Microsoft tries, fails to crush 'gender bias' lawsuit brought by its own women engineers Microsoft has failed in a bid to shoot down a lawsuit alleging that its employee rating system was biased against women. A US district court in Washington has tossed out [PDF] the Redmond giant's motion to dismiss a complaint lobbed at it by three women engineers, who allege the system for evaluating engineering and technical …
Sure, it could be true. But it could also be an issue of people who got it into their head that they 'deserve' a positive promotion and will now start blaming everything except themselves.
I think people sometimes get it into their head that they're entitled for things like a promotion, but fact of the matter is that you're entitled to getting paid to perform the job for which you've been hired. It is awesome if you get promoted and rise up in the ranks, but it's most certainly not the companies obligation to even do so.
And I think that last part is what some people tend to forget sometimes.
The issue is whether Slurp has enough real evidence to have case tossed. The judge ruled they so far have not shown enough. Given the arrogance of Slurp and their incompetent (more accurately criminal) HR systems there is a very real possibility these suits have merit.
"Sure, it could be true. But it could also be an issue of people who got it into their head that they 'deserve' a positive promotion and will now start blaming everything except themselves."
This is all fine until you start looking at how groups work and, in particular how covert or hidden bullying works. A new member of staff doesn't even need to have done anything wrong to be subject to covert bullying. Their only "mistake" is say to have been appointed to a team where the manager, irrespective as to how effective they are, happens to be managing a project that has internal political pressure on it and divided opinions as to the usefulness of the project. Those who dislike that project or that manager will tend to transfer those feelings onto anyone assigned to that team.
Sad to say it doesn't take much to get otherwise reasonable and social people to join in with such back biting. As it is covert the object of the attacks will struggle to work out what is actually going on. This type of situation will inevitably skew any peer review system against the person concerned without them knowing that anything is afoot. Even if that person is then placed in another team.
Sure, it can be down to whining women and gender politics but then again maybe that Trump - Billy Bush video indicates that maybe women have a point. There *are* plenty of ways to exclude women from the type of male bonding that takes place in all-male gatherings. Gatherings that can take place in work time when women on the team are not present. Gatherings that strengthen a male bias in a team by encouraging the airing of negative comments about women.
You've completely missed the point. The plaintiffs are alleging that statistically speaking women are invariably ranked downwards in the their performance reviews compared to men. If its true that can only be the product of bias either conscious or unconscious. "sides of the story" dont come into it in this case as its potentially about all females not the expecations of one or two.
Of course that doesnt rule out demographics and other measures impacting the underlying populations but its certainly not worth dismissing it out of hand.
It also assumes the women you hired are averagely as good as the men you hired.
But - This isn't the same as saying are women aren't as good as men.
If the prestigious university CS depts you recruit from are 92% male and the small percentage of women are being grabbed by academic programs and your competitors who are all trying to make their own gender balance look better then you may have to look for more non-traditional candidates to increase the number of women. These might not perform as well
This is just statistics, nothing to do with women. If you were ordered to make your own company/group/football team 50% Welsh speaking by the end of the year then you might have difficulty filling those roles at the same level of qualification you normally demand. However good the Welsh are at IT/football. The long term solution would be to encourage more children to learn Welsh - but there isn't much your dept can do about that this quarter.
"The plaintiffs are alleging that statistically speaking women are invariably ranked downwards in the their performance reviews compared to men."
That was nowhere in the article.
The plaintiffs are saying that the system is biased because the management AND peers are male, and there is no way for them, as women, to get a fair review. It is much the same argument as a black person being tried in front of a jury of white people, lawyers and judge. I can see their point, but exactly how do you fix it?
It is much the same argument as a black person being tried in front of a jury of white people, lawyers and judge. I can see their point, but exactly how do you fix it?
For IT people and for juries, I don't know. But you can at least try. For example, professional orchestras now routinely audition for new members behind a screen, to try to prevent any unintentional bias. And there are now more women and ethnic minorities in orchestras.
