THE GENDER GAP IN THE DANISH IT SECTOR

(1)

THE GENDER GAP IN THE DANISH IT SECTOR

Addressing the Underrepresentation of Women in the IT Sector

MASTER'S THESIS

Mathias Christian Steffensen Student Number: 110473

17. January 2022

Copenhagen Business School Master of Science Business Language and Culture Diversity and Change Management

Supervisor: Nicole Ferry Page Count: 78 STU Count: 177.903

(2)

1

Abstract

This paper contributes to gender equality studies through the investigation of the underrepresentation of women in IT by examining factors that women in the field perceive to influence their career development, how and why they arise, and how they are experienced. A phenomenological study was conducted with the aim of gaining an understanding for what causes the phenomenon by analyzing the experiences and perceptions of 11 women working in the field.

By exploring the individual experiences of women in the field, I found that women all have unique

experiences that are influenced by various factors. The main factors of influence include the degree

of masculine domination in the workplace, which relates back to the culture of the organization,

and the individual identities of the women, which, in turn, relate to their individual backgrounds,

upbringing and values. The study concludes that the underrepresentation of women in IT can be

explained through the masculine domination of the field and the misconceptions about women in

IT that follow. The lack of opportunities for career development, the unableness to enact femininity

due to the masculine domination of the field, and the social barriers that stem from the inherent

misconceptions about women in IT result in women’s negative experiences in the field, which

constitute this study’s coverage of the factors that cause women to deselect and opt out of the field.

(3)

2

1. Introduction

Although the physical number of women in Science, Technology, Engineering, and Mathematics (STEM) fields is continuously growing, the percentage of women employed in said fields remains lower than the percentage of men employed. According to Hill et al. (2010), the underrepresentation of women in STEM first becomes noticeable at the university level. On Average, girls and boys take the same science courses in elementary, middle, a nd high school and fare equally well. Therefore, girls and boys should be equally prepared to pursue science and technology university educations. Although girls and boys share the same prerequisite skill level, significantly fewer girls than boys end up pursuing STEM educations (Hill et al., 2010). Not only are women less likely to enroll in technology-specific bachelors, their representation in science and technology declines further at graduate level and even further in the transition to the workplace (Hill et al., 2010). Specifically for Information Technology (IT) educations in Denmark, the average share of female admissions in higher education between 2015 and 2019 has been around 25.5% and the average share of female completions in the same time period has been 23% (Danmarks Statistik, 2021).

In the recruitment and retention of women throughout the field of IT and over the last four decades

an imbalance has persisted between men and women, from the first IT experiences girls make at

school, to the absence of women on boards, in corporate management, and in academic positions

(Ramsey et al., 2005). Not only is this a major issue of gender equality, equity, and inclusion, but

scholars have begun studying the phenomenon of a gender gap in IT with the purpose of averting a

forthcoming societal crisis: the crisis of an ever-shrinking IT workforce. Ramsey et al. (2005)

described how the proportion of women in IT has declined from 40% of the workforce in 1986 to

29% of the workforce by 1999. By 2018, 25% of jobs in the global IT workforce were held by women

according to global talent network Adeva IT. This is despite women making up almost 50% of the

total global workforce (Luenendonk, 2020). From this data, it can be concluded that female

participation in IT work has been in constant decline for nearly 40 years. Ramsey et al. (2005) state

that many fear the peril that the U.S. IT industry is facing because of the shrinking workforce crisis.

(6)

5

In the U.S., it has been proposed that the crisis could be mitigated by increasing the participation of underrepresented groups, such as women (Ramsey et al., 2005). However, the same crisis is also threatening the Danish IT workforce. Recent studies by the Danish Business Authorities predict that Denmark will be missing around 19.000 IT specialists before 2030. If this tendency is not corrected, the Danish IT industry will not be able to follow the digital development necessary to compete and may therefore compromise the future economic prosperity of Denmark (Askjær, 2016).

1.1 Problem background

Ahuja (2002), a notable academic in information system management, emphasizes the importance of promoting women’s entry and advancement in the IT workforce. She argues this is not only because of the imminent labor shortages in the IT industry, but also because women may prove to be a key resource of skilled technological workers. However, shifting the focus away from the contributions to the IT industry that women could constitute, the question at hand is why women make up such a small percentage of the IT workforce. Ahuja argues that under the circumstances, identifying the factors affecting women in IT careers is of the utmost importance. She proposes that future research should be directed towards the exploration of the barriers facing women and the reasons for said barriers' existence.

