Current Events > Male uber drivers make more than female drivers with gender blind algorithm

(+) Yeah! (+)

http://freakonomics.com/podcast/what-can-uber-teach-us-about-the-gender-pay-gap/

Get ready for this study to be used to reinforce peoples own beliefs

ive used uber tons of times and i dont think ive ever had a female uber driver before
---

Wow what a sexist article we need to ban all algorithms
---

why not quote the study

https://web.stanford.edu/~diamondr/UberPayGap.pdf

The growth of the "gig" economy generates worker flexibility that, some have speculated, will
favor women. We explore one facet of the gig economy by examining labor supply choices
and earnings among more than a million rideshare drivers on Uber in the U.S. Perhaps most
surprisingly, we find that there is a roughly 7% gender earnings gap amongst drivers. The
uniqueness of our dataknowing exactly the production and compensation functionspermits
us to completely unpack the underlying determinants of the gender earnings gap. We find that
the entire gender gap is caused by three factors: experience on the platform (learning-by-doing),
preferences over where/when to work, and preferences for driving speed. This suggests that,
as the gig economy grows and brings more flexibility in employment, womens relatively high
opportunity cost of non-paid-work time and gender-based preference differences can perpetuate
a gender earnings gap even in the absence of discrimination.

---

Actual paper:
https://web.stanford.edu/~diamondr/UberPayGap.pdf

uber driver income is gender-blind. Their income is the result of miles driven and availability/surge pricing. Men make more on average because they're more willing to work during surge times (ie, weekends at ~2am), because the average male uber driver has more experience than an average female uber driver (so they're able to take shortcuts/"local routes"), and be more likely to speed which is also higher money over time.

Not surprising. Also females only make up <30% of the driver work force.

Edit - ninja'd~
---

https://twitter.com/bv/status/963077450379988993
https://twitter.com/bv/status/963077454771445761
And. Here. We. Go.

Tag
---

Women may actually be making less for doing their job better, from a customer's point of view

what a ridiculous claim
---

So according to those twitter posts, this has nothing to do with gender and everything to do with individual people

ok
---

mostly it has to do with customers priorities

high priorities:

driver available (faster driving = more pickups = more available = more $$$)
driver available at odd hours (throttled pricing = more $$$ at 2AM than 2PM)

lower priorities:

safe driving
---

I've never had a female uber driver
---

CruelBuffalo posted...

https://twitter.com/bv/status/963077450379988993
https://twitter.com/bv/status/963077454771445761
And. Here. We. Go.

whatever gets em clicks.

If we posit that "driving the speed limit for safety purposes" is "driving better", then, sure. They're right.
---

philsov posted...

If we posit that "driving the speed limit for safety purposes" is "driving better", then, sure. They're right.

but is the driver's safety the primary metric for 'doing their job better' in the customer's point of view? that was assumed by bloomberg, so they could draw that conclusion.

in reality, the biggest reward comes from availability at odd hours, which would be implied to be the best metric of 'doing your job better'
---

The company's gender-blind algorithm is falling short: Men still earn about 7% more than women do

I fail to see how this is necessarily a bad thing as long as the gap is strictly the result of personal choices.
---

The company's gender-blind algorithm is falling short: Men still earn about 7% more than women do

How can someone type "the bling algorithm is falling short" with a straight face? A more accurate statement would be:

"Despite the gender-blind algorithm, women are falling short and earning 7% less than men"

I guess that wouldn't go over well though.
---

What a shock, a "wage gap" study that finds that men earn more for the same job by working longer hours during harder periods with more experience is being used to argue for equality of outcome, not equality of opportunity.
---

s0nicfan posted...

What a shock, a "wage gap" study that finds that men earn more for the same job by working longer hours during harder periods with more experience is being used to argue for equality of outcome, not equality of opportunity.

yep. by bloomberg, a source that is focusing its market on specific customer bases, such as 'angry men who respond to ragebait' and 'angry women who are looking for fulfillment of their confirmation bias'
---

I've used Uber in 3 different cities now, ~20 rides total, and have yet to get a female driver.

Also, I hope Uber is ready for the angry feminist mob heading their way.
---

voldothegr8 posted...

I've used Uber in 3 different cities now, ~20 rides total, and have yet to get a female driver.

Also, I hope Uber is ready for the angry feminist mob heading their way.

I had one female driver, lol she also showed she had a gun after chatting with her a bit. That was interesting lol. (We were asking how she deals with assholes)

Darkman124 posted...

s0nicfan posted...

What a shock, a "wage gap" study that finds that men earn more for the same job by working longer hours during harder periods with more experience is being used to argue for equality of outcome, not equality of opportunity.

yep. by bloomberg, a source that is focusing its market on specific customer bases, such as 'angry men who respond to ragebait' and 'angry women who are looking for fulfillment of their confirmation bias'

Lol this

I'm an Uber driver and have no fucking idea how any one person can make more or less than another based on race

Nobody tips

You can't increase the fare manually

You drive the hours you want

Fricking dumb
---

CruelBuffalo posted...

voldothegr8 posted...

I've used Uber in 3 different cities now, ~20 rides total, and have yet to get a female driver.

