Gender Bias in Tech: Examining Evolution & Persistence of Stereotypes

Speaker 1: Thank you. Ciao. Hello everyone. I'm Esther. I'm a web developer at Torchbox. And Torchbox who Those who don't know is a digital agency specializing in creating a website, web application, or digital marketing strategy. And we have also create a very very handy GMS for Django called Wagtail. But I am also one of the organizers of Pyconitalia and one of the Strawberry GraphQL code dev So today I wanted to talk about a little story. So early this year

Speaker 1: I was in Florence to meet one of my best friends, Sabrina, and we went in a bookshop and she suggested me a very interesting reading. Invisible Woman Explorer Data Bias in a World Designed for Men, written by Callorin Perez. Is a very interesting reading. Um talks about How gender bias affected the use of data in many areas like healthcare, education, public policy, etc. So yeah, spoiler, we have a long way to go to fix gender gap. But today I don't wanna talk about this book itself. itself but about a particular chapter, chapter four, that focus specifically on the tech industry, so our

Speaker 1: industry. And I was particularly shocked to know that during the 40s and the 50 woman women, no men, were the dominant sex in programming. So the opposite of what we have today. And of course, like many of you, I think, uh I have seen this picture of women working with big computer, old picture black and white, uh with many cable and stuff, but I never realized how many of them they so they they were the majority Um so some there were the majority so much that were article encouraged women into programming and like magazine like Cosmopolitan, so

Speaker 1: a typical feminine magazine And this to me seems images from another world, another era, is how is even possible? How did we get From this situation to what we have today, that we are struggling to fix gender gaps and increasing women participation in tech So yeah, um for let's divin in some historical reason. So during the Second War, men were called to arms, so many women work in technical roles. including operating uh early computer like INIAC in the US and CLOSSUS in the UK. So those experiences

Speaker 1: paved away for women to continue to work in technical field also after the war. And the second point is that the perception of programming was different. In the early days of computing, programming was seemed like a less prestigious than hardware engineering because hardware uh was morr considered more challenging so maskulin oli och stad programming was compared to clerical work, so like typing, filling, those traditional more female dominated fields. So this perception made it possible for women to enter in this field.

Speaker 1: Um yeah, so there was already a gender bias here. that led to underestimate the complexity of women work. So the term programming was not yet my stream so there were many assumptions around that. In in fact, uh one of the uh women that uh work at ENIAC, Ruth Licherman, is in an interview is trying to clarify it what is what is programming. So the interviewing is uh asking her a question and um assuming that yeah programming is easy what you thought was just plug and play And root start uh correct him

Speaker 1: no it actually involves more than that. It involves take the problem, spit into pieces, take pen and paper, start to do diagrams and actually uh sometimes there were women specifying in a part of the software because it was too big And only after all this process they went to the machine and start to programming or in this case plug in the cables So yeah, that sound a bit familiar to me, but yeah, no , do not you why. Um so So, we're going to be able to do that.

Speaker 1: So, every programming technique must be work it out for the first time and to resume that INYAC made a lot of innovation in programming methods. For example, Betty Hoberton invent has invented a special technique that involves stop the machine in the middle of the process and look at the partial results. So I think we call now breakpoints. So yeah, thank to her the next time you are debugging your application. So thank Betty for this one But yeah, to going back to our story, what happened next? So the computer became commercial

Speaker 1: starting from the fifty because before was only used for scientific studies or government programs. But at the end of the decade computer became mainstream and uh company like EBM started to sell them massively. So uh any company wanted to buy one and start to build his custom software, so they will start looking for skiller programmers and quickly to develop their software. And good programmers were hard to find because no one knew how to program yet. So yeah, that's when the first software crisis was declared. So the industry starts

Speaker 1: to use some attribute tests and personality profiles to rank the candidates, the potential programmers. because nobody knows how to program yet, so was a brand new thing. So the company has to find out the potentially good employees and train them after. So it was a huge, huge investment. So they start looking to only those skills and characteristics that can lead to be successful in programming. So the underlying assumption here was that there were some innate qualities that can be linked to a good programming performance performance. So

Speaker 1: this tests aim to evaluate specific skill like verbal meaning reasoning with mathematical trivia logic puzzle. or other qualities like emotional stabilities and stuff like that. This is the picture is actually an example of one of the EBM aptitud tests. So they were already criticized, they were already making jokes of that because study proved that there was no significant coloration between um a good test and the subsequent job performance. So yeah, but this that they continue to use that.

