Tuesday, June 21, 2011

Prey to Biases

It's not as if I thought the current Supreme Court would actually produce a majority decision that recognized the overwhelming statistical evidence for systematic, nationwide discrimination against the female employees of Wal-Mart in salary and promotion. And, since they did not in fact do so, we can at least seek solace in the text of the decision, in which the perpetual arguments about discrimination and unconscious bias are dramatically displayed.

All but one of the men of the US Supreme Court decided that the female employees of Wal-Mart did not have enough in common to represent a class that could bring suit because Wal-Mart gives its individual managers (>65% of whom are men) so much individual discretion in personnel decisions. Also, Wal-Mart forbids discrimination -- they have a policy! Never mind what the data show. As long as individual managers are discriminating against individual women without specifically saying that they are doing so, and as long as Wal-Mart doesn't have a formal policy that endorses discrimination, female employees as a group don't have much in common.

Ruth Bader Ginsburg's dissent makes this point:

"The practice of delegating to supervisors large discretion to make personnel decisions, uncontrolled by formal standards, has long been known to have the potential to produce disparate effects," she writes. "Managers, like all humankind, may be prey to biases of which they are unaware."

At least 5 Supreme Court justices are prey to biases of which they and everyone else are well aware, with dire consequences for real people (but not major corporations) in this country.

PS - In recognition of this major defeat to more than 1.5 million women, this is the headline that The New York Times came up with:

Wal-Mart Case Is a Blow for Big Cases and Their Lawyers

Well, some of those lawyers may well be women, so it's not as if that headline totally misses the mark.


Anonymous said...

Come on FSP, have a little sense. They presented THREE ... just THREE women who had personal stories of "discrimination". WalMart employs 1.6 million women. Just 3 disgruntled workers are evidence of companywide discrimination? You are a scientist and a journal editor and a reviewer. Would you accept a paper that drew a conclusion about a population of 1.6 million based on 3 samples?

Curt F. said...

The decision to reject the certification of all female Walmart employees as "class" for the purposes of a class-action lawsuit was unanimous. All the justices of the supreme court concurred with the judgement.

The rationale used to reject the certification was not unanimous, but was split 5-4. However, the decision itself was unanimous.

The Supreme Court overruled the 9th Circuit in this case. The dissent to the now-overruled 9th Circuit decision by Judge Sandra Ikuta is also interesting reading.

Anonymous said...

From Slate:

"Of course, the lack of a formal policy doesn't always derail gender discrimination claims. And the plaintiffs had amassed statistical evidence from a pair of sociologists to show that Wal-Mart paid men more than women, promoted males over females, and did so in numbers that could not be readily explained away. As Ginsburg writes in her dissent: "Women fill 70 percent of the hourly jobs in the retailer's stores but make up only 33 percent of management employees. The higher one looks in the organization the lower the percentage of women." The plaintiffs' uncontested statistics, she notes, show that women working in the company's stores "are paid less than men in every region" and "that the salary gap widens over time even for men and women hired into the same jobs at the same time.""

That sounds like more than 3 women to me.

Anonymous said...

that Ginsburg quote in your post totally sums up my dept. head.

He didn't give me as much of a raise as my male counterparts this year because I had a baby last year and only submitted 3 proposals (one of which got funded by the way).

But his major comment on my review: "not enough proposals" and I found out later my raise was not as high as the mens'.

that ruling is a tragic day for women

EliRabett said...

Perhaps it is time for women not to shop at WalMart?

EliRabett said...

If you are submitting more than three proposals you are not submitting enough papers and you are certainly not doing enough research and work with your students. man or woman

Anonymous said...

Anonymous at 4:38 (first post) perhaps does not have enough knowledge of the legal system to understand what a Class Action suit is, or the concept of "Lead Plaintiffs."

Anonymous at 9:26 quotes from Slate to explain how the lead plaintiffs represent the 1.6 million women working for WalMart. Typically, a class action suit would involve exploration of how the lead plaintiffs are representative of the class.

The unanimous part of the decision had to do with under which part of the law the plaintiff's sued. This involves damages (that's an oversimplification), rather than the question of whether there was discrimination. This is legal esoterica that is typically not explained well, if at all, by the popular media. So, we know almost nothing about the legal concept about which the justices were unanimous.

