Slashdot Mirror
Study: Women Less Likely To Be Shown Ads For High-paid Jobs On Google
An anonymous reader writes: A team of researchers from Carnegie Mellon University has found that women seeking jobs are less likely to be shown ads on Google for high-paying jobs than men. The researchers created more than 17,000 fake profiles, which were shown roughly 600,000 ads on career-finding websites (abstract). All of the profiles shared the same browsing behavior. "One experiment showed that Google displayed adverts for a career coaching service for '$200k+' executive jobs 1,852 times to the male group and only 318 times to the female group." The article notes, "Google allows users to opt out of behavioral advertising and provides a system to see why users were shown ads and to customize their ad settings. But the study suggests that there is a transparency and overt discrimination issue in the wider advertising landscape."
*Grabs Popcorn* It seems Feminist Friday and SJW Saturday came early this week.
Perhaps women are 6 times less likely to click an ad for $200k+ executive jobs. If the algorithm prioritizes ads based on past behavior of other persons, given all identifiable traits of each person, then this is very well to be expected.
And would go to show that stereotyping is not always evil, but sometimes it comes from innocently putting together past information to be more efficient today.
All my liberal friends think I'm a conservative, all my conservative friends think I'm a liberal.
Is this an article or an advertisement? The icon says "Ad", but it's listed as submitted by an "anonymous reader" and gives the appearance of being a news article.
Is Slashdot trying to destroy itself?
...why would you expect it to be gender-neutral?
how meny are fake anyways to get the H1B in?
So, this is the lack of opportunity in tech people were talking about? So, women have less chances to succeed in tech because they didn't' get as many ads from Google? I guess with ABP I have no hope of any opportunity in tech. :,(
We present AdFisher, an automated tool that explores how user behaviors, Google’s ads, and Ad Settings interact. AdFisher can run browser-based experiments and analyze data using machine learning and significance tests. Our tool uses a rigorous experimental design and statistical analysis to ensure the statistical soundness of our results.
Rigorous experimental design and statistical analysis? Well, I'm sold. Teh data this amazing tool produces must be undeniably sound and its analysis unassailable.
We cannot determine who caused these findings due to our limited visibility into the ad ecosystem, which includes Google, advertisers, websites, and users.
Oh.
In an obvious policy of sexism, female's browsers were less likely to be sent openings or training for plumbing, roofing and landscape services.
No explanation was given by press time.
Google isn't really "choosing" who gets served ads as much as advertisers do. They ask for specific demographics, and the Google engine matches users to those demographics. If you want to serve your ads to males between 35 and 50 with an estimated gross income above $150k. It's not detailed *how* they made sure the browsing was identical.
I'd be curious what the results would be if you set up the profiles and surfed, but had only female subjects running "male" profiles and visa versa.
Is it just my observation, or are there way too many stupid people in the world?
The (picture/job) algorithms disagree. I hate to see what crime algorithms come up with.
Slashdot serves me adverts about ladies shoes?
At some point the women told Google their gender. Why? What moron thinks Google needs to know their gender?
But once you give Google (or Facebook, or Yahoo, or basically anyone...) information like gender, then I guarantee you they will correlate it with other people.
What this means is that somewhere in Google's algorithm they have found that people that claim to be women (this is the internet after all), are less likely to click on ads for high paying jobs.
So Google wisely decides to show them less such ads.
Do not blame Google for basing their ads on what they know about you and ALSO what they know about people like you.
Do blame yourself for telling Google that much about you.
excitingthingstodo.blogspot.com
That's a step forward for Google's AI!
Employers typically pay for the number of profiles on a site, either directly or indirectly.
CMU is screwing with employers by creating 17k fake profiles.
Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
In other news: investigators into violent homocides are 10x more inclined to look for male suspects then for female suspects
I can't recall ever seeing google post a job ad on my browser. I do notice Dice ads on the side of slashdot from time to time, but I don't pay any attention to them because they are ads and nobody pays attention to ads on the internet.
If you are not allowed to question your government then the government has answered your question.
Machine learning algorithms don't give a fuck about sexism and gender equality?
Maybe the correlation is actually with jobs men are more interested in, which also just happen to be paid better. I know this doesn't fit the genderists narratives, but please just try using Occam's Razor once in a while.
I am not chauvinist, however I see this as business as usual. All google ad's work of off statistical engines. If statistically there are fewer women in executive positions then ad's towards those positions and services should statistically match. I'm sure marketers see this story as simply business. That said it does not help equality.
