How The CONSAD Report On The Wage Gap Masks Sexism, Instead Of Measuring It

[Crossposted at Alas and TADA.]

A recent comment by Robert cited the same Bush-administration-commissioned study that Ballgame of  Feminist Critics once cited. Ballgame wrote:

Except that they’re all pimping the Big Lie by misusing statistics. If you take motherhood (and time away from work) by itself that doesn’t explain the Gender Gap. And if you take professional field by itself that doesn’t explain the Gender Gap. And if you take the fact that women work in vastly safer occupations by itself that won’t explain the Gender Gap, either. But if you take all those factors together (along with some others, like men working longer hours on the job)? Well, then you find that women make at least 93%-95% of what men make, according to statistical analysis by the CONSAD Research Corporation.

In other words, no substantive evidence for a Gender Gap! Christina Hoff Sommers has more. (And yes, as a progressive I’m deeply embarrassed by the fact that the fucking American Enterprise Institute has straighter talk about this issue than many of my progressive sisters.)

There’s a lot wrong with Ballgame’s argument — and not just his belief that an anti-feminist like him is in any meaningful sense a progressive.1

1) Typically, neither Robert or Ballgame mentioned that CONSAD used a radically different sample of workers than virtually all other wage gap studies. I don’t think either of them were being deliberately dishonest; rather, I suspect they both lack basic knowledge of what they’re talking about.

The standard wage gap figure only includes full-time, year-round workers,2 but CONSAD included part-time (and, I suspect, short-term) workers, which means it’s looking at a significantly different population. So Ballgame’s comparison of CONSAD’s results to a standard figure from Newsweek is apples and oranges — or, as Ballgame puts it, “misusing statistics.”

Why does this matter?

2) The single largest factor that CONSAD found “explained” the wage gap is the difference in hours worked. Since women are more likely to work part-time, and since CONSAD (unlike standard wage gap studies) included part-time workers in their sample, in effect CONSAD is comparing mostly female part-time workers to mostly male full-time workers. Then — what a surprise! — they determined that the difference in hours worked accounts for a huge portion of the wage gap they measured.

3) Ballgame’s argument is that CONSAD study shows that the wage gap is not caused by sexism — or at least, that no more than 5-7% of the wage gap is caused by sexism.

But that’s not a reasonable interpretation of CONSAD’s results.

First of all, there are important kinds of direct employer discrimination which CONSAD’s methods cannot measure or disprove. For example, some employers are more likely to hire women to lower-paid positions and men to higher-paid positions. (Empirical testing – by sending male and female testers to apply for the same jobs — has proven that this sort of sexist occupational sorting sometimes happens.)

This sort of occupational segregation leads to women’s average work experience not being as good as men’s — which CONSAD’s methodology would classify as an “explained” difference in wage gap that has nothing to do with discrimination. It would be more accurate to conclude that the differences in women’s and men’s resumes may be partly caused by employer discrimination, and CONSAD’s methods cannot account for this.

Similarly, if employers are less likely to promote women (all else held equal), that would contribute to women being paid less overall — but would CONSAD’s study would, again, consider that explained and therefore not discrimination.

4)3 For me, probably the most important kind of sexism going into the wage gap is the sexism of unquestioned assumptions; unquestioned assumptions about who does the housework, unquestioned assumptions about who does the child-rearing, unquestioned assumptions about innate ability, and most of all, unquestioned assumptions about how jobs are designed for people with wives at home.

I call this last factor the “Father Knows Best” economy; most jobs implicitly assume that workers have wives at home who are taking care of the kids and house, so that these responsibilities never need to be accommodated by employers. Maybe that assumption made sense half a century ago, but it doesn’t make sense now; and by continuing to implicitly make this assumption, our economy is making it unfairly difficult for caretakers (who are usually women) to have careers.

Ballgame’s big mistake is assuming that sexism in the wage gap (if it exists at all, which he denies) is entirely a matter of women being paid less than men for identical jobs. But most economists who study the wage gap believe that it’s caused, to a significant extent, by occupational segregation, which means women and men are sorted by the market into different jobs – and the women’s jobs, on average, pay less.

Arguments like Ballgame’s implicitly see the wage gap as all or nothing; either the wage gap is caused by employers hiring women at lower wages for the identical job, or else sexism and discrimination have nothing to do with the wage gap. But this is such a foolish and unsupportable model of how sexism works in the labor market, that there’s no reason at all to use it, unless one is either totally ignorant of real-world labor economics, or seeking a way to rationalize away sexism against women.

When discussing direct employer discrimination, it’s more realistic to discuss elements like selective hiring, training, promotion ladders, and other things that are a good deal more complex than CONSAD’s vision of the labor market allows for. Given two equally able applicants for a $40,000 job, one male, one female – which one will employers tend to prefer? Once hired, who is more likely to get mentored? Who is more likely to be given the assignments that lead to promotion? Who is more likely to be perceived as doing good work, all else held equal? And if these factors mean that women are rewarded less than men for identical labor market participation, to what degree does that reduce women’s incentive to participate equally in the labor market? All of these are ways that sex discrimination actually happens in the marketplace — and none of them are detectable by by CONSAD’s methods.

5) Ballgame claims that if women are paid “at least” 93-97% of what men are paid (as CONSAD’s biased study found), that means there’s “no substantive evidence” of a gender wage gap. Even if we accept CONSAD’s results — and we shouldn’t — I don’t believe for a moment that if one out of every twenty dollars Ballgame earned were withheld from him because of discrimination, Ballgame would consider that irrelevant.

6) Ballgame implies that “the Big Lie” is that studies that account for multiple factors (occupational difference, danger4, motherhood, experience, hours worked, etc) would not find a significant gender wage gap, and only studies that fail to control for multiple factors find a significant gender wage gap.

This is breathtakingly ignorant; if Ballgame knew the first thing about wage gap studies, or had bothered to do even ten minutes of research, he’d realize his argument is ridiculous.

I’ll list just a few of the many gender wage gap studies that disprove Ballgame’s “Big Lie”: Wood, Corcoran & Courant (1993), Journal of Labor Economics; Dey & Hill (April 2007), American Association of University Women Educational Foundation; “Women’s Earnings” (Oct 2003), United States General Accounting Office; Blau & Kahn (June 2006), Industrial and Labor Relations Review; Mandel & Semyonov (Dec 2005), American Sociological Review; Boraas & Rodgers (March 2003), Monthly Labor Review; Johnson & Solon (Dec 1986), American Economic Review; Mulligan & Rubinstein (August 2008), Quarterly Journal of Economics; Fields & Wolft (Oct 1995), Industrial and Labor Review.

7) Robert allows that in the 1980s, there was actual discrimination against women (unlike today). But conservatives and anti-feminists in the 1980s were making the same arguments denying discrimination that Robert and Ballgame make today, except that in the 1980s they would have said that discrimination existed in the 1960s (but not today). Twenty years from now, conservatives and anti-feminists like Robert and Ballgame will be saying that discrimination existed in the 2010s, but certainly not now in the 2030s.

  1. Imagine someone saying “I’m a progressive, I just think that big business is right about workplace economics, not labor unions” or “I’m a progressive, I just think that “The Bell Curve” has straighter talk about race than activists of color do.” []
  2. For example, this page is currently the top google search result for “gender wage gap.” Note that all the graphs specifically look at “full time year round” workers. []
  3. I recycled some of this from a post I wrote in 2005. []
  4. Ballgame apparently doesn’t realize that if there is a danger pay premium, which there isn’t, then it would be included within occupational difference and thus accounting for it separately would be double-counting. []
This entry posted in crossposted on TADA, Gender and the Economy. Bookmark the permalink. 

50 Responses to How The CONSAD Report On The Wage Gap Masks Sexism, Instead Of Measuring It

  1. 1
    Robert says:

    The standard wage gap figure only includes full-time, year-round workers,2 but CONSAD included part-time (and, I suspect, short-term) workers, which means it’s looking at a significantly different population.

    So the standard wage gap figure is discussing only a subset of women, while the CONSAD study is discussing the actually existing population of workers? I’ll grant you that it’s comparing apples to oranges…but if I’m concerned about the state of women’s employment, doesn’t it make sense to look at all women? Including the ones working part-time or seasonally?

    I agree with you about the many social factors that go into discrimination, and how these factors combine and conspire to lower women’s vocational performance and outcomes. But my comments were made in the context of a critique of a specific law, a law that basically requires equivalence of outcome without equivalence of input. Yes, it is unfair and sexist that Sally got screwed over by her guidance counselor and that she couldn’t get the same jobs I got in our respective work histories, and as a result, now our salaries are different.

    The fix is not to require me and Sally get the same wage, the fix is to go back to the roots and work on the origins of the problem, going forward.

    I’m NOT a progressive, and one of the big reasons for that is progressives so often jump to the end of long, contingent processes and decide that the outcomes should be adjusted by decree. However well-intentioned, the social outcomes of this are almost always disastrous. (Problem: racially identifiable groups of mostly-poor people have bad primary and secondary schools, leading to low admission rates at high-end colleges and universities. Progressive solution: quotas and preferences to require unprepared students be admitted to places where their failure is virtually guaranteed! Don’t worry, though, when people start noticing this, we’ll just call them racists so they’ll shut up.)

    That aside, maybe you could do people like Ballgame, who do want to be thought of as progressive, a solid and put together a list of all the dogmas one must endorse to be progressive. You already mentioned that you’re not allowed to question the economic genius of the UAW, or measure human intelligence. What are the other thoughtcrimes?

