Gender and STEM‎ > ‎

Nerds, Geeks & Barbies: A Social Systems Perspective on the Impact of Stereotypes in Computer Science Education

The Domination and Partnership Social Systems Continuum

In The Chalice and the Blade: Our History, Our Future, Riane Eisler (1987) examines a “cataclysmic turning point during the prehistory of Western civilization, when the direction of our cultural evolution” shifted from thousands of years of peaceful coexistence to a more violent social system (p. xvii). Eisler’s systems-level analysis documents thousands of years of peaceful and prosperous civilization—a partnership social system—featuring an egalitarian social structure that emphasized linking and was founded on care and respect. Several thousand years ago, we shifted to a predominantly hierarchical, authoritarian social structure—a domination social system—that emphasizes rigid rankings and is founded on power and fear. The following table lists some core characteristics of domination and partnership social systems (Eisler, 1987, 2000, 2002, 2007).

Core Characteristics of Domination and Partnership Social Systems



Power- and control-based

Care- and respect-based

Emphasis on ranking

Emphasis on linking

Win/lose orientation

Win/win orientation

High degree of fear, abuse, violence

Low degree of fear, abuse, violence, (not required to maintain rankings)

Value so-called “male” traits (such as control and conquest) over so-called “female” traits

Value traits that promote human development such as nonviolence and caregiving

Whether we create a social system that is more oriented to domination than partnership influences which human behaviors are fostered or inhibited. A partnership system supports our capacities for caring and connectedness while a domination system fosters our tendencies towards power and conquest (Eisler, 2007, p. 95). To co-create inclusive computer science classrooms, it is critical that more of us understand the attitudes and beliefs embedded in domination computer science education, and how effectively this shapes who participates, who succeeds, and teacher perceptions and expectations of students. We can all contribute to the shift away from domination and back towards partnership simply by listening to the inner voice that honors our essential, empathic, caring nature as human beings.

Stereotypes: A Tool of Domination Systems

Stereotypes are a powerful tool for laying the foundation of domination systems. Stereotypes circumscribe the boundaries around where we “belong” and what is “possible” for us in our lives. We learn both about how to view each other (which teaches us to “discriminate” and rank by category), how to view ourselves (which teaches us to internalize views of being “less than” in relation to gender, race, class, and other systems of ranking), and how to organize our society (which teaches us who belongs where). These representations have a powerful impact on the possibilities that people perceive for themselves and influence the daily behaviors through which they manifest these possibilities. Therefore, an in-depth understanding of stereotypes and their influence is critical to beginning to understand how we all continue to participate in recreating domination computer science education.

The Gender Boxes

Although stereotypes of race, gender, class and other social categories have similar effects, here I focus on gender. We learn the boundaries of the gender boxes from a variety of sources—family, culture and community, and our social institutions. Although family and culture are the most intimate teachers, what we learn from social institutions about gender is sometimes more significant because it is so pervasive. As I will demonstrate later, this has profoundly negative impacts on computer science education.
The gender boxes to which common human traits have been assigned in the U.S. are:
  • Maleness is associated with science, hard, strong, rational, active, assertive, competitive, task-oriented, and the primary measure of self-worth is based on accomplishments
  • Femaleness is associated with nature, soft, weak, emotional, passive, submissive, cooperative, relationship-oriented, and the primary measure of self-worth is based on appearance.
Certainly, neither all women nor all men necessarily view themselves these ways, but these are the gender boxes that are persistently purveyed by larger social institutions.
Based on our sex, we are expected to demonstrate the traits of the “appropriate” gender box. To verify the boundaries of the gender boxes, just observe when women or men are censured for being “unfeminine” or “unmasculine.” For example, women may be criticized for being too “aggressive” because “assertiveness” belongs in the male box, or men may be criticized for being too “emotional” because that belongs in the female box. Further, when men are chastised for being emotional, it’s usually done by derogatory references to the female such as “sissy” or “girly.”As active participants in this domination culture, both men and women learn to reassert the boundaries of the gender boxes for both men and women.
The first fallacy with the gender boxes is that these aren’t gendered traits, they’re human ones. Few of us completely fit into the gender box of either set of so-called “male” or “female” characteristics. Unfortunately, most of us learn that we are supposed to (degrading our confidence in our identity), and most of us learn to reinforce the boundaries of the gender box for others (leading to a kind of perpetual social abuse). This hampers women and men from fully manifesting their authentic selves.
The second problem is that we don’t just rigidly define one gender as separate from the other; we also consistently rank one set of gendered characteristics over the other—reflecting core characteristics of domination systems. Our social institutions—including education—are largely organized around “maleness” and so-called “male” characteristics as the standard of behavior. This means that the human characteristics of the “male” gender box are the traits that are valued in our social institutions. This has very specific repercussions with regard to computer science education.