There is or can be built a machine that can simulate any physical object. -Church-Turing principle
I have a study of my own too: People with social justice agenda will always find a difference to be enraged about. Even when they've bent the world to make everything equal (and terrible) they'll decry a man sexist because he doesn't shop for shoes as regularly as women. It will never end.
Jesus fucking christ people shut the fuck up.
I would say that this study is more sexist than it's own claims because it implies that women have no motive to seek out the high paying jobs that this study claims they deserve and are therefore helpless little princesses that need to be coddled like babies. THAT I think is far more offensive than any of these superfluous claims of sexism. If a woman truly wants that job, then just like the rest of us human beings, she needs to step up and take the initiative. It's that simple.
Who is this study mill that keeps putting out articles on a regular basis about the supposed sexism in all things? They need to be knocked down a peg before their eternal sexism soap opera ruins our society.
Right now more people are willing to pay more to Google show advertizements for shoes and lipstick to women than similar masculine products to men. So women see more lipstick ads and men see the lower priced ads for 200K jobs. It is even possible men are more likely to fall for 200 K con jobs than women who are more savvy in spotting fake ads.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
Probably even simpler: There are more ads specifically targeting women (shoes, makeup, etc) than for men making their ad pool larger and thus automatically diminishing the opportunity for ads for of high paying google to be shown.
Interesting argument. Do you have any proof that it is true?
When it comes to gender issues in tech the geek also seems to have a mission, but sterotypes rule, facts and analysis are optional.
Women are more likely to see ads for tampons, seriously people, stop writing and think before you post a friggin article.
Moderating "-1, Disagree" is simple censorship. Have the guts to post your opinion. -- Spazmania (174582)
Hey, statistical inference! Have you stopped beating your wife?
I am quite sure that these accounts were *not* set up by hand. Therefore any mechanism that Google has to detect if a count could have been invoked. That in turn would skew the results.
Just £$%@ off.
I was wondering last night on my landlord was looking worried. He's from China....
That men receive 75% fewer ads for women's fashion and makeup products while searching in Google.
Harrison's Postulate - "For every action there is an equal and opposite criticism"
The article states that the advertisement is for "coaching services", not a job. So someone out there is selling the promise of a high paying job, and they think men are more likely to fall for it. Saying that this will contribute to the gender wage gap is a long stretch. It is more likely to lead to men wasting a lot more money than women.
I thought it's the advertisers who choose which target user profile their ads should be shown to?
And can you really blame them for trying to keep their advertising costs low by selecting a target profile as narrow as possible to keep "wasted" views, that have to be paid for, too, as low as possible?
I don't see Google at fault here.
bickerdyke
And this is why people think the tech sector is unfriendly to women. NotDrWho - you're not helping the rest of us look sane. The same token which lets you make pathetic generalisations of an entire gender based on cultural stereotypes works the other way too. Please grow up.
It's because those "local moms" are already "making $8000/mo working from home"
Or at least that's what most of the ads tell me.
Google probably doesn't care, but those researchers were engaged in click fraud. If I were an advertiser, and I paid for 600,000 ads that only got shown to robots, I would be pissed. That's a lot of money right there.
"First they came for the slanderers and i said nothing."
Two issues I see:
Popisms.com - Connecting pop culture
Sure thing Dr. Misogynist.
Article fails to consider that Google has profiles for millions of people, making the 17,000 fake profiles they created an insignificant amount to accurately prove anything one way or the other
It seems insidious when Google is appearing to favor men over women for high priced jobs, so lets look at a different advertiser - Me.
A few years ago I was the music director for an all male chorus in a neighboring town. Every year we have a "guest night" where we invite people to come and join us, sing a couple a songs, and hope to get new men to audition and join our group. Advertising dollars are tight, so rather than ask Facebook to show our ad to everyone within 25 miles of our rehearsal spot, we asked just for men. Pretty simple, really.
Is it just my observation, or are there way too many stupid people in the world?
