Ethnic diversity is the expression of diversity one finds within a single entity (a business, a team, a community, or a country). A person’s ethnicity usually includes beliefs, nationality, and language and gives the person a distinct sense of belongingness among a group [17]. Diversity means including or acknowledging people from a wide range of backgrounds. The existence of people from various ethnic and cultural backgrounds or identities is referred to as ethnic diversity. Ethnic-racial diversity is important because people should have a commitment to evaluate their personal biases and stereotypes in order to facilitate collaboration and cooperation[1]. If someone only cares about their core beliefs and cultural identity, people will not work together to benefit the common good. There must be a common understanding and willingness to push society forward for the benefit of everyone. For this reason, diversity in IT and software development teams is a topic that has attracted the attention of researchers and organizations, who seek to understand the benefits and challenges that diversity can provide [21].

There is an increasing debate about ethnic diversity in IT and a broad body of research demonstrating the importance and effectiveness of diverse teams as being more innovative and more productive and directly impacting the business rules of companies [Chapter 21, “Bringing Diversity in Software Engineering Education from the Middle East and Africa,” Chapter 29, “Strategies for Reporting and Centering Marginalized Developer Experiences” [7]]. For example, studies on diversity in software engineering indicate that the efficient use of talent from diversity groups can play an active role and benefit organizations in terms of competitiveness, performance, economic growth, and social development [8, 10, 13, 21, 24, 26]. According to Patrick and Kumar [18] and Mujtaba [15], diversity among employees tends to make the business environment conducive to serving external customers, in addition to promoting positive publicity for the company.

In terms of benefits for companies, other indicators can be mentioned such as the use of different experiences to add a different perspective, increasing the number of solutions to be developed [23]. In addition, diverse environments tend to promote the development of more robust final products and develop solutions to more complex problems [4, 25]. It is noticed that software development companies tend to increasingly observe the importance of diversity and inclusion in software development teams. As a result, they seek to implement diversity management models to enhance the positive effects of this inclusion. However, the benefits and challenges of including diversity in IT are not only related to organizations, universities, and companies; they also permeate among the people who make up the diversity groups. Through the accounts of diversity groups, such as ethnic-racial diversity, we can observe that social issues (e.g., the way people interact with each other based on their culture), for example, have a profound impact on the inclusion process, whether in the educational environment or in the industry.

The purpose of this chapter is to share how ethnic-racial diversity has been explored in software engineering literature and related areas. To achieve this objective, we carried out a systematic mapping of the literature. According to our results, we observed that the field of research on ethnic-racial diversity in IT is rich and complex. The number of papers found demonstrates that this is a research topic of recent interest in the field of computing. The papers found are divided into reports of personal experiences lived in academia or industry, reports of experiences in educational programs aimed at teaching programming, and the management model of inclusion of ethnic-racial diversity in the industry.

We also observed that few papers are dedicated to carrying out a broader discussion on the origin of the problems and challenges that impact ethnic-racial diversity groups, and there are no definitions about ethnic-racial diversity groups or even consensus on which terms define such groups.

Ethnic-Racial Diversity

Boukreris et al. [3] observed that the idea of diversity conveys the notion of variety, difference, and opposition. According to Page [17], diversity can be understood in two groups: identity and cognitive diversity. Identity diversity is characterized by other subgroups such as gender, ethnicity, religion, sexual orientation, culture, and other social markers. Cognitive diversity differentiates individuals based on their intellectual capacity.

Debating only about diversity in its general sense can result in distancing or even erasing the various groups that help compose this diversity. Therefore, identifying which diversity groups we are referring to and discussing the aspects that characterize them is essential to obtain a more accurate understanding of the data that are raised in relation to the diversity groups in question.

Within the identity diversity groups, we can infer ethnic-racial diversity. Ethnic-racial diversity according to Gomes [9] is the way in which some intellectuals refer to the black segment (formed in Brazil by black and brown people). According to the author, the term ethnic-racial comprises the multiplicity of dimensions and issues that involve the history, culture, and life of blacks in the African diasporaFootnote 1 (like Brazil), in addition to physical characteristics and racial classification.

