Jobs for the girls

robp

Victor wrote: »

No, because a fair proportion of women defer to men and discriminate against other women - you get stuff like this: https://twitter.com/BiellaColeman/status/1060947727692021761

I support going gender blind. A blind algorithm is the way to go but it is not always possible. It would work fine as a way to promote a lecturer to professor but you couldn't use it hire lecturers.

That been said in Australia going gender blind has reduced the numbers of women in some jobs.
https://www.abc.net.au/news/2017-06-30/bilnd-recruitment-trial-to-improve-gender-equality-failing-study/8664888

Another example is the famous Uber study. Female Uber drivers earn less even though pay is based on an entirely gender blind formula. It couldn't be more gender blind. They earn less as they prefer to drive more safely by going at a slower speed, while men driver fast, incur risk to themselves and gain more pay as a result.
http://fortune.com/2018/02/06/uber-gender-pay-gap-study/

SterlingArcher

GM228 wrote: »

Indeed everyone in AH is a self proclaimed expert on just about everything at times.

Mind you the expertise of the experts on research, policy and law at the Gender Equality Taskforce is questionable in light of the report Peregrinus linked to, may as well of handed it over to AH for quality expertise

I would go further and say that the Gender Equality Taskforce was wholly compromised.

At least with the AH expertise there were no personal or subsidised financial gains or promise's to be made.

We are talking about the installation of systematic discrimination into academia.

Nobody in their right mind could consider this a wise move and the fact it is being ham fisted in speaks volumes.

Peregrinus

robp wrote: »

. . . Another example is the famous Uber study. Female Uber drivers earn less even though pay is based on an entirely gender blind formula. It couldn't be more gender blind. They earn less as they prefer to drive more safely by going at a slower speed, while men driver fast, incur risk to themselves and gain more pay as a result.
http://fortune.com/2018/02/06/uber-gender-pay-gap-study/

Which just goes to show that a "gender-blind" formula isn't necessarily truly gender-blind. What seems to be at work here is that the formula implicitly endorses the more typically male prioritisation of speed over safety, and therefore pays more money to drivers who share this priority, who are mostly men.

It's not that different from saying that an employment requirement of "must be at least 1.80 metres in height" is gender-blind. Formally it's gender-blind, but in practical operation it's anything but; it will result in an overwhelmingly male workforce.

robp

Peregrinus wrote: »

Which just goes to show that a "gender-blind" formula isn't necessarily truly gender-blind. What seems to be at work here is that the formula implicitly endorses the more typically male prioritisation of speed over safety, and therefore pays more money to drivers who share this priority, who are mostly men.

It's not that different from saying that an employment requirement of "must be at least 1.80 metres in height" is gender-blind. Formally it's gender-blind, but in practical operation it's anything but; it will result in an overwhelmingly male workforce.

There is nothing stopping women driving fast. Many women do drive fast and do so to a very high level. I'm a male and several female friends drive much faster and safer than me. Amongst US female Uber drivers, females on average choose to drive slightly slower. But is a choice. To imply women can't drive faster is patronising.

The Uber workforce exists to service a purpose. So do professors. And if either sex is not willing to do service that purpose than they don't deserve a job.

Look either men and women are essentially identical and should be equally represented or not.

Peregrinus

robp wrote: »

There is nothing stopping women driving fast. Many women do drive fast and do so to a very high level. I'm a male and several female friends drive much faster and safer than me. Amongst US female Uber drivers, females on average choose to drive slightly slower. But is a choice. To imply women can't drive faster is patronising.

The Uber workforce exists to service a purpose. So do professors. And if either sex is not willing to do service that purpose than they don't deserve a job.

Look either men and women are essentially identical and should be equally represented or not.

I'm not saying that they can't drive faster; I'm saying that, as reported, they typically don't drive faster. They balance the speed/safety tension differently to men.