  2. 2
    Robert says:

    Also, I do not say (and I don’t think that CONSAD says) that only 5 to 7 percent of the gap is explained by sexism.

    Rather, only 5 to 7 percent points are putatively identifiable as being caused by direct discrimination of the sort that is susceptible to remedy via ex-post-facto adjustment.

    I don’t think one dollar in twenty is so inconsiderable as to be ignored, and you’re quite right that I would howl to the heavens for my dollar – but we’re measuring social costs and benefits from specific legislation.

    It might well make sense to upend settled expectations of proof in the face of severe direct discrimination – if women are getting paid half the wage of identical men or something like that. The case for that kind of radical change diminishes as the severity of the direct discrimination diminishes.

  3. 3
    Ampersand says:

    Before I consider answering that, Robert, let me ask — is there any belief a person could hold which would cause you to not think of that person as a “conservative”? Or is it your view that there are no beliefs incompatible with conservatism as you understand it?

  4. 4
    S.K. says:

    Before I took a cushy office job getting paid at least ten percentage points less than my male colleagues, some of whom have less education and experience in their fields than I do, I worked with a group of mostly men, in a hazardous/dangerous work environment. We used to make jokes about the ten-cent-per-hour pay differential for our hazardous work environment. It was funny because it didn’t exist. We were, in fact, paid less than our colleagues working “safer” jobs. The corporate logic is such that if you are desperate enough to take a dangerous job, you are desperate enough to accept lower pay. Capitalism does not reward people who work harder, or under more hazardous circumstances (like people who work in slaughterhouses, in restaurants, in sewers, etc.). That’s a lie your parents told you.

  5. 5
    gin-and-whiskey says:

    Typically, neither Robert or Ballgame mentioned that CONSAD used a radically different sample of workers than virtually all other wage gap studies….The standard wage gap figure only includes full-time, year-round workers,2 but CONSAD included part-time (and, I suspect, short-term) workers, which means it’s looking at a significantly different population.

    Huh. I didn’t know that either.

    But I guess we have three different reactions. Conservatives say “see? no wage gap!” Progressives say “well, if it doesn’t show a wage gap it’s wrong.” I ask “Which sample type is better, and why? “Because it’s the one we always used” isn’t a good reason to choose a particular method unless you need to make a long term comparison.

    The single largest factor that CONSAD found “explained” the wage gap is the difference in hours worked.

    Now hold on one gosh darn moment. Are you saying that “wage gap” shouldn’t take account of hours worked? Yeah: work 20 hours, and you make less than working 25 hours. Why would that be sexist?

    For example, some employers are more likely to hire women to lower-paid positions and men to higher-paid positions. (Empirical testing – by sending male and female testers to apply for the same jobs — has proven that this sort of sexist occupational sorting sometimes happens.)

    That is employment discrimination. It’s not, however, necessarily employment WAGE discrimination.

    This sort of occupational segregation leads to women’s average work experience not being as good as men’s

    Hey, what happened to comparing apples and apples? We’re not talking about work experience here,we’re talking about pay. Work experience and pay are different things.

    Again: worse experiences would be discrimination. But noy PAY discrimination. “The world is sexist” can’t be justification for “…therefore we can use whatever method we want to improve the lot of women.” The discrimination you’re citing and the solution you’re proposing should match.

    For me, probably the most important kind of sexism going into the wage gap is the sexism of unquestioned assumptions; unquestioned assumptions about who does the housework, unquestioned assumptions about who does the child-rearing, unquestioned assumptions about innate ability, and most of all, unquestioned assumptions about how jobs are designed for people with wives at home.

    Sure. And these things have, generally speaking, everything to do with society and very little to do with employer wage discrimination.

    I call this last factor the “Father Knows Best” economy; most jobs implicitly assume that workers have wives at home who are taking care of the kids and house, so that these responsibilities never need to be accommodated by employers.

    That’s total BS. There are a host of reasons why employers would act like they do, only one of which is what you propose. You’re imputing motives to employers which are ridiculous. Sure, most jobs simply don’t offer the worker-centric stance that you have projected would “need” to be adopted if your idea wasn’t followed. So what?

    Ballgame’s big mistake is assuming that sexism in the wage gap (if it exists at all, which he denies) is entirely a matter of women being paid less than men for identical jobs.

    Wow. How on earth could ballgame have made that assumption?…..

    Oh, *I* know. It must have been all the “equal pay for equal work!” stuff. And the “paycheck fairness” stuff! And the “don’t get paid as much for doing the same work as a man” stuff!

    Wait a minute… are those lies? Are you saying that the complaint was “equal pay for some other different kind of work which women get steered into due to societal sexism?”

    Because that is true, and relevant, and important. But it’s sure as hell not what people SAY. In a post discussing deceit, that seems pretty relevant, dont’cha think?

    But most economists who study the wage gap believe that it’s caused, to a significant extent, by occupational segregation, which means women and men are sorted by the market into different jobs – and the women’s jobs, on average, pay less.

    A problem to be sure, but NOT a problem of “paycheck fairness” to be sure.

    And of course the degree to which it is a problem is tricky to find out. Some of it is in fact a choice; jobs have a lot of different costs and benefits. Working part time sucks (you make less) and also rules (you work less) Etc. And some of it is a choice which is made in school, long before work is even an issue. I’m all for expanding women’s studies departments if that is what people want, but in terms of future paychecks those WS majors would probably be better off majoring in biochem or engineering.

    When discussing direct employer discrimination, it’s more realistic to discuss elements like selective hiring, training, promotion ladders, and other things that are a good deal more complex than CONSAD’s vision of the labor market allows for.

    …or that the paycheck fairness act has a chance in hell of addressing. Yes? Limited studies have limited uses. Limited solutions, likewise.

  6. 6
    Robert says:

    Don’t shift. You didn’t say “believes” in your dig at Ballgame, you said “thinks”.

    There are lots of beliefs that go into a political philosophy, and yes, there are plenty of beliefs that would be incompatible with being a conservative (or a progressive, or a liberal, or a Jainist). “I believe the government ought to be the provider of most goods and services” is a non-conservative belief, for example. “I believe that women ought to stay home and raise babies and cook” is a non-progressive belief.

    “Thinks” goes to analysis. Regardless of your beliefs, analysis is either valid or invalid, depending on the logic and the correctness of any empirical facts asserted or relied on. I can believe “the government ought not to be involved in setting wage rates” and also think “studies have demonstrated that minimum wage rates at a certain level do not damage economic growth.”

    Offhand I can’t think of an analytical conclusion, based on correct data and functional logic, that a person could reach which would disqualify them as being a conservative. I’m tentatively willing to say that there are no such conclusions; data is data.

    You seem to be saying something different, however. Whether the Bell Curve’s analysis is correct or incorrect isn’t material; it’s objectionable for a progressive to think it’s correct. Whether the UAW or Toyota management’s economic theories produce superior or inferior results isn’t material; it’s objectionable for a progressive to “side” with management, a priori.

  7. 7
    Robert says:

    Shorter me: I think if I convince a progressive, through rigorous analysis, that market prices lead to better efficiency and more jobs than government-set prices, he can still be a progressive.

    If I convince a progressive, through emotional rhetoric, that government is evil and taxation is theft and women should stay home and cook, he isn’t a progressive anymore.

  8. 8
    Schala says:

    For example, some employers are more likely to hire women to lower-paid positions and men to higher-paid positions. (Empirical testing – by sending male and female testers to apply for the same jobs — has proven that this sort of sexist occupational sorting sometimes happens.)

    From my experience in no-promotion-possible-entry-level-jobs (warehouses and data-entry, jobs in retail etc), men are preferred overall, in more physical jobs, while women are preferred overall, in more clerical or other less physical jobs (cashier, client service).

    For example, say I apply for an office job in a small-medium business that owns a big warehouse to supply its branch stores and the ones it recently bought to add to its already existing chain (of then totalling 35 stores). They don’t list it in the add, but they’re also looking for warehouse help, at the same wage. Both jobs require no experience and are at the same physical building (the office is just above the one store that is there, while the warehouse is besides/behind it).

    If I apply as male, they’ll strongly insist that I instead take the warehouse job, without having seen me in person, assuming that I can defacto lift 40+ lbs stuff on a daily full-time basis without problem or much effort (I got a work accident due to my back, while lifting 50 lbs stuff – they tried to fire me for it). I’m not making more money, I’m not being promoted (can’t be), and I’m doing work I find difficult (given my size and strength) instead of work I find more intuitive (like working on computers).

    This also applies in big chain retail stores that have big backstores or warehouses attached to them. They won’t hire you as cashier, but they’d love to have you work the backstore. Same wage, more physical and dirty.

    Even if a woman is hired in a warehouse, she’ll typically be given the lighter kind of work, that other workers generally find “too easy”. Involving little to no heavy lifting, for the same wage. I worked 2½ years in that warehouse (I quit to transition, the wage was pretty good for what I can normally expect, with an union), and men did that work only if there wasn’t any woman.

    I’d say this is also sexism and discriminatory, and perpetrated by many an employer. Just notice how many male cashiers you see in grocery stores vs male in-store workers (not the specialists or managers since this isn’t entry level – they require experience, degrees etc), who work the same wage and hours. And that’s applying without selecting a specific position (maximizing your chances of being hired).

    I don’t have the degrees or experience, or contacts if it helped, to get a job above entry-level and I’ll be lucky if it’s not minimum wage. Working class people are more numerous than middle class people, even if people try to define middle class as “anything above homeless status”, it doesn’t make sense.