For Gomes [9] the ethnic-racial expression has been adopted within the theoretical and political contexts, trying to end the impasse and dichotomy between the concepts of race and ethnicity to refer to the black segment. The author also adds that, for an in-depth understanding of ethnic-racial relations, one must consider the identity processes experienced by the subjects, that is, the way in which they observe themselves, identify themselves, and talk about themselves and their ethnic-racial belonging.

In this work, we chose to use the term ethnic-racial diversity to refer to the black population, whether in African countries or in their diasporas, regardless of the form of composition of this group in each country of the diaspora, for example, in Brazil it is composed of black and brown people. We also use the term black people, when we see it necessary to give greater emphasis to the ethnic-racial diversity group we are referring to. Thus, in the next section, we will reflect on the need to debate the nuances of ethnic-racial diversity within IT.

The Challenges of Ethnic-Racial Diversity: A Necessary Conversation

In Brazilian society, there exists the “myth of racial democracy,” an idea that suggests that all white people and non-white enjoy exactly the same rights and privileges. Besides that, this does not consider the currently embattled for the social rights of the non-white population. However, there is some progress about the matters of increasing public politics to fight against racial inequality. According to Ribeiro [22] (translated by the author)

The path to change is complex, but reality is not immutable. The strengthening racial consciousness has been an important element in this construction. Ethnic-racial education is an important aspect for the realization of democracy. In the 21st century, the State must accelerate the process of inclusion, based on racial policies, as a focus of social justice.

—Matilde Ribeiro

Among the issues that are deepened by the myth of racial democracy is racism. The issues born of racism exist and are huge. They are related to all structures of society: economic, political, social, cultural, historical, and religious. However, racism does not act only in the field of structures; it creates ways of being and thinking, it is systemic, and, therefore, it determines the actions of individuals insofar as it defines and impregnates culture. Among the problems intensified by racism are the few or inexistent opportunities for blacks in certain sectors of society and the prejudices, arising from stereotypes, built on black bodies and reinforced by structural racismFootnote 2:

The stereotypes are mental plans from the cultural process of classification of pieces of information. But, this categorising is formed by the perception of social rules explicit and non-explicit, and about this point there is a correlation between the inconscient and the racism.

—Adilson José Moreira [14] (translated by the author)

The technology sector also has some prejudices about why certain bodies are more or less frequent within this space, for example, some gender and ethnic-racial prejudices try to explain that women are naturally less interested in the area of technology or that certain cultures and countries have more aptitude for the sector [5, 12]. These biases exclude from the analysis the necessary incentive for people to decide for a certain area and link the justification of the natural to the choices made [6, 11].

According to the Stack Overflow annual survey conducted in 2022, with more than 69,000 users working in the technology sector in the programming area, about 1.49% of people declared themselves to be black in the question about race and ethnicity. However, the response related to the white option represents about 40% of users, followed by the European option, about 37%. The other options appear below 10%. Given these results, it seems that the population within the programming area is mostly white, according to Stack Overflow. It is therefore observed that analyzing the low participation of black people in the technology sector through the lens of meritocracy is a mistake. Because to understand that racism is the main illness that prevents the access of black people to the spaces that they can occupy, mainly from the filter of culture, is to understand the dynamic, renewable, and conflictive character of racism.

Therefore, reflecting on these and other challenges faced by the black population is essential to devise intentional actions to face these challenges and promote abundant ethnic-racial inclusion in the area of technology, because from these reflections, we no longer understand the number of black people in technology as a matter of aptitude and choices, but as the result of a series of combined factors that resulted in few opportunities for a population that, only in Brazil, constitutes themselves like the majority of the population. In the next section, we present the methodology used for the development of this work.

Research Methodology

For the development of this systematic mapping, we used the guidelines provided by Petersen et al. [19]. The objective of this mapping is to identify how ethnic-racial diversity is being considered in software engineering research. As a result, the following research question (RQ) was defined:

  • (RQ) What studies have been published on ethnic-racial diversity in SE, and what are their challenges?