This post has been deleted by its author
I really hope that's just a few little typos and not an accurate portrayal of how long it takes the Reg to post a story. :)
I think you misunderstood the article - the dates quoted are, if you like, the "statute of limitations" on the case. In other words, the Judge has said that any evidence dated from Sept 14 2012 onwards will be admissable in the eventual trial (whether the evidence is valid, is of course, for the actual trial to decide). The plaintiffs had requested an earlier cut-off date, which makes me suspect that in at least one of their cases, the "I didn't get the promotion" incident occurred sometime soon after May 2011.
Shame on you for thinking that the Reg would ever be late to file a story! Repent, O ye of little faith!
as the BOFH said once. You can do the job? You are in. For me this whole age/sex/orientation/whatever is not interesting in a professional context (and if people behave professionally).
I do admit: it is a serious problem if people are discriminated against - no matter what. Unless they are, as Simon put it, "a thicko" and should not be allowed an opinion. But do not think you are better than anybody just because you are white/black/female/middle-aged/christian/pastafarian/whatever.
I know I am too thick in some fields, I make an effort of not pushing my thoughts on those who have a clue (I will ask and try to learn, though).
On the other hand: one of my former female colleagues at university made a point that she was milking the system, and that she was able to get funds etc. just because she was female... this was totally against her "unequal opportunity" policy.
"as the BOFH said once. You can do the job? You are in. For me this whole age/sex/orientation/whatever is not interesting in a professional context (and if people behave professionally)."
I feel the same. If you have the experience, and can do the job... no matter your sex/age/whatever, as long as the job get done and is done properly.
The entire case is based on the assumption that all men are misogynist and therefore their peers will discriminate against them in these reviews. This is patently absurd.
A peer based system is far more likely to benefit these women because any misogynists will have their opinions balanced out by normal guys who will respect merit. If the female members of staff are competent.
But I am guessing any women who are more concerned with gender politics than doing their jobs well aren't going to stack up well against their peers when judged on merit.
The entire case is based on the assumption that all men are misogynist and therefore their peers will discriminate against them in these reviews
Flawed assumptions indeed.
The case is based on the assumption that all other things being equal, people tend to prefer members of the same social group. Depending on how and where you were brought up and lived, the specific group boundaries may differ - however, they will always be present. This is a perfectly natural and often useful bias, and it takes a conscious effort to suppress it in situations where it is not approapriate. This effort can be made easier or harder (and therefore more or less effective) by organizing the evaluation in different ways.
The plaintiffs in this case clearly feel that their employer did not do enough to make the process fair. Without knowing the details of the case, it is impossible to tell whether they are right - however, the trial judge, who is in the position to assess all evidence, seems to agree that there is a case to answer.
"The case is based on the assumption that all other things being equal, people tend to prefer members of the same social group."
In a neutral environment where people can choose their social circle I may have conceeded this point. In a forced work environment I do not believe this to be the case.
At work virtually everyone I know - with a few bigotted exceptions - everyone I know will choose to be biased towards the person who is easy to get along with and who makes the teams job easier and therefore their lives easier. Merit is the driving factor for most.
Women who focus on gender issues instead of work, who cause friction in the workplace will be passed over for promotion for good reason.
No, I would have thought the case boils down to this simple statement from their lawyer.
"because as a group they received, on average, lower rankings despite equal or better performance"
And if they can prove that as true then clearly MS are in the frame and it's not just the case of a few uppity women "focusing on gender issues".
"because as a group they received, on average, lower rankings despite equal or better performance"
This is clearly the heart of the matter. However, there is the obvious problem of who decided that their performance was equal or better... particularly as it would appear that their peer review decided that they weren't. Though obviously this leads to a circuitous argument.
Do they have an independent evaluation of their performance, and of all of their peers? Or is it their own assessment of their performance?
I'm sure most white, middle-aged, middle-class men will have seen other people being promoted and wondered why, and maybe even thought... "But I'm better than them!".
Doesn't make it so...
"A peer based system is far more likely to benefit these women because any misogynists will have their opinions balanced out by normal guys who will respect merit."
