Could your recruitment process be discouraging female developers?

Charlotte on October 17, 2019

There’s a huge push for technology companies to hire more women but as female developer myself, I really haven’t seen much change in the past few y... [Read Full]
markdown guide
 

Beautifully written.

I will just add that as a white male, I would also trust much more a company that is doing things like you suggest.

Maybe, instead of complaining that it's so hard to find good developers, company should start fixing their recruitment process!

 

Thanks so much. I agree! the current process is demoralizing to everyone and rejects lots of great developers.

 

I might share an unpopular opinion here, but I question the relation of gender and hiring processes. In general, the improvements to hiring methodology you are suggesting are good but I don't see any reason why the mentioned problems would specifically target women. Being through the process of hiring on both sides (as making hiring decisions and as a candidate) many times, I agree that the vast majority of assessments and interviews are simply inefficient to judge whether a candidate is a good fit.

About the industry:

Now, we are certainly not perfect when it comes to inclusion and there is a lot to be learned in the years to come. However, I am always a little irritated about the emphasis of gender in our industry. What other industry has such a high rate of female influencers while having less than 12% of its members being female? What other industry has such a diversity on other levels (heritage, race etc.)? Again, that does not mean the industry is perfect and that there aren't points worth discussing, but I simply don't believe that this particular level of discussion is helpful.

About job requirements:

It’s been found that women will only apply for a job if they meet 100% of the job requirements but men will apply if they meet 60%

I'd be interested in the source of this finding. Assuming that is accurate, what do you think the appropriate approach to this problem would be? This certainly points to a difference in self-assessment and probably reflects more on perceived chances than reality. This is certainly worth looking into, but unfortunately tells us nothing about actual hiring. We would at least have to consider the success rate of interviews to make any assumptions at all (because that fact wouldn't matter if e.g. men are rejected proportionately more than women. In other words, if the x percent more men get rejected at x percent, we would only have to wonder why men apply for jobs they are not qualified for)

The idea of professional gender neutrality:

Now, please be honest with yourself. I know that from your position as someone coming in with a political view this might be a little different, but I can assure you that 80% of the time I have no idea what gender I am communicating with on GitHub nor do I care (unless I have to talk about someone in the third person and suddenly have to wonder if I refer to accounts like "lowdash85" as he or she). Why is that? Because of a bias of a completely different nature: Personal profiles. Only about 40% of the people I interact with have a profile picture that is an actual photo. And you know what? That is because most capable coders tend to be more conscious about privacy or at least consider it ideal to remain some form of anonymity and out of principle generate a persona that they refer to depending on the platform (e.g. avatar, handle etc) that is intentionally "a different entity" than their real life personality. People with profile pictures nearly always want to build up a brand of some sort. And these people are either rightfully showing their face as they have gained some recognition in the community or simply have a higher likability to be noobs. I am therefore not surprised at all that the accounts not revealing their gender (whether or not they are female) have a higher acceptance rate in general. So in my opinion we would need to compare accounts that are identifiable as men with accounts that are identifiable as women (so only accounts with profile pictures) to study gender biases. Because what this study actually showed is that my personal feeling is justified: accounts that don't make a point about who they are have a higher acceptance rate.

In conclusion, you are right about your main points:

  • Diversity is beneficial for the work environment and growing out of the male dominance is an important factor for that.

  • The relationship of the hiring process and job requirements is a catastrophe (and I would go far beyond your careful wording here as I personally feel that over 90% of all tech companies hire the wrong people for the wrong reasons)

Now, I also believe that we should protect us from political opportunism. And I must state it as it is in the hope you are not personally attacked: All brilliant female coders I know have advanced without ever playing the gender card. I don't know who you are, but you are certainly trying to build up a brand (you appeared out of nowhere here and on Medium this month and your website has links to your publications, Twitter and Instagram but no repositories or previous work) and ride the #WomenWhoCode wave either for personal or political reasons. I don't have a particular problem with smart marketing, but simply am afraid that the strides we made towards equality cannot be advanced by identity politics.

 

As I've mentioned in the post, it isn't just women that this applies to, but all diversities, however I'm writing from a female developer point of view. The reason why this process affects diversity more, is because you have a small pool of potential diverse candidates, so then you either put them off before they even apply, or you cut them out due to the process being too harsh.

Now GOOD men get knocked out of this process too, but this is simply pointing out to companies, who let's be honest really do not care if they knock out a good white male when they have a tonne applying, that it does matter that you're putting off good people AND that's why your diversity is so poor.

I don't know why some males get so wound up by any article that mentions gender. We're not trying to harm males in the industry, we're trying to get more women into the industry.

With regards to the GitHub issue you are talking from personal experience too, you say you don't judge but you can't speak for everyone. The point that the statistics are showing is that when gender is revealed then female code is rejected at a higher rate.

And as for the unnecessary personal attack; I've actually not appeared out of nowhere. I've been working for over 11 years in industry as a team lead, architect, senior developer. I've decided to start sharing my knowledge in blog posts since I left my last permanent position, which I was busy working on all hours so apologies if I don't have open source code for you to judge me on. If you look back through my twitter you'll see I've actually been posting for years in the Drupal community.