    As for higher jobs, white collar jobs (I don’t get the distinction between pink and blue collar as helping any…it’s probably compounding the issue by making people self-segregate, lest they be accused of being atypical for their sex), well, people tend to choose this at the tertiary education level, themselves.

    There probably are social, peer and other pressures making women more likely to study in nursing or teaching and more men in mechanic, but at the base, the ratio is still not 50/50 (it’s definitely not 95/5 like nursing actually is now, though).

    People choose what they want to study in, and eventually work in, based on multiple factors, including:

    Potential schedule flexibility (4-days weeks, possibility of going off if an emergency occurs),

    Importance of wage for self (it’s not like teachers or nurses don’t know it’s paid less than engineering),

    Peer-pressure (much as some guys might like esthetics and hair-dressing, they might avoid it to keep their peer popularity, or simply to not be taken as defacto gay – women avoid mechanic for much the same “it’s masculine” reason),

    Self-view of gender roles (if you’d shame yourself for doing something),

    Hiring being less likely because of a characteristic (even if a man studies in caretaking, his chances of being hired at the local daycares are much lesser than a woman’s, regardless of his qualifications)

    and how they see their future (career focused? wanting a family soon? more quality time with children over bigger paycheck? something you really like, but might pay less (artists, book writers)? or just a job that pays the bill?).

    It ends up with women working on average 7 hours less a week, for full-time workers only (32 vs 39). And the #1 reason given is to have a better quality of life. Same reason given for the majority of women part-time workers.

    Priorities may be different, especially if one isn’t assumed to have to provide for 2, 3+ people, but can focus only on sustaining one’s self (or even be just an extra wage for the household). Someone having to provide for many would probably pick the career-focus, with the bigger paycheck jobs (lawyer, doctor, engineer) – over doing what he or she likes but pays less, or doing less hours.

    It also seems there’s something about (many) men having a sense of pride and not wanting to be provided for (and (many) women believing this is the way it should be). My boyfriend would never request alimony for himself, or even go at the food bank to take some food, he’d view it as a failure to take care of himself. Probably a few men and women alike who think the same about being stay-at-home (that a man should be self-sustaining financially, not supported by a girlfriend or wife – hence the bad rep stay-at-home dads get, being accused of having a free ride and what not). A legacy of the past; men are ‘less of a man’ if they don’t work to at least sustain themselves and that’s also according to most women (who don’t want a stay-at-home-husband for the most).

    Women rarely have this sense they must provide for their family, unless single mother, and then it’s usually too late to do studies leading to that big paycheck job (what with being busy with the kids and older than new grads – can’t say if they have less desire for it though). So prioritizing quality of life, even if they do work full-time, is something they take for granted, for many. Getting overstressed, dirty or working 80 hours a week shifts (truck drivers) is for the few who really want to, they can afford not to. And humans (as a majority) generally prefer the least-demanding way as long as it works (hence the dreams of the 60s science fiction about the future being much less work and much more entertainment and leisure).

    I’m not sure if women are more or less intent on having promotions than men. Personally I avoided promotions in my last line of work (videogame testing) because I didn’t want to be in a position to fire anyone (can’t bring myself to) or have to do overtime a lot mandatory (I do value my quality of life above wage, by far). Higher “ranks” also had less to do with testing, and more with managing testers’ tasks. I much prefer the testing, even if less paid.

    It’s quite possible that some business use sexism to prefer men in higher positions. I’ve been in too low-wage low-status jobs to see that firsthand (or to even anecdotally go against it), but I can believe it happens. I just doubt it accounts for a 24% difference in wage over all full-time workers.

  9. 9
    Sebastian H says:

    This seems to be getting heated, probably over a history I missed. But as to the statistics:

    “The standard wage gap figure only includes full-time, year-round workers,2 but CONSAD included part-time (and, I suspect, short-term) workers, which means it’s looking at a significantly different population.”

    Which to say a significantly *more comprehensive* survey of actual workers? I can almost see you writing that sentence as an opening to bash the standard figures.

    “The single largest factor that CONSAD found “explained” the wage gap is the difference in hours worked. Since women are more likely to work part-time, and since CONSAD (unlike standard wage gap studies) included part-time workers in their sample, in effect CONSAD is comparing mostly female part-time workers to mostly male full-time workers. Then — what a surprise! — they determined that the difference in hours worked accounts for a huge portion of the wage gap they measured.”

    The tone makes it sound like a criticism, but this is good statistics. It is a fact that more women are part time workers. If that fact explains a huge portion of the wage gap, then that fact explains a huge portion of the wage gap. Now if you want to say that the fact is caused by sexism, you can do so. But it probably isn’t caused by employer-acted-upon sexism.

    “For me, probably the most important kind of sexism going into the wage gap is the sexism of unquestioned assumptions; unquestioned assumptions about who does the housework, unquestioned assumptions about who does the child-rearing, unquestioned assumptions about innate ability, and most of all, unquestioned assumptions about how jobs are designed for people with wives at home.”

    This is fine, and is very much what I believe. But the problem with this is that so far as common progressive measures against the wage gap this ends up mostly being things that are well outside the control of the employer. If you are going to restrict the wage gap question to broad societal changes it makes sense. But when you start talking about enforcing post-facto wage adjustments from employers or other methods of attacking employers–which is by far the most common substantive remedy talked about–then if you believe that outside expectations shape most of the wage gap you are attacking it at the wrong point.

    This is the kind of thing that drives me crazy in politics all the time (on both sides). If you believe that a very large portion of the wage gap is explained by ‘choosing’ to remove yourself from the job market for on average 5 of the most key advancement years (the middle ones) [and by 'choosing' I'm completely open to the understanding that the 'choice' is heavily freighted by pressures from the husband, grandparents, and friends in what may or may not be appropriate ways], focusing on the employer for legal remedies seems a really unjust way to attack the problem.

    It seems to me that on some level you are talking past each other. Most of the attacks on the wage gap seem to be in the context of government laws about employer remedies. In those contexts, the statistics cited seem to be appropriate. It really does appear from the statistics that to the extent that there is pay differential, almost all of it is explained by things well outside the employer’s control. It appears that ‘choice’ factors are very prominent. [Again with the understanding that 'choice' in this context may be heavily influenced by societal pressures which may very well be sexist]. It appears that to a very large extent, it does not exist due to employer-side discrimination. So to the extent that employer-side remedies are being talked about (which is to say a large part of the conversation when it comes to government enforced remedies), it makes sense to note that the employers are no longer a big part of the problem.

    I think part of the problem in this conversation is the way the larger wage gap question (which from your post we seem to agree consists largely of non-employer factors) is used as leverage in a debate about employer-side remedies which seem (to some at least) out of proportion to its contribution to the problem.

    “Given two equally able applicants for a $40,000 job, one male, one female – which one will employers tend to prefer? Once hired, who is more likely to get mentored? Who is more likely to be given the assignments that lead to promotion? Who is more likely to be perceived as doing good work, all else held equal? And if these factors mean that women are rewarded less than men for identical labor market participation, to what degree does that reduce women’s incentive to participate equally in the labor market? All of these are ways that sex discrimination actually happens in the marketplace — and none of them are detectable by by CONSAD’s methods.”

    Actually *most* of these are very much detectable by CONSAD’s methods so long as they actually lead to higher pay. The only one that isn’t is the first because then you don’t get in the door. Most of the others seem well tracked by looking at people with the same level of experience. On any one on one comparison, you might not be able to narrow it down, but given 100 men and 100 women with 15 years of uninterrupted experience in a field, the promotions, mentoring, assignments, and attributions of good work, tend to show up in actual pay.

  10. 10
    Schala says:

    This is the definition of antifeminism which I found on 2 “online dictionary” sites:

    adj.
    Characterized by ideas or behavior reflecting a disbelief in the economic, political, and social equality of the sexes.

    And this is the definition of Michael Kimmel:

    According to pro-feminist sociologist Michael Kimmel the term antifeminism means

    opposition to women’s equality. Antifeminists oppose women’s entry into the public sphere, the re-organization of the private sphere, women’s control of their bodies, and women’s rights generally. Often this is justified by recourse to religious and cultural norms, and sometimes it is justified in the name of “saving” masculinity from pollution and invasion. Antifeminism often promotes a frightened or nostalgic retreat to traditional gender arrangements, supported ideologically by religious or pseudoscientific notions of “natural law.” Antifeminists typically accept the traditional gender division of labor as natural and inevitable, perhaps also divinely sanctioned.

    While Michael Flood has a definition more aligned with yours (Barry):

    Michael Flood argues that antifeminism denies one or more of three general principles of feminism:

    * The social arrangements among men and women are neither natural nor divinely determined.
    * The social arrangements among men and women favor men.
    * There are collective actions that can and should be taken to transform these arrangements into more just and equitable arrangements.

    Especially the 2nd point.

    Ballgame and I and others labeled anti-feminists believe in point 1 and 3, definitely. But our disbelief or scepticism for point 2 doesn’t make us believe any of the Kimmel definition or the 2 dictionary definitions above.

    I’m all for equality of the sexes, social, economic and political. I just don’t believe that it’s a given that, in North America, men are always, or on the balance, more advantaged than women*. That doesn’t make me opposed to equality, because I think many things need fixing, just not on a single side of the divide.

    *(I’m simply agnostic on the idea – a “we can’t know” stance)

  11. 11
    Robert says:

    Hmm, I’m inclined to weakly endorse point one, endorse point two with the addition of the words “on balance”, but argue against point three in either direction on conceptual grounds. Collective action is usually the source of problems, not the solution.