1.2 Purpose of the study and research question

The purpose of this study is to investigate the underrepresentation of women in IT by examining

factors that women in the field perceive to influence their career development, how and why they

arise, and how they are experienced. The goal is to gain an understanding for what causes the

phenomenon by analyzing the experiences and perceptions of women in the field. Besides the

national economic perspective of securing the future of Denmark as a technological frontrunner,

the relevancy of the study stems from its contribution to gender equality studies by exploring the

factors that compromise gender equality in the field of IT. These considerations have led to the

development of the following research question for this study:

(7)

6

How can the underrepresentation of women in IT be explained through the experiences of women in the field?

In order to answer the research question I am going to explore what the factors that influence the experiences of women in IT are, how women are perceived by members of the IT field, and how the experiences of women in IT are influenced by the preconceptions of the field.

1.3 Delimitation

As established, this study is exploring the underrepresentation of women in IT in Denmark. Thus, the framework of the study consists of the diversity category of gender, the field category of IT, and the geographical focus on the country of Denmark. Although it would be interesting to explore whether underrepresentation in IT as a phenomenon also occurs across ethnicities, between ages or other categories of diversity, the scope of my study is restrained by time and resources.

Therefore, other categories of diversity will not be addressed or explored. Furthermore, the underrepresentation of women is not just a phenomenon relating to the field of IT, but is specifically prominent across most STEM fields (Hill et al., 2010). As with the reason for choosing gender as the diversity category focus, IT was selected as the field category to narrow the scope of the study. Due to cultural, economic, and political differences between countries, I deemed covering multiple countries in this study too broad as well. Lastly, the study does not distinguish between women working in the private or public sector, nor the size of the companies in which the women are employed.

1.4 Disposition

This section will account for the structure of the study. First, I present a review of literature on the

concepts of gender, and the relationship between gender and IT, to establish prior knowledge on

the subject field of the study and to establish the relevancy of the study. Hereafter, I account for

the methodological considerations that form the study’s framework for understanding. T hese

considerations include the philosophical approach to the research, the chosen methods for

(8)

7

research, and the choices made with regard to the collection of data. Subsequently, I present the empirical findings of the study, divided into categories based on the experiences of interview respondents. Furthermore, the experiences of the respondents are analyzed. Lastly, I discuss the empirical findings of the study against the broader literature of the field and present the final result of the study by answering the research question based on the implications of the findings and the discussion.

2. Literature review

In the following chapter, I justify the choices for my study and the theoretical approach in this research. This study seeks to understand the underrepresentation of women in IT through individual experiences of women in the field. As expressed in my methodology section, hermeneutic phenomenology is my focal point for understanding and interpreting this phenomena. With a basis in hermeneutics, theory and literature are used as tools for data interpretation to deepen the analytical dimension of this study. In other words, research on gender and IT can aid in making sense of this study’s collected empirical data and also explaining the social phenomena found in the data.

Thus, to further understand what it means to be a woman in a male-dominated workplace, as is the general case in the field of IT, it is important to outline the current literature and theories that may help explain how the phenomenon of the underrepresentation of women in IT is constituted and/or perpetuated. Additionally, how stereotypes, bias, prejudice, discrimination, and other elements of gender diversity theories could influence this underrepresentation.

To start the literature review, I will provide a general introduction to women's representation in IT

to lay out the specifics, of what scholars believe to be the reasons for their underrepresentation in

the field. Since the goal of the study is to understand women’s experiences in the context of IT work,

I will proceed to review literature on gender and related social phenomena such as social

constructions of gender, gendered categorizations, and, organizational culture, and how these

social phenomena influence the current perception of women in the field of IT. Furthermore, I will

review literature that provides insights on women’s own perception of the IT field. The framework

of the first part of this chapter is more broadly set on gender and STEM. The purpose of using STEM

(9)

8

literature as a starting point is mainly to gain insight on social phenomena and concepts that affect women's career development in more general terms. Literature on gender and STEM is largely more available as mentioned in the methodology section, however, I will link these phenomena and concepts to gender and IT more specifically once established in the second part of the chapter.

2.1 What is known about gender

In order to explore the gender gap in IT and the barriers that women face, the concept and role of gender must first be illuminated. Gender is one of the main diversity categories taken into consideration in diversity literature and is often part of organizations’ diversity programs (Mensi - Klarbach et al., 2019). Although gender and sex are often conflated, the two distinct but interdependent phenomena of gender and sex need to be clarified. Sexes are based on biological attributes such as chromosomes and sexual organs, whereas gender can be defined as a social category (Mensi-Klarbach et al., 2019). In this thesis, participants in the study biologically belong to the female sex and identify themselves as women.

