Recently I've been thinking about inherent biased towards applicants based on their prior work / educational history, in terms of organizations they have been to.
I get that it acts as a moderate signal of how tenacious and inherently able that applicant might have been in the past if she went to XYZ school or ABC company. But there will be so many false negatives!
A lot of folks who probably would be otherwise capable of the job at hand would be dismissed for not having the right pedigree. Also, past performance is not always the best Intuit if future success.
And then of course is the issue of how do you even define success, especially in case if university graduates. How does a Yale or Harvard define our measure success for its students... Probably in terms of potential impact they have on the world or how good their names may look on the rolls 10 years down the line.
A lot of the issues, especially in terms of job applicators but also applicable in admissions process, ultimately comes down to information and numeric asymmetry between the organisation and the applicant.
If there was a way to somehow capture the readings - however brief - for rejection or selection of applicants or a documented scoring methodology, that itself would bring a lot of change.
I would love to know if some resources to further explore the alternative models of applicant assessment which does not place as much importance on the past organisations / schools they have been to.
Professor Iris Bohnet of Harvard wrote a fascinating book called "What Works" that you might find interesting. What I really like about the book is right from the jump Bohnet basically says all this "diversity training" and "PC culture" stuff is a waste of time and an uphill battle. We all like to pretend we're unbiased but Bohnet tells the truth from the outset: everyone has biases. The book focuses on systems that emphasize equal opportunity and remove some of those inherent biases we all have.[1] Bohnet's premise is that we'll get better results if we stop designing systems with the faulty assumption that most people/systems/institutions aren't biased in some way or another.
There's a chapter about "blind auditions" that Orchestras began to implement in the 1970's that you might find interesting.[2] After reading that chapter I remember wondering how much we're overthinking a lot of these issues and whether we might be overlooking some pretty easy changes that could have a significant impact.
One interesting fact from the blind auditions study was that even without being able to see the musician they were able to demonstrate bias (likely unconscious) when women wore high heels to their audition because the sound gave away their gender. It's really fascinating some of the ways we unconsciously "judge" those around us.
As Prof. Bohnet herself points out, the HR dept. should do more A/B testing of its practices instead of relying on the gut feel.
Hence the gradual increase in the prevalence of "people analytics" (apart from allure of sucking in all the data-points you can).
But there are a few problems:
1. A/B testing in talent mgmt practices is costly in terms of time-investment.
- Final, long term impact of such experiments are not visible to the org for, well, a long while.
Sure, may be some short term impacts like increased diversity - may be visible in the short term.
But is that worth all the experimentation for all the firms except the largest ones like FANG?
The experiment cost is probably not worth the value for the smaller orgs.
2. Vetting people for skillsets and qualifications only, masking the personal details - like in the OP on Helsinki, is a good start.
But it does not address the biases on the historical disadvantages like not coming from a pedigreed organization because of not having been to a pedigreed school.
Think of being ranked lower in the applications to Netflix because you are not coming from Google because you didn't go to Stanford.
Blind -audition technique would be the best in situations like these as the focus would be on demonstrating the skill needed and not the "qualifications" (read - pedigree).
Such skill tests can probably be done easily for programming and other situations but a lot of roles do not have a strictly defined evaluation criteria as the KPIs are...well... fuzzy.
Or at least not something which can be demonstrated in a short "test".
Think sales / pro-management. You can not demonstrate in a test if you are a good enough sales person who can meet the revenue projections ("sell this pencil to me" doesn't count).
Issue here is lack of a easily reproducible and reliable signal for your future success for such roles, thereby increasing the dependency on past history - and by association, your pedigree.
Of course there will be other issues like isolating performance factors into "chance" and "skill" in a noisy dataset that is usually available, if at all.
Or hiring mindset which focuses on the required skillset (what is needed to get the job done) but the "best person" for the job which muddies the waters by adding evaluation factors which are really not material to performance on the job at hand.
But if the cost of experimentation can be reduced by devising quick tests for generating reliable signals independent from historical background, it would be awesome!
I know that as a result of diversity training and guidelines I am 10x as cautious about my behavior at work in front of women as with just men. That means there is a camaraderie that the men have that the women are left out of. When I worked on Wall Street this divide didn’t exist but there was definitely a culture of sexually hostile work environment and sexual hatrassment.
I used to recruit for tech start ups and can confirm that school is just about always the most important thing to get you in the door, particularly early on in your career.
It's something that I fought against (I have a technical background and had the advantage of being able to evaluate technical skill somewhat independently), and did well because of it.
There were a few candidates that I brought in that I was reprimanded for - physically brought into our partners' office to be told off - only to have that candidate receive multiple job offers from our clients. The partners weren't qualified to really evaluate technical skill, but knew what candidate qualities that a start up is willing to pay a recruiter's fee for.
At the end of the day, my candidates were exceptional for having risen above their alma maters' reputations, and the Ivy league grads are more consistently good/qualified.