  12. 12
    Schala says:

    I think the Kimmel definition, if accepted by more than a few, is putting people in a double-bind.

    They’re either feminist or pro-feminist, or ultra-conservative people who refuse women any rights and are deeply religious.

    Having anti-feminist defined as such puts people in the middle as being extreme. I’m at the extreme left politically (I love my free healthcare, and welfare as a last-resort income, and I’d welcome government intervention to lower millionnaire-wages of certain people (who obviously don’t need THAT MUCH to live or get by) – in favor of raising minimal wage to liveable-alone levels).

    Sounds communist or socialist no?

    But I’m painted as being at the extreme right, because I’m against totally-ignoring problems men have (which political feminist organizations do, around here anyways). Funny that.

    If we’re going to go Marxist about it, to me it’s being against ignoring half the proletariat, or painting a full half of the proletariat as being the bourgeoisie (when in fact the rich/powerful/no-problem people are a minority).

  13. 13
    Ampersand says:

    So the standard wage gap figure is discussing only a subset of women, while the CONSAD study is discussing the actually existing population of workers?…

    No, that’s not correct. They’re both discussing only a subset of workers; the “actually existing” population of (female) workers would be larger than either of the samples described above, and would include full-time, part-time, off-book, and unemployed workers.

    Similarly, Sebastian wrote:

    Which to say a significantly *more comprehensive* survey of actual workers? I can almost see you writing that sentence as an opening to bash the standard figures.

    Why is more comprehensive better than more focused? It seems clear to me that both focus and comprehension have advantages.

    The tone makes it sound like a criticism, but this is good statistics. It is a fact that more women are part time workers. If that fact explains a huge portion of the wage gap, then that fact explains a huge portion of the wage gap.

    It doesn’t explain a huge portion of “the wage gap” as that term has been most frequently used in the economic literature.

    My point is that “the wage gap” is a specific, standard figure, calculated in a specific way. (It’s not calculated in the way I’d choose if I were starting from scratch, but I’m not starting from scratch.) So I have to either choose to use a wage gap measure that’s comparable to the standard wage gap, or if I choose another measure, make it clear I’m using a non-standard measure and avoid making inappropriate apples-to-oranges comparisons. Ballgame’s post did neither.

    If you are going to restrict the wage gap question to broad societal changes it makes sense. But when you start talking about enforcing post-facto wage adjustments from employers or other methods of attacking employers–which is by far the most common substantive remedy talked about–then if you believe that outside expectations shape most of the wage gap you are attacking it at the wrong point.

    First of all, I feel that you’re implicitly putting words into my mouth. I didn’t talk about solutions at all in my post, let alone talk exclusively about post-facto wage adjustments. So I think your argument here is pretty much a strawman.

    Second of all, suppose that of the wage gap that’s caused by sexism, 66% is caused by social factors and 34% is caused by direct and indirect effects of employer discrimination (numbers in this sentence pulled directly out of my ear, btw). It doesn’t follow from this that we should ignore employer discrimination, or even that we shouldn’t put most of our current effort into fighting employer discrimination. It might make more sense to focus on employer discrimination because a vast reduction of employer discrimination is a necessary step towards reforming social factors; or it may be that employer discrimination is relatively low-hanging fruit.

    Thirdly, feminists have spent decades doing a lot to reform social roles of the sexes, so it’s not clear that hasn’t been getting much (if not most) of the attention. These reforms tend to be extralegal and social, rather than making changes in policy, but that doesn’t mean they don’t exist.

    Actually *most* of these are very much detectable by CONSAD’s methods so long as they actually lead to higher pay. The only one that isn’t is the first because then you don’t get in the door. Most of the others seem well tracked by looking at people with the same level of experience. On any one on one comparison, you might not be able to narrow it down, but given 100 men and 100 women with 15 years of uninterrupted experience in a field, the promotions, mentoring, assignments, and attributions of good work, tend to show up in actual pay.

    Yes, but CONSAD will count much of that as “explained,” which means — in CONSAD’s model — that it can’t have anything to do with discrimination. For instance, if I promote Steve over Sally because Steve really likes hanging out in strip clubs with me, then by CONSAD’s measure Steve’s higher pay is caused by a different job title, and sexism has nothing to do with it. If Steve gets a promotion (and raise) because I’m a sexist boss who constantly sends better opportunities Steve’s way and mentors him more, CONSAD doesn’t see any sexism there. If I give the conferences to Steve to handle because I’m sexist — which inevitably means Steve works more hours than Sally — CONSAD doesn’t see any sexism there, since Steve’s higher pay is “explained” by hours worked.

    And, of course, if after five years of this treatment Sally decides her career is going nowhere so she’d might as well take 2 years off to have kids, CONSAD will consider that “explained” and say it has nothing to do with sexism.

  14. 14
    Ampersand says:

    There are lots of beliefs that go into a political philosophy, and yes, there are plenty of beliefs that would be incompatible with being a conservative (or a progressive, or a liberal, or a Jainist). “I believe the government ought to be the provider of most goods and services” is a non-conservative belief, for example. “I believe that women ought to stay home and raise babies and cook” is a non-progressive belief.

    “Thinks” goes to analysis.

    So if someone says “I believe the government ought to be the provider of most goods and services,” then you’d consider that a thoughtcrime and say they can’t be a conservative; but if someone says ““I think the government ought to be the provider of most goods and services,” then that’s completely different and you won’t doubt that they’re a conservative. Got it.

    Whether the Bell Curve’s analysis is correct or incorrect isn’t material; it’s objectionable for a progressive to think it’s correct.

    I find the belief that black people are inherently and genetically stupider than white people -which, boiled away from the excuses and weasel-words, is one of the main points of The Bell Curve – both offensive and empirically wrong. More to the point, it’s contrary to progressivism. Anti-racism is a core progressive belief, in my view. If you believe racist analysis which says white people are inherently, genetically superior to black people is true, then as far as I’m concerned you’re no progressive.

    Whether the UAW or Toyota management’s economic theories produce superior or inferior results isn’t material; it’s objectionable for a progressive to “side” with management, a priori.

    No, I didn’t say that; please try not to put words in my mouth. I can easily imagine a progressive thinking that a particular management is right, and a particular union wrong, about a zillion possible issues. However, someone who as a general rule is always on management’s side (not just one company or industry, but management in general) whenever there’s a conflict of management and union on virtually all issues… is someone who, it’s safe to bet, has core conflicts with progressive beliefs.

    I think your point is that I’m an Orwellian fascist who wants to have everyone who disagrees with me on any little issue shot in the back of the head, right? I don’t think that’s correct. I can think of a zillion progressives who have disagreed with me on this or that issue, who I still think of as progressives; similarly for feminists who I still consider feminists.

    But I’m not going to consider Rush Limbaugh a feminist tomorrow, not even if he says “I think women are equal to men and I self-identify as a feminist,” if his hundreds of other substantially anti-feminist views are unchanged. At some point, unless you’re arguing that to avoid being a Stalinist Thought-Policeman I have to jettison common sense altogether, I have to admit that someone who is virtually always posting to attack feminism while praising anti-feminists is probably an anti-feminist.

  15. 15
    Silenced is Foo says:

    Okay, now I’m going to get all harsh realpolitik libertarian here:

    I think the difference is that with dangerous, manual-labour jobs, the guys are getting a per-hour differential. It exists. That differential, however, is between their job and a job that is below minimum wage, and therefore does not exist.

    Dangerous minimum-wage manual labour jobs happen to men who don’t have any other options. Therefore, they *are* getting a premium for their willingness to take that kind of miserable working condition – their alternatives are so bad that they don’t exist thanks to current labour laws.

    Why don’t women end up in these jobs? They do. Their jobs simply have a different “the alternatives to this situation don’t exist” set-up. Bottom-rung sex-workers. Unemployed (and unumployable) partners to men that they have no alternative but to be financially dependant upon (leading to the possibility of abuse, since they are trapped). And so on.

    There’s a classic author the MRAs love to quote… I forget who… who described being an impoverished bum living on the street. The author noticed the complete absence of women at the absolute bottom of society. That women don’t work the absolute crap jobs, and live on the street, and join these dregs. And so the men would pay for prostitutes because those were the only women they could see.

    The cognitive dissonance was staggering. To say “there are no women at the bottom” and mention prostitutes who turn tricks for bums in the same sentence.

    Sorry, I’m kind of rambling aimlessly.

  16. 16
    Ampersand says:

    I think Kimmel’s definition is too blunt and extreme to be very useful, and I think the dictionary definition (like a lot of dictionary definitions) is too un-nuanced.

    Michael Flood’s definition seems pretty good to me, although it’s not exactly how I’d define it, perhaps.

    I’m at the extreme left politically (I love my free healthcare, and welfare as a last-resort income, and I’d welcome government intervention to lower millionnaire-wages of certain people (who obviously don’t need THAT MUCH to live or get by) – in favor of raising minimal wage to liveable-alone levels).

    Of course, in most of the world those beliefs would make you a moderate. In the US I’d say they’d make you a leftist, but not really an “extreme” leftist.

  17. 17
    Schala says:

    Of course, in most of the world those beliefs would make you a moderate. In the US I’d say they’d make you a leftist, but not really an “extreme” leftist.

    I believe in other leftist things, and I don’t think those beliefs make me only Center or Center-Left for say, Canada, who is very left compared to the US (abortion and separation of church and state are a given, same-sex marriage is behind us and won’t decide the result of an election, gay and trans people can openly serve in the military, we have universal healthcare).