2.1.1 Gender as a social construction

Mensi-Klarbach et al. (2019) present Calas and Smircich’s 1996 definition of gender as: “a social construction that describes male and female roles in the given social and cultural context, a product of socialization and experience” (p. 98). This corresponds with the World Health Organization’s (WHO, 2011) definition, where the socially constructed characteristics of men and women include norms, roles, and relationships of and between groups of men and women . Gender roles may vary from society to society and across cultures. When a person is born, their sex will determine what appropriate norms and behaviors they are taught, including how they should interact with people of both sexes within households, communities, and places of work (WHO, 2011). For the last 20 years, researchers have been preoccupied with studying sexual differences, seeing these differences as the natural result of sex-based categories. However, Mensi-Klarbach et al. (2019) argue that researchers would do better by shifting their research towards investigating contextual restraints.

The reason for this is that biological sex differences do not explain the underrepresentation of

(10)

9

women, but rather societal and organizational practices favoring men over women. Societies where men are seen as superior to women are called patriarchal regimes (Hill et al., 2010). In these societies, various structures and manifestations that constitute societal culture are founded on skewed power dynamics that favor men and oppress women (Hunnicutt, 2009).

Nordic countries, like Denmark, are generally viewed as gender equity leaders, with Denmark ranking 29th in the 2021 Global Gender Gap report (World Economic Forum, 2021). However, this does not mean that segregation is not present in Denmark. The report clearly depicts imparity on the Global Gender Gap Index. The Global Gender Gap Index is measured on four major dimensions:

Economic participation and opportunity, educational attainment, health and survival, and political empowerment. On a score from 0 to 1, where 0 symbolizes complete imparity between men and women and 1 symbolizes parity, Denmark scores 0,75 for the first, 1 for the second, 0,97 for the third and 0,3 for the last of the abovementioned dimensions, averaging at 0,77 (WEF, 2021, p. 173).

What is especially noteworthy for this study is the distinct segregation in STEM skill attainment and

education. While educational attainment in Denmark reaches almost 100% parity between the

genders, the parity value for STEM education attainment is only at 0,4 (WEF, 2021, p. 173-174). In

an analysis presented in the report, it is clear that the challenge of the number of women who study

in STEM fields can be seen as a symptom of wider biases and not as a ‘supply problem,’ which it has

often been termed. The report states, “Gendered signals from the labour market—the social

experience of learning in STEM classes and working in technology fields—go a long way toward shaping the potential employee base of the professions making them distinctively male” (WEF, 2021, p. 64). Referring to the challenge of the number of women who study in STEM as a 'supply problem' suggests that the problem at hand concerns a shortage of talent. However, the Association of Women in Science illustrate how there is a 35% drop between women earning a degree in STEM and women working in STEM (AWIS, 2019). Thus, a fully tapped talent pool cannot be the prima ry issue. Instead, the problem may be one of retention. Specifically, the Global Gender Gap report analysis suggests that the inherent biases of STEM fields generate a demand for male over female talent. This is in line with Mensi-Klarbach et al.'s (2010) argument on how the underrepresentation of women can be explained by societal and organizational practices favoring men over women. To gain an understanding of the roles ascribed to women in Danish society, and why men are favored

(11)

10 over women in STEM fields, I examine the socially constructed characteristics of men and women and sociocultural gender norms such as stereotypes and biases in the following sections.

2.1.2 Masculinity versus femininity and categorization

Similar to the describtion of the construction of gender by Mensi-Klarbach et al. (2019), Lorber et al. (1991) explain how the construction of gender starts with assigning an individual to a sex category based on the genitalia they were born with. These sex categories become gendered through naming, attire, and other gender markers, and once gendered, peoples’ behavior towards the individual will adapt accordingly (Lorber et al., 1991). Thus, gender identity unfolds based on how one is perceived and treated. Once categorized into a specific gender group, different expectations, roles, and norms will apply to an individual throughout their lives. Due to the social construction of gender, it is ,therefore, socio-cultural processes that shape gender identities, and not biology. Femininities and masculinities are socio-cultural categories that describe gender identities (Gendered Innovations, n.d.). However, femininities and masculinities do not map onto biological sex, as certian behaviors or practises can be recognized as either feminine or masculine across cultures, irrespective of whether they are adopted by men or women (Gendered Innovations, n.d.).

The ‘women in STEM’ discourse follows a binary categorization of femininity and masculinity. Here, women are viewed as domestic, passive, and emotional, while men are viewed as rational, individualistic, competitive, and technically skilled. Figure 1. shows how the ‘women in STEM’

discourse is deconstructed (Phipps, 2007). It should be noted that this binary categorization of

gender is not limited to the STEM discourse, but is assumed in most sociological research designs

(Lorber, 1996).

