2

Barriers and Supports for Learners in Computing

As computing has become increasingly prevalent in the many facets of our daily lives, efforts have focused on engaging children and youth in computing. A number of initiatives, in and out of school, seek to engage learners in a variety of different computing experiences—many with the goal of fulfilling the needs of the future workforce. As part of these expanding opportunities, there is strong emphasis on increasing the participation of learners from groups that are historically underrepresented in computing compared to their representation in the U.S. population, namely, woman, people of color (i.e., Black, Latinx, and Indigenous learners), individuals from rural communities, and those with perceived differences in ability (e.g., students with learning disabilities). Reaching representational parity is a goal that still remains despite these national efforts to increase participation of learners from underrepresented groups.

The causes of the patterns of underrepresentation in computing are complex (Charleston et al., 2014; National Academies of Sciences, Engineering, and Medicine [NASEM], 2018a; 2020; Scott, Sheridan, and Clark, 2014). A number of factors have been identified, including structural barriers such as access to courses and to technology, cultural barriers related to the norms and practices in the discipline, and stereotypes and implicit biases. Many of these barriers are rooted in racism and sexism and assumptions about who is capable of succeeding in computing.

In this chapter, we examine patterns of participation in computing and discuss the barriers to participation. While it is beyond the scope of the committee’s charge to provide a detailed analysis of the historical, cultural, and social factors at play in the patterns of underrepresentation, they are

important to understand at a broad level as they shape the learners’ perceptions of and experiences in computing. To illustrate how these barriers impact the trajectories of individual learners, we weave personal narratives from youth throughout the discussion.¹

REPRESENTATION IN AND ACCESS TO COMPUTING

Computing, along with other STEM fields, is a discipline that has long been perceived as being biased toward White men (Charleston et al., 2014). As stated above, there are a number of ongoing efforts to increase the representation of women and people of color. Despite these initiatives, the trends in representation of individuals in computing jobs throughout the past decade have remained relatively stable (see Figure 2-1). In 2019, the total U.S. workforce for all occupations included about 157.5 million workers; 47 percent are women, 12.3 percent are Black, 6.5 percent are Asian, and 17.6 percent are Hispanic. In contrast, as can be seen in Figure 2-1, in the computing workforce 25.3 percent of the workers are women, 10 percent are Black, 16.5 percent are Asian, and 8.9 percent are Hispanic. These data mirror the known patterns of overrepresentation of Asians and underrepresentation of women and Black and Hispanic people (Debusschere, 2018; Funk and Parker, 2018). It is important to acknowledge that Asian is an incredibly broad category with different histories of immigration and exclusion from U.S. society and economies. Although Asians are overrepresented in computing, they are underrepresented in the United States and are subjected to many of the same experiences described throughout this chapter.

These patterns of representation (and underrepresentation) start early. For example, as described in more detail in Chapter 6, access to computing varies by grade, rates of poverty at a school, ethnic make-up of a school, and school size (Banilower et al., 2018). Further, the patterns of who takes computing courses are comparable to the patterns observed in the profession; girls and people of color are less likely to take advanced placement computer science courses (see the section on Equity in AP Computer Science Courses). Additionally, as described in Chapter 5, there are differences in the availability of computing experiences in out-of-school settings (Afterschool Alliance, 2016).

Access to computers and the Internet at home is another factor that may contribute to patterns of underrepresentation. The term “digital divide” is used to refer to differences in access to, or use of, technologies

___________________

¹ As noted in Chapter 1, these cases are illustrative and reflect the experiences of learners who have chosen to persist in computing. These cases do not highlight the many other examples in which learners do not persist either because they drop out or feel forced out (as discussed throughout this chapter).

across individuals or schools based on race, gender, socioeconomic status, geography, education level, disability status, and first or primary language (Gorski, 2005). One common indicator of this “divide” is inequality in access to computers and the Internet (Gorski, 2002; 2005). Figure 2-2 shows the access to computers in the home for learners broken down by race and ethnicity. Figure 2-3 shows Internet use in the home. During the 7-year period from 2010 to 2017, levels of access and use have remained relatively stable for learners who are White or Asian, while access and use among learners from other racial and ethnic groups has increased. Despite these increases, gaps still remain, particularly between learners who are Native American/Alaska Native and their White and Asian counterparts.