    The Conservative party of Canada is more left than the Republican party of the US, and that’s saying much since it’s the right-most party of Canada. Liberals are center-right for us, but center or maybe center-left for the US – they’re like Obama and Clinton.

  18. 18
    ballgame says:

    Amp, I would like to respond to your post. Could you provide the following pieces of information?

    1. What, in your view, is the “standard” definition of “the wage gap”? (Or “the gender gap” … I assume these things are the same in your view but if not please explain how you define them differently if you do.)

    2. What is your understanding of what that “wage gap” is, mathematically (i.e. is it 77¢ on the dollar, 95¢ on the dollar, whatever)?

    3. Do you believe that 100% of that numeric value is accurately attributed to anti-female sexism in the workplace? If not, what numeric value best expresses the extent of such sexism?

  19. 19
    Ampersand says:

    1. What, in your view, is the “standard” definition of “the wage gap”?

    The most common formula I’ve seen is median annual wages of full-time, year-round female workers expressed as a percentage of the median annual wages of full-time, year-round male workers.

    It’s also very common to see the wage gap calculated using the median weekly earnings of full-time workers; this is very handy because sometimes more recent numbers are available for median weekly earnings than for median annual earnings.

    As far as I can tell, these two ways of measuring the wage gap became so common because those are the numbers that have been the most easily available for decades.

    Note that the wage gap doesn’t have to refer to all workers in the USA, although in practice it virtually always means that unless specified otherwise. But I’ve certainly read studies that did specify otherwise, calculating the gender wage gap among a specific demographic group, or in a specific industry, for instance.

    (Since several people seem to be confused by this point, let me emphasize that I’m NOT arguing this is the best imaginable method of calculating the wage gap. I am arguing that it’s the most common one in government and in scholarly articles, and the one that should be used in order to avoid apples-and-oranges problems when comparing certain statistics related to the wage gap.)

    (Or “the gender gap” … I assume these things are the same in your view but if not please explain how you define them differently if you do.)

    Arguably, it should be “the gender wage gap in the USA,” since there are other wage gaps in existence (such as the racial wage gap) and other countries. However, in the context of this discussion, I believe everyone understands “wage gap” to refer to the gender wage gap in the USA.

    I don’t use the term “the gender gap” to refer to the wage gap, since I frequently hear the term used to refer to other things (such as the partisan voting differences between the sexes). However, if you want to use the term in this discussion to mean the same thing as “the wage gap,” that’s fine — in this context, your meaning is clear.

    2. What is your understanding of what that “wage gap” is, mathematically (i.e. is it 77¢ on the dollar, 95¢ on the dollar, whatever)?

    For workers over age 16 in the US in the year 2009, 77% for year-round, full-time workers, or 80% for weekly wages of full-time workers. (As 77% vs 80% demonstrates, the exact number can vary a bit based on exactly what sample you’re using.)

    (I’ve also seen people express this as 23% and 20%, instead of 77% and 80%).

    3. Do you believe that 100% of that numeric value is accurately attributed to anti-female sexism in the workplace?

    No, of course not.

    If not, what numeric value best expresses the extent of such sexism?

    Impossible to say — measuring the extent of an activity that is deliberately kept secret is always guesswork. (Measuring rape prevalence runs into a similar problem — and rape prevalence is, imo, probably a lot easier to measure). There is no study I could point to and say “this proves it, the number is absolutely no higher/lower than X,” and to think that it’s possible to do so shows a fundamental misunderstanding of the issue. And in addition, the studies that exist are mostly only studies of one particular form of sexism in the workplace, leaving other forms unmeasured.

    But what the hell, this is only a blog. My seat of the pants guess is that if all workplace sexism could be eliminated (which by definition would mean eliminating a great deal of social sexism too, since the two are interdependent), we’d see around half to two-thirds of the wage gap disappear. But that’s only a guess.

  20. 20
    mythago says:

    Do you believe that 100% of that numeric value is accurately attributed to anti-female sexism in the workplace?

    Does it matter what the numeric value is, if it’s not 0%? And by ‘anti-female sexism in the workplace’ are we including only direct, overt sexism – say, managers who deliberately only promote males? Are we including less overt sexism – say, male supervisors who don’t invest in female employees because they just aren’t as comfortable around women, or a hiring manager’s fantasy-baseball group or his college fraternity having networking access to him in a way that females don’t? Are we including sexism that isn’t necessarily anti-female but that impacts women’s ability to earn money – such as the harsh judgment of men who spend time on family responsibilities, so that a married couple might decide it makes more sense in the long term for Mommy rather than Daddy to be the primary caretaker of their children?

  21. 21
    Schala says:

    The most common formula I’ve seen is median annual wages of full-time, year-round female workers expressed as a percentage of the median annual wages of full-time, year-round male workers.

    Adjusted for number of hours worked, and overtime?

  22. 22
    Sebastian H says:

    “First of all, I feel that you’re implicitly putting words into my mouth. I didn’t talk about solutions at all in my post, let alone talk exclusively about post-facto wage adjustments. So I think your argument here is pretty much a strawman.”

    No I’m not. The argument takes place in a context. Pointing out that context is in no way strawmanning. Just like pointing out the existence of a rape culture when someone isn’t explicitly referencing it isn’t strawmanning.

    “Yes, but CONSAD will count much of that as “explained,” which means — in CONSAD’s model — that it can’t have anything to do with discrimination. For instance, if I promote Steve over Sally because Steve really likes hanging out in strip clubs with me, then by CONSAD’s measure Steve’s higher pay is caused by a different job title, and sexism has nothing to do with it. If Steve gets a promotion (and raise) because I’m a sexist boss who constantly sends better opportunities Steve’s way and mentors him more, CONSAD doesn’t see any sexism there. If I give the conferences to Steve to handle because I’m sexist — which inevitably means Steve works more hours than Sally — CONSAD doesn’t see any sexism there, since Steve’s higher pay is “explained” by hours worked.”

    No, CONSAD will pick up almost all of those because it factors in based on years worked in the field, and while individual cases may vary, on average if women aren’t getting promotions, high level work, and good credit for the work they do, they will have lower pay when compared to men who have a similar number of years of experience. And for the higher paying jobs where most of the difference exists, it won’t be explained by hours worked either because both will be salaried.

    “My point is that “the wage gap” is a specific, standard figure, calculated in a specific way. (It’s not calculated in the way I’d choose if I were starting from scratch, but I’m not starting from scratch.) So I have to either choose to use a wage gap measure that’s comparable to the standard wage gap, or if I choose another measure, make it clear I’m using a non-standard measure and avoid making inappropriate apples-to-oranges comparisons. ”

    You’re appealing to authority here in a very uncharacteristic way. First it isn’t an overwhelming convention, in the sense of driving on a particular side of the road in the US, or performing exponents before adding. Second, look at the title of your post. It isn’t “How the CONSAD report on the wage gap uses a different measure.” It is “How the CONSAD report on the wage gap masks sexism….” That isn’t just about convention, that is saying that it is a worse way of measuring it. But so far as I can tell it isn’t a worse way of measuring it, everything you’ve brought up makes it sound like a more precise (though of course still not perfect) way of measuring it, at least *in* the workplace. Which may or may not be important to you, but clearly could be important to employers.

    One of the key problems in these measures is that they neither method is capable of discriminating between free choices (such as they exist) which end up with lower pay as a consequence, and sexist nudged choices (either externally forced upon a woman through direct pressure, or internally enforced through cultural norms). But NEITHER what you call the standard measure nor the CONSAD measure pick up the difference between those in any way.

    Now, to be clear, he definitely overstates the case when he says “In other words, no substantive evidence for a Gender Gap!”. But he would certainly be safe in saying that the workplace side of the gender gap appears to be less than 1/5 the generally advertised gender gap. (Which is an enormous markdown).

  23. 23
    Ampersand says:
    The most common formula I’ve seen is median annual wages of full-time, year-round female workers expressed as a percentage of the median annual wages of full-time, year-round male workers.

    Adjusted for number of hours worked, and overtime?

    Typically, no, not adjusted. That comes later. When economists talk about what percentage of the wage gap is or isn’t “explained,” that’s exactly the kind of thing they’re referring to — what you’re calling “adjustments.”

    I’ve also seen the government release hourly wage figures (rather than weekly or annual), but for some reason I haven’t run across that as often; I don’t know if they don’t release it as often, or if it’s just not as publicized or repeated.

  24. 24
    Robert says:

    So if someone says “I believe the government ought to be the provider of most goods and services,” then you’d consider that a thoughtcrime and say they can’t be a conservative; but if someone says ““I think the government ought to be the provider of most goods and services,” then that’s completely different and you won’t doubt that they’re a conservative. Got it.

    I think I made the distinction between belief and thought/analysis pretty clear, and you’re being rhetorically obtuse. But for clarity: the examples you give are the same thing. Replace your second case with someone who says “the analysis shows that for some goods, the government can provide them more efficiently than the private sector, for reasons [x] [y] and [z]” – then that guy can indeed be a conservative – indeed, that person is a classic libertarian, advancing the public good argument for some government services.

    I find the belief that black people are inherently and genetically stupider than white people -which, boiled away from the excuses and weasel-words, is one of the main points of The Bell Curve – both offensive and empirically wrong. More to the point, it’s contrary to progressivism. Anti-racism is a core progressive belief, in my view.

    “Human equality is a contingent fact of history.” – Stephen Jay Gould

    Dr. Gould, of course, was correct – both about the empirical truth (large human groups are all broadly equal in capabilities, though with interesting variations the more granular you get with your analysis) and its contingent nature (there is absolutely nothing that requires this to be true; it simply happens to be true.)

