This chapter presents the results of a qualitative study that aimed to understand how people in software development teams feel about gender diversity in software engineering, the perceived benefits, and the perceived difficulties. To achieve that, we applied a qualitative survey. We found out that gender-diverse workplaces are prone to have better ideas sharing, better decision-making, creativity, and innovation. Women inspire other women, and some men reported being touched by the subject and diligently are deconstructing their prejudice and misconceptions about women in technology. However, it is still common to not see women in teams or to see just a few. A white, cisgender man is the pattern most of the time. The same pattern repeats itself in leadership positions leading to male protectionism and privileges. Additionally, other dimensions of diversity pervaded the answers, like intersectionality and race/ethnicity.

The Survey

Software engineering involves real people in real environments. People create software, people maintain software, and people evolve software. Accordingly, to truly understand software engineering, it is imperative to study people – software practitioners as they solve real software engineering problems in real environments. It means conducting studies in field settings [13].

This study aims to understand how people in software development teams feel about gender diversity in software engineering. So we applied a qualitative, unsupervised, largely open-ended survey, using a web self-administered questionnaire. Our target population consists of people who identify themselves as part of a software development team worldwide.

The survey questionsFootnote 1 for this work were formulated based on the Empathy Map Canvas, created by Dave Gray (2009) as a tool that helps teams develop deep, shared understanding and empathy for other people [7]. Henschel et al. [8] say empathy corresponds to the ability to understand others’ minds, feel their emotions outside our own, and respond with kindness, concern, and care to their emotions.

The participation was voluntary and confidential. Using the mechanisms of the social networks, we estimated the survey link reached around 1,280 people, and 149 answered totally or partially. We evaluated that 60 respondents answered less than 20% of the questions, mostly only the demographics. We could not use these answers to extract information. In the end, we used answers from 89 respondents, corresponding to 6.88% of the people impacted by the post/link to the survey.

We had 44 respondents self-declared as man (49.44%) and 43 as woman (47.19%). Only one respondent self-declared as non-binary. One respondent self-declared as bisexual, which is a sexual orientation, not gender. Once it was self-declared and we did not have a way to identify gender, we considered it important to keep the information. About the age of respondents, on average, the respondents were 35 years old (36.09 for men, 35.15 for women). The self-declared non-binary was 31 years old and the self-declared bisexual 24 years old. The oldest respondent was 61 and the youngest 23 years old. Race and ethnicity was also a self-declared question. We had 65 respondents self-declaring white (73.86%), 5 brown (5.68%), 5 black (5.68%), and 13 diluted between Greek, Latin, Brazilian, Chinese, Swedish, Indian/Asian, and not informed (14.79%).

We also asked some professional demographics. About the role the respondent had in the software development team, we had 80 responses where 40 said they were software engineers. The other 40 answers were spread in a spectrum of scrum masters, product managers, engineering managers, DevOps, business analysts, data scientists, directors, professors, quality engineers, researchers, software architects, systems analysts, and trainees. About years in technology, we had 79 answers for this question, and, on average, respondents worked in technology for around 13 years (max, 30 years; min, 0.3 years).

To analyze the survey data and gain deeper understanding of the data content, we used thematic analysis [2]. Through the process, we created 429 codes for 709 segments of the answers of the 88 respondents. We grouped the codes into 18 themes, and for five of them, we split them in sub-themes, better supporting to answer our research questions. Table 11-1 presents a summary of the themes and sub-themes.

Table 11-1 Summary of themes and sub-themes
Full size table

The following sections we present the themes we defined during the thematic analysis and the perceived benefits and perceived difficulties of gender diversity on software development teams reported by individuals.

Perceived Benefits of Gender Diversity on Software Development Teams

Diversity is good beyond ethical reasons; it is recognized as valuable, and a lot of studies have been done about it [16]. Large technology companies have been increasing their efforts to have a more diverse workforce, increasing minority numbers through recruiting, working to minimize unconscious bias, and also investing in programs to increase representativeness [12].

From the Benefits of Diversity theme, respondents see gender-diverse environments as more prone to innovation, creativity, and better decision-making. Nonaka and Takeuchi [15] emphasize that a team made of members with different backgrounds, perspectives, and motivations is critical for organizational knowledge creation to take place. Knowledge creation is the basis of repeatable innovation in companies. “An organization’s internal diversity must match the variety and complexity of the environment in order to deal with the challenges posed” [20].

That having a healthy gender mix results in better team working, better decisions, and a better product. A lot of people are tired of the “bro” culture around startups, and having a healthy gender mix ensures you don’t end up with a laddish/pizza and beer feel to the team.