Also, I hope Uber is ready for the angry feminist mob heading their way.

I had one female driver, lol she also showed she had a gun after chatting with her a bit. That was interesting lol. (We were asking how she deals with assholes)

Actually has multiple in Vegas. They were all good

GregShmedley posted...

CruelBuffalo posted...

voldothegr8 posted...

I've used Uber in 3 different cities now, ~20 rides total, and have yet to get a female driver.

Also, I hope Uber is ready for the angry feminist mob heading their way.

I had one female driver, lol she also showed she had a gun after chatting with her a bit. That was interesting lol. (We were asking how she deals with assholes)

That was stupid of her.

Yeah, but if shes working late hours and if she has a permit....cant blame her

Some one should archive all that crap Bloomberg posted as an example of how the media can spread false information with out actually saying some thing that is incorrect. It's really amazing.
---

Reading about any study from the general media is mind numbing. I recommend everyone disengage any time a journalist says "study says..."
---

You should chang it to women get paid less for being slow

lmao
---

COVxy posted...

Reading about any study from the general media is mind numbing. I recommend everyone disengage any time a journalist says "study says..."

Especially on hot button issues like this

Just because you get to a destination faster than someone else doesn't necessarily mean you did it in a more unsafe way... where is Bloomberg getting the "women drive safer" stat? Was it in the study?
---

This study quite clearly illuminates a structural bias:
- Men tend to be more available than women during odd hours. Why? Supposedly because women have to spend more time doing unpaid labor at home, such as chores, taking care of kids, etc.
- Men tend to prefer driving faster. This is a whole can of nature/nurture worms on its own.

So even when it looks equal on a surface level, because the algorithm doesn't discriminate by gender, these structures result in an unequal distribution. This is the part that one side cries itself hoarse over, and the other handwaves away as "personal choices".
---

scar the 1 posted...

So even when it looks equal on a surface level, because the algorithm doesn't discriminate by gender, these structures result in an unequal distribution. This is the part that one side cries itself hoarse over, and the other handwaves away as "personal choices".

OTOH, the algorithm's purpose isn't to account for structural gender problems, it's to reward the drivers who earn Uber the most profit, thereby encouraging more behavior by them that earns them the most profit

the structural biases working against women--their driving habits resulting in less fares/hour and their domestic responsibilities resulting in less of the most needed hours--means they earn uber less money.

it wouldn't make sense for uber to pay employees who earn them less the same as employees who earn them more. they'd lose their highest-earning employees to a competitor who did not try to correct for these biases. quits are at all-time highs right now, so this is a real concern.
---

AngelsNAirwav3s posted...

Wow what a sexist article we need to ban all algorithms

writing unbiased algorithms is a super fucking difficult task

there's some good reports on how FB used code from new hires who were isolated from the rest of the team to prevent subconscious biases on content distribution

...and then it still ended up being biased in favor of left wing news

Nuanced study that mindless journalists can only see in black and white. Just another day ending in y
---

Darkman124 posted...

scar the 1 posted...

So even when it looks equal on a surface level, because the algorithm doesn't discriminate by gender, these structures result in an unequal distribution. This is the part that one side cries itself hoarse over, and the other handwaves away as "personal choices".

OTOH, the algorithm's purpose isn't to account for structural gender problems, it's to reward the drivers who earn Uber the most profit, thereby encouraging more behavior by them that earns them the most profit

the structural biases working against women--their driving habits resulting in less fares/hour and their domestic responsibilities resulting in less of the most needed hours--means they earn uber less money.

it wouldn't make sense for uber to pay employees who earn them less the same as employees who earn them more. they'd lose their highest-earning employees to a competitor who did not try to correct for these biases. quits are at all-time highs right now, so this is a real concern.

Would be good to see how this would change for single women and single men though idk if Uber has that information on its riders. Because for taxes you may be single, but you could still be living with someone/dating

CruelBuffalo posted...

Would be good to see how this would change for single women and single men though idk if Uber has that information on its riders. Because for taxes you may be single, but you could still be living with someone/dating

that was also my thought on the unpaid household labor bit tbh
---

I've used Uber in at least 5 or 6 cities across three states. I probably have ridden in Uber at least 50+ times. I literally have never had a female driver.
---

scar the 1 posted...

So even when it looks equal on a surface level, because the algorithm doesn't discriminate by gender, these structures result in an unequal distribution. This is the part that one side cries itself hoarse over, and the other handwaves away as "personal choices".

It is just personal choices.
---

scar the 1 posted...

This study quite clearly illuminates a structural bias:
- Men tend to be more available than women during odd hours. Why? Supposedly because women have to spend more time doing unpaid labor at home, such as chores, taking care of kids, etc.
- Men tend to prefer driving faster. This is a whole can of nature/nurture worms on its own.

So even when it looks equal on a surface level, because the algorithm doesn't discriminate by gender, these structures result in an unequal distribution. This is the part that one side cries itself hoarse over, and the other handwaves away as "personal choices".

It doesn't illuminate ANY bias by default. Like you said, whether it's "bias" or "choice" is a matter of perspective and debate, so to start with the claim that it is showing bias is to put the cart before the horse.