    If you believe racist analysis which says white people are inherently, genetically superior to black people is true, then as far as I’m concerned you’re no progressive.

    If you accept an analysis which is true, then you are rational. If you reject an analysis which is true, or accept an analysis which is false, then you are irrational. You’re hinging your ideology on requiring the rejection of an analysis; you can get away with it in the broad picture because you happen to be lucky and the analysis happens to be false.

    What happens tomorrow, when a shocked and dismayed Dr. Hypothesis uncovers new genetic evidence that proves “OMG, it turns out that people from Sweden, and only then, have haplogroup BS-123, which confers total immunity to a newly epidemically spreading viral disease whose main impact is to mildly reduce the cognitive function of people who catch it.”? Now you’re forced to choose between science and ideology – there’s a group of white people who are about to be smarter than everybody else. What if it’s Mongolians who have BS-123? Indigenous Australians?

    A more plausible example – what if homo floresiensis (the “hobbits”) had lived into the present day? They were possibly human. They were also, very likely, considerably dumber than the modern average (teeny little brains). And they may have survived up until 12,000 years ago – that’s practically into historical time. That’s a long string of likelies and maybes, I know – I don’t advance it as a likely hypothesis, but as a what if? – but does being a good liberal mean I have to pretend that my hobbit neighbor with the 500cc brain is just as smart as my black African neighbor with the 2000cc brain?

    Much better, in my view, to hinge your anti-racism on the Jeffersonian formulation that a people’s capabilities are no measure of their rights or their worth. Ronnie Retard and Ginny Genius are both fully human and have full rights – and the group distribution of this trait (or any other) is *immaterial* to political and civil rights. Black (and brown and white and hobbit) people get to vote and own property and run for office and serve on juries and go to jail for crimes and pay taxes and get felt up by pervy TSA agents same as everybody else – whether the median black (or brown or white or hobbit) IQ is 65, 100, or 135. It just don’t matter, chief.

    I think your point is that I’m an Orwellian fascist who wants to have everyone who disagrees with me on any little issue shot in the back of the head, right?

    No. For one thing, you Orwellian fascists lack the guts to pop anyone, so you’d want them sent to a re-education camp.

    My point was that you apply the wrong type of test (empirical rather than ideological). Your examples of Rush et al are examples of someone arguing in bad faith or making a deceptive claim to hold an ideology; no, I don’t require you to put your common sense in abeyance when it comes to bad actors. I think that your approach “reads out” too many people who in fact are ideological allies but whose experience and knowledge means they know a different subset of facts than you do.

  25. 25
    Robert says:

    Getting back on topic -

    The other thing I disagree with you here is that you’re characterizing the CONSAD study as “masking” sexism. It’s not masking sexism, it’s being agnostic about the exact cause of some forms of discrimination – while leaving sexism as a very obvious and reasonable default explanation. They’re trying to answer the question “how much wage differential is explained by intentional gender discrimination”, not “what is the root cause of all the components of wage discrimination”. Indeed, “sexism” seems a very reasonable first explanation for many, if not all, of those components.

    It’s important to know what’s intentional and what’s structural or social, because it gives us better information about where to target efforts at reform. If all the discrimination is intentional, but we ignore that and focus on ensuring that girls get technical educations and great mentoring in college, we’re wasting our time. If none of the discrimination is intentional, but we ignore that and focus on making it easy to sue for intentional discrimination, we’re again wasting our time (and hobbling our economy, to boot). Obviously the truth is some mixture of these factors, but it’s well worth knowing what the relative weight is.

    Your criticism of the study seems more focused on the fact that it doesn’t prove what you would like it to prove, than on any real flaws or failure of design. But in fact the study provides a lot of support for your way of viewing the issue – there are all these “choices” and social factors going into pay equity that are very important issues and a large source of causality. I.E., it’s worth our time to encourage girls to follow lucrative career paths and to tell them it’s OK to put family behind vocation, because it turns out those things are really relevant.

  26. 26
    Elusis says:

    Amp…

    Reading this latest back-and-forth over the past couple of days caused me to wonder why you keep having the same conversation with the same people. Particularly when you’re giving pointers to earlier writings you’ve done on the same subject, which your detractors obviously haven’t read because they keep coming back with “well what about taking time off work?” and “well what about women mommy-tracking themselves?” which you’ve addressed time and again.

    My conclusion is that trying to talk about the wage gap in this situation is like trying to talk about the existence of a god or gods to an atheist.

    As I’m sure you and most of the folks here know, there is a crucial distinction between agnostics and atheists. Agnostics do not believe in a deity(ies) because they believe such a thing is either totally unknowable, or that the evidence we have at this point in time is lacking. “Soft,” “open,” or “weak” agnosticism in particular holds that while belief is not possible given the evidence we have now, at some point in the future it is theoretically possible we could get some sort of better evidence that would fully confirm or deny the existence of a supreme being(s).

    On the other hand, atheists reject belief in the existence of one or many deities. It is the position that there are no deities.

    Your attempts to argue the existence of a wage gap suggests that you are pitching your efforts to wage gap agnostics – those who do not believe it exists, but who might at some point deem some piece of evidence sufficiently convincing to change their mind.

    I think you are arguing with wage gap atheists.

    I think any evidence provided to them that seems to support the existence of a wage gap will be immediately regarded as suspect, inadequate, insufficient, or otherwise flawed simply by virtue of the conclusion it supports.

    What I wish I had time and energy (and propinquity) for is to ask the kind of question that facilitator Lee Mun Wah asks when he hosts “difficult dialogues” about various diversity issues: “What would it mean to you if ___ did exist?” (Racism, sexism, a wage gap, whatever.) “What would it mean about the world if this were true? What would the implications for you be?” But I am not him, and I am not face to face with the wage gap atheists, and I have stopped trying to have many of those interactions online because they are often too much cost for too little result.

  27. 27
    Mandolin says:

    I do not agree with your definition of either atheist or agnostic.

  28. 28
    Robert says:

    I would be very interested to know your definitions, Mandolin. It’s not like you’re going to be the one who dragged us off topic :)

  29. 29
    Cessen says:

    Agnostics do not believe in a deity(ies) because they believe such a thing is either totally unknowable, or that the evidence we have at this point in time is lacking. “Soft,” “open,” or “weak” agnosticism in particular holds that while belief is not possible given the evidence we have now, at some point in the future it is theoretically possible we could get some sort of better evidence that would fully confirm or deny the existence of a supreme being(s).

    This sounds closer to the definition of atheist that I’m familiar with.

    I consider myself an atheist, and the way I generally explain my views to people is thusly:
    1. I don’t believe that Santa Claus (as in the modern myth) exists.
    2. The conviction and extent to which I don’t believe this could, indeed, be expressed quite accurately as “I believe that Santa Claus does not exist”.
    3. If compelling evidence (sufficiently compelling for such an extraordinary claim) were to present itself that he did exist, I would seriously reconsider my belief.

    Replace “Santa Claus” with “God” or “gods”, and there you have it.

    In other words, it’s not an irrational, un-changable belief that God doesn’t exist. It’s a rational belief based on lack of evidence, that is open to change.

    I think this is generally the position of most people that identify as atheist.

    My understanding of agnosticism is more along the lines of “Due to lack of evidence, I’m substantially uncertain of Santa Claus’ existence.”

  30. 30
    Schala says:

    My understanding of agnosticism is more along the lines of “Due to lack of evidence, I’m substantially uncertain of Santa Claus’ existence.”

    But it ends up going with Occam’s Razor, where the non-existence of God or gods is more likely than their existence without proof of any kind (and citing the existence of humans as proof is tautological). So agnostics could generally go with a “I don’t believe gods exist” too.

    In Star Ocean 3, humans find weird artifacts on ruined planets (still breathable though) that don’t look like they have anything to do with the level of civilization present. They have no real explanation for it, but assume it’s from an advanced civilization (more advanced than them, and humans can do at least intra-galactic travel easy then). Only later do they find out it’s “Tools used by the gods”…and then meet those gods. Only then can they revise their judgment that the gods do in fact exist. Even if they end up being Matrix-like game developers from the 4th dimension (doesn’t make them any less powerful in our universe). Ironically, those gods could only be known to exist by having directly intervened in our universe (which they have, 20 years before the game).

  31. 31
    Jake Squid says:

    An important thing to define when discussing the definitions of atheism and agnosticism is the term “god/gods.”

    So, for example, what definition of god makes the Matrix-like game developers from the 4th dimension accepted as gods?

  32. 32
    ballgame says:

    There’s a lot wrong with Ballgame’s argument — and not just his belief that an anti-feminist like him is in any meaningful sense a progressive.

    Amp, I’m not surprised that you start off what could be an interesting and substantive discussion with your usual false ad hominem smear against me.

    Saddened, but not surprised.

    It’s not terribly relevant, but since you brought it up, I’m a pro-choice, pro-gay marriage, pro-equal-pay-for-equal-work, Democrat-endorsing, high-income-group-tax-raising, civil-liberties-loving, single-payer-advocating, democratic socialist, and I’ve been a Leftie gender egalitarian since my late teens. Apparently in your world that doesn’t count as feminist or progressive.

    Whatever. Given your insulting misrepresentation of my views, it’s equally unsurprising that you misrepresent the CONSAD study as well.

    Let’s start off with a biggie:

    6) Ballgame implies that “the Big Lie” is that studies that account for multiple factors (occupational difference, danger, motherhood, experience, hours worked, etc) would not find a significant gender wage gap, and only studies that fail to control for multiple factors find a significant gender wage gap.