Stereotypes in Computing: A Tale of Three Barbies

Stereotypes also project specific limitations in terms of women’s perceived access to, interest in, and capabilities in computing, as developers, users and beneficiaries of technology. Many of these stereotypes are perpetuated in mass media that focus on the computing industry, and they influence perceptions of who belongs. “Media images more frequently depict computer programmers and developers as males, and women as users. For example, in advertisements of technology products, women are often presented as passive and inexpert users . . . Men . . . are characterized as deep thinkers concerned with the future” (Barker & Aspray, 2006, p. 38).
One doesn’t have to look far to observe how this influences actual behavior—girls are major texters and social networkers while women remain underrepresented as developers of computing. Barbie, a popular toy in the U.S. for over four decades, serves as an interesting site upon which to explore gender stereotypes in relation to women and computing since she is such a cultural icon. Here is a tale of three Barbies.
In 1992, Mattel introduced Teen Talk Barbie and one of her many phrases was “Math class is tough!” The American Association of University Women (AAUW) led what became a large public outcry over the negative message that this sent to girls. Within three months of the doll’s release, Mattel announced that they would reprogram the computer chip that stored Barbie’s 270 phrases to only include 269—removing the offending phrase about math. Although it’s great that the AAUW was able to organize for positive change, it’s not so great that no one at Mattel thought about this before they made the decision to include such a phrase. How could this happen? The answer is: gender stereotypes about women in math.
In 1999, Mattel licensed their Barbie and Hot Wheels logos to Toronto-based Patriot Computer to develop Barbie and Hot Wheels PCs. The Barbie PC was a silver box with the classic Barbie-pink plastered all over in the form of giant daisies. The Hot Wheels PC was royal blue with bright yellow and orange flames. The problem was not just their “gendered” appearance; the problem was their “gendered” software. “Among the software titles offered with the Hot Wheels PC but not the Barbie PC were BodyWorks, a program that teaches human anatomy and three-dimensional visualization, and a thinking game called Logical Journey of the Zoombinis” (Headlam, 2000, p. 1). The Barbie PC featured more design software than educational software—disadvantaging girls in terms of developing the knowledge that might lead them to study computing.
However, a story from a student in one of my courses highlights how this gendered approach also harms boys. My student was in Toys R Us with her seven-year old son, a budding artist, who really wanted the Barbie PC since it had more design software. Although my student was willing to buy it for him, her sobbing son was already gender socialized enough to understand that it was meant for girls, and he refused to accept it. Mercifully, Patriot Computer filed for bankruptcy in 2000, and Mattel quit making these computers (Kanellos, 2000).
In 2010, Mattel released Computer Engineer Barbie. She has a Ph.D., wears a Bluetooth ear piece, and carries a pink laptop. So, let’s see, I can be a computer engineer as long as I maintain my “girly” pink appearance? Hmm. Need I say more? I’m not sure this is real progress, but I guess it is a step in the right direction.

The problem remains that if we are ever to break down stereotypes, we need images of a wide variety of people who actually look like us engaging in many occupations. Without them, we’ll remain stuck in a domination system where we continue to limit ourselves and each other based on learned stereotypes. A recent story in The New York Times demonstrates my point. It began with a listing of Candace Fleming’s resume (including a double major in industrial engineering and English from Stanford, an M.B.A. from Harvard, a management position at Hewlett-Packard, and experience as president of a small software company). However, she had two explicitly sexist encounters while raising money for a start-up company she co-founded—one venture capitalist suggested her business cards should just say  “Mom” and another “potential backer invited her for a weekend yachting excursion by showing her a picture of himself on the boat — without clothes” (Miller, 2010).

Don’t get me wrong, as much as I’d like to place all of the blame on Barbie, stories like this are not all her fault. Stereotypes are pervasive and ubiquitous. One doesn’t have to look far to find other examples of gender stereotypes in computing. For example, popular television programs like The IT Crowd (in the U.K.) and The Big Bang Theory (in the U.S.) feature nerdy (mostly white) geeky men and attractive (mostly blond) but technologically inept women.
These influences add up, and they are all the more powerful because they seem so innocuous. Hey, it’s just a toy. Hey, it’s just a joke. Hey, it’s just a painfully persistent reminder of who belongs in technology and who does not.