Those women you mention who are trying to be influencers; they're trying to influence the younger generation of females. Why do you see that as such a bad thing and why does it mean that because they're on social media that they aren't skilled?

 

Those women you mention who are trying to be influencers; they're trying to influence the younger generation of females.Why do you see that as such a bad thing and why does it mean that because they're on social media that they aren't skilled?

No, you misunderstood me here. I mentioned that as one of the indicators of our industry being relatively advanced when it comes to diversity. Maybe "respected individuals" would have been a better term than influencers. I meant influencing technology, not influencing in the sense of being a role model (which they are, or course, but that wasn't my point).

I don't know why some males get so wound up by any article that mentions gender. We're not trying to harm males in the industry, we're trying to get more women into the industry.

I don't know which of my words let you come to that conclusion. And I also don't know which "we" are you referring to here in particular. I am certainly not motivated by "protecting male dominance", I would love to live in a world where my peers can be whoever they are and from whatever background without that being a basis of judgement or evaluation. As such, I evaluate opinions and actions based on whether I see them as beneficial or counterproductive towards this goal. So assuming your intentions are honest, we do have the same goal. I just don't find it productive to separate humans into groups and identities while having the declared goal of overcoming exactly that.

The point that the statistics are showing is that when gender is revealed then female code is rejected at a higher rate.

Well, that is the question, isn't it? I assume you read the actual publication, so I'll get into detail:

We have one dataset that examines pull requests based on gender only. The identification of gender was generated by external accounts (namely Google+).
In this scenario we found that female coders had a higher acceptance rate.
Now, statistically speaking, this is already a problem as we can only work with a subset of participants, as contributers needed to provide their gender in the first place while requiring additional platform memberships in the second place, potentially missing a relationship between competency and social media membership here. As pointed out in the peer reviewed version, this alone already questions whether the dataset is meaningful. But it get's worse when we start looking at a subset of people who are identifiable as being a member of a certain gender based on GitHub profile alone. Only that did not happen. Rather, a subset of that group (identifiable as female) was compared against the rest. But what they should have done is using their first dataset as a control group and then compare the differences within each gender (so does it make a difference for both genders whether or not their gender is easily revealed) and then compare the outcome. And personally, I would also be interested in statistics about what gender distribution exists in that realm (e.g. are men more likely to reject PRs from women than women themselves.)

This does not mean I cannot imagine that such a bias exists or manifests in such a way. But this study is not suited to isolate what your source derives from it.

And as for the unnecessary personal attack;

I did not mean to attack you, but am aware that this impression was unavoidable as my eloquence in this language has its limits.

so apologies if I don't have open source code for you to judge me on.

Yes. I know you probably meant that differently but that is indeed the problem. And I fully admit that this is exactly what I would have done as I believe that is the basis a developer should be judged on. Not on gender, ideology or political affiliation. But that does not mean that my intend would have been to identify flaws in your code. It would have given me the opportunity to evaluate whether you are in this community as a valuable member or as a political agitator. And again, sorry if this sounds like a personal attack, but I don't know you or anything about you and can therefore only judge you based on what you share.

I just feel as though you are arguing unnecessarily about something you've never experienced. Had I have come in as a less experienced developer, it still shouldn't require judgement of my code in order to write an article about the recruitment process. I'm sorry if you feel it's your right to be able to judge me, but I have no motives apart from to get to know like minded people, share and improve my knowledge. You should probably also look up Women Who Code too as it's not just a hashtag, it's an organisation that many of us are members of that is doing amazing work.

Well, your post is titled

Could your recruitment process be discouraging female developers?

and I happen to look for developers. So naturally, I was interested in finding out whether you have valuable insight into that BECAUSE I have not personally experienced it from a female perspective. In other words, I am here to find answers to the question your article is about.

I'm sorry if you feel it's your right to be able to judge me

Yes. Let me make one thing absolutely clear to avoid misunderstandings: It is my right to evaluate any information I process. It is an intellectual necessity to judge and it is correct to judge. We just all need to be very cautions about what we base our judgments on as we are terribly flawed in that respect and must therefore be able to change our judgement when additional information comes in (you know, fighting one's own confirmation bias). And I think we would agree on what judgement should not be based on. As a matter of fact, you have responded very judgmental to all my remarks and I ignored it as it is your right to do so. You assumed my gender, my motivation and my intent and talked about "men" vs. "we". And that's all fine, that's all human. But I can only have a meaningful conversation with you if we accept that undeniable fact: Yes, I judge you and yes, you judge me. Let's go from there.

I know Women Who Code is not just a hashtag. But that does not mean that everyone using the hashtag is affiliated in any way. But since they are globally active and you mentioned it: When you say "amazing work", what do you mean by that? And there is probably many things, but besides the mission, what does the existence of this organisations do for female coders in your mind? Or, maybe females to become coders, if you think that is their strong suit. And I am not asking to provoke you, I am really curious as it seems hard to carve out what they are about. And BTW, I think that would interest a broader audience as well, should you want to make that a post rather than a comment.