According to Petersen et al. [19], to start the systematic mapping, it is necessary to define the scope of the research, which includes the definition of research questions and identification of keywords. The keywords defined were software engineering, software development, software developer, race, ethnicity, geography, socioeconomic status, ethnic-racial, and racism. We also tested the search string using keywords such as people of color and black people, but the results did not change.

We chose to search in electronic databases that met the following source selection criteria: databases that include articles from related journals and conferences in the respective context of this research, databases with an advanced search engine that allows you to filter the results by keywords that address the search questions, and databases that provide access to full articles written in English.

Based on these criteria, the following databases were selected: ACM Digital Library (DL), IEEE Xplore, and Scopus. The search string has been adapted for each database according to the search functionality offered by it. In parallel, a list of control articles was generated, used as a way of validating and guaranteeing reliability and relevance and evaluating the research sequences.

The search string was formulated according to combinations, variations between keywords, and some characteristics of these words. The end result was as follows:

(“Software Engineer” OR “Software Developer” OR “Software Development”) AND ( “Ethnicity” OR “Racial” OR “Ethnic” OR “Socioeconomic Status” OR “Ethnic-Racial” OR “Racism”)

For this string, no time restriction was applied, since the focus was to understand when ethnic-racial diversity is a subject for research in the area of software engineering. The keyword “socioeconomic status” was selected due to the observation in previous papers that some articles, when analyzing the socioeconomic issue of people in technology, have the black population as a research group.

The inclusion and exclusion criteria were defined to optimize the filtering of the results obtained, executing the search sequence in the chosen databases, such as follows:

Inclusion criteria

  • The terms defined for the search or by similarity with the subject

  • Articles from academic journals, conferences, and workshops

  • Works written in English

Exclusion criteria

  • Keywords and abstracts that are not focused on software engineering or related areas

  • Articles that are not focused on software engineering or related areas

  • Conference proceedings, courses, standards, panels

  • Study format not in pdf or not available

  • Articles that do not deal directly with the black population, that is, that do not analyze this population and its nuances

  • Articles that only mention ethnic-racial diversity among existing diversity groups

After the initial query, we reviewed the results to determine which results should be included in the analysis based on the inclusion and exclusion criteria. A two-step process was followed to identify the set of papers in scope of the review: The abstract of each paper was reviewed to determine whether it should be considered further. This resulted in a further reduction in the number of papers to consider further. In some cases, it was unclear from the abstract as to whether the paper met the inclusion or exclusion criteria in which case it was categorized as a Maybe and taken forward to the next step. Subsequently, the full paper was briefly read to consider whether it should be considered further. In some cases, a discussion took place between us two authors to decide whether to include the paper.

Each paper was then read in its entirety and classified according to the target audience and search field. After that, we identified common themes for each paper and summarized the main results. In the next section, we present the results obtained from the implementation of this methodology.

Findings

In this section, we present the results of the systematic literature mapping. We selected 11 papers that we accepted based on the acceptance criteria defined for this work. The string used to find these articles was applied in August 2022 to the following search databases: ACM Digital Library, IEEE Xplore, and Scopus. The number of papers found through the string and selected can be visualized in Table 3-1. We analyzed the general themes of each paper, the target audience within the ethnic-racial population, whether the field of research was academia and/or industry, as well as the territories on which the papers were addressed, that is, if the reported experiences were about countries on the African continent or its diasporas. The accepted articles can be observed in Table 3-2.

Table 3-1 Papers per electronic database
Full size table

During the information extraction phase, we read the 11 papers to analyze and classify them according to the research question. We note that most of the papers are recent, and therefore it is suggested that it is a topic on the rise. We created a classification scheme to present the results according to extracted data. Such classifications were territory, target audience, and research field. Due to the number of papers found, we believe that the analysis of articles based on these indicators will help us obtain a greater understanding of how research on ethnic-racial diversity in the area of computing is being conducted.