Geek, Hacker Stereotypes and Academic Performance

Although I’ve only offered a few examples of stereotypes in relation to technology, the fact is that their persistent and pervasive presence has dangerous consequences. It teaches us to limit ourselves and each other; but, perhaps the deadliest cost is the ways in which stereotypes teach us to internalize negative messages about our capabilities. This has a seriously detrimental impact on computer science education.
Nearly 50 years of data from the Draw-a-Scientist-Test (DAST) shows the remarkable persistence of stereotypes about who belongs in science and technology. Researchers have now tested many populations including elementary students, college students, and teachers of math and science in the U.S. and internationally with woefully consistent results (Fung, 2002; Rubin, 2003; Thomas, 2006). DAST participants have repeatedly imaged “a scientist as a middle-aged or older man wearing glasses and a white coat and working alone in a lab” (Sadker & Sadker, 1995, p. 123).
These types of stereotypes begin to influence attitudes about who belongs in science and technology at very early ages and their costs are exponential over time. One group of seventh graders drew scientists before and after a visit to Fermilab (Bardeen, 2000). Their images demonstrate both the persistence of stereotypes and the power of seeing diverse people doing working in technology. One student (Beth’s) “before” image in is a classic representation of the DAST stereotype.
Joanna Goode, Rachel Estrella, and Jane Margolis conducted interviews about participation in computer science with over two hundred high school students and teachers in the racially diverse Los Angeles Unified School District (Goode, 2006). Their research corroborated the power and source of stereotypes; for many students “their images of who works in computer science comes largely from popular culture” (Goode, 2006, p. 99). Students mentioned that in media ranging from magazines and books to television and film, the most persistent stereotype is that of an anti-social, lone programmer, staring at a computer screen 24 hours a day, 7 days a week—the computer geek. It isn’t hard to understand how girls who’ve been gender-socialized to be relational would find this image, and the lifestyle it suggests, more unappealing than boys. Of course, like all stereotypes, this uni-dimensional depiction of a lone computer geek is inaccurate since most computer science professionals must work in teams to develop ideas and products.
Dale Spender corroborates that what most girls turn away from isn’t the technology; what “they turn away from is the image of the scientist or the computer hacker” (Spender, 1995, p. 173). In comparison to boys, this leads to many girls being underprepared in math and science by the time they’re ready for college. Multiple scholars have documented the predictable self-esteem slide that occurs for most girls during adolescence as they begin to feel increasing social pressure to be “feminine” (Brumberg, 1997; Pipher, 1994; Sadker & Sadker, 1995). Since girls shy away from the image of “scientist” as “unfeminine” in those pivotal adolescent years, this leads them to take fewer advanced math and science courses in junior high and high school (Goode, 2006; Sadker & Sadker, 1995). In addition, “girls are significantly underrepresented in after-school computer clubs, as computer participants, at free-access times using the computers, and in advanced computer electives” (Rosser, 1995, p. 147). This leads even fewer girls to make successful transitions from high school to college in terms of being either users or developers of technology.
Zarrett and Malanchuk conducted a longitudinal study of 1,482 adolescents, of whom 61% were African American and 35% were European American, over a nine-year period (1991–2000) to examine socio-psychological factors that influence computer-related occupational choices inclusive of race, gender, and socio-economic class (Zarrett, etal, 2006). Their findings showed that individuals’ choices to pursue technology careers are related to their “perceived ability or mastery of the field,” their experiences with the subject (classes in math and computer programming), and “cultural norms and stereotypes” (Wilson, 2003, p. 75-76). Black males and females, and White females shared one key predictor of interest—self-concept. For White males, what mattered most “was others’ encouragement and, importantly, valuing math at an early age” (Zarrett, etal, 2006, p. 75). Stereotypes are likely to negatively influence the self-concept of women and people of color, especially in relation to technology. Margolis found similar results in their 2008 study of three Los Angeles public high schools. Nia, an African American 11th grader, expresses a common perspective: “I think minorities are . . . scared, you know, to jump into the future because what it looks like is only Caucasians should be in that industry” (Margolis, 2008, p.3).
Fear of being perceived negatively according to stereotypes can also diminish academic performance—social psychologists call this “stereotype threat.” Stereotype threat is likely to operate regardless of the individual’s level of confidence, because it has to do with the effect of an individual’s expectation of how they are perceived by others more than their own sense of identity. Studies on race and gender stereotypes in relation to test performance have documented the power of stereotype threat (Cooper & Weaver, 2003). Steele & Aronson gave a 30-question verbal test (designed to be similar to the verbal portion of the SAT) to Black and White Stanford undergraduate students who were told that their test was diagnostic (measuring their verbal “strengths and weaknesses”) versus non-diagnostic (no mention of measuring ability) (Steele & Aronson, 1995). Since one stereotype that African Americans face is that of “poor academic performance,” the researchers supposed that students who were told that their ability would be measured would perform more poorly. “Blacks in the diagnostic condition performed significantly worse than Blacks in the non-diagnostic condition,” and than Whites in either the diagnostic or non-diagnostic condition (Steele & Aronson, 1995, p. 8). Blacks in the diagnostic condition also had lower accuracy (number correct over the number attempted) and completed fewer items than Whites (Steele & Aronson, 1995, p. 8).