—#5, Man, Software Engineer

From the Diversity as a Whole theme, respondents believe diversity can create a more democratic workplace. They try to give visibility to unconscious bias (the social stereotypes about certain groups of people that individuals form outside their own conscious awareness [14]) and have a neutral workplace. Judy [10] says in a performing team, each member relates to the other as equals. The team will inspect its behavior and continuously improve the social skills required to communicate. It is a requirement for trust and the kind of collaboration that leads to cohesion and self-direction.

I believe that greater diversity would make the environment richer in ideas, dynamic, and democratic.

—#4, Man, Software Engineer

From Companies Trying to Be Diverse and Inclusive, we had respondents mentioning the cultural codes of the companies as important tools to address ambiguous situations. Most codes of conduct aim to protect members from harassment (aggressive pressure or intimidation) [21]. As such, workplace harassment is any offensive, belittling, or threatening behavior toward an individual worker or group of workers. It results in an unpleasant, humiliating, or intimidating environment employees feel uncomfortable in and consequently damages the effective work and productivity of employees [21].

Our company has a beautiful culture code that encourages us to be the best person every day. If someone feels bad about a situation, it should be addressed to an anonymous channel, and HR will take care of it.

—#36, Woman, Scrum Master

The theme Men and the sub-theme Deconstruct Prejudice helped us understand that some men are not only empathetic with the subject but they are also to identify their gaps to support better work environments. Viana et al. [23] analyzed the stereotypes attributed to “egalitarian men,” men who support gender equality in relation to domestic and family responsibilities as well as inclusion in the workforce. They found out the egalitarian man is perceived as fragile, sensitive, incompetent, and feminine. On the other hand, he is also seen as more competent and social than egalitarian women and traditional men [23].

I am always willing to learn more and more. In the end, I’m a cisgender man, and with that, everything I say or think is projected through the search for information. I’m not in my place of talking about gender diversity, so I always try to put myself in a learning position, deconstruction of my prejudices and construction of a space for diversity.

—#29, Man, Software Engineer

The theme Women and the sub-themes Women Characteristics and Women Supporting Women helped to identify some characteristics that are more prominent in women, such as empathy, flexibility, and collaboration. Turley and Bieman [22] identified specific competencies of knowledge, personality, and attitudes as significant factors influencing software engineers’ performance. Darley and Smith [6] say females pay more attention to details and disparate, multiple cues for information processing in simple and complex tasks.

Empathetic, flexible, collaborative.

—#28, Woman, Engineering Manager

Also, respondents emphasized the importance of women supporting other women. Research indicates that the social aspect of Agile practice, particularly routine face-to-face meetings and pair programming, reduces women developers’ sense of isolation and raises their satisfaction and confidence [10]. It also reduces feelings of internal competition and builds trust [10]. Studies have shown that mentors can help women motivate them and improve their self-confidence to achieve their goals [24].

I always try to share and praise the women I admire and who motivated me to continue in the software development area.

—#14, Woman, Business Analyst

Perceived Difficulties of Gender Diversity on Software Development Teams

The first theme is No Diversity where respondents mentioned few women in the teams and no other gender than man. The pattern repeats itself in leadership positions, where there are, most of the time, white men, who are also cisgender (a person whose gender identity matches their sex assigned at birth), linking to the theme Leadership and Management Roles where respondents mentioned the difficulties for women to foster their careers to leadership positions.

Our findings are aligned with the previous work of other researchers. The stereotype of a computer scientist is a middle-class white man who is often geeky and antisocial [17]. While stereotypes are not necessarily representative of the general population, they do impact the perception of who belongs in the field and can act as exclusionary forces for people who do not fit the stereotype [17]. However, respondents identify characteristics in women considered leadership ones: collaboration, conciliation, determination, empathy, flexibility, organized way of working, and will to win. Jetter and Walker [9] say, on average, women are more likely to avoid competition, underperform in competitive environments, and exhibit higher risk aversion than men. Persistent social phenomena, such as the gender wage gap or the underrepresentation of women in highly competitive occupations and job positions, have been linked to such observations. One prominent hypothesis to explain this phenomenon relates to the idea that the gender of one’s opposition could influence competitive behavior. More generally, people may behave differently when competing against adversaries from the opposite sex. Women are especially underrepresented in jobs generally associated with high-pressure environments and large stakes and also technology where women usually occupy less than 20% of positions [9].