When hiring, time is such an important factor. Many engineering teams are unwilling to take the chance on an a candidate that graduated from unfamiliar school because their hiring process is weak and/or inefficient. Or they waited too long to hire, and need someone "yesterday". At least they're more likely to end up with a good candidate by using the school as the qualifier, even if it costs them more because of the candidate's demand.
TL;DR : The 'top' candidates' values are artificially inflated as a result of high demand, which itself is due to ill-conceived and insufficient hiring processes across most companies.
Even assuming that there is some small enough set of great schools to use them for that signaling as Google traditionally did -- I do not think that's a rational assumption, but that's another discussion -- then at least in the USA you should also know the candidate's parents' financial situation and whether their parents went to the same elite schools.
For starters, because it's a lot easier to "get in" if you have no trouble paying for it, and you have your "legacy admissions."
If you really think XYZ is that great, you should really want to hire the people who fought to get in, fought to succeed, and did.
I agree with the cynicism about choosing based on school. But it does work to some extent. If everyone agrees that XYZ/ABC schools/companies are the best, then rationally students will try hard to get in to those schools. Therefore you can get most of the best by filtering for that history. Sucks if you didn't care about grades when younger but actually turned out smart, but most works.
That works for some cases, but not where a candidate's parents also went to the same school or where the parents can and do donate a lot to the school. At most of the schools we're discussing, both situations give the student a boost in admission and in passing classes regardless of effort.
(Disclosure: I went to a well-known school, but without either of the parental benefits I just described. My parents helped in other more indirect ways throughout my upbringing, but they didn't come from a wealthy or connected family background.)
No suggestion, but an anecdote that illustrates a false negative like this.*
Stack Overflow's got an Azure SRE spot open, asking for experience with PowerShell DSC and Terraform, among others. The main goal for the role appears to be scaling their enterprise offering.
I've worked as both a sysadmin and a dev for years now, and used both to manage sizable deployments in public clouds, both Azure and others, mostly for enterprise software companies. This is a mildly specialized skillset (though getting less specialized by the day), and I figured I'd get at least a screening call.
I didn't even make it through the automated resume screening, and if I had to guess it's because I don't have a degree, much less a CS degree.
* On the off-chance my employer reads this, this was more of an exploratory application than me actively hunting for other opportunities.
The problem with bias is that it has affected peoples opportunities in the past. This will be reflected in a CV.
I'm not sure the inclusion of name / age / etc will solve this problem, so Helsinki might still be on the right track.
It feels like the solution is to intentionally introduce a reverse bias towards these candidates. But that idea does sit well... Also interested in any other suggestions here.
Yes, The other information is a job application, at least you can argue that somebody could make a decision based on those information. Whether decision process is just is obviously arguable. If name, age is totally irrelevant to a job, at best it will not be used for selection, while there is much risk that it will be used for discrimination consciously or not.
I've long suspected this was the only remotely feasible way to remove bias from recruiting (and possibly interviewing itself) -- I'm interested to see how this will go.
Politics about bias and diversity aside, if your goal was to objectively disable people from being biased, the simplest solution always seemed to be to remove their ability to apply any of their biases.
It was tried in France without great success. The law was passed in 2006 but the application decree (which states how/when the law comes into effect) was never published. Consequently, this law was never upheld. Why? Field trials between 2006 and 2009 showed it didn't have a measurable effect and in some cases ended up discriminating more. The experiments' results are controversial (small sample size) but despite pushes in 2011 and 2014 the law seems effectively dead in the water.
Anecdotally, I've been on the recruiting end of "anonimized" CVs pushed by outsourcing companies. I was never fully clear on whether this was a legal requirements for them, or if it was a gesture to show inclusiveness and social responsibility on their part (take this with a pinch of salt, most outsourcing companies in France are little better than horse-dealers and will push any warm body your way if they think they can turn a profit). At any rate, most of these were hilariously badly done: using initials for the last name but leaving the first name in full, leaving telling initials (1), or other subtle or less subtle ways to hint at the contractor's origin and gender.
(1) e.g. a first name initialed J.L., J.M. or some other variation points to a Jean-something (Jean-Louis, Jean-Michel, etc.) composite first name, which is a clear indication of a "native" Frenchman (and a hint as to the person's age as well)
I really wish there was more published about why it did/didn't work -- I saw the other posts about similar studies that all seemed to fail, but the details seem really scant.
It's still unclear whether the problem was with the implementation (the experiment) or the method.
If there's spooky action/unintuitive reality at play here it's the job of a good scientific study to pull it out and lay it bare.
IME, outsourcing companies do this so you have to go through them to talk to the individual attached to the resume, not for any moral purpose. I've been on the receiving end of that one - a recruiting firm was sending out a (poorly) anonymized copy of my resume without my knowledge.