Media attention is usually given to the headline - in this case - that discrimination against women was dismissed. While Scalia was dismissive of the possibility of discrimination, that was not the basis for the unanimous part of the ruling.

The part of the suit regarding the possibility of discrimination was NOT unanimous - that was 4-5. Ginsburg's quote (true to her roots, gosh love her) expresses this disagreement very well.

Thus, it is incorrect to say that the court was unanimous in dismissing the possibility of discrimination.

Scalia's (typically) tortured logic here (based on the quotes I read in the newspaper) are at least as much about protecting Corporate Financial Interests as it is about dismissing women, in particular, or plaintiffs, generally.

Class action suits are important because they are one way ordinary people have of standing up to Corporate Giants. If your credit card company, for example, rips you off for $200, that's wrong, but it's not enough money to get a lawyer and sue. But if a thousand people like you get together, you can hire a lawyer. And the possible payoff will be worth the lawyer's time. Yes, plaintiff's lawyers get a bad reputation from doing this, but this is the system we have and in it, this is the way "the little gal or guy" can seek redress.

If a company dumps toxic waste on the ground and contaminates the water, a class action may be the only means people have of seeking compensation. (Actual case, about which a book was written and a movie made.) One person against a giant corporation is at a huge disadvantage.

So, this change in the law (which is arguably activist, since it went well beyond the issues in the case presented before the court) making it much harder to get a class certified, is blow to the women in this suit, or any suit in which they allege discrimination, but it is also a blow to anyone who is ever wronged in a way that is, by itself, financially (because the legal system cares vastly more about money than people or justice) too small to justify seeking legal recourse. It means that it is harder for people to unite to right wrongs and seek redress.

This is the legacy of the newest (as well as older) justices chosen for their ideology rather than their wisdom.

Anonymous said...

I already avoid shopping at Walmart- one more reason

@EliRabett, whoa, that's a broad statement. Not necessarily true. Ask FSP for example ;-)

Anonymous said...

Just FYI - according to NPR this morning smaller cases can still be brought in state courts. I believe the reason the ny times had this headline is that while the impact of this case is on women, the impact of the decision is to set precedent for class action suits in general.

Anonymous said...

Scalia in the majority opinion:

"Left to their own devices most managers in any corporation--and surely most managers in a corporation that forbids sex discrimination--would select sex-neutral, performance-based criteria for hiring and promotion that produce no actionable disparity at all."


Theorem: Sexual discrimination is not prevalent in hiring and promotion decisions.

Proof: Employers want to hire the best people available. Best is defined through "performance-based, sex-neutral criteria." Most "performance-based criteria" are "sex-neutral," so that really isn't a problem. Moreover, there is only one way to judge according to those criteria, so bias cannot enter there. And even if there are a few biased jerks, the sample size is really, really large and they represent only a small proportion. QED.

Now that sexual discrimination has been solved, onto the next problem.

EliRabett said...

WRT the number of grants, unless you are submitting the same proposals to a lot of agencies and even then because you have to tailor it, every proposal takes 3-4 weeks out of your life if you are going to win. It's easy to submit losing proposals.

Female Science Professor said...

Eli, I think your statement is too broad. In some fields, you have to submit >3 proposals to have any hope of supporting your research group. I submitted >3 proposals in the past academic year, all but one was funded, and I'm doing fine with papers and advising and all the rest. I am certainly not alone in this. Writing (excellent) proposals takes a lot of time, but it doesn't take every second of your life for weeks. Well, one of the ones I wrote this past year did, but that was unusual.

Anonymous said...

What, you mean the Supreme Court is made up some excuse to cover up and protect institutionalized discrimination? I'm shocked!

... wait, no, I'm not.

Although, if it makes you feel better (it won't), the Supreme Court of Canada endorsed the use of violence against children a few years ago, so this isn't an isolated problem.

EliRabett said...

Let us consider the life cycle. Writing a grant takes a considerable amount of time 2-4 weeks of "scientist time" (not 40 hours a week) is not unreasonable. We can argue about 2 or 4. We then come to the question of what the hit rate is. If you are under 10% you are spending most of your time spinning your wheels.