    This is breathtakingly ignorant; if Ballgame knew the first thing about wage gap studies, or had bothered to do even ten minutes of research, he’d realize his argument is ridiculous.

    This is a breathtaking misrepresentation of what I said. My exact words (which you quote earlier) were, “But if you take all those factors together (along with some others, like men working longer hours on the job)? Well, then you find that women make at least 93%-95% of what men make, according to statistical analysis by the CONSAD Research Corporation.” (The key word, of course, being “all.”)

    My actual quote is accurate; your paraphrase is not. Let’s take a look. You follow that paraphrase with this:

    I’ll list just a few of the many gender wage gap studies that disprove Ballgame’s “Big Lie”: Wood, Corcoran & Courant (1993), Journal of Labor Economics; Dey & Hill (April 2007), American Association of University Women Educational Foundation; “Women’s Earnings” (Oct 2003), United States General Accounting Office; Blau & Kahn (June 2006), Industrial and Labor Relations Review; Mandel & Semyonov (Dec 2005), American Sociological Review; Boraas & Rodgers (March 2003), Monthly Labor Review; Johnson & Solon (Dec 1986), American Economic Review; Mulligan & Rubinstein (August 2008), Quarterly Journal of Economics; Fields & Wolft (Oct 1995), Industrial and Labor Review.

    The problem with your point here, Amp, is that the CONSAD report incorporated ALL but one of the studies you just cited! Let’s take a look at them. The CONSAD study specifies 14 factors that affect the difference in earnings between men and women. So how many factors do those studies you just cited analyze? The CONSAD study indicates how many on most of them:

    Blau & Kahn (June 2006), Industrial and Labor Relations Review: 2/14

    Mandel & Semyonov (Dec 2005), American Sociological Review: 4/14

    Boraas & Rodgers (March 2003), Monthly Labor Review: 10/14

    Johnson & Solon (Dec 1986), American Economic Review: 12/14

    Mulligan & Rubinstein (August 2008), Quarterly Journal of Economics: 9/14

    Fields & Wolft (Oct 1995), Industrial and Labor Review: 9/14

    So those studies 1) looked at multiple factors, but 2) did not look at all the factors, and 3) did not come to the same conclusion about how much of the gender gap could be attributed to factors other than the gender of the worker. This, of course, is entirely consistent with what I said, despite my being “breathtakingly ignorant.”

    Moreover, it’s consistent with what at least one of the studies you just cited says (“Women’s Earnings,” Oct 2003, US GAO):

    Even after accounting for key factors that affect earnings, our model could not explain all of the difference in earnings between men and women. Due to inherent limitations in the survey data and in statistical analysis, we cannot determine whether this remaining difference is due to discrimination or other factors that may affect earnings. For example, some experts said that some women trade off career advancement or higher earnings for a job that offers flexibility to manage work and family responsibilities.

    (Emphasis mine.) So even one of the studies you cite admits their analysis of the factors they looked at doesn’t provide definitive evidence of discrimination.

    Returning to your assertions:

    2) The single largest factor that CONSAD found “explained” the wage gap is the difference in hours worked. Since women are more likely to work part-time, and since CONSAD (unlike standard wage gap studies) included part-time workers in their sample, in effect CONSAD is comparing mostly female part-time workers to mostly male full-time workers. Then — what a surprise! — they determined that the difference in hours worked accounts for a huge portion of the wage gap they measured.

    Your point here doesn’t make sense, Amp. The CONSAD study was looking at the hourly wage rate, and isolating the impact of part time vs. full time work was one of many factors that they controlled for.

    Anyway, why don’t we look at exactly what the CONSAD study says? What do they tell us about what accounts for “most” of the gender gap?

    Specifically, variables have been developed to represent career interruption among workers with specific gender, age, and number of children. Statistical analysis that includes those variables has produced results that collectively account for between 65.1 and 76.4 percent of a raw gender wage gap of 20.4 percent, and thereby leave an adjusted gender wage gap that is between 4.8 and 7.1 percent.

    So CONSAD is clearly positing “career interruption” as the leading cause of the gender gap. In fact, because they used a more comprehensive sample than other studies, they were able to determine something more:

    The coherence of the quantitative results obtained from the analysis of cross-sectional data in this study and the quantitative results derived in analysis of longitudinal data in previous studies strongly suggests that the percentage of the raw gender wage gap for which the percentages of similar workers who are not in the labor force or are working part-time statistically account are actually describing, in large part, adjustments in the wage rates of people who have interrupted their careers by temporarily withdrawing from the labor force or switching from full-time to part-time work.

    In other words, it is the “career interruption” (which is occasionally characterized by switching from full time to part time work) and not working part time per se which is the greater explanatory factor for much of the raw gender gap.

    Jumping back to your OP:

    Typically, neither Robert or Ballgame mentioned that CONSAD used a radically different sample of workers than virtually all other wage gap studies.

    The question is to what extent discrimination against women causes them to earn less than men. You tacitly admit there isn’t one standard definition of the term in your response to me above. I’ve seen a variety of worker samples used. I’ve seen studies that look at extremely narrow samples and I’ve seen other studies that look at very broad samples. Whether the CONSAD focus is more appropriate than the focus of other studies is a legitimate topic for discussion; implying that the CONSAD study is “wrong” because they took a more comprehensive approach than most other studies is absurd.

    The comparison of the CONSAD conclusion to the conclusions of others looking at less complete statistical analyses is not “apples to oranges.” I am not using the CONSAD study to dispute the numeric value of those who claim that the difference in the median full time wages between men and women is whatever Newsweek claims it is. I’m citing the study to dispute the implication that this difference is due to workplace gender discrimination.

    Additionally, the hourly wage rate differential CONSAD worked with is, in fact, just about the same as the figures you cite: +24% (M/F) or -19% (F/M), so the notion that their methodology caused them to somehow overlook the extent of ‘real’ discrimination appears dubious in the absence of any genuine analysis demonstrating how.

    There are a number of other serious errors in your OP, Amp, but this partial rebuttal will have to do for the moment.

    [Plaudits for fixing the editing module, though!]

  33. 33
    Robert says:

    Apparently in your world that doesn’t count as feminist or progressive.

    Tell it to your idol Sarah Palin, you Tea Party neanderthal MRA right-wing thug!

  34. 34
    Elusis says:

    Mandolin – I was going straight off the Wikipedia versions, which appeared to me to jibe with what I learned when I was first trying to figure out what to label myself (what was that website with all the incredibly useful definitions, summaries, and links of assorted world religions and sects? I thought it was Beliefnet but either the website has undergone a major re-focus along with a re-design, or I’m thinking of another site. It’s not Teaching Tolerance either, apparently. Damn.) The soft/hard agnostic distinction has been particularly salient for me personally.

  35. 35
    Schala says:

    So, for example, what definition of god makes the Matrix-like game developers from the 4th dimension accepted as gods?

    1) They created our world (to them it’s a very high-tech MMO, for fun). The fact that people became sentient is an artefact of their extremely complex personality algorithms (basically, to them we’re all NPCs).

    2) Magic is something inherent in their world, and something they initially taught to some people in our world a la Game Master (called various things: heraldry, runology, symbology), except not to humans…humans came about people who had those abilities (Star Ocean 1 and 2 – in the 26th century, the 3rd game is set in the 29th century) and made a science out of it (Symbological science).

    3) They can modify or even outright “delete” our world (the final boss, the head dev, does this).

    Ironically, the gods intervened in our world to warn humans that they had trespassed on the domain of the gods, by having a science called Symbological Genetics (which could theoretically allow people in our world to go in theirs, and affect them – sort of like if Mario Bros came in your living room).

    They announced the destruction of the Milky Way to scientists…who decided they had nothing to lose, and applied their science (genetically engineered humans with special symbological capacities who could enter the 4th dimension world and interact with people there, as if they were 4D themselves – those happen to be 3 of the heroes, 2 of them being children of the scientists, out of 10 playable characters).

    They call our world “The Eternal Sphere”, and its a massively multiplayer game on multiple time axis (they are beyond time, to them it’s normal). The protagonists call the dev’s world “4D Space”.

    Comparatively, in Star Ocean 2, you meet the Nedians, who live on a large meteor shielded by a strong force-field (allows breathing and vegetal and animal life to live) and have mastered time travel, at least 700 million years before our time (one of the main characters is the daughter of a scientist from that time – sent forward 700 million years by her mother). They’re not gods though, in the sense that they don’t have direct life-or-death control over the existence of the universe.

  36. 36
    gin-and-whiskey says:

    Elusis says:
    11/26/2010 at 10:04 pm

    Amp…

    Reading this latest back-and-forth over the past couple of days caused me to wonder why you keep having the same conversation with the same people. Particularly when you’re giving pointers to earlier writings you’ve done on the same subject, which your detractors obviously haven’t read because they keep coming back with “well what about taking time off work?” and “well what about women mommy-tracking themselves?” which you’ve addressed time and again.

    That’s ridiculous.

    Saying that something has been “addressed” doesn’t mean it’s been conclusively shown; or even that it’s been addressed in a fashion that people can (or should) believe. I hear that all the time on this board: “yeah, that’s been addressed by Favorite Source” and I’m wondering “how did they skip over the question of whether Favorite Source is the proper authority and/or wrong?”

    I’ve read plenty of things. I disagree with plenty of them. And since I believe my analytical skills to be at least as good as Amp’s or yours, I also sometimes disagree with the proper interpretation of sources even when i agree on whether the source itself is valid.