>the simplest solution always seemed to be to remove their ability to apply any of their biases
Bias will exist regardless of which information you present, the most you can hope for is shifting and limiting the types of bias that come into play to eliminate useless or harmful bias. Even if you grant anonymity to the applicant, recruiters can still be biased against or in favor of the school somebody went to, what they studied, the organizations they were involved in, etc.
For thia reason I hire based on anonymized data that also discards all place name locations and institutions. The result is quite effective, as you are presented with an applications that show only the bare bones of someone's experiences without context that might bias you.
Recently I've been thinking about inherent biased towards applicants based on their prior work / educational history, in terms of organizations they have been to.
I get that it acts as a moderate signal of how tenacious and inherently able that applicant might have been in the past if she went to XYZ school or ABC company. But there will be so many false negatives!
A lot of folks who probably would be otherwise capable of the job at hand would be dismissed for not having the right pedigree. Also, past performance is not always the best Intuit if future success.
And then of course is the issue of how do you even define success, especially in case if university graduates. How does a Yale or Harvard define our measure success for its students... Probably in terms of potential impact they have on the world or how good their names may look on the rolls 10 years down the line.
A lot of the issues, especially in terms of job applicators but also applicable in admissions process, ultimately comes down to information and numeric asymmetry between the organisation and the applicant.
If there was a way to somehow capture the readings - however brief - for rejection or selection of applicants or a documented scoring methodology, that itself would bring a lot of change.
I would love to know if some resources to further explore the alternative models of applicant assessment which does not place as much importance on the past organisations / schools they have been to.
I think that's more of a problem of degree -- it's hard to pin down where harmful bias starts, but if you can do that, then this makes the anonymization solution even more applicable.
The question is whether the school is offering a harmful bias -- people get screened by school, GPA, etc all the time but at the point that society agrees these are harmful then you can just anonymize them out. It gets murkier on the moral/ethical side but I think that's the nature of the problem, not a problem with the solution.
I wonder how far it goes. In the USA, there are universities that have very distinctive ethnic compositions: UC Irvine is very different from Howard is very different from BYU. Do they anonymize the university name as well? Perhaps translating it to "large research university" and "liberal arts college." Though probably you can get some signals from there as well.
Personally I've never understood the importance put on degrees in the tech field. Sure it is important for your first job, but after 2 years, who cares. When reviewing resumes, if the person has been in the industry for a bit, the degree is irrelevant. And the fancy expensive school, big whoop. I rarely even read that section. I only care about their recent accomplishments and hopefully they have a good number of past ones.
Where are the results of this Australian study? That news article doesn't list them.
Following through on their source [1], it says that the study outcome actually favors women and minority candidates, but it doesn't say how much, or talk about statistical signifance. It seems like the experiment design has been criticized on this one study too, and not releasing the results seems suspect.
The study also cites that blind auditions were a success for Orchestras [2]:
>In a well-known study analysing data on auditions and hiring by orchestras over this period, Goldin & Rouse (2000) found that the use of blind auditions had a major impact on gender bias in orchestras, increasing the likelihood of female musicians being selected by 25-40%.
The study states that after anonymization of resumes women were around 3% less likely to get picked and men around 3% more likely. Data were statistically significant at the 99% level.
>> The trial found assigning a male name to a candidate made them 3.2 per cent less likely to get a job interview. Adding a woman's name to a CV made the candidate 2.9 per cent more likely to get a foot in the door.
Seems like it did the job of removing existing bias when details were blind.
Don't you mean, succeeded spectacularly? If the composition of candidates was changed when less biased/irrelevant information was displayed, arguably the selection could only be more meritocratic, not less.
This matches my experience in IT: sometimes recruiters prefer to hire a girl, even if less qualified, just for sake of having a girl in the team, because it is "improving the morale" :)
There are tons of studies out there showing monocultures are bad for business. In your case, I'm sure your IT department had to interact with women on occasion. Having women on the team will allow the group to be aware of social considerations specific to women that they may otherwise miss.
The people who reviewed and approved the designs for our office space were all males. Guess what they forgot to put in the women's restrooms? When women complained, they put an open trash can next to the toilet with no lid...
The problem with the whole conversation around this topic is everyone pretending that women and men on average have the same level of skills and experience as each other. Then they act shocked when there are so few qualified male nurses or female mechanics.
Businesses only care about this topic, because it helps them design products and provide services that address a wider market. So, maybe they should just admit they don't always hire the absolute technically best candidate as the goal is to improve the social structure of the team.
> Anonymous recruiting entails that the ... date of birth ... are not included in the job application
If I write on my application that I have 25 years of experience, does that have to be excluded due to it revealing that my date of birth is likely more than 40 years ago? Can they really believe that the amount of experience isn't important?
I'm 46, and people my age talk about leaving off earlier experience, so as not to look too old. I condensed my earlier experience (the first 10 years) into a couple sentences describing the roles--not to look younger, but because with an average stint of 2.5 years, there are just too many jobs to list.