SOCIAL AND CULTURAL BARRIERS IMPACTING PARTICIPATION

While access and opportunity likely contribute to the historical patterns of underrepresentation, social and cultural barriers rooted in racism and sexism play a significant role. Rodriguez and Lehman (2018) stress

the importance of understanding who is being privileged or marginalized in the system, and how individuals from privileged groups have access to opportunities for success and possess power over marginalized groups. Computing, as a field, has been historically composed of White men, and the values, norms, and practices of the field have been shaped by them (Björkman, 2005; Faulkner, 2001; Rodriguez and Lehman, 2018). These norms, values, and practices are reinforced and perpetuated by those in the discipline with the most power (Rasmussen and Håpnes, 1991) and

are communicated to prospective learners (Rodriguez and Lehman, 2018). Thus, in both in-school and out-of-school learning environments, learners are exposed to the cultural norms and assumptions of the field, and this shapes their perception of computing and their own relationship to the discipline (Barkey, Hovey, and Thompson, 2015; Hansen et al., 2017; Rodriguez and Lehman, 2018; Schulte and Knobelsdork, 2007; Wong, 2016; Zander et al., 2009). In some cases, the norms and values of computing may be unfamiliar to individuals who are not White men, and in some cases, they may even be at odds with the norms and values learners bring from their own homes and communities (NASEM, 2016).

When individuals experience overt or implicit racism and sexism, they may perceive an environment as oppressive and hostile (Beasley and Fischer, 2012; McGee, 2016; Naphan-Kingery et al., 2019). Moreover, they may come to feel as if they are “tokens” or representatives of their social group (Kanter, 1977), which has been associated with depression and anxiety (Jackson, Thoits, and Taylor, 1995) and declining performance due to stereotype threat (Steele and Aronson, 1995).

Sense of Belonging and Identity

Across the varied settings of everyday life, as well as in media and popular culture, young people encounter representations of STEM and computing culture and identities. This includes encounters with role models, media depictions, and instructional content that influence how learners value and identify with different activities, practices, communities, and potential occupations. Learners begin to develop mental models about vocational pathways and identities and also develop a self-concept in relation to potential occupational pathways. This influences a learner’s sense of affiliation with a professionally authentic computing community of practice (see Chapter 3 for additional discussion of communities of practice). Box 2-1 presents the case of Nathan, who first started out playing Minecraft with friends and, through his continued computing experiences, felt “competent and empowered” as he began to develop an identity as a computing professional.²

Research has suggested that one of the reasons women and people of color choose not to participate in computing is a sense that they do not belong in the field (Barker, McDowell, and Kalahar, 2009; Cohoon, 2002; Margolis and Fisher, 2002; Sax et al., 2018). Sense of belonging

___________________

² Four individuals—Nathan (Box 2-1), Raven (Box 2-2), Hermione (Box 2-3), and Antonio (Box 2-4)—and their self-reported formative experiences of authentic experiences for computing are described. Out of consideration of these individuals’ privacy, proper names have been changed. Throughout the report, where appropriate, the experiences of these individuals are referenced to illustrate various aspects of the committee’s framework (see Chapter 3) with respect to authentic experiences and their potential impact for outcomes related to computing.

refers to the extent to which individuals feel like they belong or fit in a given environment and is a fundamental need that drive learners’ behaviors (Baumeister and Leary, 1995; Sax et al., 2018; Strayhorn, 2012). As described in the previous section, learners from groups that are traditionally underrepresented in computing (based on gender, race, ethnicity, or perceived ability) experience learning environments in computing differently (Barker, McDowell, and Kalahar, 2009; Margolis and Fisher, 2002; Margolis et al., 2008; Strayhorn, 2012). For example, women may be

discouraged from participation in STEM because the typical discourse promotes the narrative that these fields are better suited to men (Blickenstaff, 2005; Haaken, 1996; Raffaelli and Ontai, 2004; Reilly, Rackley, and Awad, 2017). These messages can be conveyed by parents, teachers, and others who have this assumption about computing careers (Eccles, Jacobs, and Harold, 1990; Sadker and Sadker, 1994).