Unfortunately not. Normal guys who fully respect women will rate on ability alone, not favouring women over men, hence there would be an underlying trend favouring men. It only takes one misogynist to start skewing the figures.
Guys like champions. We respond well to them. We raise them on our shoulders. We are socialized to admire those that achieve greatness. This works well with genuine egalitarian female professionals. They get treated like another professional and actually get less grief from men than from other women.
In practice, it's often lower ranked females that would seek to sabotage a female professional.
Your maths is wrong there.
Negative inputs by misogynists are not 'balanced out' by normal guys, but merely diluted by them.
If you have two identical candidates, a man and a woman, who are reviewed by ten peers, and one of those peers is a misogynist, then all else being equal the man will get the better review because the misogynist poisons the well, however slightly. If those two candidates were competing for a post, then that slight disadvantage becomes a binary loss of opportunity.
Microsoft may have had a tricky problem in that in an overwhelmingly male environment, inevitably the few females were reviewed by mostly male peers, and the above problem would (not could) manifest itself. However Microsoft could (at least after being alerted to the problem) have taken reasonable steps to reduce it, such as ensuring feedback was not anonymous, including 'dotted line' review processes where people of similar skills but in different teams participated in review processes, etc.
Such an approach would help all bullying, not just sexism, so isn't positive discrimination or the like.
Hopefully it's clear that all of the above applies regardless of the actual merits of these three individual women.
based on valid and reliable performance measures
Such as:
* Machines borked per mandatory update?
* Gigabytes of private data per Windows 10 beta tester alpha tester victim user?
* Number of disabled data slurp settings reset to defaults?
* Most third-party devices rendered unusable?
That on average men can't be expected to evaluate women fairly, in a peer review setting.
So you need to be certified unbiased for peer reviewing? Or women should only be reviewed by other women.
Expand that out into other Minorities. This could end with really perverse outcomes.
Oh hang on its the US! Their legal system always gets it right.
Never mind
> "Furthermore, Plaintiffs allege that the 'vast majority (over 80 per cent)' of the managers at the calibration meetings and people discussions were men, and Plaintiffs posit that 'female technical employees were systematically undervalued compared to their male peers because as a group they received, on average, lower rankings despite equal or better performance'," the court order continues"
Over 80% male. Say we round it down to be exactly 80% male. That means, given that it was based on peer reviews, that should they have done equal quality of work to their male counterparts, they had a 20% chance of getting the promotion. Not good odds to start with, as betting agencies make a fortune out of fools who think a 20% chance is a "sure thing".
So because this 20% chance doesnt come to fruition for 3 people (which doesnt even meet the minimum number required for a 20% chance to actually pay off once), it's automatically gender bias, rather than skills bias.
> "The engineers allege that the review system relies on manager and peer input from a group that is overwhelmingly male and, as a result, the female employees they evaluated may have missed out on raises and promotions."
Perhaps they should be suing their female ex-school peers, who decided against going into STEM jobs and thus perpetuating the very "gender bias" they are now claiming to be fighting.
> "Plaintiffs allege these performance evaluation methods are 'invalid' because they 'set arbitrary cutoffs among performers with similar performance' and are 'not based on valid and reliable performance measures'," the court's ruling, dated October 14, reads."
So it's gender bias because everyone who gave the same level of performance, were treated the same?
>So because this 20% chance doesnt come to fruition for 3 people (which doesnt even meet the minimum number required for a 20% chance to actually pay off once), it's automatically gender bias, rather than skills bias.
That is assuming that the selection is entirely random, or all are equally skilled and productive. If one of the female engineers where indeed better than their male colleagues, then this would bring her into contention for the role. That again is assuming that the promotion is solely based on their technical ability and ability to deliver. Itmay well be that the position that they will be promoted to will need other skills that they may not have exhibited.
Where employing a woman will be seen as high risk due to the excessive chance of gender-specific litigation that will follow when the new employee isn't treated as the special snowflake she believes herself to be.
Decades of sacrifice, suffering progress and astonishing work by generations of women undone in a few years by the current breed of whining non-thinking over-entitled third wave feminists. Well done all of you.