We observed that, regarding the territory indicator, 100% of the articles dealt with cases and experiences in African diaspora countries, such as Brazil and the United States of America. Only P11 spoke of experiences in African countries beyond their diasporas. The other classifications can be observed in Tables 3-3 and 3-4. Some themes appeared in more than one article in different ways, such as the insertion of black people in the area of computing, whether in academia or industry, challenges related to stereotypes within the area of computing, opportunities and challenges to remain in the area, and racism. In the next section, we will analyze the data extracted from these articles and reflect on the challenges and opportunities they present.

Table 3-2 List of accepted articles
Full size table
Table 3-3 Classification by target audience
Full size table
Table 3-4 Classification by search field
Full size table

Analysis of the Results

According to the results presented in the previous section, we obtained three classifications of indicators: territory, target audience, and research field. However, there was an intersection of categories as we analyzed the data. Regarding education and the academic environment, the papers found (P2, P3, P7, P8, P9) observe issues related to access to higher education in computer courses, challenges related to student permanence such as dropout rates, training rate, and entry into the labor market. The paper P2 observes the entry of women and black people in the field of computing after their experiences in other fields of studies, due to not seeing themselves as computer scientists given questions of stereotypes about the area and their identities. The papers P3 highlights the violence that black academics suffer. Still on the field of education, some works, in addition to mapping the number of black people in computing, make suggestions for educational actions to increase the insertion of black people in computing or their permanence in the area. This is the case of studies P7 and P8, which suggest summer programming courses for African-American students or specifically for black men.

As for the studies that refer to the research field of industry (P1, P6, P10, P11), we observed that the issues related to insertion and permanence reported to the academic research field remain. The paper P10 looks at the nuances related to accepting contributions to OSS projects in relation to ethnic groups such as black people. The authors state that specific diversity factors related to race, personality, gender, geographic location, and other social markers can affect the acceptance or rejection of pull requests. Also according to the authors, there are differences in the acceptance rate of work pull requests between groups of different ethnicities. P11 also notes that the field of computing in recent years, through higher education and code bootcamps, has realized the promise of social mobility, which for a historically marginalized population is a breath of hope about individual achievements.

The papers P4 and P5 note that whether in academia or industry, the proportion of black people in technology is staggeringly low. P4 highlights that black men in the United States of America have a strong inclination toward technology during their youth, but this identification does not advance toward professionalization and performance in the technology market. This mismatch may occur because black students are historically considered academically inferior when compared with white students and are more likely to attend schools that are poorly equipped with resources that contribute to providing a quality education. The paper P5 highlights that, although there is an effort to include women in the field of computing, the number of black women who benefit from these policies is still low.

A common aspect found in the articles selected through the mapping is the report of the low insertion of black people in the computing area. The authors of P10 highlight the experience within the OSS community regarding submitting and correcting code for OSS projects. Among their results, they note that the number of black people developing software is 0.1% out of a total of 6.8% of developers on the GitHub platform. They found that 0.13% of contributions sent to OSS projects on GitHub were made by black people and that nontechnical factors related to code quality such as developer ethnicity influence the evaluation result, for example, raising the acceptance rate of contributions from white people between 6 and 10% and that contributions from developers in Asia and Africa are about 7 and 16% less likely to be accepted when compared with contributions from senders in North America. Also according to the authors, people of the same ethnicity are more empathetic in correcting the codes of their peers.

In certain studies (P2, P4), it is observed that the authors do not understand the insertion of black people either in the field of academia or industry, related only to factors of personal interest or better aptitude (reinforcing the idea of merit). Some works reinforce that there are social factors that influence the participation of black people in the field of computing, such as the issue of stereotypes. According to the study P4, among the many stereotypes aimed at black people are those that black men are aggressive and that black people in general are superstitious, lazy, carefree, aggressive, intellectually inferior, ostentatious, active in sports, artistic, and poor academic performers. In this sense, it is vital to combat this racism and stereotype for the broad participation of black people in various spaces of society, including computing. According to P8 people are more likely to identify with something if they perceive it to be the norm in their social group. However, the stereotype of the dominant group that makes up the field of computing is non-black people.

