Rethinking Gender Diversity and Inclusion Initiatives for CS and SE in a University Setting

Abstract

Jocelyn Simmonds, María Cecilia Bastarrica, Nancy Hitschfeld-Kahler, Pablo Villar Mascaró, Diego S. Wistuba La Torre, Jo Muñoz Montenegro, and Millaray Cárdenas

In this chapter, we summarize lessons learned from gender-related initiatives conducted at the Computer Science Department (CSD) of the University of Chile, located in Santiago, Chile.

Introduction

The lack of gender diversity in the software engineering (SE) industry in South America is well documented [4, 7], which is usually attributed to the low number of female students that apply to undergraduate computer science (CS) and SE programs. In fact, out of the Science, Technology, Engineering, and Math (STEM) fields, CS and SE are the least gender-diverse, with representation hovering around 10–15%. Acknowledging this situation, there has been an explosion of programs and activities promoting women in computing (WiC), seeking to create spaces for women to network among themselves and increase their confidence in their own work. We have previously reported about the impact of one such initiative, an admissions program for women at the University of Chile, which has one of the most competitive science and engineering programs in Chile [1, 9]. Thanks to this program, a third of our undergraduate students are now women, up from 19% in 2014, and several local universities have adopted similar programs.

Recognizing that diversity in STEM is important, our campus now has a Diversity and Gender Office, which is in charge of making crosscutting changes to improve diversity on campus regarding gender, sexual orientation, and indigenous peoples, mainly. This office has recently been awarded the United Nations Development Programme’s Gender Equality Seal.¹ Obtaining this seal meant defining dimensions, actions, and metrics to help promote gender equality, like creating gender taskforces at a department level. Since our department (CSD) has historically had one of the lowest percentages of female students, the first two authors of this chapter were asked to pilot the creation of one of these taskforces. We invited current and former students, administrative staff, and faculty members to participate in the taskforce, given the broad nature of the issue that we were asked to tackle.

The CSD Gender Taskforce started by gathering information from different sources, trying to understand where we stood in terms of diversity. This led to the discovery of a hidden reality: our students are much more diverse regarding gender and sexual orientation than we had originally imagined. The undergraduate CS student union carried out a survey about gender identity and sexual orientation, with almost 180 responses (approximately 30% of our students). Almost 8% of the respondents identified themselves as non-binary or trans and 33% as non-heterosexual. In other words, although the department’s work on improving the participation of women in computing is crucial, it is too narrow and ignores other sources of discrimination, like gender nonconformity and sexual orientation.

Rodriguez-Perez et al. [8] carried out a systematic literature review about perceived diversity in software engineering, identifying five dimensions of diversity that have been studied by the SE community: gender, nationality, age, race, and combinations of the previous four. Gender was by far the most studied dimension, but as we had also done until this study, most of these works assumed gender as a binary construct. The only exceptions are the works published by Ford et al. [2] and Prado et al. [6], which focus on the experiences of transgender developers. Ford et al. [2] studied whether transgender developers really felt more included in the development process under remote working conditions, and Prado et al. [6] developed a set of recommendations for hackathons that want to be more inclusive of transgender people.

In order to rethink the diversity and inclusion initiatives at a department level, we carried out a qualitative study, looking to identify issues that hinder gender diversity. The contributions of this study (and chapter) are the following:

• 1. Provide evidence that, although focusing on improving female participation is great, there is a broader, more systemic problem of lack of representation in our field, and any diversity and inclusion efforts undertaken by universities must be intersectional.

• 2. Serve as a local wake-up call. Students that identify as diverse use different strategies to fit in on campus, like creating peer support groups. However, we are not addressing the basic issues that may affect their learning experiences and that can put them at a disadvantage compared with their peers.

• 3. Provide recommendations for CS and SE departments that are looking to become more diverse and inclusive, based on the experiences and issues that affect students that identify as part of the LGBTIQ+ community, as well as our previous experiences.