(12)

11 Figure 1. Deconstructing ‘women in STEM’ discourse.

Phipps (2007) argues that the ‘women in STEM’ discourse has voluntarist and individualistic underpinnings, suggesting that a lack of women in STEM fields is based on notions of access and choice. Thus, girls and women deselect educational and career opportunities in STEM as the masculine image of STEM fields conflict with prevailing stereotypes of femininity. In other words , one potential reason for the underrepresentation of women in STEM could be how gender norms shape women’s self selection into STEM or IT careers. However, it must be kept in mind that this framework posits that young boys and girls are placed in predetermined sex roles throughout their educational and social lives. Therefore, they arguably develop sex-specific skills and interests, which is why girls are often driven away from STEM fields (Phipps, 2007).

Having established that genders are social constructs, Mensi-Klarbach et al. (2019) argue that

genders are actively created or ‘done’, rather than being an assemblage of individuals’ traits and

fixed characteristics. As an example of how gender is done, Lorber et al. (1991) describe how people

assume the gender of a small child based on its attire. Earrings would suggest the child to be a girl,

whereas a baseball cap would suggest the child to be a boy. Identity-relevant categories such as

gender are produced and reproduced through social practices. The gendered categorization

(13)

12

organizes identities in binary ways that reinforce differences in terms of privilege and power.

However, two women or two men are not necessarily identical just because they share a gender.

Binary gender classification evokes stereotypes of female and male traits as if they were universally shared and applicable (Mensi-Klarbach et al., 2019). ‘Doing’ gender is an unavoidable routine accomplishment. Due to social structures, interactions, and organizational practices, people will always ‘do’ gender. For this reason, gender-based inequality persists as long as the masculine is more valued than the feminine. It is also the reason why specific jobs and roles in society are ascribed to specific genders. Jobs that require feminine traits such as nurturance, sensitivity, and empathy will enable the enactment of a female gender identity. However, by working such a job, the female gender identity will also stabilize, meaning that the link between the enactment of femininity and a female gender identity is reinforced and remains unchanged (Mensi-Klarbach et al., 2019). Thus, “individuals reflexively produce and stabilize the gendered job through their performance of gender identity - and simultaneously produce and stabilize the identity through performing the job” (Mensi-Klarbach et al., 2019, p. 129).

However, when people sometimes work jobs that are associated with different gender identities.

Male gender identities are not associated with jobs such as kindergarten teachers as caring for others is a feminine trait (Mikkola, 2019). Thus, when a man works as a kindergarten teacher, he engages in boundary work, for example by playing football with the boys. Engaging in boundary work allows him to enact masculinity by differentiating between female and male aspects of the job. He then draws boundaries between the different aspects and engages only in activities associated with masculinity (Mensi-Klarbach et al. 2019). Examples of boundary work in the field of IT is how women can draw boundaries between hard and soft aspects of IT work. Soft skills are attributed feminine gender identities (DeAngelis, 2021), so women can enact femininity through aspects of IT work such as creativity and communication, which are typically considered soft skills (Moser, 2013).

When examining women’s career development, it would, therefore, be relevant to first understand how women form their identities in response to their perceived reality. According to Alvesson et al.

(2008), identity can be viewed as the key to understanding the relationship between self, work, and

(14)

13

organization. Villeséche et al. (2018), argue that when individuals reflect on identity in an organizational context, they ask the questions: “who am I?”, “who am I not?”, and “who do I want to be?”.

2.1.3 Gender and institutions

Now that the sociocultural categorization of men and women has been examined, the focus of this next section will be on how those categories affect and have been affected by societal and organizational structures. According to McDonald (2013), the development of female and male careers and the differences between the two is rooted in a historical, familial division of labor, and interaction at home. Women have always been the caregivers and men have always been the providers of the household. Therefore, gendered segregation comes into sharp relief in the unpaid domestic labor versus paid employment. Since then, this gendered segregation has been reproduced throughout history, resulting in the sociocultural categorization of men and women of contemporary times (McDonald, 2013). Thus, as men formed the majority workforce, their reality became the standpoint from which the world was viewed, resulting in the establishment of the male norm (Smith, 1988). In other words, male organizational norms laid the foundation for organizational reality as men formed the majority of labor forces. In that way, the typical worker image became associated with masculine traits and women remained primary caregivers responsible for domestic work.

The link between gender and organizations has been the focal point of a significant portion of diversity research (Mensi-Klarbach et al., 2019). However, the aspect of male dominance has sometimes been overlooked. Acker (1990) argues that researchers have failed to acknowledge male dominance as a factor because genders are not visible in masculine environments, where only the masculine is present. Therefore, organizations have for a long time been classified as gender neutral.