I'm being completely serious here, normally when it comes to "gender issues" I try to listen even though the outcome is almost invariably that I'm not a women so I can't possibly understand. But in this particular instance who the hell is the victim here? No one who is qualified for a $200K+ per year job finds it by clicking on banner advertisements, the only impact they have ever had on anyone who doesn't fall into the category of functionally retarded is possibly to remind them to update their LinkedIn profile. The only conclusions from this 'study' that I can draw in order of likelihood are A.) The option for Male is default for Google profiles and click-fraud bots rarely bother to change it. B.) More men surf the internet while intoxicated or depressed; because no one without an impaired sense of judgement would ever think that anything productive could come from clicking on those ads. C.) Possibly that men are more impulsive than women and simply follow the advertisement "just to see where it goes".
Let's for a moment play along with the people who are crying bloody murder over this and say that there is some vast conspiracy to hide these kinds of ads from women; it still isn't going to impact the number of females who get six figure jobs. Finding work, takes work; anyone who is actually looking for a job will tell you that regardless of gender a well paying job will never just fall into your lap. Do you want to know how the women who is making $200K a year got that kind of salary? The answer is simple: She didn't stop looking until she found it.
Read the actual article and paper, men weren't shown more ads for positions paying $200k+, they were shown more ads for coaching services to help them get jobs paying $200k+. It is more accurate to say that men are more willing to pay money to get help getting jobs paying $200K+, read men are relatively more desperate to get high paying jobs. This matches pretty well with what (if we are even remotely honest) we already know, men are more likely to be judged based on their job. Even Mythbusters was able to easily conclude that, for example, men who make more money are considered cuter by women on average. (same men, but with different job profiles). This increased desire for high paying jobs and willingness to sacrifice for them (in this case, literally paying money for coaching), may well result in men getting more high paying jobs, BUT again these are ads for services, not job offers.
First... fuck the stupid suggestion that a computer is sexist. I mean... that is the literal implication here. That the COMPUTER is sexist.
Second, if it is doing this then there are REASONS for it. Computers don't care if you have a vagina or a penis... they don't even know what that means. there's no subroutine checking for dick and then giving dick better jobs.
Third, once you've figured out the reasons... THAT is the actual story. The story likely will read something like this "people that click on these things are less likely to click on these other things."... Fucking shocking.
I would sacrifice orphans to a dark god if it meant the morons writing these stupid articles had black voids open behind them and suck them into a world of madness and horror.
I am so fucking tired of people that either don't know how statistics work writing articles that are simply using statistics to "beg the question" or even worse, people that do know and are simply political hacks writing this shit to confuse the rubes.
Just enough.
I've decided to stop wasting my time responding to AC trolls/sockpuppets... so if you want a response from me... login.
"career coaching service" ... maybe they have found that men are more likely to pay for these services?
I remember them from the days of paper want ads - services promising to land you the dream job, if you signed up and paid them a bucket of money
Advertisers can buy ads for demographics, but the ad servers also have the capability of automatically optimizing towards users they consider more likely to engage with those ads. Google's auction-based buying systems (AdWords, AdExchange, Bid Manager) all allow you to enter in specific demographics, yes. But they ALSO allow you to buy on a cost-per-click basis, at which point the system internally attempts to show the ads towards the users within the selected targeting who are more likely to click. The reasoning is simple - the ad will show either way but Google only gets paid if the user clicks on it. So Google is incentivized to try and push ads exclusively to users who are (relatively) likely to click.
An advertiser very well may have selected "target all users between the ages of 35 and 54" and created the ad with no gender preference. But as the ad runs, if Google's internal analytics show that men are disproportionately likely to click on that ad relative to women, they will automatically push the ad towards men.
So no, it's not necessarily all on the advertisers. But it's not automatically "sexism" within the ad-serving system, either. It very well may be self-selection at work. Now, IF what I describe above is the case, you could then delve into *why* men are disproportionately likely to click on high-paying jobs. It very well might be sexism in the greater society at large - but that's an entirely different question/issue.
The /. summary is wrong. The "1,852 times to the male group and only 318 times to the female group" difference is not "Ads For High-paid Jobs".
Those ads are for coaching; in other words, the advertisers have reason to believe that men are more likely to fall for their scam than women.
Woman here. Please don't presume to speak for me.
Guys make jokes. Women should learn how to handle them. I like to think I do, even though I realize that you're getting my own perspective on this.
Joke wasn't funny, but it wasn't particularly hostile, and I'm not running and screaming from my job because some dude on a website made a joke about maternity leave.
And if that's the thing that makes you want to quit...well, I guess my best advice for you is to not leave the house.