Research on gender stereotypes confirms a similar pattern. The gender stereotype that women face is that they are “not as competent as men at technology, science or math” (Steele & Aronson, 1995, p. 96). Figure 3 shows the results when Spencer et al. (1999) gave a computerized math test to men and women; one group was told that the tests had reliably proved gender differences in the past. No mention of gender was made to the control group. Men and women who were not reminded of gender prior to the test performed at the same level. Women who were reminded of gender prior to the test solved 1/4th as many problems correctly compared to men; while men who were reminded of gender prior to the test solved even more problems correctly than the men in the control group. Women performed down and men performed up according to gendered expectations with regard to their math abilities.

Clearly, stereotypes matter in significant ways. They influence how individuals view themselves, how individuals expect others to view them, and how we view each other. Although the degree of impact that stereotypes have on an individual’s life will be ameliorated by the messages that they receive about their identity from their family and culture of origin, the messages they receive from domination social institutions remain a powerful factor. It’s not hard to understand how the weight of negative stereotypes may take a toll on many women and other marginalized groups with regard to their success in computer science education. This underlines the damage caused by the domination computer science education I describe in the next section, as well as the importance of the partnership computer science education strategies I recommend later.

Domination Computer Science Education

Domination systems feature these core characteristics: 1) power-, and control-based; 2) emphasis on ranking; 3) win/lose orientation; 4) high degree of fear, abuse, violence; and 5) value so-called “male” traits (such as control and conquest) over so-called “female” traits (Eisler, 1987, 2000, 2002, 2007). Like other social institutions in a domination system, education reflects these usually unnamed core characteristics. Since most of us have learned to think that this is “just the way things are,” these domination characteristics implicitly influence the content, process, and structure of educational institutions (Eisler, 2007). This section explores a few examples of how domination characteristics are reflected in the content, process, and structure of computer science education.
Few computer science curricula honor the ways in which students may have very different preparation for success due to systemic social barriers that are beyond individual control. A rigid curriculum that draws a line in the sand and offers neither flexible points of entry nor courses that help students fill gaps in knowledge is a classic symptom of domination education. Another curricular concern is that books (and other learning resources) rarely highlight the contributions of women and people of color in computing and technology, thus keeping the white geeky nerd stereotype unchallenged.
The competitive, weed-out, you’re-probably-not-smart-enough-to-succeed environment in many programs also reflects domination characteristics of fear and control, ranking, win/lose, and conquest. This is often exemplified by teaching styles that are based on power, fear, and competition—the throw-them-into-the-deep-end-of-the-pool style. Students who don’t drown get to stay, and in a competitive climate, most students focus only on saving themselves leaving even more of their peers likely to drown.
Barker and Garvin-Doxas’ (2004) ethnographic study of computer science classrooms documented the negative impacts of domination learning environments where patterns of communication breed an impersonal environment, guarded behavior, and the creation of hierarchies resulting in competitive behavior. The authors also showed how alternative communication choices that foster collaborative supportive learning environments could lead to a better climate for learning.
Those who have internalized negative stereotypes about their capabilities or about who belongs in the world of computer science face even more barriers to success in this climate. Indeed, research on why students leave computer science has documented this fact. One recent survey of randomly chosen students (275 women) at a large public university reported that about one-third of the women expected not to experience a welcoming atmosphere if they pursued computing careers, while none of the men expressed this concern. Further, nearly 20 percent of the women believed they would not fit in with coworkers in a computing career, and more than 80 percent felt that they would not enjoy such a career (Bartol & Aspray, 2007).

Other studies name so-called “masculine” values and practices as an important contributor to the lack of women in computing education and employment. Further, many women reject computing in the first place because they consider it a “masculine” domain (Bartol & Aspray, 2007; Faulkner, 2000; Wilson, 2003). Johnson explains that in a domination social system most high-status occupations “are organized around qualities culturally associated with masculinity, such as aggression, competitiveness, emotional detachment (except for anger), and control” (Johnson, 2006, p. 98-99). This is further evidence of the oppressive weight of the geek, nerd, and Barbie stereotypes.

Indeed, male-identified language and dominance metaphors are pervasive in science and technology. In Secrets of Life, Secrets of Death: Essays on Language, Gender and Science, Evelyn Fox Keller (1992) describes the significance of metaphors in terms of how they influence our perceptions as well as the questions we ask: “Different metaphors of mind, nature, and the relation between them, reflect different psychological stances of observer to observed; these, in turn, give rise to different cognitive perspectives—to different aims, questions, and even to different methodological and explanatory preferences” (p. 31).
Domination characteristics of power, fear, and violence are embodied in the metaphors of computing. Computer jargon—the language of the daily discourse—is one example of this, featuring such terms as: boot, crash, abort, kill, hacking, blue screen of death, brute force, killer app, and number crunching (Cohoon & Aspray, 2006, p. 145; Spender, 1995, p. 200). Although violence is a predictable element of domination systems, the particular prevalence of violence against women results “in patterns of chronic fear and avoidance as women and girls learn to circumscribe their lives in order to reduce the odds of being singled out for harassment or attack” (Johnson, 2006, p. 58). In that broader social context, it should not be surprising that a computing culture that fosters violence metaphors is a less hospitable one for most women and some men.
We created domination education systems and we can change them, but not until more of us recognize the significance of their damaging characteristics. From this new perspective, we can begin to co-create partnership education with our students.

