fact, aren't MBAs by and large trained to push this sort of pay inequity all the way down? (I'm asking, I'm asking.)

Yet a control economy that pays X for job Y with transparency is hardly efficient.

Fwiw, pay in companies is often unequal based on all sorts of intangibles that can't be controlled for. I know it's anecdata, but in two companies I worked in, management & marketing personnel recruited from a set of colleges preferred by the president or senior management were paid more than people doing the same job that came from other schools or who rose from the factory or office ranks. Male or female. Hardly fair. But capitalism does not easily allow for pay equity. To suggest that there is something that can be done short of tearing it all down and starting over.
posted by beelzbubba at 4:39 PM on May 9, 2010


But capitalism does not easily allow for pay equity. To suggest that there is something that can be done short of tearing it all down and starting over.

In today's world, or even simply the blue for example, when one aspect of the excesses of an obviously dysfunctional and seriously need of overhaul system or the other is brought up in as disparate threads as oil spills and women MBA's salaries, perhaps its time to step back far enough to take a critical look?
posted by infini at 4:48 PM on May 9, 2010


perhaps its time to step back far enough to take a critical look?

I'd say long past time. I'm almost 60, and I'v been urging that critical look for about 40 years now. I'll keep urging it as long as I can. I'd also say that the front lines of that critical look ought to be those who are getting those MBAs--I'd like to think that they'll focus on something other than greed.
posted by beelzbubba at 6:38 PM on May 9, 2010


ferdinand: I am also curious about your proposals regarding the elimination of these unconscious factors you posit are interfering with women closing the income inequality gap. How are they to be identified and what measures do you suggest be implemented to cure them?

beelzbubba:I do not see any way in our crypto-capitalist economy that this can be legislated or enforced. As has been noted upthread, it is in a business's best interest to hire the best employees for the least pay. There certainly are cultural factors that have influenced the apparent facts that men are offered more and that the gap increases over time. Missing above is any real process for righting this wrong.

I can't help but agree, beelzbubba. I feel hopeless. So many people tried so hard for so long.

This thread is timely. From the 1970s-90s, my mother worked her hairsprayed, white collared, power suited ass off in hopes of some bullshit title like "Executive Vice President in Charge of Money" in her Forbes 500 company. She worked full time and then came home and cried because she wanted see us more but she knew going part time would end the dream of a corner office that kept her going through years of jobs filled with assholes who acted like her MBA meant Master of Barista Administration. So she stood firm and stayed full time, until she got passed over for a promotion because she was told they were afraid she'd go part time. You know, because she had kids. Could she have sued? She didn't care. She was tired of feeling like a bad mother at home and a humorless shrew (i.e feminist) at work, so she went part time and took a disproportionally steep pay cut (because allowing her to work part time was "doing her a favor") and then came home and still cried because now she felt like she had betrayed a gender's-worth of people. Her lack of face time with us in our early years was for naught. Should she have been demanding raises? Hell yes. But she likes to be liked and she was tired of being considered a bitch at work and at home (we were teenagers...). Our family didn't need the money (so this doesn't even begin to touch how hard it is for poorer women, or women without partners like my dad who help out at home). She'd already lost her chance at the brass ring. She took the layoff option they offered her a few years later.

This is as anecdotal as you can get (though common at her company), and I realize that, as well as the fact that my mother is not the perfect test case or example of the very real income gap that persists when all other factors are held equal. Frankly, I think the system is fucked and has been for a long time. Yes, it has gotten better, since we started from a place of not being considered people. My gender didn't prevent me from becoming a lawyer, like it would have in the past. Woo! But it feels like more and more women are being pressed up against that glass ceiling and suffocating more quietly, because things are supposed to be better.

I have no clue how to fix it, ferdinand. But it's there, just waiting for me to rise high enough, so I'm sure as hell not going to stop pointing it out.
posted by sallybrown at 7:26 PM on May 9, 2010 [4 favorites]


Just a thought:

While we might debate how the current circumstance has come about I don't find anyone here advocating the status quo.

It's a good thing that we try and diagnose this, it profits us I think. If we don't identify the actual cause then similar to arresting someone innocent of a crime we double the error by arresting the wrong criminal. The problem goes unaddressed while the cause evades.

Difficult sure, and not angry free. But I'm not seeing anyone here arguing that the status quo should remain.

Signed,
Middle Aged White Man
posted by vapidave at 7:51 PM on May 9, 2010 [1 favorite]


I had an eye opening experience early in my working life. I was asked to sit in on a round of interviews for a job opening for someone that would be working closely with me. There were 11 candidates, three of them male, all of them were just out of undergrad. After the interviews, three of us, myself, another female coworker, and my male boss discussed all the candidates. We all agreed on our first choice, one of the male applicants. He had a lot of experience and probably was the best choice. He was offered the job but turned it down because he had a better offer. We went back and forth over all the candidates, with our boss liking one male candidate and myself and my coworker liking another male candidate. Eventually the guy we liked was hired and it turned out he sucked at the job.

But the really important thing was WHY all the male applicants were seriously considered and none of the female applicants were. All eight of the female applicants were well dressed, polite, answered questions well, and behaved very professionally. The problem was that we couldn't identify anything about them to distinguish them from one another. They all seemingly lacked the self-confidence to relax and be themselves. Because what we were really trying to determine was not how well qualified someone was, but what their personality was and whether they would be good to work with. The men on the other hand, had no qualms about advertising their strengths, making jokes, and in other ways putting themselves forward. We were able to get a sense of who they were. But the women seemed to be too focused on being polite and modest that we passed over a bunch of people who would almost certainly been better at the job, in favor of someone who showed more of their personality in interviews.