But let us say you get funded. Then, of course you have bought into annual reports, administratriva without end and more. Even if you have help there are logjams only the PI can clear.

And, of course, there are the annual calls from the program manager inviting you oh so politely to sit on this or that review panel. A huge time sink and we could go on.

The points are that

a. Proposal writing is a huge time sink. Eli has seek over the years that the quality of proposals has gone down as many people try and play the odds by submitting large numbers of proposals.

b. Proposal winning is a huge time sink even without the research that goes with it.

Anonymous said...

Eli, it's me again. Usually I write between 4 and 6 proposals a year; I have a "batting average" of 1-2 a year. What I try to do is to have 1-2 where I am the only PI, and the others are collaborative, either with me as lead or as a co-PI. I find those a lot easier to write, and often very stimulating. I agree with you that it is not helpful to send a lot of proposals in at the expense of quality, but again, that's not always the case.
BTW it's funny that I'm getting into this with you. Today I lectured my co-PI, who's getting desperate about tenure, to stop writing so many proposals and stop saying yes to everybody, and just focus on 1-2 things in the next year where she is the PI (she's got a fairly good amount of funding, but it's all collaborative).
What I'm trying to say is that there are different situations, and different strategies, and different ways to do well.

Anonymous said...

Eli, you seem to be saying there is no point in writing proposals. Both successful and unsuccessful proposals are "time sinks". Sure they are, but given that the NSF budget for basic science etc. research isn't going to increase in a major way anytime soon, do you think we should all just give up and not write proposals? Or are you saying we should write fewer proposals but make sure they are so good that funding is guaranteed? I wish it were so simple.

GMP said...

I'm seconding what Anons at 8:36 and 9:31 said. I write certainly 5+ grants every year (probably not more than 8 though), with 2-3 as a single PI and the others collaborative. How much money you need to bring in scales with the size of the group: e.g. for 10 people in the group, 1-2 people per grant (obviously I am assuming NSF-like budgets), you have to have about 5-8 active awards at any point in time which means between 2 and 3 new awards every year (assuming 3-year grants).

Another aspect is the field: in some fields, money brought in directly translates into advancement (good fundraiser = beloved by college administration) whereas in many others money is viewed purely as a necessity to get research done (should be sufficient but needn't be excessive). Then there's the career stage -- must have single PI grants on TT, afterwards you can go more collaborative...

I don't think anybody submits crappy grants precisely because it's such a huge time sink. At the NSF, often it really makes your heart sink what doesn't get funded, there's so much good science proposed; but with 10% paylines there is simply no guarantee.

Anonymous said...

This is off-topic for the main post, but a follow-up to Eli's comment.

I've wasted too much time over the last few years writing unsuccessful grant proposals. I'm seriously considering giving up on grant proposals entirely and working without grad students. My research productivity would probably go up.

I wish that grad students were funded more directly and not through the NSF/NIH lottery.

Anonymous said...

It is ironic that no one notices how all 3 women on the Supreme Court nodded to the gender discrimination charge, while the men were split 5-1.

Who are the closed minded ones here?

Would you be outraged if all the Catholics on the Supreme Court voted to have Catholic prayer made compulsory in public schools? Or would you accuse them of bias?

If you think that women can never do anything wrong by definition, well that is like saying: "my religious scriptures are faultless, your prophet is a liar".

Anonymous said...

Re: "It is ironic that no one notices how all 3 women on the Supreme Court nodded to the gender discrimination charge, while the men were split 5-1. "


Being a woman in physics, I'm acutely aware of how sexist many men are, especially the older ones or the ones who grew up in other countries. The men, however, are completely blind to it. If you ask them, they would report that there is no discrimination towards women. But if you ask any woman in the same department, the vast majority of them will quietly acknowledge discrimination, even if they never formally reported events to the University.

Anonymous said...

How does anyone get to the conclusion that "women never do anything wrong by definition" (Anon 3:19) from this post or anything else written here? Unlike the flawed analogy about Catholics, this conclusion is like my concluding that this Anon has 5 heads, just based on the bizarre comment and nonsensical "logic".

My interpretation of the male vs. female split by the Supreme Court is that the women actually know that it is absurd to conclude that there is no discrimination just because a company has a formal policy forbidding it, whereas only one of the male SC justices knows this.