Partnership Computer Science Education

Partnership systems feature these core characteristics:
  • care-, and respect-based;
  • emphasis on linking;
  • win/win orientation;
  • low degree of fear, abuse, violence; and
  • value traits that promote human development such as nonviolence, empathy, and caregiving (Eisler, 1987, 2000, 2002, 2007).
This section shares the following strategies for co-creating partnership computer science education:
  • partnership ways of knowing;
  • partnership teacher-student relationships;
  • co-creating collaborative learning; and
  • partnership evaluation measures.
Many of these strategies have been shared by scholars who are exploring ways to retain more women and students of color in computer science education. However, I hope that as you consider them with a better understanding of the enduring damage of stereotypes and through the broader lens of partnership characteristics, you’ll begin to see why they are such effective strategies for change. Therefore, these ideas are not meant to be prescriptive, but to offer a few sign posts on the new road to partnership that I hope we can build together.

Partnership Ways of Knowing

We all have different ways of learning that may be influenced by a variety of factors, but in domination education we have emphasized one learning style to the near complete exclusion of others. Traditional domination education emphasizes a kind of “abstract theoretical knowing, divorced from the real world” (Bucciarelli, 2004, p. 136-137). The “separate knowing” characteristic of traditional disciplinary ways of thinking features a concerted effort to be “objective” by separating and suppressing the self, “taking as impersonal a stance as possible toward the object” under investigation (Belenky, et. al, 1986, p. 109). This is the type of “decontextualized, either-or thinking” that predominates in scientific and technical education today (Bucciarelli, 2004, p. 140). Not only does this privilege certain types of learners and learning styles over others, it often means that “the emotional and moral aspects of a problem are typically disregarded as irrelevant and are thought to get in the way of an adequate solution” (Bucciarelli, 2004, p. 137). We cannot afford to ignore the moral aspects of technology development, most especially as its global impacts are rapidly growing.
The answer may lie in “connected knowing,” which was first described by Belenky (1986). As the phrase suggests, “connected knowing” focuses on using connection via empathy and care to understand the object of study more deeply and is “rooted in everyday experience, intuitions, and feelings” (Belenky,, 1986, p. 112; Bucciarelli, 2004, p. 141). It offers a partnership perspective.

Partnership Teacher-Student Relationships

Teachers must liberate themselves from traditional, hierarchical visions of how classrooms and learning outcomes should be structured—that is with an almost exclusive emphasis on the rational, usually one-way communication of ideas, as an educator’s primary job. This domination education practice reflects the privileging of the rational (associated with maleness) over the emotional (associated with femaleness). Therefore, to make the shift from domination to partnership, teacher-student relationships must honor emotion, as well as rationality. Teachers can begin by cultivating their own intuitive knowing—allowing themselves to listen to students with their hearts, as well as their heads.
A primary key of partnership education is for teachers and students to establish healthy relationships. Creating an emotionally-safe learning environment is critical to empowering students. Partnership teachers create an environment where students feel safe to: “feel and know what they feel; tolerate confusion, uncertainty; express what they feel and think; ask questions that feel ‘dumb’ or ‘have no answers’; take risks, make mistakes, and grow and forgive” (Kessler, 2004, p. 65). One way to reframe student expectations of teacher-as-dominator is to make sure that students know that you genuinely care about their learning experience. In Teaching Community: Pedagogy of Hope, bell hooks (2003) says: “Committed acts of caring let all students know that the purpose of education is not to dominate, or prepare them to be dominators, but rather to create the conditions for freedom” (p. 92).
Creating a climate in which students feel safe to ask questions about what they don’t already know is critical to any effective learning environment. To create a climate in which students begin to feel free to ask questions, teachers must consciously work to deconstruct their position as “the expert with all of the answers.” Given the dominator educational training that most teachers have had, this can be difficult. One place to begin is to step out of the role of expert-with-all-the-answers (even when you have them) and turn student questions back over to the class to answer—a partnership practice.