I truly think that the different ways women are socialized play a major role in holding back women's professional achievement.
posted by threeturtles at 8:17 AM on May 10, 2010 [1 favorite]


"[...]you can be either thought of as a cutthroat bitch who cares about nothing but lining her own pockets, or you can seethingly settle for $5K a year less than your colleagues are making."
posted by KathrynT at 2:45 AM on May 9 [10 favorites +]


Chose cutthroat bitch who cares about nothing but lining her own pockets.

The corporate world / office politics = full contact blood sports.
posted by ruelle at 9:02 PM on May 10, 2010


beelzbubba: it is in a business's best interest to hire the best employees for the least pay

In the short term. In the long term (from my What Managers Can Do link above),
Turnover is really expensive, says Steve Sanger, chairman of General Mills. "If we've invested in recruiting and developing good people, then we want them to stay." On average, replacing an hourly worker costs an organization 50% of that worker's annual salary; replacing a professional worker costs 150% of her annual salary. Why so much? Add up screening and hiring costs, opportunity costs for the employees doing the hiring, and lost productivity until the replacement worker gets up to speed, and you're talking about a lot of money.

Factor in the low morale of employees who have to pitch in while replacements are found and trained, and the costs skyrocket. Our calculations show that turnover costs can have a huge impact on the bottom line of a typical midsize company--costing as much as 3.4% of revenues and an astounding 45% of profits.
Implementing Babcock's suggestions from that link would involve altering workplace culture norms (eg, mentoring including specific negotiation coaching for all aspects of the mentee's job, not just salary; fostering a workplace culture that exposes unfounded stereotypes of "pushy" women vs "go-getting" men, rather than letting them slide). That would have its own difficulties, but perhaps not as entrenched as what would be involved in salary transparency.

eg, Deloitte's proactive efforts to retain and advance women into corporate leadership seem to be paying off. OK, the links don't mention "salary negotiation skills" specifically, but I would think that would be covered in their mentoring and coaching programs.
posted by cybercoitus interruptus at 11:24 PM on May 10, 2010


Chose cutthroat bitch who cares about nothing but lining her own pockets.

But here's the point your missing. That option doesn't get you the same advancements that the same behaviors get men. It gets you passed over for promotions, dinged in your reviews for being "not a team player," and often hustled right out of the power circles entirely. Every time? No, but a lot of the time. In many workplaces, there's simply no way for women to win.
posted by KathrynT at 11:54 PM on May 10, 2010 [1 favorite]


transona5: The study controlled for aspiration levels.

My point depends on this, in fact (at least where this particular study is concerned). Assuming that men and women aspired to the same goal in the workplace, I wished to bring up the fact that if their motivations might generally be different, that would probably affect their eventual success.

It should be noted that the study's authors' idea of controlling for aspiration levels regarding salary seems to be to consider how men and women who aspired to "the CEO/ senior executive level" fared against each other in terms of compensation. I believe that means that if a certain group were to have more modest expectations of earning for the same position, then the control could be essentially misleading. I would think that if you want to control for aspiration in a study that looks at disparities in eventual salary, you have to ensure that your respondents have the same target salary rather than the same target job.
posted by millions at 12:12 AM on May 11, 2010


beelzbubba: it is in a business's best interest to hire the best employees for the least pay

In the short term. In the long term (from my What Managers Can Do link above),


You are right, of course. In the finest tradition of GW Bush, I misquoted myself. What I should have said was, businesses, and managers, and HR people believe it is in their best interest to...etc.
posted by beelzbubba at 8:13 AM on May 11, 2010


Ooops, mislinked within my own damn post. Sorry cybercoitus interruptus for attributing your remarks to millions.
posted by beelzbubba at 8:15 AM on May 11, 2010


My first really profesional job made me $27,000. At the same time I had a coworker making the same amount. She indicated to me that she should make more because her job had more responsibility or something (it didn't...I even had employees under me, etc.)

Anyway, I was in the office one day working on my budget for the next fiscal, and we were looking at bonuses and raises and etc. Our insurance premium had just recycled and gone up right at 20%. So she's asking for a 6% raise for the year, we had been promised I think 3%. Anyway, so I'm looking at my insurance info, and my insurance cost our organization right at $4,000 a year. She was a 25 year old female---hers cost just under $13,000.

So right off the bat, her annual compensation package was $40,000, whereas mine was $31,000. I realize that we all deserve health care or whatever, but my point is valid I think. I'm interested to know if the alarmist numbers on salary disparity include things like insurance offsets and whatnot.
posted by TomMelee at 9:26 AM on May 11, 2010


Thanks for the (part of an) anecdote.
posted by Pope Guilty at 10:12 AM on May 11, 2010


"But here's the point your missing. That option doesn't get you the same advancements that the same behaviors get men. It gets you passed over for promotions, dinged in your reviews for being "not a team player," and often hustled right out of the power circles entirely. Every time? No, but a lot of the time. In many workplaces, there's simply no way for women to win."
posted by KathrynT at 8:54 AM on May 11


In my humble experience, in corporate life, what gets a woman the right promotions and into "power circles" is that she is ready to engage in office politics:
- 1: giving favors;
- 2: investing in long term personal relationships in order to gain allies in the highest corporate offices;
- 3: making her name reknown beyond her immediate corporate environment (both to the competition and the media).

Essentially, it all boils down to your ambition. It's OK not to want to want to be that person that seeks friends in high places for future favors. However, if you want to be a member of the board or its chairman, you have to -not only- "play the game" but to throw yourself into office politics.
posted by ruelle at 3:26 PM on May 11, 2010


« Older C.W. Roberts Contracting in Florida has come up wi...  |  Everybody Edits... Newer »


This thread has been archived and is closed to new comments