With a lot of effort, we managed to advance some steps in terms of representation. However, in more strategic positions, the participation of women is almost nil.

—#25, Woman, Engineering Manager

The themes Sexism and Prejudice, Men, and Professional Insecurities continue the previous ones. Respondents reported male protectionism and privileges preventing women from advancing in their careers. Together, they lead to a chain of difficulties for women:

  • Women suffer from sexism and prejudice that can be veiled and also can lead to harassment.

    Some time ago, my wife reviewed a code, and even though she was a senior, their colleague said that he did not accept her review because she was a woman. She complained, and the company did nothing. There was also the case that his manager called her to dinner at a one-on-one meeting.

—#18, Man, Software Engineer

  • There are two sub-themes from the theme Men, related to attitudes observed in men: the Men Characteristics and the “Bro” Culture. Both are related with what the respondents called the male protectionism.

    A network of male protectionism, which prevents women from advancing in strategic positions (or advancing with great difficulty, even requiring a certain masculinization to do so).

—#24, Woman, Engineering Manager

  • Women go through Professional Insecurities. They feel they need to prove they are as capable as men, and it generates the feeling of never being competent enough, and their opinions are worthless and disrespected.

    I often feel that I need to go beyond my peers concerning training like I’m never competent enough.

—#27, Woman, Engineering Manager

The “brogrammer culture” is a term that acts as shorthand for pointing to sexism in the tech industry. The term “brogrammer” began as a satirical term to refer to a man who can code and succeeds with the behaviors of a stereotypical “frat boy” and ambition to become rich fast [17]. By definition, women are excluded from this group and often objectified and pushed out of workplaces because of the fraternity-like environments created as a result of brogrammers being in the space. Women who succeed in these environments may also face difficulties due to their gender, as being better than men can be seen as threatening. In addition, how women perceive being gendered in male-dominated spaces indicates how they relate their experience to gender. Women’s beliefs about the reason for the lack of women at high levels in companies, for example, influence whether they are motivated to fight structural barriers for other women or if they reify glass ceilings [17].

Themes Companies Do Not Support Diversity and Inclusion and Missing Affirmative Actions and Initiatives relate to one another once companies seem to have difficulty supporting diversity and inclusion and affirmative initiatives for diversity are scarce: too much talking and no action. More than that, the perception is that companies are hidden behind a facade of employers’ branding initiatives that happen only from the door out. Simmonds et al. [18] observe that affirmative action programs to boost female enrollment in programs can have positive effects for science, technology, engineering, and mathematics undergrad programs; however, the initiatives yield weaker results for computer science/software engineering majors.

Too much talk and little action, nonrecognition of women in development teams, the lack of looking at different aspects of diversity.

—#20, Woman, Software Engineer

The theme Intersectionality permeates diversity. Intersectionality [5] is an analytical framework for understanding how aspects of a person’s social and political identities combine to create different modes of discrimination and privilege [5]. For some women, in particular women of color who are doubly and sometimes triply excluded because of their skin color and language practices, the discourse on gender may itself be harmful [4].

Intersectionality is a complex and relevant approach that is little known in the scope of the software engineering research field. Respondents brought to the light they see the number of opportunities increasing for women; however, the initiatives seem to be focused on white women. More than that, they mentioned they have few black and transgender workmates. Another face of diversity mentioned was ageism, the process of systematic stereotyping or discrimination against people because they are old, just as racism and sexism accomplish with skin color and gender [3].

In the previous company, I was the only woman on the team I was on, there was only one black man, and diversity was not encouraged; not everyone had a voice in the company.

—#16, Woman, Software Engineer

Gordó n and Palacios [19] say the diversity crisis is not limited to women it is about social identities that go beyond gender and race, but it is mainly about power. There are cases in which software workers belong to two or more underrepresented groups. It is clear that systems of privilege and oppression often converge for underrepresented groups, that is, there are organizational power dynamics that have historically privileged some groups and marginalized others in the software engineering field [19]. They also say an intersectional approach invites software engineering researchers to read data in different ways and ask other questions that increasingly demonstrate the flaws of a race-only or gender-only approach.

Also, Meritocracy and Elitism are also understood as problems related to diversity. Meritocracy is a social system in which advancement in society is based on an individual’s capabilities and merits rather than based on family, wealth, or social background [11]. Nowadays, meritocracy is a subject with a lot of attention; however, the “if you want you can” speech opposes the lack of financial resources some individuals have to enter the software engineering field. The area is defined as elitist once equipment and training are usually expensive. If an individual does not have the means to be trained as a software developer, the discourse of meritocracy does not apply.