Co-Creating Collaborative Learning Experiences

Partnership education necessitates a redefinition of the learning environment to move from “the single expert view to a more collaborative and engaging classroom” one that facilitates student understanding of complex issues from a variety of viewpoints (Dakers, 2007, p. 16). Teachers need to co-create learning with students by actively linking ideas and theories with feelings and experiences in meaningful ways. When teachers cultivate an environment where students are empowered to co-create knowledge, they become better critical thinkers, self-reflective knowers, and life-long learners—just the kind of educated citizens our world needs in the 21st Century.
Spender offers additional reasons for moving towards this type of partnership model of teacher-student relationships. She suggests that in a computer-based world where information is more broadly available, the concept of teacher as the “knowledgable expert” becomes more questionable (Spender, 1995). Since it is increasingly easy for savvy students to seek out the facts that are of interest to them, this raises serious questions about how teachers must redefine their role in the learning experience. It also suggests the increasing necessity for training students to be better critical thinkers and to be more skilled at scrutinizing, evaluating, and synthesizing ideas independently. Spender (1995) says that the “graded curriculum where students are to study a specific period or problem one year, and move on to another the next, looks increasingly absurd as kids dial up databases on whatever takes their interest, and become independent learners” ( p.103). As we move towards a model of education based on information technology, Spender (1995) sees subject lines crumbling, students becoming learners/doers, and teachers becoming “teachers of human beings, instead of teachers of a particular subject” ( p. 115).
If we are to become teachers of human beings, not simply teachers of ideas, we must create a climate for collaborative learning that allows students to co-create knowledge—to think for themselves, not for the teacher. A competitive learning environment that fosters ranking is not well-suited to all learners. A collaborative learning environment fosters the partnership characteristics of connecting and linking, and may better serve the needs of more learners. While many math, physics, and programming classes feature “timed tests and competitions to see who can solve the problem first at the blackboard,” Rosser (1997) shows how “encouraging cooperative problem solving where everyone ‘wins’” is more attractive to more students ( p. 15). Further, Namenwirth points to the emphasis on competition in science as problematic in terms of preparing students for careers: “While competition often is effective in augmenting motivation and dedication to one’s scientific career, it is antithetical to a fundamental characteristic of science—the need to share one’s methods and results” (Namenwirth, 1991, p. 24).
Kirk and Zander (2002) outline strategies for moving from competition to collaboration: “1) guiding students toward collaborative problem solving in class; 2) supporting students toward success with accessible non-violent examples; and 3) creating a positive climate for student questions in and out of the classroom” (p. 120). Cohoon and Aspray (2007) also include collaborative methods on their “things that work” list: collaborative methods such as pair programming where students take turns writing code, and structured labs that emphasize hands-on experience ( p. 168). Cohoon and Aspray share the results of McDowell (presented in 2003) who studied the effects of paired programming on 555 students (25% women) in an introductory programming course. They found that women’s confidence increased with pairing (but remained lower than paired men’s) and that both male and female paired students “were significantly more likely than unpaired students to declare a CSE major” (p. 169).
Pedagogical methods that allow students more opportunities to collaborate rather than compete are also more likely to result in the understanding of cross-cultural diversity that will be critical in the global technology industry. Trajkovski says that just “as management theory in IT has experimented with Taiwanese guanxi networks and various strains of Japanese-inspired ‘quality circles,’ science education has much to gain through experimentation in the transformation of pedagogies with the inclusion of cross-cultural diversity” (Trajkovski, 2006, p. 282).

Partnership Evaluation Measures

Evaluation of student learning is another important way to cultivate a partnership education experience. One transformational practice that supports student learning (and implicitly battles stereotypes) is affirmative written responses on student work. Many students’ previous education experiences are full of so many “no”s that for most of them the “shame” button is easily activated. Once that emotional state is engaged, no real knowledge can sink in.
In fact, students are far more likely to find their way to what they are ready to learn when teachers affirm their tentative steps towards that new knowledge. The power of shame to shut down learning, and the power of affirmation to transform is well-documented. In New Vision, New Reality: A Guide to Unleashing Energy, Joy and Creativity in Your Life, psychologist Donald Klein (2001) describes how shame and humiliation shut down our creative capacity and how appreciation opens up spontaneous channels of creative energy. Although Klein is speaking largely in terms of our inner emotional lives, these ideas apply equally well to learning environments.
Criticism shuts down creativity, while praise inspires creativity. For partnership educators, the emphasis should be on affirming and appreciating who students are and where they are in their learning. However, that doesn’t mean that teachers should never “correct” students. They should just adopt a partnership attitude of respect while asking questions that guide students away from places where they are stuck and begin to illuminate new perspectives. Partnership educators try to serve as a signpost in the road that guides students to new paths, rather than the domination approach of handing them a map with only one path highlighted (implicitly suggesting that this is the only “right” path).

Tests and Journals

Partnership evaluation measures also necessitate a richer blending of so-called objective measures (such as tests) with so-called subjective measures (such as journals). In fact, some scholars have raised questions about the usefulness of testing in specific relation to 21st Century education. Spender (1995) challenges the validity of teaching students to store information in their heads, which she sees as less useful in the computer era than facilitating their development as critical thinkers. As an alternative, Spender suggests that “we move from content-based exams to the daily activity of doing research, learning, thinking, using . . . no matter how you label it, it’s a method which doesn’t lead to ‘correct’ answers or standardised [sic] responses” (p. 109). This also helps co-create a collaborative learning environment.
One of the best informal methods for evaluating student learning is the use of journals. Journals benefit the teaching and learning experience in a rich variety of ways. Students benefit from journals because they can: better prepare for in-class discussions; show what they know (versus being tested and found “deficient”); explore ideas and clarify thinking informally while it’s still under development; get regular feedback/guidance from the teacher; and build a relationship of trust with the teacher. Teachers benefit from reflective journals because they can: demonstrate the new power relationship via affirmative, guiding comments; track student growth and development over time; incorporate topic questions linked to major themes as part of journal requirements; regularly assess where students are individually and as a group in terms of their intellectual and emotional status; and adapt the class along the way according to student needs. Lewandowski (2003) demonstrated the benefits of journals for beginning programmers.