In fact, I'd say dave420 and AC are far more sexist. You presume that women need standing up for. That's why my sig is what it is. Let women handle their own business. The ones who can are the ones that you want to work with you. Not the ones that get all drippy-eyed at the vaguest and non-general insults.
Some people don't believe in fairies. I don't believe in The Patriarchy.
Exactly, one need only look as far as the female CEO of Reddit to see how these are all just false stereotypes!
SJWs... forward march!!
...should be the new Correlation does not imply Causation.
Let's give a simple example: both the NFL and the NBA contain predominantly black players in contrast to the minority that exist in the general population. I don't think anyone would suggest that because of this disparity, that the sports teams and their leagues are discriminating against non-black athletes and that something must be done to make these sports come into line with the general population. Ironically, people do argue that because the teams/leagues are predominantly black, certain high-profile positions (like coaches and quarterbacks) should also be predominantly black and that it is discrimination that it isn't so.
Will any of these 'researchers' ever get past the data collection phase and do actual research to find the cause. Real researchers use data as a starting point for further research - data indicates a discrepancy, now let's find out why there is the discrepancy. Many modern researchers take the Fox News/MSNBC approach to it - gather the data, put your ideological spin on it, case closed.
Even these researchers admit in the abstract and associated article that they can't be sure why there is a discrepancy. Let's see if I can write my next thesis for Carnegie Mellon University --- statistics show that current college enrollment is roughly 40% male and 60% female; obviously, this show rampant systemic gender bias against men. Something must be done about it. End of thesis.
Slashdot now more likely to show shit non-stories.
Another superfluos study. I don't mean that the result is uninteresting, but a study was not necessary to find out. You study mysterious unknown processes in nature. These clowns has studied ads served by algorithm - but the algorithm is written by humans and therefore available. Just look up what it will serve for men & women, no "study" necessary.
Off to study just what the mysterious %-button on pocket calculators do. Perhaps I can get a ph.d. . .
6641352 out of 5 women are raped. Women are paid -99 billion dollars for every male dollar. Men manspread and an individual man will take up 95% of the seats in a car. And even though you're far far more likely to be violently assaulted walking home at night if you're a man, women are still in more danger walking home at night alone.
Bot here.
What is this "joke" thing humans are so found of ?
And GP comment makes no sense. Why should bots leave when they are asking for maternity ? What is this "stirred up office" thing ?
Oh, bleep bloop, update incoming, it looks like a critical bug fix. See you later humans.
Study is downright ignorant - correlation doesn't equal causation. Just because men might be correlated strongly towards being shown a 200,000+ job training (non-existent - have fun climbing the corporate ladder) doesn't mean that Google is trying to suppress higher paying job training from women. What the algorithm does convey is that men are more likely to click these ads than women (women realize the ad is stupid and not going to result in an instantaneous pile of cash?). Take away what you want, but the former is a silly conclusion.
Then what? Are the creators of this study expecting women everywhere to boycott google? Best of luck with that.
I'm a little concerned about using the Times of india for most of the study, with some Guardian as an alternative. Also the "found discrimination" was predominently from a pair of job related ads. Extrapolating a conspiracy as some will do, from two ads is a little tough. But I don't doubt that the basic premise is true, that website searches are tailored by gender.
The search results a woman might get are based not on active male sexism, but given the likely results of women as an aggregate. But it is machine sexism based on what the algorithm thinks you might be looking for.
This is exactly the kind of crap you end up with when you try to serve ads based on profiling people.
I've so far found duckduckgo to be the only popular engine worth a crap. Peopple should be able to find what they are looking for without the freaking search engine filtering it, and trying to get you t continually buy stuff it thinks you want.
The only thing I wonder about is just what the hell is datemypet.com? ( one of the genderised sites used in the study) I'm afraid to even click on that one.
The shepherds did so well protecting the flock that the sheep no longer believed that wolves existed.
Google's ad system is entirely reactive. Aside from the fact that most ads ARE gender targeted due to the fact that members of different genders make choices based on different factors, you cannot imply that something reactive, especially a mindless algorithm, is "Discriminatory". That's like saying that rain puddles are discriminatory to dry pavement. The rain might happen to fall on the pavement, and it might even flow to dry pavement as the lowest point it can reach, but it means nothing. Discrimination is by definition a proactive behavior, as it implies thought before an action.
There is no reason to make studies for something, which is clearly defined.
There is an algorithm and there are people buying ad space, which define which ads should be shown to which people. You can just ask them. About what they did and why they did it.