Sense of belonging is important for developing an identity as someone who can succeed in computing. Dempsey et al. (2015) defined a computing identity as the extent to which the learner sees themselves as a computing professional (Rodriguez and Lehman, 2018). There is a strong relationship between having a computing identity, intending to pursue computing as a major (Dempset et al., 2015), and feeling a sense of fit with the field (Lewis, Anderson, and Yasuhara, 2016; Lewis, Yasuhara, and Anderson, 2011). For example, in school settings, learners are less likely to be exposed to teachers of color, which is a missed opportunity to expose learners to role models and pedagogies that are informed by minoritized perspectives and experiences (see Chapter 6, Equity and Access in Computing in Schools section).

In understanding how learners experience computing and how they see themselves fitting into the field, it is important to consider the learner’s intersecting identities (Coles, 2009; Crenshaw, 1994; Donovan, 2011; Settles, 2006). That is, for example, how being a woman and being Black or Latina might together create a unique lens through which an individual experiences and perceives the world. In addition, the ways that other people perceive an individual’s intersecting identities may contribute to a learner’s experience of marginalization (Rodriguez and Lehman, 2018). Despite the importance of intersectionality for understanding learners’ experiences and the development of their identities in computing, current research predominately focuses on gender or race and ethnicity.

The Role of Stereotypes and Implicit Biases

Stereotypes create associations between occupations, fields, and identity categories such as class, race, and gender. Stereotypes suggesting science and other high-prestige occupations are the province of people who are privileged in society—upper-middle class, White, and male—are pervasive even in school settings (Allen and Eisenhart, 2017; Bang et al., 2013; Carlone and Johnson, 2007; Carlone, Scott, and Lowder, 2014; Tan et al., 2013). For example, researchers have found that stereotypes about the gendered nature of a discipline (e.g., computer science is for men) have a negative effect on whether people not of that gender (e.g., women) have a sense of belonging to that field (Cundiff et al., 2013; Martin, 2004).

Stereotypes about computer science start young and may be more prevalent than stereotypes about other STEM disciplines, such as science and

math (for additional discussion, see Chapter 6). For example, researchers found that learners as young as first grade believed boys were better than girls at robotics and programming but did not believe the same about math and science. They also found that girls with stronger stereotyped beliefs had lower interest and self-efficacy in computer science (Master et al., 2017). However, providing girls with experience in programming at this age seemed to confront these trends and reduce gender gaps by increasing learner interest and self-efficacy in computer science.

Models do not come solely in the form of adults. Eckert and McConnell-Ginet (1995) describe learners as making behavioral judgments based on what near peers are doing and what those in the next stage are doing. For example, sixth graders look toward what eighth graders do, and this may then shape the degree to which sixth graders participate in computing experiences.

At the high school level, one study found that negative stereotypes about computer scientists have a detrimental effect on girls’ interest and sense of belonging in computer science (Master, Cheryan, and Meltzoff, 2016). Further, these researchers discovered that stereotypes can be communicated by the physical environment (such as images that convey the notion of computer science as geeky on the wall of a classroom) and that physical environments that do not convey these images can have a positive effect on girls’ interest. As described above, the tendency to reiterate the lack of women in computer science activates a stereotype threat that negatively influences women’s involvement (Naphan-Kingery et al., 2019). As aspirations are impacted by stereotype threat, an effort to change recruiting verbiage and attitude is required at all levels (Shapiro and Williams, 2012).

Box 2-2 provides a case example of Raven, who first developed an interest in computing in elementary school as she used graphic editing tools while working on the school’s yearbook.

Strategies to Address Cultural Barriers

Lack of socially similar (defined as being of the same gender, race, or socioeconomic class) role models has frequently been cited as an important reason why there are few women in STEM and IT (Teague, 2002; Townsend, 1996). Exposure to alternative representations in the media, instructional materials, or through atypical role models exhibiting vocationally relevant dispositions and skills is a way of combating stereotypes (Carter-Black, 2008; O’Keeffe, 2013). Several scholars have found that exposing learners to women in computer science who can serve as role models that are similar to those learners can help them see themselves in the computer scientist role (Asgari, Dasgupta, and Stout, 2012). For example, young women exposed to computer science instructors who are women in high school were more likely to major in computer science in college (Beyer and Haller, 2006).