Sharing Authority with Students

Another important way to communicate that you are co-creating a partnership learning environment with students is via assessment techniques that place more authority with students. Combine assessment methods that allow students to show you what they know (usually more informal methods) with methods that test them on what you think they should know (more formal methods).
Rosser offers an innovative type of partnership evaluation measure in the design of tests: “students generate (and answer) their own [test] questions . . . It would provide practice in posing questions, involve students in their own learning, generate ideas for course content, and, for us instructors, provide a window into the world of our students” (Rosser, 1997, p. 94). Implementing this idea has many benefits. It challenges students to thoughtfully reflect on what they consider the most significant components of their courses. It might help instructors recognize how well they’ve conveyed what they feel is important, and give them tools for redesigning their courses. It allows students to show what they’ve learned, not just what teachers think they should have learned. It gives students a way to demonstrate their capacity for complex synthesis of ideas.
In the end, developing the test questions becomes a valuable assessment measure on its own as has been demonstrated by PeerWise. PeerWise, a system that allows students in computer programming courses to submit multiple-choice questions with associated explanations, has been demonstrated to support students in reinforcing learning and better understanding course concepts. These questions are then available to other students in the course and can be answered for study purposes, critiqued and discussed, and rated for difficulty and quality. (Denny,, 2008).
Another strategy for evaluation of more formal assessment measures (such as research papers), is to ask students to write self-evaluations. Teachers can develop a self-evaluation form that contains a grading rubric and specific questions for students to address. Providing space for narrative commentary about what they’ve learned and for sharing what has influenced their learning experience becomes another opportunity both for students to learn and for teachers to gauge where students are in their development.


Addressing the underrepresentation of women and some communities of color in computing requires us to understand how pervasively stereotypes impact us, how computer science education embodies the characteristics of a domination system, and how to adopt a partnership perspective. Co-creating systemic change is a large-scale project that requires a long-term commitment by a group of well-informed teachers dedicated to weaving the new threads of partnership computer science education. Adopting partnership perspectives on computer science education will not only help us improve the success rates of women and other underrepresented groups, it will also contribute to a more diverse and innovative computing industry that is better prepared to address 21st Century challenges—fostering a real digital revolution.


Bardeen, M. (2000). Who’s the Scientist? Retrieved April 18, 2010 from index.html.
Barker, L. J., & Aspray, W. (2006). The state of research on girls and IT. In J. M. Cohoon & W. Aspray (Eds.), Women and information technology: Research on underrepresentation (pp. 3-54). Cambridge, MA: MIT Press.

Barker, L.J., & Garvin-Doxas, K. (2004). Making visible the behaviors that influence learning environment: A qualitative exploration of computer science classrooms. Computer Science Education, 14.2, 119-145.

Bartol, K. M., & Aspray, W. (2006). The transition of women from the academic world to the IT workplace: A review of the relevant research. In J. M. Cohoon & W. Aspray (Eds.), Women and information technology: Research on underrepresentation (pp. 377-419). Cambridge, MA: MIT Press.

Belenky, M.F., et. al. (1986). Women’s ways of knowing: The development of self, voice, and mind. New York: Basic.

Brumberg, J. J. (1997). The body project: An intimate history of American girls. New York: Random House.

Bucciarelli, D. (2004). If We Could Really Feel: The Need for Emotions of Care Within the Disciplines. In R. Eisler & R. Miller (Eds.), Educating for a culture of peace (pp. 136-159). Portsmouth, NH: Heinemann.

Cohoon, J. M., & Aspray, W. (2006). A critical review of the research on women’s participation in postsecondary computing education. In J. M. Cohoon & W. Aspray (Eds.), Women and information technology: Research on underrepresentation (pp. 137-180). Cambridge, MA: MIT Press.

Cooper, J., & Weaver, K.D. (2003). Gender and computers: Understanding the digital divide. Mahwah, NJ: Erlbaum.

Dakers, J. R. (2006). Dialectic argumentation for promoting dialogue in IT education: An epistemological framework for considering the social impacts of IT. In G. Trajkovski (Ed.), Diversity in information technology education: Issues and controversies (pp. 15-37). Hershey, PA: Information Science Publishing.