The point here is that the formula pays more to faster drivers, therefore rewarding those drivers (mostly men) who prioritise speed. But this may simply reflect the fact that the formula was developed by men, or that it was developed by studying data from a pool of mostly male Uber drivers, and so it reflects the typically male prioritisation. It doesn't follow that the formula is in some objective sense "right", or that it is optimal.

7aubzxk43m2sni

Peregrinus wrote: »

I'm not saying that they can't drive faster; I'm saying that, as reported, they typically don't drive faster. They balance the speed/safety tension differently to men.

The point here is that the formula pays more to faster drivers, therefore rewarding those drivers (mostly men) who prioritise speed. But this may simply reflect the fact that the formula was developed by men, or that it was developed by studying data from a pool of mostly male Uber drivers, and so it reflects the typically male prioritisation. It doesn't follow that the formula is in some objective sense "right", or that it is optimal.

Do you think that it should be readjusted?

Peregrinus

I don't know; I just don't start from the assumption that it is presently optimal.

Basically, you can't defend the outcomes this formula produces by pointing to the fact that these are the outcomes it produces. My view would be that a formula that works to pay more to men than to women (or vice versa) requires investigation, and possibly revision. But you can't say that it definitely requires revision; that judgement can only be made after the investigation.

7aubzxk43m2sni

Peregrinus wrote: »

I don't know; I just don't start from the assumption that it is presently optimal.

Basically, you can't defend the outcomes this formula produces by pointing to the fact that these are the outcomes it produces. My view would be that a formula that works to pay more to men than to women (or vice versa) requires investigation, and possibly revision. But you can't say that it definitely requires revision; that judgement can only be made after the investigation.

It appears to me that the algorithm is as gender blind as an algorithm can be, and that equality of opportunity to earn equal amounts has been afforded to both genders.

Peregrinus

Pierce Colossal Pocket wrote: »

It appears to me that the algorithm is as gender blind as an algorithm can be, and that equality of opportunity to earn equal amounts has been afforded to both genders.

It depends on what you mean by "gender-blind".

To go back to an earlier example, a job requirement of "must be at least 1.80m in height" is formally gender-blind, but of course will exclude far more women than men. In its impact it's not gender-neutral at all.

Suppose we adopt this requirement for, say, a retail sales assistant position. Leaving aside the question of justice/fairness to applicants, there's an issue from the employer's point of view; lots of otherwise well-qualified applicants are going to be excluded by this. And, in other respects, some of them might make better sales assistants that some of the people who meet the criterion - they might be better at communicating with customers, for example, or more creative in meeting customers needs. And this doesn't just affect the employers, of course, but also the customers.

So we should ask ourselves "Why is this requirement here? Just how important is it that the assistants should be at least 1.80m tall?" And we might find that the requirement was included, say, because of a need to reach goods stored on the top shelf. In which case we should at least consider whether a better approach - better for employers, better for customers, fairer to jobseekers - might be to drop the requirement and instead provide footstools, or rearrange the shelving. Everyone's a winner!

Right: The Uber case. According to the article, the gender disparity was the result of a number of factors:

- Experience: 77% of women quity driving for Uber within 6 months, as compared with 65% of men. Longer-serving drivers tend to earn more.

- Speed: The male drivers tend to drive faster. The algorithm rewards this.

- Pitch: The male drivers are more likely to ply for hire in areas of higher demand, and more frequent surge pricing.

OK: Without thinking too deeply about it, two things leap out. First, the higher dropout rate for female drivers is in itself concerning, no? The gig economy is supposed to be flexible, to suit those who seek control over their work /life balance, etc. It's supposed to address the problems of inflexibility and demands for a 40-hour commitment that work to women's disadvantage in more traditionally-structured employments. But that doesn't seem to be working here. Why? There have already been reports of toxic and abusive culture at Uber; are these two things connected? It certainly seems worth looking into. This doesn't necessarily suggest that the algorithm needs changing, but it does point to other action that may need to be taken.