Speaker 1: So what came out is some beliefs. So for example, with stuff like the relationship between programming and musical ability. Or yeah, that programming uh enjoy their work, dislike routine, um and they were particularly interested in puzzle solving activities, they disinterest people, programmer dislike closed interaction, so they prefer to work rather than people Um yeah, so those personality profiles formalize the programmer stereotype that we

Speaker 1: all know. A male typically wearing a hoodie, always wearing a hoodie, does spend all of this time behind the computer which may not leave much time for socializing. Unfortunately, this stereotype discouraged women from pursuing uh programming careers and uh the feel that uh you must be obsessed with computer to be uh succeed in programming. So yeah, all all of this so the from like 1981 the woman participation in the tech industry and start to decrease So both in occupation

Speaker 1: and in computer science. Plus males usually have more experience in computer because they tend to have one at home So and of course they were using all the time to play and stuff. So which gave them the privilege in classes because as a result, they were more confident and wan var this ting empates the stereotyping woman were less kapable of programming basis So stereotypes and bias continue to affect the tech industry today and leading to lack of diversity and inclusion in the tech teams. So addressing those bias

Speaker 1: and interrupting them is crucial to create a more inclusive industry. And yeah, there are several factors that contribute to gender bias. unconscious bias. So from when we are young we we get our culturally based mental models that help us to dry the world, but they also um lead us to unconscious bias and comes in um as to miss sam sträng, sam talent, or karatteristik in people who doesn't fit vårt of what is a god leader, a god teknikal person. And yeah, these

Speaker 1: bias are naturally for human psychology and we all have them in some degrees. The important thing is to recognize them and take action to interrupt it. So it's not about blaming individuals or grupps. Just focus uh on identify them and work together to to create uh a more inclusive environment. Gender stereotype, for example when women may be characterized as too aggressive, why the same behavior is acceptable for men. And this bias influence hiring and promotion decision, so

Speaker 1: resulting in disparity in compensation. Is estimated that women women globally uh are paid l about twenty percent less than men. So um this is data getting worse if you're speaking of women of color. So this means that uh wo women are working for free two months a year. Um And yeah, also gender bias create an hostile environment, make it difficult for women to pursue their career We have to think about also the gender bias in our algorithm, for example, some years ago.

Speaker 1: A study fied out a linkering were suggested suggesting more male candidates than female candidates So they later hired a diversity team to fix this problem, but it still has it in some ways. And plus These algorithms are not open source, so we rely only on some independent studies. Um and Later these big companies start to lay off many, many um teams and and employees. So I hope that this will not the case. Det may be också the diversity team will be reduced and hoffly not. But we can

Speaker 1: the consequences we'll know only in the years in the future. So yeah. Another another study find out that when company and in their core values emphasize bureaucracy At men we offer more bonus, higher bonuses and uh more more uh respect women. So this is because the meritocratic system, the evaluation of performance is often based on subjective judgment and assumptions, which can be influenced by gender bias. So, this study hajar the importance of beja of the intentional konsekvens

Speaker 1: av meritocracy and create a truly equitable and objective evaluation process. So the consequences. Recent here we uh we know that Um there was a gender bias in in some software, for example in the val the recon the um The reconnizing of the face for example when you go to the hypers uh and you scan your passport uh the software uh is failing when uh the person is a pi is a person of color more than the other because the software is not training enough

Speaker 1: to um to recognize all the people. And the same is uh for like for digital technology like trackers that are more efficient for men than women. So um this lack of diversity in tech industry has led to lack of perspective and problem solving which can lead to less innovation, less creativity in the tech industry. Okay. to fix these issues. So because these bias are are natural as we

Speaker 1: say and it's important to recognize and and stop them. Why did it matter? Because inclusion is not a nice to have, it's a priority, it's a must. The company profitability will benefit from investing in diversity team and uh brings innovation will help to understand better the client from a different perspective And yeah, we need to fix just because it's more profitable, but just because discriminating people is wrong. Yeah, it's not fair. Um I want to show like an example here

Speaker 1: of this brilliant mathematician called Diana Tamila. who uh helped to solve a problem, the hyperbolic plane, for centuries we were not able to represent it And this is he helped the mathematician to solve this problem. It's a very good example that show how the uh innovation comes from different skills they combine in different skin and disciplines. So as organization, what we should do to actually practical fix this problem. Let's find down some hints. Let's call it this. So

Speaker 1: from the job description, the words would use that are considered more masculine. Women tend to uh think about uh uh more men work it here so and they find job less appealing. So develop a job advertisement guidelines that uh uh for example that use more masculine words like competitive, asserting, ambitious , we should change them uh in order to appeal more women We have to um know that um research has shown that women apply for a job if they don't apply for a job if they don't reach one hundred percent the qualification.

Speaker 1: So we should think about uh revising our job description and um rather than have a lice nice to have that may disencourage the qualified woman and here there is a really nice reel, very short, hope the audio work that showed this problem.

Speaker 2: Five years experience with Excel. I only have four years and three months.

Speaker 3: Well, I heard of Excel ten years ago, so that's basically ten years experience.

Speaker 2: Attention to detail. I did make a typo last year.

Speaker 3: I always know when my My mom moves my stuff.

Speaker 2: Great communicator. I've been working on my communication issues in therapy.

Speaker 3: Yeah, I talk good.

Speaker 2: Works well with others. They're gonna find out I had a friendship ruined over a project in fifth grade.

Speaker 3: Others love

Speaker 2: Startup environment. Ugh I hate playing ping pong.

Speaker 3: Startups, no HR.