 

Ok, let me see if i get this right. :)

The bias you are indicating is that some people have too much, or too little, confidence while in the recruitment process, and so we need to change the quality checks to include more people, because those people are secretly the shit, and our process cannot see it?

Dehumanizing interview processes are a problem across the culture spectrum last i checked. At least i have tried them a few times.

Why not encourage higher confidence levels, that's also a bias?

I'm really trying not' to be an ass-hat here, that's my bias i guess.
So please be gentle. :)

 

No, you're totally missing the point here. The current process isn't a quality check, it's actually a rather lazy robotic process that doesn't require much effort from the companies but requires lots of preparation from candidates. This is why it's not at all about lowering the bar; of course companies want to make sure to hire talented people, but this process simply hires one type of person.

Take this example, a skilled person comes in with years of experience and skills beyond coding and messes up one algorithm question because they've spent years doing actual work so that's it, they're out. A graduate comes in who has literally just memorised algorithms and they ace it. Now the companies involved don't care, they've got tonnes of applicants... but they aren't considering that they don't have tonnes of diverse applicants. They are making that pool even smaller because they don't have the human element to their processes.

I get that lots of males don't understand the problem and think it's just minorities complaining. But we experience bias on a daily basis. We don't just go into interviews with the standard worry that everyone has, but the added worry that the interviewer might hate the idea of a woman going into this role. You can call that lack of confidence, maybe it is a little, but that's because of how women are treated in the industry.

 

I think we agree on the first part, that the process is robotic. I’ve been in the exact problem you describe.

But the latter i will have to take your word on. I still think a mitigating action could be to work on confidence. Especially because of the biases you mention.

Even if’ the recruiter did all the things mentioned, in the current state of things, wouldn’t a female still be loaded with the same bias of less confidence?

How would we solve for that?

OK apologies I must have misunderstood :)

I don't think it's a purely confidence thing. In fact most women are totally confident in their abilities to do their jobs. So if the process was less demoralizing I think you'd get the best out of everyone.

No problem, I am not always the best at expressing my self. :)

Yeah, It’s lack of confidence in the hiring managers/process etc. right? at least that was my assumption.

I was thinking that might be cause for anxiety, meaning poorer interview performance than someone without it.

 

This is definitely a good article. Companies often put down a wish list of skills and attributes that they want which can discourage applications. I think that the bullet points should be more focused.

It's simple math as well. If your interview process drives people away, then you're reducing an already small candidate pool of women developers.

 

Don’t expect every candidate to possess the same skills during interviews.
Test the candidates how they’ll be tested in their jobs.
Treat candidates like humans.

I would say those rules would help everyone, no matter what they are.

With this i disagree:

Don’t be so black and white with job descriptions.

I want to know exactly what they are looking for.

Partly so i can avoid companies that require some things that i dont intend to learn.

Partly because the alternative is fuzziness, subjectiveness and "good will" of the interviewer - which is a playground for biases and discrimination.

 

I can understand that you want to know exactly who you are looking for but you have to also be realistic here. I've seen job descriptions that list every language and skill going. If you have a few things that are necessary then yes put them down, but don't put down every nice to have because that will put off people who might pick up those few things on the job really quickly.

 

Yeah, clear label for "bonus points" is necessary.

Its often easy to misinterpret demands from wishful thinking ;)

 

Great post, I agree with those improvements. Companies recruiting often do not follow the golden rule of "treat people the way you want to be treated" which is disappointing. I get that it is more effort for them and a company exists to make money, but leading someone on only to completely ghost them is a common occurrence and unacceptable in my book. At least send a canned response.

I especially like the part about not being black and white with job descriptions. A candidate could be the most experienced JavaScript/TypeScript developer on the planet and a great fit in other areas, but if the job posting says Angular and they never touched Angular, they are rejected. Hire the person that has the background and qualities you are looking for, not only those experienced in a very specific tool. They can pick up that tool and contribute in many different ways if they are the right fit.

 

Thank you! that's very true. It's really frustrating when you know you're a good developer and can do the job but they don't give you the chance to show just how quickly you'd be able to pick up a new technology. But then again it shows it's probably not the sort of place that would be nice to work!

 

I can totally relate to not applying for a job if I don't meet all the requirements. Especially when the posting says stuff like "Must have experience in (insert technology here)".

 

👏🏼👏🏼👏🏼

I guess you've most likely already seen this, but Ronna Steinberg has a really good talk on the subject published on youtube:

youtube.com/watch?v=f0pHdBnKHuQ

 

Thank you, I've actually not seen this, I'll go and watch it now :)

 

You don't apply to large corporations because of their very strict code tests? I just got hired at a fortune 100 company, and across my 3 interviews I never had to whiteboard, take a code test, or submit a code sample. Nothing like that.

It would be unfortunate if pre-conceived notions are holding you back from decent opportunities

 

No, at companies I know for a fact use this process. For FAANG companies it's a given, even if the recruiter contacts you first. I was told I'd have 5-6 interviews from them consisting of whiteboard algorithms and system design. I'd be intrigued to know which one didn't do this. Is there any way to DM on here if you don't feel comfortable posting?

 

Yeah, if we both follow each other we can DM

code of conduct - report abuse