Other steps people my age take are to stop listing dates--either select dates like university graduation, or--in extreme cases--all employment dates.
Keep in mind that there are different biases at different screening points along the way. The manager with the open position may want someone older and more experienced, but the people recruiting, screening, and filtering the candidates may be biased or simply not familiar with the value of experience that doesn't match today's buzzwords.
I've got about 14 years of development experience and I also leave off some of my earlier experience on my resume. I do it not to conceal my age, but because the stuff I was doing 10 years ago is so much less interesting and relevant to my applications today.
Does my outlook not resonate with you/other older employees?
To be fair, the hiring signal you can get from a description of 25 years of experience is probably comparable to the signal you can get from a description of the most recent 5 or maybe 10 years of experience.
I hire with anonymous data. I would look at a statement like '25 years experience' on a CV and completely ignore it, as I have no idea how competant the candidate is based on this. I would look for the key technical words and impact statements in the CV to show depth of experience rather than general statements like this.
People are going to fake a lot more degrees and other qualifications.
When they come in for an in-person interview racial/gender/age/whatever bias is still going to be there.
Only way to prevent this would be to forbid in-person interviews before hiring/signing a conract, but then companies would presumably simply stop hiring from applications, because the qualifications are unreliable and they can't get a character impression of the applicant.
Hiring by reference from existing employees would become the norm (maybe it already is).
Why would someone do that if they know the faked degrees or qualifications won't pass the background check eventually? Just to get an onsite interview?
Yes, just to get an onsite interview. I saw this many times and it's very sad.
Also, I relate this to culture. Not to offend anyone, but from my personal experience lying in CV is much more common for some countries, that for others.
Scale maybe terrible, like multiple certifications in some field with no actual basic knowledge or skills. Like Certified Senior (sic!) Software Developer who can't tell you difference between stack and queue, or CCIE certified network engineer who has no idea what is DNS and how it works (real case).
As soon as I walk in the door I will be eliminated. But how can you interview someone without actually seeing/talking with them? Slack? Would anyone hire based only on a Slack interview?
Even if you get hired, you still have to show up for work and can be fired later due to the same bias.
If both the application and the work is completely remote and anonymous it might be possible to avoid most bias, but that's only feasible in IT and a couple other sectors and even there it's not common.
I hire on anonymized data. I can tell you that it works well. The only flaw is that if your pool of application sifters is not diverse, bias will start to creep back in. For instance, in a group of several application sifters of a single gender who are collectively deciding which applications to progress to interview stage, one may frequently encounter presumptions of gender and age of the applicants due to implicit bias by the sifters.
The organisation must have a very respectful culture in which colleagues feel comfortable calling out others on their biases when this happens.
Economist here. This will bake in bias that affirmative action and other programs seek to counter. Inequality of opportunity unfortunately isn't countered by this approach. Why? The same group(s) facing discrimination at the employment point also face discrimination earlier in life, or have baked in social discrimination (such as the US justice system overpenalizing blacks relative to whites/Asians, or Ivy League admission rates for Asians versus other groups).
It would be nice to build a group-blind world, and the more we adopt these specific approaches the better can can iterate towards. However, care should be taken to ensure the baked in biases don't further entrench.
Correct. However, the problem is the existing discrimination is not remedied, and is instead perpetuated, under a fully blind approach.
Inequality is another discussion than bias and discrimination, in my view. Inequality is solved by implementing encouraging measures. Discrimination is resolved by removing entrenched barriers.
Explain how inequality of opportunity is not countered by this approach?
Of course people are a sum of everything that happened in their life, but this is a way to make a particular job application much more fair, based on capability to do the job rather than irrelevant physical details. This is not meant to solve everything at once.
The very examples you post are about situations that would be helped if there was blind judgement.
Yes, it would be much better if we are able to remove discrimination earlier in life.
But this is not a either/or solution between affirmative action and equal opportunity.
But if the assumption that positive reinforcement make a better just society for future generation then doesn't it mean equal opportunity will make also have the same positive effect albeit a little less ?
I think what the GP is stating is that if we only assess accomplishments and we do it at this stage (later in life after many different phases) those systematically disadvantaged will not be competitive (example: If you went to a crummy public school you have a harder time getting into an ivy league, harder time getting a prestige internship, getting a premium job, etc)
Affirmative action laws were introduced to address this, but the trend toward meritocracy counters them.
My guess; we end up with the worst of both worlds: One stream blind to these factors and a bunch of regulated exceptions.
I'm sorry, but this sounds absurd. Correct me if I'm wrong, but what you're saying implies that employers should hire people that are less qualified for a job just because they belong to a group, members of which on average are more likely to be discriminated earlier in their lives than others?
Engineer here, and no, blind hiring will not bake-in any biases, because that's exactly the problem this approach solves - it aims to hire the best candidates regardless of their race, origin, religion, accent and other irrelevant characteristics.