And they don't result in unequal distribution. They result in equal distribution for the labor performed. IF a woman worked the same hours during the same shifts, she would receive the same pay. It'd be like if I claimed a part time worker making less than a full time worker was "unequal distribution."
---

I don't think that Uber needs to change anything about their payment plan. As far as the company is concerned, the interesting question for them might be "Why do more women choose to stop driving for us?". My friend is a part-time Lyft driver, and if I were in her shoes, I definitely would have stopped driving based on how terrible some of her customers sound.

I think this does indicate something that's important to me especially with regards to blind algorithms or machine learning, which is that an algorithm is only as blind as the data that goes into it.
---

The very last thing we need is "feminism in machine learning." ML is already on the verge of collapse (again) after like a 30 year hiatus. Having to weave politics into it will nuke it entirely for another few decades.
---

s0nicfan posted...

The very last thing we need is "feminism in machine learning." ML is already on the verge of collapse (again) after like a 30 year hiatus. Having to weave politics into it will nuke it entirely for another few decades.

What are you talking about.

Machine learning is not on the verge of collapse.
---

I don't see how it's an unequal distribution if it is a blind algorithm that pays X salary for Y work. Men doing the work at more demanding hours and doing it more efficiently doesn't change that.

I also refuse to call the domestic maintenance people do at home "unpaid labor". Who the fuck is going to pay you for taking care of your kids and doing your dishes? Should I bill my wife next time I fix the A.C., take care of the pest control, or take out daughter to school? Get outta here.
---

Anteaterking posted...

s0nicfan posted...

The very last thing we need is "feminism in machine learning." ML is already on the verge of collapse (again) after like a 30 year hiatus. Having to weave politics into it will nuke it entirely for another few decades.

What are you talking about.

Machine learning is not on the verge of collapse.

A number of companies, but primarily google, have shown that through some simple tricks you can abuse the fundamental structure of neural networks to trick them into classifying things to whatever answer you want (among those it provides). It's a huge blow to ML because it touches on the foundation of the theory and casts serious doubts as to whether it can be "patched" or whether ML for classification is a dead end.

http://www.slate.com/blogs/future_tense/2018/01/04/google_researchers_tricked_an_a_i_into_thinking_a_banana_was_a_toaster.html
https://gizmodo.com/this-simple-sticker-can-trick-neural-networks-into-thin-1821735479
---

So men are more likely to work at rush hour (dangerous conditions), and drive faster to get passengers to their destinations, picking up more riders in the process (efficiency) ...but women drivers in general should get paid the same despite putting in less work?

3rd wave feminism need to go to hell.
---

s0nicfan posted...

Anteaterking posted...

s0nicfan posted...

The very last thing we need is "feminism in machine learning." ML is already on the verge of collapse (again) after like a 30 year hiatus. Having to weave politics into it will nuke it entirely for another few decades.

What are you talking about.

Machine learning is not on the verge of collapse.

A number of companies, but primarily google, have shown that through some simple tricks you can abuse the fundamental structure of neural networks to trick them into classifying things to whatever answer you want (among those it provides). It's a huge blow to ML because it touches on the foundation of the theory and casts serious doubts as to whether it can be "patched" or whether ML for classification is a dead end.

http://www.slate.com/blogs/future_tense/2018/01/04/google_researchers_tricked_an_a_i_into_thinking_a_banana_was_a_toaster.html
https://gizmodo.com/this-simple-sticker-can-trick-neural-networks-into-thin-1821735479

I think you're vastly oversimplifying the field to think that all or even most of machine learning is in image recognition or on data sets where such "sabotage" is a concern.
---

Men make most of that computery algorithmy stuff so no surprise it's sexist. More proof we need more feminism.

Anteaterking posted...

I think you're vastly oversimplifying the field to think that all or even most of machine learning is in image recognition or on data sets where such "sabotage" is a concern.

Ehh... that's fair. Although I think you're under-emphasizing the importance of "sabotage" in classification, since ANY classification algorithm that uses DL or NN would be susceptible to abuse. I was speaking more from experience where its getting harder to sell customers on ML as a solution because of stories like those, and if the money dries up so does the research in that area.
---

s0nicfan posted...

Ehh... that's fair. Although I think you're under-emphasizing the importance of "sabotage" in classification, since ANY classification algorithm that uses DL or NN would be susceptible to abuse. I was speaking more from experience where its getting harder to sell customers on ML as a solution because of stories like those, and if the money dries up so does the research in that area.

I just don't think your concern about "feminism in ML" is well founded. We have laws against certain types of discrimination in this country, so it makes sense that we don't give companies carte blanche to say "I'm not being discriminatory, the blind algorithm is!" for making decisions that we wouldn't let a human make.

I don't think it really applies to the detriment of Uber here. I'm just saying, if your takeaway from this is "See, I guess there isn't structural bias in our society" but if the reason women are working less is because of societal pressures to be the person who "gives up their career" to take care of children or if they are leaving the service because of sexual harassment by customers, that doesn't erase bias, it just means it isn't at the feet of a sexist boss.
---