In the following, the link between gender and male dominated fields will be explored, in order to

assess whether and how male dominated fields constitute oppressive forces on female

employment.

(15)

14

2.1.4 Gender in male dominated fields

It has been established that Denmark is considered a feminine society, however, this does not mean that all fields of industry have achieved a perfectly balanced workforce. As previously mentioned, many of the STEM fields, including IT, suffer from a significant gender gap in their workforces, thus, making them male dominated (Luenendonk, 2020). To gain an understanding of how gender roles have arisen and continue to be upheld, and the rise masculine domination, the theory of “symbolic violence” by Pierre Bourdieu will be examined in this thesis. The symbolic violence is a theory published in Bourdieu’s 1999 book: Masculine Domination. It addresses how gender roles have affected the perception of man and woman as seen from a socially constructed perspective (Bourdieu, 2002).

Bourdieu (2002) describes how society has bred a masculine domination, which presides in subconscious and underlying structures and norms. This affects the way men and women perceive, think, and act. The symbolic violence occurs as society accepts the masculine domination. The word

‘symbolic’ refers to the non-material including language, perception, and communication. Thus, the symbolic violence is not an act of physical violence, but rather a non-physical exercise of power, where the dominated agent (the women) accepts the perception of reality that the dominati ng agent (the man) presents. The dominated agent accepts this perception of reality as it seems natural and obvious. The masculine domination is not easily altered, as it is cause for historical and cultural arbitrariness, which is rooted in our society. An example of this arbitrariness is the gender segregation mentioned above, where the role of women is as caregivers and the role of men is as providers. Bourdieu (2002) argues that this could count as one of the reasons why women are excluded from masculine fields and jobs.

2.1.5 Token women

In a societal context, gender issues cannot be seen as issues of majorities versus minorities as neither men nor women constitute a minority in modern day society (Mensi-Klarbach et al., 2019).

However, the same cannot be said for workplace contexts. Kanter asserted in her 1977 book Men

and Women of the Corporation that minority standings in the workplace exist due to organizational

(16)

15

behavioral patterns, for instance in organizations in which women are underrepresented. These behavioral patterns are based on the dynamic interaction between power and opportunity. In organizations where women constitute a minority, they are subject to high visibility among employees, assimilating behaviors, and exclusion. Typically, underrepresented women will work under skewed power/opportunity conditions compared to a majority of male workers. These women are defined as token employees and are expected to behave in a stereotypical manner (Kanter, 1993). Due to the underrepresentedness of the women in a given organizational context, the individual token employee will often be deprecated by the dominant group, which in this case would be a male-consisting majority. Token women are viewed more as a symbol for their category than they are acclaimed for their individual performance. Therefore, these women are measured differently than their male counterparts and their performance has to be exceptional for them to gain attention (Kanter, 1993).

Fairhurst et al. (1983) challenges Kanter's research. They believe that Kanter is correct in her posit that token employees face a different set of work conditions than the majority group, however, they disagree on the significance of numerical imbalances when it comes to power relationships. They suggest that power is more completely determined in organizations. Kanter’s theory is based on the premise that role variation is not likely to occur, however, Fairhurst et al. (1983) believe that it has been shown how individual responses to the token- and majority member’s role may vary in relation to sources of power that are not based solely on numerical abundance or scarcity. A status inconsistency may arise when token- and majority members possess powers from different dimensions of social or occupational stratifications. Thus, the power acquired through numerical majority may not always be exercised if a token member is uniquely qualified and hence, has achieved a higher status in terms of expertise although holding a lower status in terms of majority.

In other words, relationship rules may be negotiated between tokens and majority members if said

rules are based on abilities rather than class membership. Nevertheless, however qualified a token

member may be, their power will not easily become viable if majority members keep reminding

them of their lower societal/organizational status (Fairhurst et al., 1983).

(17)

16

Fairhurst et al.'s (1983) challenge of Kanter’s research is founded in Kanter’s methodology. Although her work may be insightful and she succeeds in proving her point about control being based on sex, her study is very limited when it comes to participating individuals and contexts, thus rendering it a premature generalization. They argue that when performing studies like this in the future, researchers have to pay attention to demographic and political factors, along with the backgrounds of token members. In Kanter’s study, successful token women were older, had technical backgrounds, and were politically wise as a result of previous token experiences. On the contrary, newly employed token women straight out of college would be expected to have a more difficult time exercising power and resisting majority group assertions of control. This is due to their unlikely political awareness, experience, confidence, and knowledge on how to take advantage of the power acquired through dependencies (Fairhurst et al., 1983).