Establishing a Departmental Gender Taskforce

Figure 23-1 shows a timeline of the main actions the CSD has been involved in to improve the participation of women in computing in Chile and Latin America. From 2011 to 2016, we focused on bringing together women in computing, with panels and conferences,^2,3 helping set up affirmative action programs on campus,^4,5 as well as establishing outreach activities to empower girls and young women.⁶ From 2017 the focus has been on consolidating these initiatives and helping establish the Diversity and Gender Office⁷ on campus. Work on the Gender Equality Seal began in 2020,⁸ and the CSD Gender Taskforce is a part of this initiative.

Figure 23-1

Timeline of the gender-related initiatives organized by or involving CSD members

The goal of this taskforce is to recommend actions that the CSD can take to become more diverse and inclusive. Participants include current and former students, administrative staff, and faculty members, and meetings are held on a regular basis every trimester. The taskforce was asked to collect data related to gender in the CSD and then propose actions to improve on this baseline diagnostic. We collected gendered data about our undergraduate and graduate programs, teaching teams, etc. The undergraduate student union carried out the survey mentioned in the introduction, finding that 8% of the almost 180 respondents identified as non-cisgender and 33% as non-heterosexual. Since our campus has been a historically male-dominated space in a socially conservative country, we wanted to explore the issues that these students experience on campus. To do this, we organized a focus group, so as to draw upon respondents’ attitudes, feelings, beliefs, experiences, and reactions in a way where other methods are not applicable.

Study Design

All the authors of this chapter participated in the organization of the focus group. The first three authors are faculty members. Also, the second and third authors are recent department chairs and as such are highly familiarized with the departmental academic and administrative processes. They also helped design and implement the campus Gender Equality Admissions and Faculty programs that appear in Figure 23-1. The first author co-founded the WiC conferences that appear in the same timeline. The remaining authors are undergraduate and graduate students that volunteered to help carry out this study.

We defined two main topics that we wanted to explore with the participants, both with a gender perspective: (1) the learning environment on campus and (2) the use of campus spaces. The first topic was further broken down into three subtopics – classroom experience, courses, and internships – and the second topic into two – the use of classrooms and labs and the common areas, like cafeterias and restrooms. We posted a call for participation on the CSD student forum, inviting students that identified as members of the diverse CSD community to sign up. We got 11 responses, but only 6 were available on the date of the focus group.

The focus group was moderated by one of the authors, assisted by another author, a CSD student that identifies as a part of the diverse community, who observed, took notes, and asked clarifying questions. The focus group was carried out in person and lasted two hours, where participants first signed an informed consent, and we then asked them for permission to record (only audio). We began the focus group by asking participants to introduce themselves. They wrote their preferred pronouns on their place cards and told us how long they have been students at the CSD, telling us what they liked about computing and whether they have been teaching assistants (TAs) in CSD courses. We then moved over to the focus group topics, asking them to talk about their experiences as gender- and/or sexually diverse students on campus.

The focus group transcript and observation notes were coded by one of the authors, following general open coding guidelines. The rest of the authors revised the coding process and the later translation (as the focus group was carried out in Spanish). The participants were divided evenly by the number of years spent at the CSD (two participants in their first year, two in their second year, and two in their third year). Two of the participants identify as non-binary, and the rest identify themselves as women. None of them have been TAs in CSD courses.

Findings

We now report the main findings of our focus group, grouped by main topic. Positive experiences and perceptions are labeled with a plus sign and negative ones with a minus sign.

Regarding the Learning Environment

LE1 – Faculty attitudes and behaviors: The participants indicated that they have a good relationship with the CSD professors. Their experience with CSD courses is starkly different from the initial math and physics courses, where they felt that professors want to show off how much they know. CSD professors answer student questions outside the classroom and try to build a trusting environment in the classroom. For example, they ask students for their preferred pronouns and use inclusive language (+). Professors correct themselves if they make mistakes, so the participants perceive an effort to improve (+). They did not feel that teaching assistants (TAs) made the same effort (−). Member checking revealed various possible reasons, ranging from ignorance/disinterest to a lack of empowerment on behalf of the TAs.