Box 2-3 describes the case of Hermione, who had exposure to role models who reflected her identity, in multiple settings, all of which allowed her to persist in computing while recognizing that in the field of computing she is “unique” as a Black women. Studies in engineering have also shown a positive relationship between exposure to women role models and persistence in the major (Amelink and Creamer, 2010). However, other scholars have found that role models can have a negative influence if they conform to negative stereotypes. For example, Cheryan et al. (2011) found that when role models conform to the stereotype that computer scientists are “geeks,” they did not have the expected positive influence on women’s self-efficacy.

Culturally relevant, responsive, and sustaining pedagogical approaches (described in more detail in Chapter 7) have highlighted the importance of settings that recognize the varied backgrounds and unequal resources that young learners bring to STEM experiences (Ladson-Billings, 1995; Scott and Zhang, 2014; Scott, Aist, and Hood, 2009; Scott, Sheridan, and Clark, 2015). For example, research with young women of color in STEM

programs has highlighted the importance of creating programs grounded in learners’ identities and life experiences, equipping learners from minoritized groups to be change agents who can challenge dominant practices and cultures (Ashcraft, Eger, and Scott 2017; Ashford et al., 2017; Scott and Garcia, 2016). These counter spaces are a way of making STEM and computing relevant and authentic to learners who are marginalized in the dominant cultures of STEM, and have been shown to increase their engagement and interest (Barron et al., 2014; Erete, Martin, and Pinkard, 2017; Lee et al., 2015). Box 2-4 presents the case of Antonio, who had multiple opportunities with computing in a number of different settings; through these experiences he had opportunities that allowed him to further develop his interest in programming and robotics.

ROLE OF AUTHENTIC EXPERIENCES

The committee sees authenticity as central to learners’ creation of a computing identity and authentic experiences as a vehicle for addressing the structural and cultural barriers inherent in computing. Throughout this report, the committee puts forth two senses of authenticity: personal authenticity, which describes activities in which learners find personal meaning and interest, and professional authenticity, which describes practices that are like those used by professionals in computing. Learning environments designed for professional authenticity exhibit features of problem solving, creation, experimentation, and inquiry that mirror or are directly connected to the culture, practices, and communities of computing professionals. We use the term “personally authentic” when the activity is personally or culturally meaningful in the mind of the learner.

Authentic experiences in computing come in many variations from coding an app for a service-learning project to crafting/sewing e-textiles with a caregiver to playing video games with friends. All of these might be considered authentic experiences, but they might vary in the degree to which they are personally or professionally authentic.

The committee recognizes that there might be a perceived tension between personal authenticity and professional authenticity. But this tension dissipates when personal and professional authenticity are understood as separate dimensions rather than as oppositional to each other, or as two ends of a single continuum. Instead, the committee considers personal and professional as each representing their own continuum, and an experience can be more or less personally and/or professionally authentic on each spectrum. An experience that learners find personally motivating may not be grounded in professional practice, just as an experience that a learner recognizes as reflecting professional practice can also be grounded in per-

sonal interests. But experiences can be designed to be both professionally authentic and personally authentic.

In it important to recognize, however, that professional practice in computing reflects a history and disciplinary culture that tends to exclude women and other marginalized groups such as Black, Latinx, and Indigenous learners. Therefore, an experience that is “professionally authenticity” may inadvertently exclude some learners. Although educators can work to create learning experiences in computing that foster personal authenticity by incorporating cultural elements and practices of groups traditionally underrepresented in the field, the professional world may also need to evaluate and address policies and practices that create barriers to diversity and inclusion.

SUMMARY

Many authentic learning environments in computing have strong associations with a culture of science and technology that has historically been dominated by White men. What feels authentic to those who reflect the dominant culture of computing may feel exclusionary to women, people of color, and those with differences in perceived ability. As articulated throughout this chapter, women and minoritized learners experience multiple barriers to participation in computing. Some barriers are technological and economical, while others are social and cultural, including racism, sexism, stereotypes, and implicit bias. These barriers can in turn influence computing identity and a sense of belonging to STEM and computing in particular. Providing learners with authentic experiences that attend to both professional and personal authenticity may be a mechanism for addressing some of these barriers.