There will always be those that have an over-inflated view of themselves that feel hard done by when overlooked for opportunities. If you are in a minority in a pool of equally suitable applicants, then the chances of you being given the job is low. That's how equality works. Trying to induce gender balance in senior positions where the pool is not in balance is a positive discrimination scenario - unless you are lucky to have your minorities being more suitable for the role. But would you then have that niggling thought that you only got the role to satisfy a quota?
Okay my query is, if you can't use peer and managerial review/input to decide promotions/rises (which I assume is taken alongside objective performance data) how else do you decide who gets the up and who doesn't?
I say we go back to nepotism and length of service with basic pyramids.
Note I go ballistic when I just get a flat department wide payrise, nothing more vexxing than getting the same rise as some card whose done nothing all year (well there is, but not for the purpose of that sentence). Normally a sign that it's time to move on to a new role.
Anyway, as to this specific instance, dunno, not involved, if the systems broke it's worth sorting out for all involved. If it's not - oh well more change for the lawyers.
> if you can't use peer and managerial review/input to decide promotions/rises.... how else do you decide who gets the up and who doesn't?
Algorithms, big-data, data-science.Then you can't be blamed for anything because it's SCIENCE.
ie you invent a bunch of meaningless metrics (lines of code, bugs closed, features implemented) which are then gamed by the same weasels who downrate women - but this time you cant complain because SCIENCE.
Then we can get a good look at exactly what metrics being used are. If there is indeed significant wiggle room in assessing performance, and clear indications that better candidates were passed over, then there should be penalties. If the problem is just a perceived one, the perception being that men can't be unbiased in their assessments, then the case should fail.
Trying to decide the outcome before the evidence has been gathered and assessed just reveals your own, liberal or authoritarian, prejudice.
I think that should be "unbiased based on gender / ethnicity / religion / sexual orientation"
Given if we're asserting that a person will be biased based on their gender then everything else is up for questioning too.
However everyone is biased about things to some degree or another the goal is to have the correct set of checks and balances to ensure a fair enough system. I mean I'm biased against people that don't investigate things for themselves and ask me stupid questions, have you not heard of google people! Also I can't help but distrust anyone that mentions crossing your arms is defensive, and anyone that likes physical contact as a form of communication. There is a long long list of other things as well.
I certainly wouldn't imagine anyone suggesting that you can only ever be assessed by someone who perfectly matches the assessed's various cultural, genetic, social and religious parameters. Isn't that called a self assessment? (sorry posing myself a question)
Anyway the outcome should be interesting I assume the answer will be an expert computer system of some sort.
'Tis a shame I only have but one upvote to give you.
That is the whole reason for going to court - to assemble the evidence, look at it all together, and then determine guilt/innocence. Pre-judging the case based on soundbites and opening statements is an excellent way to display your own bias.
@sabroni - well said! I'd add that corporate culture can make a huge difference in how fairly any minority group is treated in the workplace, irrespective of bias in the culture in general. I've seen this in my own career.
The issue centres around whether Micorsofts' method of assessing performance was objective or not. Without knowing the precise details of how they assess their employees, it's rather difficult to comment sensibly on that.
It has been interesting though, to see the reaction of commentards here, some of whom clearly do appear to have bigoted attitudes towards women and the feminist movement, and seem to forget that they, like any other group (like, ooh, I dunno - men in general, say?) do number amongst them the flawed, the narrow-minded and the whiny ones. But that doesn't mean that all of them are so, and indeed, the vast majority of them are not. Well done to the rest of you for your more balanced comments!
I think people on the internet have a tendency to over react and to generally behave in ways they wouldn't ever imagine in person. So I kind of imagine that most of the things said on the internet are the angry 4 year old inside our mind that fly off the handle at a moments notice and if we were all having a nice discussion around a table and a tasty meal everyone would be far less shouty.
Undoubtedly also partly true that golf buddies will boost the ratings of their buddies, Battlefield players will give points to those who also play over World of Warcrafters. Etc etc
Problem is compunded if they have anti-discrimitaion policies in the HR dept where those who get employed in the first place are given unwarranted priority due to gender.