LE2 – Peer attitudes in the classroom: CSD has almost doubled in enrollment in a short time, but participants perceive that it is a horizontally interrelated community. The gender-diverse subcommunity is organized, and they expressed strong feelings of friendship and camaraderie, even if they do not know every member of the subcommunity personally (+). However, they have identified a group of cisgender heterosexual (cishet) male students that prefer to avoid social interactions with the CSD community in general, since they seem to be uncomfortable with or not interested in diversity (−). For example, some of these students have made fun of people for using inclusive language (−).

LE3 – General perspectives about CSD courses: Unlike math and physics courses, CSD courses focus more on homework and teamwork (+). Support material from previous semesters is available (+), but not always of the best quality (−). Some professors could make better use of the Learning Management System (LMS) used on campus (−), and professors should avoid making unilateral changes to course evaluations, like adding or removing an assignment (−).

LE4 – Gender-diverse perspective in CSD courses: Assignments and midterms are written in a neutral manner (+), using animals or groups of people when characters are needed for illustration (+). Professors avoid using only traditionally male names in examples (+). However, courses do not usually include content created by members of any historically marginalized groups, not only gender, nor are their contributions highlighted (−). For example, one participant indicated that they knew more about WiC from volunteering at a coding camp for girls than from the CSD.

LE5 – Perceptions about their own learning abilities: The participants indicated that, as part of the diverse community at CSD, they felt like they cannot fail, so they must make a larger effort than their cishet male peers to show that they are just as good as students (−). Since most of the CSD students are cishet men, the participants also felt the constant need to compare themselves or compete with the cishet men students (−). They feel an implicit pressure in being a woman or part of the diverse community (−).

LE6 – General perspectives about internships: The COVID-19 pandemic delayed the internship process (−). The recent changes to procedures for reporting internships to receive credit are also confusing/unclear (−). They feel like the CSD provides little guidance on how/where to look for internships, how to prepare a CV, etc. (−). They know professors can help them with the internship search process (+), but shyness prevents them from contacting them (−).

LE7 – Gender and sexual diversity at internships: The transition from the CSD to the work environment is difficult (−). They have to move from a diverse community to an environment where they are usually the only person that identifies as gender- and/or sexually diverse (−). If there are more women at the company, they are usually not engineers (−). Some participants mentioned that they perceive that they are treated differently than the rest of the people at the company (−) and that they feel that they have to show they know more than a cishet male intern (−). One participant was more empowered and asked interviewers about their company’s diversity policies and statistics (+). Interviewers usually show little understanding of gender identities and sexual orientation (−). Another participant mentioned a case where the technical interview turned into a conversation about the gender transitioning process (−).

Regarding the Use of Campus Spaces

CS1 – General perspectives about common spaces: The participants shared various reasons why they were sometimes uncomfortable in common spaces. They feel intellectually insecure and want to avoid being “wrong” in public, like making coding mistakes (−). Since they feel watched when working on a computer, they prefer to use computers that face the wall (−). They are also afraid or embarrassed to use whiteboards to work out problems (−). They also get unsolicited advice or help when coding or solving problems (−). They find it hard to shut down these unsolicited offers; some of those that offer help are oblivious to the discomfort caused by these interactions, while others assume that their help is needed (e.g., “mansplaining”) (−). Some of the older participants indicated they gained confidence over time (+) and that they now feel that they can ask for help or even offer help themselves (+). The general consensus was that cishet men do not seem to be afraid of making mistakes in public, so they like to flaunt what they know in common spaces like study halls and labs (−). These cishet students are also comfortable explaining course topics to their peers and frequently speak over women and students that identify as diverse (−). Women and other students that have a hard time with a course prefer to study in the library, since it has cubicles where they can work in a more sheltered manner (−). The participants that identify as women reiterated that they felt watched in the computer labs (−).