2.1.6 Gender stereotypes and bias

Heilman (2012) defines stereotypes as “a generalization about groups that are applied to individual group members simply because they belong to that group” (p. 114), where gender stereotypes more specifically are generalizations about the attributes of men and women. Heilman (2012) suggests that stereotyping can be viewed as a sort of subconscious coping mechanism to process new information faster. By generalizing, the brain spends fewer cognitive resources than it would by making explicit observations at every new encounter. It happens automatically and is not easily disregarded.

As previously mentioned, stereotyping can and has affected children from a very young age.

According to Hill et al. (2010), there are two prevalent stereotypes in STEM, the first being that girls are not as good at scientific subjects as boys, and the second being that scientific work is better suited to boys and men. Children become aware of these stereotypes as early as elementary school, and can express stereotypical beliefs about which scientific courses are suitable for girls and boys.

These stereotypes have been proven to persist among adults as well, and negatively affect the

stereotypical image of scientists being men (Hill et al., 2010). These negative stereotypes affect

women’s and girl’s aspirations and performances in science - a phenomenon called stereotype

(18)

17 threat. This stereotype threat could explain girls’ lacking interest and aspirations for careers in

scientific fields. By avoiding those fields, girls may try to reduce the likelihood of judgment against them through the lenses of negative stereotypes (Hill et al., 2010). In schools, the educational material that children are exposed to and the gender of their teachers in scientific courses reinforce the stereotype of the typical worker in fields of science. Children will mainly be exposed to male figures when it comes to STEM (Dovidio et al., 2010). However, stereotypes are also transmitted between generations as demonstrated by Shapiro et al. (2011). Parents tend to hold lower expectations for their daughters’ abilities when it comes to science than they would their sons.

These reduced expectations lower the confidence of their daughters in the subjects in question and thus, affect girls’ performances. It can therefore be stated that the stereotype of a scientist being a male figure can have an influence on women’s interests for STEM subjects from an early age, in addition to the performance of women in scientific industries like the IT industry. It can be derived that this lack of interest and ability to perform is caused by sociocultural factors and is not embedded in the nature of women.

Gender stereotypes have implications for women’s career progress as gender stereotypes give rise to biased judgements and decisions, impeding women’s advancement. Gender stereotypes promote bias because of the negative performance expectations that s tem from the belief of what women are like, and the attributes perceived necessary for success in male gender-typed positions.

Furthermore, gender stereotypes promote gender bias through normative standards for behavior.

If these standards are violated, for instance if a woman is successful in a male gender-typed position, it induces disapproval and social penalties (Heilman, 2012). Biases can be either implicit or explicit.

Implicit bias implies automatic associations and reactions that happen without awareness upon encountering stimuli. These are also called unconscious biases. Explicit bias, also called conscious bias, includes preferences, beliefs, and attitudes of which people are generally aware and can identify and communicate to others (Daumeyer et al., 2019). Implicit bias is currently the prevailing form of bias in our society and continues to have an adverse effect (Hill et al., 2010).

Rowe (2008) explains how biases can lead to something he refers to as micro-inequities. These

micro-inequities are reactions that for example occur in the form of facial expressions, gestures, and

(19)

18

tone of voice and stem from unrecognized biases. This could be by assigning the role of taking notes or emptying the dishwasher to a woman. Thus, women’s self-concept and by extension their career choices may be influenced by micro-inequities over time. Women working in STEM fields, which includes IT, are assumed to be more likely to experience micro-inequities as gender-biased behavior is more widespread in these fields (Hill et al., 2010). Camacho et al. (2011), describe another form of reaction originating from gender biases. This reaction they dub as microaggressions, and they define it as small-scale discriminatory behaviors of a non-physical nature. This could be in the form of outsiders insinuating that women are not a good fit for their fields, a social isolation in situations where women are outnumbered by men, or in the form of jokes. Thus, the more male-dominated a field, the more gender bias, micro-inequities, and microaggressions occur towards women.

Although biases may only be noticeable in small amounts, both consciously and unconsciously, biased behavior accumulated may result in an increase of stress levels and exclusion.

2.2 Gender in IT

In contemporary times, the field of computer science is dominated by men, however, this has not

always been the case. A large percentage of computer pioneers throughout the 60s and 70s were

women. In the 80s, a radical shift in the IT gender balance took place, dropping the percentage of

women working in IT from around 25-30% to the 7% seen in modern workforce constellations

(Henn, 2014). The reason behind the decline in the amount of women in technology fields has not

been established with certainty. The most prominent theory among scholars is that the invention

of the personal computer has had an impact, especially how the PC was marketed. As PCs were

introduced to most households, they often ended up in the hands of male family members since it

was marketed towards boys, thus fueling the general perception that computers were ‘boy toys’. A

1990s study by Jane Margolis shows that families were more inclined to purchase computers for

their male children than their female children (Henn, 2014). Therefore, boys grew up more exposed

to technology than girls. Even though most people would think of this as an outdated mindset and

approach, the think-tank DEA suggests that the issue is ongoing. From a very young age, girls are

taught to believe that science and technology are boys’ subjects, that possible jobs in these branches

would not live up to their dreams and expectations for future career paths (DEA, 2019).