The other option is that they could be reviewed by a peer group of women.... and I bet they find they do just as poorly if not worse, because from my very relevant experience in the role I hold just now, women reviewing other women are far more likely to put the boot in so they don't face as much competition.
In some sectors, women will always be under-represented (some elements of IT, coal mines, etc) and the few women that are in the roles there are far more protective of their territory and don't like other female competition.
I reckon, from observation, that company assessments are cruelly flawed. They have no reliable metrics. They have no reliable method of knowing the metrics were properly applied. They have no metrics to decide whether a pay rise of X should be given to A rather than B. They clearly have no metrics for measuring managers against baseline competence, nor one external candidate from another.
Personally, if I have a good view of an individual's work, I can generally rank them as "yeah, more or less ok", "bloody magic" or "bloody useless". Since there aren't generally many magic or useless folk, this ranking doesn't do much for fine gradations between employees. And, of course, I don't have good knowledge of most people's performance, so much of any ranking has to be the fourth, important, category "dunno".
Not at all clear that managers have, in general, even this degree of insight into people's performance.
oh well.
And not only are they complaining, but they're bringing in a lawyer. Way to demonstrate how little faith you have in His Clairvoyant Vision And Impartial Justice. </sarcasm>
Then again, I would hardly be surprised if they were right.
Most of the comments show a deep lack of understanding about Microsoft's hiring practices and company hierarchy. Anyone who has ever been inside the company can tell you that nepotism and sexism are rampant, and most hiring decisions are made for financial reasons.
The company is not doing well, internally and externally, so they've been taking every opportunity possible to trim costs and decrease head count. This disproportionately affects the lower levels of the rank and file, since managers get quarterly bonuses where they can influence the total bonus based on "performance" and head count. It's a company culture that rewards backstabbing, and it comes as no surprise that the men are more backstab-happy than the women.
I can rubbish their products just fine. I can examine them myself. Stuff like this is nonsense, rumours, and hearsay. I'm as willing as the next guy (or 10) to bash Microsoft, but I am going to make any assumptions here or put stock in anyone that does.
The problem becomes: how/who can evaluate without bias? And if a STEM job is overwhelmingly male, how can a woman be evaluated without the system being called sexist? The same could be said for any <minority group in the workplace> that is dominated by the <majority group in the workplace>. Can we no longer take into account peers and management, if they are of a different gender/gender identification/race/nationality?
As an example, for a few years there was only ONE female software developer in our office. Does that mean that there was no way for her to get a fair review? (she is excellent, BTW).
The REAL problem here is that hideous 'peer review' system, only something that an IDIOT could have rectally extrapolated from the bowels of hell.
Other than the obvious, that a handful of idiots who use their personal bias in reviewing others are skewing the results enough to make a difference [assuming those hired in management positions are LESS likely to do so AND are in a position of ACCOUNTABILITY if they DO], with no one to blame for the (alleged) bias and result-skewing, the system ITSELF is to blame.
It's kind of like relying on up/down votes to validate your point, rather than FACTS or LOGIC. Too easy for a handful of HOWLER MONKEYS to come along and sling poo (aka downvote en masse) a particular group or opinion, thereby creating an *APPEARANCE* of unpopularity, in an attempt to shame or discredit a perfectly valid opinion or position.
So I'll blame the SYSTEM, that IDIOTIC 'peer review' system that Micro-shaft has been using, a result of the former COO who was CANNED a while back...
http://forums.theregister.co.uk/forum/1/2016/07/08/kevin_turner_and_the_new_microsoft/
It's a fair bet that THIS guy was responsible for a NUMBER of *FAILS* at Micro-shaft over the last decade or so since he's been around, *INCLUDING* GWX!!!
And so, by creating a (literal) *HOSTILE* environment within the company, they got what they asked for: A lawsuit by a couple of women who may *ACTUALLY* have a legit case, due to the HOSTILE 'peer review' system that somehow "downvoted" them OUT of promotions and raises!
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