CS2 – About specific spaces in the CSD: Participants feel like part of their insecurities are ingrained (−), but that the vibe at the CSD helps reduce their insecurities (+). Their experience during the first two common years of math and physics courses was very different, where they overheard some hallway conversations against diversity (−). At the CSD, the participants indicate that it is not clear which spaces can be used by students (−), since in-person classes only restarted in March 2022, after two years of virtual classes. Students use mainly two rooms at the CSD: (1) an open layout computer lab, which has computers along the walls, with work tables in the middle, and (2) a smaller room with lounges, a TV, and video game consoles. Both spaces are sometimes filled with cishet men, and then some people that identify as diverse prefer to not use these spaces (−). However, when they see someone who identifies as diverse using these spaces, they gain confidence and start using these spaces more (+), getting to know more people that identify as diverse in the process (+).

CS3 – Restrooms: Restrooms are a complex issue. When the original campus buildings were built (1911–1922), there were only a handful of women students on campus, so no restrooms were built for women. This situation has improved over the years, and we now have similar amount of men’s and women’s restrooms. However, today we have trans and non-binary students who do not feel safe using these binary restrooms (−). One of the women’s restrooms was first made accessible and has been recently made gender neutral (+). However, as it is in one of the oldest buildings on campus, it usually has problems with plumbing, and it is in a highly transited area, which severely limits its use (−). The non-binary participants indicated that they feel safer using women’s restrooms (+), but that they worry about offending other students when using them (−). One of the non-binary participants was stopped from using a restroom by a security guard, since their gender expression did not “match” the restroom (−). Locker rooms are also binary (−). As a result, non-binary and trans students avoid using the restrooms on campus (−). Students on another campus proposed removing the urinals from the men’s restrooms, in order to make them more gender neutral. This proposal was met with ridicule by male cishet students on our campus (−). There is no access to hygiene products for menstruating people on campus (−), while there are free condoms at the campus health center. Finally, the CSD restrooms are also binary (−).

CS4 – Virtual spaces: This topic was not planned and emerged spontaneously at the end of the focus group. On campus, we use a local LMS that has a campus-wide discussion forum, among other communication tools. The participants find that the campus forum is useful (+), but that it can also be a toxic space, where trolling, harassment, and other negative behaviors are enabled and/or tolerated (−). Cishet men dominate conversations, without necessarily making positive contributions (−). Some people use the forum to start controversies (−), and there are no moderation mechanisms in place (−). Users are only identified by name (profile photos are optional), so students with common Spanish names are anonymous in practice (−). As a result, the participants indicate that posting their opinions on this forum makes them anxious (−). They do not want to draw attention to themselves, since it may lead to harassment on their social network accounts (−), which has happened in the past.

Recommendations

Based on these findings, we propose six recommendations for CS/SE departments working on becoming more diverse and inclusive. The goal is to create safe and welcoming learning environments that promote inclusiveness, since software engineers that study in these spaces need to be aware that their potential users and future teammates are far more diverse than the demographics of their current programs. We are in the process of implementing these recommendations, together with those already identified by the CSD Gender Taskforce, like diversifying teaching teams.

R1 – Carry out awareness campaigns within the CSD: Motivated by findings LE1, LE2, and CS2. Concepts like gender identity and sexual orientation are under constant revision, so recurring campaigns about these topics for professors, teaching assistants, and students should be the norm. Spanish is a gendered language that defaults to the masculine gender, so these campaigns should also focus on the use of inclusive language. Also, since the community is constantly changing, the importance of these topics should be communicated clearly by department authorities. The community should also be familiar with protocols defined at a central level, like protocols for handling harassment cases. A study by Garvey et al. [3] found that both faculty members and administrative staff play an important role in enhancing campus climate, for example, by implementing plans to deal with incidents of harassment and/or discrimination. Students that perceived a warmer campus climate rated their academic success as higher.

R2 – Rethink CSD courses with a more inclusive lens: Motivated by findings LE1, LE3, and LE4. Contributions from historically marginalized groups are not usually highlighted in CSD courses, and professors should make an effort to include contributions from these groups in their course materials. Professors should also avoid making value judgments about course evaluations and grade averages, since certain groups of students, such as women and/or those that identify as diverse, can take these comments very personally, affecting their self-confidence and sense of belonging in CS/SE. Similar to R1, this recommendation also focuses on creating a warmer climate, now specifically in the CSD.