My first paragraph I tried very carefully to avoid normative/"should" statements.
The identification of the issue is at hand. Consider Computer Science. Historically far fewer women go through the training for a variety of personal reasons and societal pressure. So, on the margin, blind/anonymous hiring ignores that upstream delta between females in population and skilled females in hiring pool. Whether this is desirable or not is not my point, only that this is "baked in bias" perpetuating down the line.
It's noble to build a society where we value blindness on characteristics irrelevant to hiring. Doing it piecemeal can result in furthering these baked in biases.
A better approach would be to split the skills validation (blind) from the move-forward decision (adhering to whatever inter-generational correction schema the society has decided upon) where the only pool of eligible move-forwards meet the established skills validation. In this way, you only get skilled candidates, and you can satisfy whatever other characteristics that need satisfaction.
> Engineer here, and no, blind hiring will not bake-in any biases
Yes, it will: if you're born to a well-off family, you'll go to better schools before college. Due to this, you'll have a better chance at succeeding in college, have more resources to delve deeper into your field of study, better manage studying abroad, &c. All of these factors will make your list of skills more attractive to employers.
Compared to someone who was poor who went to worse schools, didn't study abroad, and couldn't afford to spend more time on their field of study due to having to work to pay for college. This will make your list of skills less attractive to an employer.
Right out of the gate, there is a bias against people who grew up poor in the hiring process.
> it aims to hire the best candidates regardless of their race, origin, religion, accent and other irrelevant characteristics.
It does so by devaluing repressed classes of people. This is a societal cost to benefit businesses.
Yeah, I don't get that approach either. Is the parent comment suggesting that employers need to assess the opportunities given to a particular applicant and then weight them accordingly? Like if they come from a poor/disadvantaged background and still have 90% the skills and experience of another candidate, well that's more impressive? How does the parent expect an interviewer to come up with these valuations of opportunity?
It is being pointed out, but it's not solely pointing out the symptom.
The goal of Helsinki here is to make the hiring decision independent of direct factors they blind themselves to (name, sex, etc.). However, the factors they condition on (education, experience) are impacted by these factors, so the hiring decision won't be independent of the factors.
The "best" thing to do is to correct the inherent bias, as you're pointing out as the real problem. We agree. This isn't something Helsinki can do in a vacuum, so if they want to achieve their goal of true "blindness" to the above factors, they have to take into account their impact on their "seen" factors somehow.
I put the formalized model I have in mind in another comment on this thread, which walks through my thought process here.[0]
> the name, date of birth, address, mother’s tongue, gender and any other personal information [are] not included
[...]
> when much of the information in the application has been hidden, it’s hard to establish a clear picture of the know-how and capability of the applicant
So they base their assessment of know-how and capabilities on age, gender, mother's tongue, and other personal information that has been hidden? I feel like I must reading this wrong.
I don't think you are reading that wrong. People use all sort of pseudo science to make inferences on people's competencies. Like if you are a woman in a male dominated field you're likely to be highly competent. Or when hiring for a leadership position an older candidate may be subconsciously preferred over a younger one with more experience. It's all generally bunk and this is imo, an admission that they have been using the wrong criteria to judge candidates.
> if you are a woman in a male dominated field you're likely to be highly competent
In my experience it's the opposite. Like with small countries: the odds they produce someone extremely good at a certain sport is just smaller than in large countries, so large countries are more likely to win some world cup or the Olympics. I know only a handful of women in IT to begin with, but most of them aren't great at it.
But that's not your point. You're saying that there are biases in hiring, and I agree. What I'm surprised by is that the article quotes them saying that they now find it very hard to judge someone. Who would say, alongside a press release about anonymous applications, that they find it difficult to base their judgement solely on only relevant information?
See also: If your mother's tongue isn't also the local native/dominant language, you must be an immigrant, which may be positive or negative depending on the circumstance.
Readit News
I get that it acts as a moderate signal of how tenacious and inherently able that applicant might have been in the past if she went to XYZ school or ABC company. But there will be so many false negatives!
A lot of folks who probably would be otherwise capable of the job at hand would be dismissed for not having the right pedigree. Also, past performance is not always the best Intuit if future success.
And then of course is the issue of how do you even define success, especially in case if university graduates. How does a Yale or Harvard define our measure success for its students... Probably in terms of potential impact they have on the world or how good their names may look on the rolls 10 years down the line.
A lot of the issues, especially in terms of job applicators but also applicable in admissions process, ultimately comes down to information and numeric asymmetry between the organisation and the applicant.
If there was a way to somehow capture the readings - however brief - for rejection or selection of applicants or a documented scoring methodology, that itself would bring a lot of change.
I would love to know if some resources to further explore the alternative models of applicant assessment which does not place as much importance on the past organisations / schools they have been to.