(20)

19

According to Linda Sax (2017), head of the research project: A Study of Gender and Racial/ethnic Diversity in Computer Science, her research suggests that girls perceive themselves as being less competent than boys when it comes to scientific subjects such as IT and mathematics. She believes that while growing up, children become convinced that STEM-subjects belong to a masculine domain through socialization. This in turn affects educational and career choices (Ringgard, 2017).

Sax rejects the proposition that women and girls are born with worse abilities than men when it comes to mathematics and technology. In fact, her research implies that women's lack of self- confidence in STEM subjects is due to media influence and how their parents raise them. This seems to be a western issue primarily, as gender differences in STEM subjects are less prominent in countries like China, Dubai, and Malaysia (Ringgard, 2017). In the 2018 Talent Gap Report by McKinsey and Innovation Fund Denmark, attention is called to the importance of fighting gender stereotypes and biases in society. The report emphasizes that parents, companies, and educationa l institutions all play a part in contributing to future developments (Mckinsey&Company, 2018).

2.2.1 Brief description of IT and IT culture

To gain an understanding of the field in which research is being done, this section is devoted to defining the concept of IT, what classifies as IT, and the culture of IT organizations. Due to a lack of a generally accepted definition of IT between scholars, Onn et al. (2013) have reviewed IT studies to establish a comprehensive definition. Based on their review and in alignment with the views of various scholars of the field, their definition of the term covers a wide range of information processing and computer application in organizations. These include information systems, communication-related technologies, computer software, information processing hardware and networks, any kind of computer manufacture or maintenance, electronic data management, and resource planning that affect corporation productivity.

According to Tsai (2011), organizational culture is the shared values, beliefs, or perceptions that

employees hold within an organization. Understanding organizational culture is important, as it

reflects values, beliefs, and behavioral norms that employees use to give meaning to encountered

(21)

20

situations. In other words, it can influence the behaviors and attitudes of employees. Ahuja (2002) explains how there are two major forms of computer culture that have emerged since the mid- 1980s. The first is the culture of calculation, which is more rigorous and ‘engineering’ in a pproach.

This is a top-down structured culture that involves hard programming. The values associated with the culture of calculation are masculinity, mastery, individualism, and nonsensuality. The second culture is the culture of simulation, which is more consistent with ‘soft’ styles that women find more comfortable. Ahuja suggests that the culture of calculation is the more dominant of the two, and that this is an instrumental reason for turning women away from the field of IT. The general culture in the IT industry is classified by long working hours, late nights, and high focus levels. This sort of work ethic could clash with the safety concerns and family responsibilities that women may have.

Furthermore, in a culture where employees are predominantly male, and where women remain on the periphery of this dominant culture, women may miss out on faculty and colleague interactions and opportunities for learning and participating. Thus, this kind of culture is producing and reproducing male domination (Ahuja, 2002).

Ramsey et al. (2005), describe IT culture in a similar manner to Ahuja’s culture of calculation. They point out that 79% of IT jobs require 24/7 dedication. Although long hours are sometimes necessary to meet a deadline, they are also considered a status symbol and a sign of machismo in IT culture, and they are typically considered more important than ‘soft work’ such as team management. Other characteristics of IT culture according to Ramsey et al. are that they are largely white, male - dominated, anti-social, individualistic, non-physical, etc. Additionally, the culture has strong in- and out-group dualisms in which intellect is valued higher than emotional, physical and sensual needs.

Typically, the dualism translates into expert and non-expert or male and female behaviors, thus

reinforcing a cultural stereotype of masculinity by which women are undervalued (Ramsey et al.,

2005). From this section, it can be concluded that biases and stereotypes against women in IT are

being reinforced through IT organizational culture.

(22)

21

2.2.2 Stereotypes and biases towards women in IT

In a study dedicated to exploring the relationship between prevailing stereotypes of the computer science field and girls’ sense of belonging, Master et al. (2016) suggest that girls ma y in fact avoid computer science because the stereotypes of the field signal them that they do not belong. They give specific examples of how computer scientists are stereotyped, such as being male, technologically oriented, and socially awkward (Master et al., 2016). However, they extensively provide STEM examples as representations for stereotyping within computer science. Thus, this study presumes that the literature previously reviewed in relation to the section on stereotypes in STEM, presented by Hill et al. (2010), Dovidio et al. (2010) and Shapiro et al. (2011), comprise a valid representation of prevailing stereotypes in the field of IT. In other words, literature on stereotypes in STEM fields is directly applicable for research in the IT field.