R3 – Foster student self-confidence and job readiness: Motivated by findings LE5, LE6, and LE7. All of the participants mentioned feelings of insecurity, anxiety, shyness, as well as a lack of self-confidence in their knowledge of course materials. This affects how they use common spaces, how they participate in group and course discussions, etc. As such, the CSD should develop workshops to help students manage these feelings and improve their self-confidence. We should also focus on job readiness, since the transition from the CSD, where diverse students have a support group, to industry, where they are usually the only person that identifies as diverse, can be tough. This involves helping them identify internship opportunities, put together a CV, etc. This is consistent with existing work, like a study carried out by Wang et al. [10] that found that female developers on GitHub have a competence-confidence gap regarding their participation in new projects. They also found that female developers first build up a reputation as helpers on projects, before attempting to contribute to the more technical parts of a project.

R4 – Revise the use of campus spaces: Motivated by findings CS1, CS2, CS3, and CS4. After two years of online courses, students are just starting to rediscover the campus. As such, we should work on clarifying and promoting the use of common spaces. This includes defining codes of conduct, helping students identify desirable and undesirable personal interactions in these spaces, both physical and virtual. We also recommend reviewing the layout of labs and classrooms, so that there are spaces where students who feel the need to work unobserved can do so. Finally, students that identify as diverse should feel safe going to the restroom on campus, so more neutral restrooms should be made available. Hygiene products for people that menstruate should also be available. The decline of “lab culture” was already an issue before the COVID-19 pandemic, as students started to use their own laptops instead of shared workstations on campus [5]. Students worked from home during the worst of the pandemic, and it is now time to focus on rebuilding supportive and collaborative learning environments, both in person and virtually.

R5 – Work with campus authorities on general issues: Motivated by findings LE1, LE2, CS1, CS3, and CS4. Several issues raised by the focus group participants should be handled by campus authorities, specifically a Gender and Diversity Office, which should tackle these issues with a gender-diverse perspective. We need to raise awareness that binary restrooms and changing rooms have a high impact on non-binary and trans students. We are also concerned about student conversations against diversity and women in common spaces, as well as professor attitudes during the first two years of common courses. These are all issues that must be strategically handled by campus authorities. In line with the recommendations made by Garvey et al. [3], best practices to improve the campus climate should be institutionalized, seeking to build a campus climate that is proactive and social justice oriented.

R6 – Work toward creating inclusive workplaces in industry: Motivated by findings LE6 and LE7. Students raised several concerns relating to internships. Although the CSD cannot directly intervene these spaces, we can raise awareness about measures that companies can take to create a more diverse and inclusive work environment. This requires a stronger commitment than just changing hiring practices; it also requires defining or examining codes of conduct, protocols for handling harassment in the workplace, etc. Also, students cannot be responsible for educating company employees on topics like gender identity and sexual orientation; they are there to improve their own technical and professional skills. The negative experiences about internships that our focus group participants described are similar to some of the challenges that have been found to push women away from the tech industry (see Chapter 4, “Breaking the Glass Floor for Women in Tech”). As such, our recommendation is aligned with the strategies that the authors of that chapter have identified to address these challenges.

Conclusions

Gender- and/or sexually diverse students face a myriad of challenges. In the case of our students, they are also studying engineering at a highly competitive science and engineering program in Latin America. Chile is a socially conservative Latin American country, and Spanish is a gendered language, where the default gender is masculine. This means that challenges that affect them need to be addressed in an intersectional manner, meaning that we have to be aware of the different types of diversity that are present in our community when designing and undertaking new diversity, equity, and inclusion policies on campus. The recommendations made in this chapter are in the process of being adopted within our department. For example, we recently created a gender-neutral restroom. We hope these recommendations may also serve as guidance for other CS and SE departments motivated to address this topic.