Any suggestions would be much appreciated
There's a chapter about "blind auditions" that Orchestras began to implement in the 1970's that you might find interesting.[2] After reading that chapter I remember wondering how much we're overthinking a lot of these issues and whether we might be overlooking some pretty easy changes that could have a significant impact.
One interesting fact from the blind auditions study was that even without being able to see the musician they were able to demonstrate bias (likely unconscious) when women wore high heels to their audition because the sound gave away their gender. It's really fascinating some of the ways we unconsciously "judge" those around us.
[1]https://hbr.org/2016/07/designing-a-bias-free-organization
[2]https://www.theguardian.com/women-in-leadership/2013/oct/14/...
But there are a few problems: 1. A/B testing in talent mgmt practices is costly in terms of time-investment. - Final, long term impact of such experiments are not visible to the org for, well, a long while. Sure, may be some short term impacts like increased diversity - may be visible in the short term. But is that worth all the experimentation for all the firms except the largest ones like FANG?
2. Vetting people for skillsets and qualifications only, masking the personal details - like in the OP on Helsinki, is a good start. But it does not address the biases on the historical disadvantages like not coming from a pedigreed organization because of not having been to a pedigreed school. Think of being ranked lower in the applications to Netflix because you are not coming from Google because you didn't go to Stanford. Blind -audition technique would be the best in situations like these as the focus would be on demonstrating the skill needed and not the "qualifications" (read - pedigree).Such skill tests can probably be done easily for programming and other situations but a lot of roles do not have a strictly defined evaluation criteria as the KPIs are...well... fuzzy. Or at least not something which can be demonstrated in a short "test". Think sales / pro-management. You can not demonstrate in a test if you are a good enough sales person who can meet the revenue projections ("sell this pencil to me" doesn't count).
Issue here is lack of a easily reproducible and reliable signal for your future success for such roles, thereby increasing the dependency on past history - and by association, your pedigree.
Of course there will be other issues like isolating performance factors into "chance" and "skill" in a noisy dataset that is usually available, if at all. Or hiring mindset which focuses on the required skillset (what is needed to get the job done) but the "best person" for the job which muddies the waters by adding evaluation factors which are really not material to performance on the job at hand.
But if the cost of experimentation can be reduced by devising quick tests for generating reliable signals independent from historical background, it would be awesome!
It's something that I fought against (I have a technical background and had the advantage of being able to evaluate technical skill somewhat independently), and did well because of it.
There were a few candidates that I brought in that I was reprimanded for - physically brought into our partners' office to be told off - only to have that candidate receive multiple job offers from our clients. The partners weren't qualified to really evaluate technical skill, but knew what candidate qualities that a start up is willing to pay a recruiter's fee for.
At the end of the day, my candidates were exceptional for having risen above their alma maters' reputations, and the Ivy league grads are more consistently good/qualified.
When hiring, time is such an important factor. Many engineering teams are unwilling to take the chance on an a candidate that graduated from unfamiliar school because their hiring process is weak and/or inefficient. Or they waited too long to hire, and need someone "yesterday". At least they're more likely to end up with a good candidate by using the school as the qualifier, even if it costs them more because of the candidate's demand.
TL;DR : The 'top' candidates' values are artificially inflated as a result of high demand, which itself is due to ill-conceived and insufficient hiring processes across most companies.
For starters, because it's a lot easier to "get in" if you have no trouble paying for it, and you have your "legacy admissions."
If you really think XYZ is that great, you should really want to hire the people who fought to get in, fought to succeed, and did.
(No I didn't go to a famous school)
(Disclosure: I went to a well-known school, but without either of the parental benefits I just described. My parents helped in other more indirect ways throughout my upbringing, but they didn't come from a wealthy or connected family background.)
Stack Overflow's got an Azure SRE spot open, asking for experience with PowerShell DSC and Terraform, among others. The main goal for the role appears to be scaling their enterprise offering.
I've worked as both a sysadmin and a dev for years now, and used both to manage sizable deployments in public clouds, both Azure and others, mostly for enterprise software companies. This is a mildly specialized skillset (though getting less specialized by the day), and I figured I'd get at least a screening call.
I didn't even make it through the automated resume screening, and if I had to guess it's because I don't have a degree, much less a CS degree.
* On the off-chance my employer reads this, this was more of an exploratory application than me actively hunting for other opportunities.
The problem with bias is that it has affected peoples opportunities in the past. This will be reflected in a CV.
I'm not sure the inclusion of name / age / etc will solve this problem, so Helsinki might still be on the right track.
It feels like the solution is to intentionally introduce a reverse bias towards these candidates. But that idea does sit well... Also interested in any other suggestions here.
Politics about bias and diversity aside, if your goal was to objectively disable people from being biased, the simplest solution always seemed to be to remove their ability to apply any of their biases.