It is an elitist area. And the elite has one color and one gender. The market would easily accept black or trans people with the appropriate technical skills, but why does the market have almost no such people? Because these people are marginalized, they are on the periphery, they are without access to our world. Equipment, courses, internet, everything is expensive, so only the same group, the same elite as ever, qualify well. Companies are more open to diversity, but to what extent will this bring a truly diversified team?

—#28, Man, Software Engineer

The last theme is Unawareness. There were respondents who reported that diversity does not matter or should not be considered when talking about software development teams. According to these respondents, people should be treated equally, no matter their gender, and what must be considered is their ability, not their identity. Also, there were comments that the career is not for women.

I’ve heard several people, mostly men (professors, college friends, lecturers), saying that [diversity], it’s not important, that this profession is not for women. There are still a lot of people who are resistant to this issue.

—#29, Woman, Software Engineer

Alba [1] says gender equality does not mean pretending that “male” and “female” do not exist. Gender equality also does not mean that males and females must always be treated the same. In some cases, what is required is not equal treatment but equitable treatment. Equity means recognizing that differences in ability mean that fairness often requires treating people differently so that they can achieve the same outcome [1].

Conclusion

We conducted this survey to qualitatively understand the perceived benefits and difficulties of gender diversity in software development teams. The answers we had aligned with different related work performed by researchers from the software engineering community. We also found studies from other knowledge fields, such as psychology, that studied gender issues in technology and corroborated our findings.

From a beneficial point of view, we have that gender-diverse workplaces are prone to have better ideas sharing, better decision-making, creativity, and innovation. Characteristics linked to women were highlighted as empathy, flexibility, and collaboration. To achieve the benefits mentioned, respondents reported that some companies worked to improve the hiring process to be more gender-inclusive. To support and guarantee inclusion and safety, solid cultural codes were created.

There is mutual support from women to women. Women support and inspire each other to remain in technology or enter the field. More than that, some men reported being touched by the issue and diligently deconstructing their prejudice and misconceptions about women in technology.

However, there are also difficulties from the point of view of gender diversity in software development teams. There are still cases where the respondents do not see diversity. It is common to see only one woman in teams or just a few. More than that, no other gender than men and women, so the white, cisgender man is the pattern most of the time. The same pattern repeats itself in leadership positions. This reported pattern leads to an issue: male protectionism and privileges. Due to that, women report being frequently disrespected and face difficulties in thriving in their careers. Sexism and prejudice used to happen together.

But there are other dimensions of diversity that pervaded the answers, and intersectionality was mentioned. Intersectionality aims to study how aspects of a person’s social and political identities combine to create different modes of discrimination and privilege. So, when racism was mentioned, we can exemplify when respondents brought to the light they see the number of opportunities increasing for women; however, the initiatives seem to be focused on white women, excluding the intersection of race and gender of black women. Ageism, the discrimination against people because they are old, was also mentioned. Some people feel they lost, or even do not have opportunities, due to their age.

A little less explored as a dimension of diversity, social vulnerability is related to meritocracy and elitism. The “if you want you can” speech opposes the lack of financial resources some individuals have to enter the software engineering field. In Brazil, equipment and training in software engineering are usually expensive, so the entry point to the field is difficult for people in social vulnerability.

We also observed a lack of awareness about the need to talk about the subject. The survey respondents reported that diversity does not matter or should not be considered when talking about software development teams. According to the respondents, people should be treated equally, no matter their gender, and what must be considered is their ability, not their identity. It would be the ideal scenario; however, due to unconscious bias that women and girls (and other dimensions of diversity) are routinely subjected to and the different other points we already mentioned in this work, we see it as important to keep an active awareness about the subject. The unconscious bias leads to people arguing that women are “biologically” and “culturally” good and trained at certain tasks such as communication, visual design, and documentation even though they do not feel interested in them.

As the main takeaways, the survey aligned with different works on the area of the importance of diversity in general, to bring more innovation and fairness to the software engineering industry. Also, the survey highlights there are still a lot of issues and prejudice, even knowing about the benefits of having more diverse work environments. More than that, research in diversity in software engineering must consider gender beyond the idea of binary genders. Additionally, intersectionality is a must. The community has been researching aspects of diversity individually. It is important that the intersection of the multiple faces of diversity be considered, considering not only gender, race, sexual identity, etc. but also social economic characteristics that put people away from software engineering positions.

Acknowledgments

Rafael Prikladnicki is partially supported by FAPERGS and CNPq in Brazil.