    ….I think any evidence provided to them that seems to support the existence of a wage gap will be immediately regarded as suspect, inadequate, insufficient, or otherwise flawed simply by virtue of the conclusion it supports.

    There are people who fundamentally deny the existence of the gap. There are people who fundamentally uphold the existence of the gap. Neither of those types will really change their views in response to conversation.

    While I agree that conversation with or between fundamentalists is generally useless, I’m curious: Do you think it’s just your opponents who are the fundamentalists? Or do you put yourself in that category as well?

  37. 37
    gin-and-whiskey says:

    To use a personal example:

    I started out not really thinking that the wage gap was a problem: not opposed to its existence, just ignorant.

    Then, I read more on it, mostly on feminist sites. I decided that it was a problem. Like many recent “converts to a cause,” I accepted most of the things I’d been told. That included an initial (rebuttable) presumption that most of the opponents were misguided and/or biased.

    Then, as I became more interested, I started reading more about it on places other than feminist sites. Law blogs. Newspapers. Magazines. Actual studies.

    As I read those additional bits of information, I came to understand that some of the more pro-legislation folks were (charitably) putting a lot of spin on things, or were (less charitably) lying or doing bad science. I had already thought this about their opponents, mind you.

    In the end, I dislike being lied to. And as it happens, it seems that most of the opponents of the act are a lot more forthcoming about reality and statistics than are most of the proponents of the act. It’s classic overreaching.

  38. 38
    MercuryChaos says:

    (Problem: racially identifiable groups of mostly-poor people have bad primary and secondary schools, leading to low admission rates at high-end colleges and universities. Progressive solution: quotas and preferences to require unprepared students be admitted to places where their failure is virtually guaranteed! Don’t worry, though, when people start noticing this, we’ll just call them racists so they’ll shut up.)

    Affirmative action does not require schools to accept unqualified students over qualified ones to meet any quotas.

  39. 39
    gin-and-whiskey says:

    Er… is that a joke? What precisely do you think AA is? Practically speaking it’s a change in qualifications to affirmatively admit students who would otherwise not be considered qualified for admission.

  40. 40
    Ampersand says:

    Every selective college routinely turns away many students that they consider “qualified for admission.” AA changes the odds about which students, among those considered qualified for admission, actually gets admitted; but it does not require admitting unqualified students.

  41. 41
    Robert says:

    Please note that I did not say unqualified. I said unprepared, which is different; if I had been writing a more thoughtful note I should have said “relatively underprepared” for maximum clarity.

    Amp is correct that AA does not require admitting unqualified students. Instead, it has the effect of setting up de facto, though not de jure, requirements that vary with the ethnic group of the prospective student. Asians who want to attend selective school [x] need higher attainments and test scores than whites, who in turn need higher attainments and test scores than blacks and Hispanics.

  42. 42
    gin-and-whiskey says:

    AA is not the only factor which incentives schools to admit less qualified students. Football is one. Alumni status is another. Being extraordinarily rich is a third. But AA (like those other examples) is at heart a program which deliberately gives weight to a known non-academic credential.

    So as with those other examples, the concept that AA is just choosing students from an equivalent pool is bunk. It’s also rendered moot by the need for AA. It exists BECAUSE it is needed.

    After all, if all the students were equally qualified, you’d be able to get appropriate admissions ratios merely by the expedient method of blind admissions, or scored process admissions, or some other factor that eliminates the ability of the school to directly influence race.* But nobody has managed to do that yet. People have done studies in which they’ve looked at the average information of students and have found that there was, as expected, a functional “head start” for AA stuff.

    That makes perfect sense. If there WASN’T a boost, it wouldn’t be AA. If you DIDN’T need to boost them, you wouldn’t need AA. That is what AA is, as distinguished from other more neutral policies like enhanced searching for applicants.

    Nothing wrong with it if you support AA, but it’s silly to pretend there’s some different mechanism.

    * or, “if all football players were geniuses, you wouldn’t need football scholarships or ‘coach preference’ admissions.” Or, “if Bobby Lee is so smart and a sure thing to admit, why does he need to list his alumni status on the application and why are you offering to build a library when he matriculates?” the proof is in the pudding.

  43. 43
    Ampersand says:

    G&W, who on earth are you responding to? Certainly not me. (e.g., Please quote where I said “all the students [are] equally qualified.”)

  44. 44
    Schala says:

    AA is not the only factor which incentives schools to admit less qualified students. Football is one.

    That is very US-centric. In Europe, this would be soccer, in Canada, this would be hockey. While the Montreal football team won the Grey cup, and seems to like, always do…most people don’t care in as much as they mostly care that their home team won. Most don’t watch it, and there won’t be a riot if they win (but there definitely is when Montreal Canadians – the hockey club – wins).

    Still, some cultures are not as sport-centric. Sport in general here isn’t the glorified Friday event where everyone goes to watch. It’s the school-paper event, where people invested in that actual team (players, coach, admirers) watch it and people from the school might take a “our school rocks” pride, to some minor extent, even if they’re not interested in the sport.

    Nothing like the US. And so, it’s also less of a “path to success” in education.

    Oh and yeah, everything below Junior pro goes unwatched. Even Junior pro has only hardcores who do. Pro gets watched, and the tickets are pricy (like 100$ for a *basic* ticket), but that’s the NHL, not high school, college or university teams (which don’t ever get watched).

  45. 45
    gin-and-whiskey says:

    I agree that if there is a large pool of well qualified preferred* individuals, then preferences don’t go below the bottom line of qualifications.

    In reality, that’s not the case, at least not everywhere. It’s especially not the case as you climb the higher education ladder.

    It’s believed that AA has a social ripple effect, in that having more purple people in boston college law school will have a beneficial effect, since BC is well regarded. But it’s not necessarily a great benefit to the actual students themselves.

    Take law school, for example. There’s a lot of competition for certain minority applicants, since the accrediting body tends to punish schools who don’t do enough to encourage minority admission. As a result, it’s a lot easier for certain tier-4-level applicants with a 3.3 GPA and a 152 LSAT to end up in a much higher-tier school. As with any preferred candidate who gets pushed up a notch, by definition the preferred students are academically out of their depth w/r/t their peers.

    Is that a good thing? From the ripple effect perspective, sure it is. From the student perspective, maybe not so much: Law schools grade on a curve, and classmates can be really intolerant of people who aren’t on the same level.

    As a result, dig into it and you’ll see that there’s a very low graduation rate for the same applicants who are given higher preferences. Again, that’s no surprise: you’d probably see it with any type of preference, but there just aren’t enough of the other ones to measure correctly, and alumni works in strange ways .

    Being a small fish in a big pond in order to make a ripple may be socially beneficial, but the result is that some unqualified students end up in the wrong spot.

    *such as aa receipients, football jocks, alumni kids, parents-build-a-library kids, etc.

  46. 46
    Chans says:

    Including hourly workers is problematic because, in my experience, hourly jobs are more likely to be filled by women (especially retail) and I believe the wage gap also needs to account for non-wage benefits like health care, vacation time, etc. Additionally, in hourly positions women will have their hours cut and be underscheduled as compared to men simply because men are subconsciously perceived to be more competent than they are.

    Yeesh, though. Ginmar was right about this place. It’s basically just one big MRA stomping ground.

  47. Pingback: Sorry, Men, You’re STILL Not Oppressed: Reexamining the Fallacies of “Misandry” | bunnika's blog

  48. 47
    Marc says:

    I would think the converse is true tue – by comparing only full time workers you create a similar bias toward men earning more since men are more likely to work full time than women. Either way, exists because men work more hours while women have more options than men to be primary parents. That is why 57% of female graduates of Stanford and Harvard left the workforce within 15 years of entry into the workforce. http://edition.cnn.com/2005/BUSINESS/03/15/optout.revolution/

    That’s why never-married childless women earn more than their male counterparts. And women between ages 21 and 30 working full-time made 117% of men’s wages. http://www.nytimes.com/2007/08/03/nyregion/03women.html?_r=1&oref=slogin

    Meanwhile, female CEOs outearned men in 2009. http://abcnews.go.com/print?id=10630664

    And female U.S. corporate directors out-earn men: study
    http://www.reuters.com/article/domesticNews/idUSN0752118220071107?feedType=R

    Being primary parent is an option few men have. Try being a single male and telling women on the first date that you want to stay home. In fact,research shows women still seek men who earn more than they do. http://www.telegraph.co.uk/news/uknews/8237298/What-women-really-want-to-marry-a-rich-man.html

    And most men have no problem with their wives outearning them. http://www.msnbc.msn.com/id/23413243

    Most working dads would quit or take a pay cut to spend more time with kids if their spouses could support the family. http://www.cnn.com/2007/US/Careers/06/13/dads.work/index.html

    Research also shows that parents share workloads more when mothers allow men to be primary parents.
    http://www.usatoday.com/news/health/2009-05-04-equal-parenting_N.htm

    In “Why Men Earn More,” Warren Farrell, Ph.D. examined 25 career/life choices men and women make (hours, commute times, etc.) that lead to men earning more and women having more balanced lives, and that showed how men in surveys prioritize money while women prioritize flexibility, shorter hours, shorter commutes, less physical risk and other factors conducive to their choice to be primary parents, an option men still largely don’t have. That is why never-married childless women outearn their male counterparts, and female corporate directors now outearn their male counterparts. http://www.reuters.com/article/domesticNews/idUSN0752118220071107?feedType=R

  49. Pingback: Könsrelaterade inkomstklyftor « Yasers hörna

  50. Pingback: The Gender Blender: Adventures in Facebook Feminism « The Teresa Jusino Experience