Secondly, the algorithm rewards speed, but does it ignore the safety/punctuality bonuses that are associated with driving at lower speed? If it rewards the one but not the other, that looks skewed, and it's a skew that impacts differently on men and women. It may be there's a good reason the relative rewards attached to speed and to safety/punctuality but, given that they impact differently on different genders, this looks like indirect discrimination unless and until that reason is identified and found to be objectively justified.

So, yeah, formal gender-blindness isn't really enough to avoid either the moral or the economic problems that result from employment/payment practices that tend to disfavour one gender.

robp

Peregrinus wrote: »

It depends on what you mean by "gender-blind".

To go back to an earlier example, a job requirement of "must be at least 1.80m in height" is formally gender-blind, but of course will exclude far more women than men. In its impact it's not gender-neutral at all.

Suppose we adopt this requirement for, say, a retail sales assistant position. Leaving aside the question of justice/fairness to applicants, there's an issue from the employer's point of view; lots of otherwise well-qualified applicants are going to be excluded by this. And, in other respects, some of them might make better sales assistants that some of the people who meet the criterion - they might be better at communicating with customers, for example, or more creative in meeting customers needs. And this doesn't just affect the employers, of course, but also the customers.

So we should ask ourselves "Why is this requirement here? Just how important is it that the assistants should be at least 1.80m tall?" And we might find that the requirement was included, say, because of a need to reach goods stored on the top shelf. In which case we should at least consider whether a better approach - better for employers, better for customers, fairer to jobseekers - might be to drop the requirement and instead provide footstools, or rearrange the shelving. Everyone's a winner!

Right: The Uber case. According to the article, the gender disparity was the result of a number of factors:

- Experience: 77% of women quity driving for Uber within 6 months, as compared with 65% of men. Longer-serving drivers tend to earn more.

- Speed: The male drivers tend to drive faster. The algorithm rewards this.

- Pitch: The male drivers are more likely to ply for hire in areas of higher demand, and more frequent surge pricing.

OK: Without thinking too deeply about it, two things leap out. First, the higher dropout rate for female drivers is in itself concerning, no? The gig economy is supposed to be flexible, to suit those who seek control over their work /life balance, etc. It's supposed to address the problems of inflexibility and demands for a 40-hour commitment that work to women's disadvantage in more traditionally-structured employments. But that doesn't seem to be working here. Why? There have already been reports of toxic and abusive culture at Uber; are these two things connected? It certainly seems worth looking into. This doesn't necessarily suggest that the algorithm needs changing, but it does point to other action that may need to be taken.

Secondly, the algorithm rewards speed, but does it ignore the safety/punctuality bonuses that are associated with driving at lower speed? If it rewards the one but not the other, that looks skewed, and it's a skew that impacts differently on men and women. It may be there's a good reason the relative rewards attached to speed and to safety/punctuality but, given that they impact differently on different genders, this looks like indirect discrimination unless and until that reason is identified and found to be objectively justified.

So, yeah, formal gender-blindness isn't really enough to avoid either the moral or the economic problems that result from employment/payment practices that tend to disfavour one gender.

Uber doesnt exist to fulfil the expectations of your ideological agenda. It doesn't exist to provide jobs for women, or men. It exists to make a a ton of money by displacing taxis and making transport easier. To argue it was designed for men by men is not only far fetched but plucked out of thin air. By your logic allowing men to take less sick days then women is indirect discrimination because it allows them to advance their careers in way lost to women.

Peregrinus

robp wrote: »

Uber doesnt exist to fulfil the expectations of your ideological agenda. It doesn't exist to provide jobs for women, or men. It exists to make a a ton of money by displacing taxis and making transport easier. To argue it was designed for men by men is not only far fetched but plucked out of thin air.

It was plucked out of thin air by you. This is not an argument I am advancing at all. Your insecurities are showing, robp. Is your safe space being threatened?