Anecdotally, I've been on the recruiting end of "anonimized" CVs pushed by outsourcing companies. I was never fully clear on whether this was a legal requirements for them, or if it was a gesture to show inclusiveness and social responsibility on their part (take this with a pinch of salt, most outsourcing companies in France are little better than horse-dealers and will push any warm body your way if they think they can turn a profit). At any rate, most of these were hilariously badly done: using initials for the last name but leaving the first name in full, leaving telling initials (1), or other subtle or less subtle ways to hint at the contractor's origin and gender.
(1) e.g. a first name initialed J.L., J.M. or some other variation points to a Jean-something (Jean-Louis, Jean-Michel, etc.) composite first name, which is a clear indication of a "native" Frenchman (and a hint as to the person's age as well)
It's still unclear whether the problem was with the implementation (the experiment) or the method.
If there's spooky action/unintuitive reality at play here it's the job of a good scientific study to pull it out and lay it bare.
Bias will exist regardless of which information you present, the most you can hope for is shifting and limiting the types of bias that come into play to eliminate useless or harmful bias. Even if you grant anonymity to the applicant, recruiters can still be biased against or in favor of the school somebody went to, what they studied, the organizations they were involved in, etc.
I get that it acts as a moderate signal of how tenacious and inherently able that applicant might have been in the past if she went to XYZ school or ABC company. But there will be so many false negatives!
A lot of folks who probably would be otherwise capable of the job at hand would be dismissed for not having the right pedigree. Also, past performance is not always the best Intuit if future success.
And then of course is the issue of how do you even define success, especially in case if university graduates. How does a Yale or Harvard define our measure success for its students... Probably in terms of potential impact they have on the world or how good their names may look on the rolls 10 years down the line.
A lot of the issues, especially in terms of job applicators but also applicable in admissions process, ultimately comes down to information and numeric asymmetry between the organisation and the applicant.
If there was a way to somehow capture the readings - however brief - for rejection or selection of applicants or a documented scoring methodology, that itself would bring a lot of change.
I would love to know if some resources to further explore the alternative models of applicant assessment which does not place as much importance on the past organisations / schools they have been to.
Any suggestions would be much appreciated
The question is whether the school is offering a harmful bias -- people get screened by school, GPA, etc all the time but at the point that society agrees these are harmful then you can just anonymize them out. It gets murkier on the moral/ethical side but I think that's the nature of the problem, not a problem with the solution.
https://www.abc.net.au/news/2017-06-30/bilnd-recruitment-tri...
Following through on their source [1], it says that the study outcome actually favors women and minority candidates, but it doesn't say how much, or talk about statistical signifance. It seems like the experiment design has been criticized on this one study too, and not releasing the results seems suspect.
The study also cites that blind auditions were a success for Orchestras [2]:
>In a well-known study analysing data on auditions and hiring by orchestras over this period, Goldin & Rouse (2000) found that the use of blind auditions had a major impact on gender bias in orchestras, increasing the likelihood of female musicians being selected by 25-40%.
[1] http://behaviouraleconomics.pmc.gov.au/projects/going-blind-...
[2] https://www.aeaweb.org/articles?id=10.1257/aer.90.4.715
http://behaviouraleconomics.pmc.gov.au/sites/default/files/p...
Page 13
>> The trial found assigning a male name to a candidate made them 3.2 per cent less likely to get a job interview. Adding a woman's name to a CV made the candidate 2.9 per cent more likely to get a foot in the door.
Seems like it did the job of removing existing bias when details were blind.
And if it's not true, arguably what you've done is removed the adjustment variables that could correct the bias, rather than the bias itself.
Deleted Comment
The people who reviewed and approved the designs for our office space were all males. Guess what they forgot to put in the women's restrooms? When women complained, they put an open trash can next to the toilet with no lid...
The problem with the whole conversation around this topic is everyone pretending that women and men on average have the same level of skills and experience as each other. Then they act shocked when there are so few qualified male nurses or female mechanics.
Businesses only care about this topic, because it helps them design products and provide services that address a wider market. So, maybe they should just admit they don't always hire the absolute technically best candidate as the goal is to improve the social structure of the team.
Deleted Comment
If I write on my application that I have 25 years of experience, does that have to be excluded due to it revealing that my date of birth is likely more than 40 years ago? Can they really believe that the amount of experience isn't important?
Other steps people my age take are to stop listing dates--either select dates like university graduation, or--in extreme cases--all employment dates.
Keep in mind that there are different biases at different screening points along the way. The manager with the open position may want someone older and more experienced, but the people recruiting, screening, and filtering the candidates may be biased or simply not familiar with the value of experience that doesn't match today's buzzwords.
Does my outlook not resonate with you/other older employees?
When they come in for an in-person interview racial/gender/age/whatever bias is still going to be there.
Only way to prevent this would be to forbid in-person interviews before hiring/signing a conract, but then companies would presumably simply stop hiring from applications, because the qualifications are unreliable and they can't get a character impression of the applicant.