Denny, P., Hamer, J., Luxton-Reilly, A., and Purchase, H. (2008). PeerWise: students sharing their multiple choice questions. In Proceedings of the Fourth International Workshop on Computing Education Research (Sydney, Australia), 51-58.

Eisler, R. (1987). The chalice and the blade: Our history, our future. San Francisco: HarperSanFrancisco.

Eisler, R. (2000). Tomorrow’s children: A blueprint for partnership education in the 21st Century. Boulder, CO: Westview.

Eisler, R. (2002). The power of partnership: Seven relationships that will change your life. Novato, CA: New World.

Eisler, R. (2007). The real wealth of nations: Creating a caring economics. San Francisco: Berrett-Koehler.

Faulkner, W. (2000). Dualisms, hierarchies and gender in engineering. Social Studies of Science, 30.5, 759-792. Retrieved on January 16, 2007 from

Fung, Y. H. (2002, November). A comparative study of primary and secondary school students' images of scientists. Research in Science & Technological Education, 20.2, 199-213.

Goode, J., Estrella, R., & Margolis, J. (2006). Lost in translation: Gender and high school computer science. In J. M. Cohoon & W. Aspray (Eds.), Women and information technology: Research on underrepresentation. (pp. 89-114). Cambridge, MA: MIT Press.

Headlam, B. (2000, January 20). Barbie PC: Fashion over logic. The New York Times. Retrieved January 20, 2000 from 20barbie.html.

hooks, b. (2003). Teaching Community: A Pedagogy of Hope. New York: Routledge.

Johnson, A. G. (2006). Privilege, power and difference. Boston: McGrawHill.

Kanellos, M. (2000). Bankruptcy crashes the Barbie PC. Cnet Retrieved December 16, 2007 from

Keller, E.F. (1992). Secrets of life, secrets of death: Essays on language, gender and science. New York: Routledge.

Kessler, R. (2004). Education for Integrity: Connection, Compassion, and Character. In R. Eisler & R. Miller (Eds.), Educating for a culture of peace (pp. 57-79). Portsmouth, NH: Heinemann.

Kirk, M., &, Zander, C. (2002, December). Bridging the digital divide by co-creating a collaborative computer science classroom. Journal of Computing in Small Colleges, 18.), 117-125.

Klein, D. (2001). New vision, new reality: A guide to unleashing energy, joy, and creativity in your life. Center City, MN: Hazelden.

Lewandowski, G. (2003, October). Using process journals to gain qualitative understanding of beginning programmers. Journal of Computing in Small Colleges, 19.1, 299-310.

Margolis, J. eds. (2008). Stuck in the shallow end: Education, race, and computing. Cambridge: MIT.

Miller, C.C. (2010, April 16). Out of the Loop in Silicon Valley. The New York Times. Retrieved April 18, 2010 from

Namenwirth, M. (1991). Science Seen Through a Feminist Prism. In R. Bleier (Ed), Feminist approaches to science (pp. 18-41). New York: Teachers College.

Pipher, M. (1994). Reviving Ophelia: Saving the selves of adolescent girls. New York: Putnam.

Rosser, S. V. (Ed.). (1995). Teaching the majority: Breaking the gender barrier in science, mathematics, and engineering. New York: Teachers College.

Rosser, S.V. (1997). Re-engineering female friendly science. New York: Teachers College.

Rubin, E., & Cohen, A. (2003, July). The images of scientists and science among Hebrew- and Arabic-speaking pre-service teachers in Israel. International Journal of Science Education, 25.7, 821-846.

Sadker, M., & Sadker, D. (1995). Failing at fairness: How our schools cheat girls. New York: Touchstone.

Spencer, S. J., & Steele, C. M., & Quinn, D. M. (1999). Stereotype threat and women’s math performance. Journal of Experimental Social Psychology, 35, 4-28.

Spender, D. (1995). Nattering on the net: Women, power and cyberspace. North Melbourne, Australia: Spinifex.

Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality & Social Psychology, 69.5, 797-811.

Thomas, M. D., Henley, T. B., & Snell, C. M. (2006, March). The draw a scientist test: A different population and a somewhat different story. College Student Journal, 40.1, 140-148.

Trajkovski, G. P. (2006). Training faculty for diversity infusion in the IT curriculum. In G. Trajkovski (Ed.), Diversity in information technology education: Issues and controversies (pp. 280-309). Hershey, PA: Information Science Publishing.

Wilson, F. (2003). Can compute, won’t compute: Women’s participation in the culture of computing. New Technology, Work, and Employment, 18.2, 127-142.

Zarrett, N., Malanchuk, O., Davis-Kean, P. E., Eccles, J. (2006). Examining the gender gap in IT by race: Young adults’ decisions to pursue an IT Career. In J. M. Cohoon & W. Aspray (Eds.), Women and Information Technology: Research on Underrepresentation (pp. 55-88). Cambridge, MA: MIT Press.