Another important analysis refers to the intersectionality of gender and racial issues. Works that address the racial issue focusing on black women deal with other elements related to female gender oppression combined with racism, which affect black women in all structures of society, whether by their gender or racial peers. This intersectionality that racism and gender oppression provoke not only pushes away the ideal of equal treatment but strengthens projects of domination, forming various social hierarchies. According to the authors of the article P3, even among black women with higher levels of education and employment, there is a wage gap, greater stress, and depression in relation to, for example, white women. According to the authors, “Being black in computing is an act of resistance to the dominant culture that denies our existence and refutes that racism exists even within the ‘meritocracy’ of computing.”

The use of self-ethnography to report personal experiences both in academia and industry by black people stands out (P3, P4, P5). According to the authors of papers P3, P4, and P5, self-ethnography and testimonial authority are methodological approaches little used in the area of computing, but which, however, are rich sources of reports and observations from an epistemic agency to witness lived experiences, because, according to the authors, there is no separation between academic or industrial life and systemic oppression, which is reflected in the absence of better opportunities and low wages. The area of software engineering is a fertile field of recognition of studies of this nature as legitimate research and a catalyst for social change. This methodological approach sees black people as subjects of rights and in the production of the study.

Other analyses are still possible from these results that could not be included in this chapter, and, mainly, new research should be carried out in order to deepen the discussion of this topic. In the next section, we present our final remarks.

Final Considerations

Discussing ethnic-racial diversity is a challenge that begins with the definition of terms, especially when the group is composed of black people. Understanding the constitution of the black, the factors that imply in self-identification are disciplines that are beyond the scope of this work, although without them it is impossible to demonstrate the complexity and richness of the subject. We observed, for example, that the paper P10 reports on the difficulty that developers have with self-declaration when it comes to the black segment, which makes it difficult to accurately express the amount of this population within the area. Even so, although there is this complexity in the definition of the term, according to the papers selected for this study, it was evident that the black population is smaller in quantitative terms within the area of software engineering compared with other ethnic groups, mainly that of white people.

We observed that there is an understanding that ethnic-racial diversity is important within the area of technology and that the factors that hinder the greater insertion of this group in the area of computing break the veil of meritocracy and personal aptitude, giving rise to other analyses such as stereotypes, racism, power relations in society, the myth of racial democracy, and meritocracy, which are grounded by structural racism.

Among the limitations of this work are the possible amounts of keywords that can return relevant papers on the topic. As it is a recent field of research, it is observed that the community is still learning to define and use the best terms to deal with ethnic-racial diversity within IT. Another limitation is the local theoretical framework. In Brazil there are renowned researchers who are exponents of research on ethnic-racial diversity that drive the debate in society and the creation of public policies, but who however politically choose to write their works in Portuguese. Therefore, the research that researchers in IT have taken on the task of carrying out, investigating in an increasingly profound way the origin of the problem of the insertion of black people in computing, the actions of reparative policies, and the generation of opportunities through the state and the industry that aim to mitigate this problem, combined with anti-racist actions, can play a very important role in the fight for more diversity in the area of computing. This can be seen already with a significant discussion on gender diversity, and we do believe that bringing the ethnic-racial diversity discussion to the same level is really important. And starting that was the main goal of this chapter.

Acknowledgments

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

A Dedication to the Memory of Michelle Miranda

The first version of this chapter was mostly written by Michelle Miranda, a PhD student supervised by me, the second author. Unfortunately and sadly, Michelle passed away on November 2, 2022, a month before we received the first reviews on this chapter. Michelle was an amazing person and researcher. Her PhD was exactly on the topic of this chapter. Most of what I’ve learned about diversity and inclusion in the past four years was by interacting with her, a black woman that made a difference and raised the awareness of this important topic. She has contributed significantly to this work, and for this reason and as a dedication to her memory, she remains as the first author of this chapter.