Hiring by reference from existing employees would become the norm (maybe it already is).
Also, I relate this to culture. Not to offend anyone, but from my personal experience lying in CV is much more common for some countries, that for others.
Scale maybe terrible, like multiple certifications in some field with no actual basic knowledge or skills. Like Certified Senior (sic!) Software Developer who can't tell you difference between stack and queue, or CCIE certified network engineer who has no idea what is DNS and how it works (real case).
If both the application and the work is completely remote and anonymous it might be possible to avoid most bias, but that's only feasible in IT and a couple other sectors and even there it's not common.
The organisation must have a very respectful culture in which colleagues feel comfortable calling out others on their biases when this happens.
"eh, midwestern state school"
"what even is an HBCU are those accredited?"
"fortran though?"
the initiatives are at least conscious of sensitivities in the space, but we have a long way to go
It would be nice to build a group-blind world, and the more we adopt these specific approaches the better can can iterate towards. However, care should be taken to ensure the baked in biases don't further entrench.
Inequality is another discussion than bias and discrimination, in my view. Inequality is solved by implementing encouraging measures. Discrimination is resolved by removing entrenched barriers.
Of course people are a sum of everything that happened in their life, but this is a way to make a particular job application much more fair, based on capability to do the job rather than irrelevant physical details. This is not meant to solve everything at once.
The very examples you post are about situations that would be helped if there was blind judgement.
Yes, it would be much better if we are able to remove discrimination earlier in life.
But this is not a either/or solution between affirmative action and equal opportunity.
But if the assumption that positive reinforcement make a better just society for future generation then doesn't it mean equal opportunity will make also have the same positive effect albeit a little less ?
Affirmative action laws were introduced to address this, but the trend toward meritocracy counters them.
My guess; we end up with the worst of both worlds: One stream blind to these factors and a bunch of regulated exceptions.
Engineer here, and no, blind hiring will not bake-in any biases, because that's exactly the problem this approach solves - it aims to hire the best candidates regardless of their race, origin, religion, accent and other irrelevant characteristics.
The identification of the issue is at hand. Consider Computer Science. Historically far fewer women go through the training for a variety of personal reasons and societal pressure. So, on the margin, blind/anonymous hiring ignores that upstream delta between females in population and skilled females in hiring pool. Whether this is desirable or not is not my point, only that this is "baked in bias" perpetuating down the line.
It's noble to build a society where we value blindness on characteristics irrelevant to hiring. Doing it piecemeal can result in furthering these baked in biases.
A better approach would be to split the skills validation (blind) from the move-forward decision (adhering to whatever inter-generational correction schema the society has decided upon) where the only pool of eligible move-forwards meet the established skills validation. In this way, you only get skilled candidates, and you can satisfy whatever other characteristics that need satisfaction.
Yes, it will: if you're born to a well-off family, you'll go to better schools before college. Due to this, you'll have a better chance at succeeding in college, have more resources to delve deeper into your field of study, better manage studying abroad, &c. All of these factors will make your list of skills more attractive to employers.
Compared to someone who was poor who went to worse schools, didn't study abroad, and couldn't afford to spend more time on their field of study due to having to work to pay for college. This will make your list of skills less attractive to an employer.
Right out of the gate, there is a bias against people who grew up poor in the hiring process.
> it aims to hire the best candidates regardless of their race, origin, religion, accent and other irrelevant characteristics.
It does so by devaluing repressed classes of people. This is a societal cost to benefit businesses.
Deleted Comment
> The same group(s) facing discrimination at the employment point also face discrimination earlier in life
The goal of Helsinki here is to make the hiring decision independent of direct factors they blind themselves to (name, sex, etc.). However, the factors they condition on (education, experience) are impacted by these factors, so the hiring decision won't be independent of the factors.
The "best" thing to do is to correct the inherent bias, as you're pointing out as the real problem. We agree. This isn't something Helsinki can do in a vacuum, so if they want to achieve their goal of true "blindness" to the above factors, they have to take into account their impact on their "seen" factors somehow.
I put the formalized model I have in mind in another comment on this thread, which walks through my thought process here.[0]
[0] https://news.ycombinator.com/item?id=18548270
[...]
> when much of the information in the application has been hidden, it’s hard to establish a clear picture of the know-how and capability of the applicant
So they base their assessment of know-how and capabilities on age, gender, mother's tongue, and other personal information that has been hidden? I feel like I must reading this wrong.
In my experience it's the opposite. Like with small countries: the odds they produce someone extremely good at a certain sport is just smaller than in large countries, so large countries are more likely to win some world cup or the Olympics. I know only a handful of women in IT to begin with, but most of them aren't great at it.
But that's not your point. You're saying that there are biases in hiring, and I agree. What I'm surprised by is that the article quotes them saying that they now find it very hard to judge someone. Who would say, alongside a press release about anonymous applications, that they find it difficult to base their judgement solely on only relevant information?