Speaker 1: Okay And the hiring process um stage start giving hiring managers blinders me that so they can focus only on the skills and qualification and track if there this practice are changing the hiring number. and uh um build um uh interview the interview think about interview process so develop a set of questions and you ask ask to all the um candidates so to help ensure fairness and consistency. And last but not least, provide encouragement bias training to the manager

Speaker 1: and hiring managers. Encourage education to young women and emphasize that there is not Only one way to be interested in computer science, so the hackerboard icon, but you don't you don't have to be a substance to be good at it and representation so our rural model and mentor play a crucial role inspiring guiding print people in their career and this is uh uh I will encourage you to watch uh a talk a keynote talk from Pyconitalia last last week the streaming is already online and uh here there is the link

Speaker 1: um the power of the presentation if you want to go deeper in in this topic Um mentorship programs are a beautiful example about how to increase women participation in in tech. finding new talents. For example in Torchbox we have the Academy but there are other programs like Google Summer of Code. So Yeah, people without having a real experience in programming can learn or switch careers. Um yeah That's are some of the resources that I find interesting. Um I will share the slide so you can

Speaker 1: And yeah, that's was actually it 's a little bit more than that.

Speaker 4: Thank you, Esther. Uh I know some of us in the room try hard to hold their tears down, uh, listening to talk. Uh really, we need to talk more and more and more and more about Gender gap. Um

Speaker 5: I have a a question about um so I've been with the same company for ten years. The IT department is kind of, you know, very much the stereotype of the nerdy guys in the hoodies. Um and recently our team has grown quite a bit. The organization as a whole is very diverse, but I'm wondering what kind of things like as a as a technical team lead I can do to you know support um a more diverse and welcoming environment.

Speaker 1: Um yeah starting hiring more women is is is a big step ensure that there is not just one woman in the room. and um also uh surveys about uh how they uh um participation and how they feel in the team so listen to them uh can be a good step ensure that um like uh there are no issue in the in the company and find a way to report if they are uh anonymously Um this is a good thing. I think feedback is

Speaker 1: a very uh good way to turn the company um culture. So

Speaker 4: I have one question myself. So no, it's a Not hard for you. Maybe more difficult to answer uh from this part of the room. Okay. What do you think about diversity in our kommunity like Python contributors, Django contributors, uh DSF or like uh other small communities like and maybe any

Speaker 1: I think we are making some progress uh recently and we are looking in in the problem and we start to see um some results and warmer participation is slightly inc improving uh in the last year I think The next step I think will be having more women or diversity of any kind in the leadership. uh that is I think is is still low uh in the representation of the company in the boards.

Speaker 4: Any other questions? Uh

Speaker 6: thank you very much for your talk. Uh my question is how would you help a woman in tech, say a woman of color, overcome uh the barrier of uh Fearing to put yourself out there like you said, women don't apply when they don't meet hundred percent of the requirements. So what advice would you give them

Speaker 1: Um to the company uh I will give advice to um uh put effort in the job description to encourage women, for example having a specific task that m encourage them and yeah to women it's a bit hard I think depends on on the person uh and It's it's kinda hard issue. Um we should be bore brave uh in in general. uh and recognize um we we always underestimate us uh I do. So try to work on on this bit

Speaker 1: uh is a really good uh first step. Um And participate to like worship on a sprint I think is a good way to be um to feel like more um competent I d I guess and confident I this is the word

Speaker 4: I think we have uh time for more questions yeah please

Speaker 7: Uh thank you for the great talk and and like for bringing up these issues. Um the st uh some some studies like a a famous Uber study show that um differences are sometimes caused uh not by biases of the companies Sorry, but by uh the general world as how it treats women differently and uh um and how how things are generally in the world harder for women sometimes. Do you think companies should try to compensate for that using affirmative action and that kind of policies?

Speaker 1: Yes, company especially big company has uh the power for change a bit the world so they should invest more in changing our mentality uh we yeah. I think so. They they have the power to do that. And once we start to think in a different way, uh then we become more normal. Hi

Speaker 8: Esther, thank you. Do you have any practical tips for trying to recognize your own unconscious biases?

Speaker 1: I can't think I'm working on it.

Speaker 8: Not yours, but mine. How do

Speaker 1: um okay. Um Um yeah, uh I I think I'm thinking um yeah it's it's It's kinda hard. I think receiving uh ask for feedback uh because It's hard to recognize my own bias. So uh I think you have to ask some opinions about other people about how they they feel what how m How did I perform not I me but in general one last question? Uh

Speaker 1: there is one

Speaker 4: okay a quick one please

Speaker 8: Great talk, thank you. Um just a question more on the educational point of view. Do you believe that we encourage females from a younger age at schools

Speaker 7: enough to step into

Speaker 1: Mm, I think we can do more. Uh I mean um if Yeah, I d I don't I think we can do do more um because um from the young age uh we still have this kind of difference between um boy toys and boy and female boys And so you usually still the issue that to girls we we give a nice kitchen to boys we need the chemical thing. This is something if we teach them for a very young age we can

Speaker 1: they will be more um confident and to go to to the school to stem education uh in computer science than when they are uh not adult but teenager.

Speaker 4: Well uh thank you Esther. Not only you gave a great talk but also try to be the answer of uh some of the difficult questions which we should be looking for answers all together. But yeah, thank you.