According to Elsbach et al. (2019), historical, social, and cultural constructions of gender have created the implicit association between men and IT. In correspondence with Heilman’s (2012) thinking, in which gender stereotypes give rise to biased judgments , Elsbach et al. (2019) argue that they have found evidence for an implicit association between IT and masculinity in a study of attitudes towards computers and technology, performed by Selwyn (2007). The study shows how gender stereotypes influence young people’s perception of IT, and that only creative applications of IT were perceived as feminine, whereas all other applications were perceived as masculine (Elsbach et al., 2019). The findings of the study are consistent with the notion of computers being boy toys, and with the stereotype of women’s inferiority in technological skills compared to men (Elsbach et al., 2019). Thus, the findings are similar to the previously mentioned posits of Henn (2014) and Hill et al. (2010).

A study by Smith et al. (2005) investigated the consequences of the implicit associations between

gender and IT on women’s attitudes towards computers (Elsbach et al., 2019). Their study finds a

link between a lower identification with IT among women and the general association between

computers and men. Furthermore, women who were still considering entering the IT field were

forced to develop strategies for overcoming gender stereotypes by either emphasizing their

masculine traits or downplaying their female status (Elsbach et al., 2019). Additionally, Elsbach et

(23)

22

al. (2019) make reference to the previously mentioned concept of stereotype threat. A study by Cooper (2006) shows how women tend to perform worse when primed with their own gender identity immediately before engaging in a computer-related task (Elsbach et al. 2019). The findings of the study are explained through the presence of stereotype threat - the pressures of negative stereotypes related to women’s technological skills (Elsbach et al., 2019).

2.2.3 Perceptions of women’s abilities in IT

As has been established, women are largely underrepresented in the field of IT, even though society has become more focused on gender equality and gender equity since the 1960’s. The rationalization of the underrepresentation has arisen from a preconception about women that historically antecedes the liberalization of the genders. According to the U.S. National Science Foundation (2007), there are five major myths about women and girls in science. These myths form three of the more prevalent preconceptions about women and their abilities in science. The first preconception is that women are biologically inferior to men in scientific subjects. The second preconception is that girls are not naturally interested in science, and the third preconception is that girls are not easily motivated to enter science fields (NSF, 2007). These preconceptions are in line with gender essentialist philosophies, by which men and women are fundamentally different due to their biology (Vinney, 2021). However, this mindset and these preconceptions have been challenged by scholars. Ceci et al. (2009) argue that the role of biology is severely overrated as biology in fact has a very small influence on performance and interests. Research shows that girls often outperform boys in scientific courses. Furthermore, Hill et al. (2010) explain some of the actual reasons why women do not pursue careers in IT according to research. The suggestion that women do not find careers in IT interesting is in some cases true. However, it has nothing to do with biology (Hill et al., 2010). Other reasons are that women either do not feel comfortable working with the people they associate with IT work (which are mostly men), or they simply do not think they would be good at it (Hill et al., 2010).

In a 2012 article, Fidelman addresses this research. He argues that women’s perceptions and

assumptions of the field are based on media coverage, stereotypes, and preconceptions they make

(24)

23

about themselves based on the preconceptions of society. In most cases, these women have not made their own experiences in the field of IT. In a 2004 study, Pollock et al. evaluated the reasons for why girls are not attracted to computer science. Based on literature findings and their own experiences as females working in computer science, they estimated that the reasons mainly included misconceptions about the field itself and about the working style of successful people in the field. Furthermore, a lack of confidence in the women's own abilities perceived necessary for success in computer science was hypothesized. The study introduced high school girls to the field of computer science through a summer school program. Resultantly, participants' perception of computer science, self-confidence in computer science and interests in computer science increased (Pollock et al., 2004). Thus, this study invalidates the essentialist preconceptions about women in IT, as it shows that women simply lack awareness, confidence, and an interest for IT. However, the study also shows that women can be motivated to enter the field. What preconceptions about women's abilities in the field of IT do, is that they create and reinforce biases and stereotypes that could discourage women from wanting to learn more about the field and, thus, become motivated (Pollock et al., 2004).

2.2.4 Token women in IT

The general concept of token women has been established previously, and I will now proceed to

address token processes in relation to the field of IT. According to Alegria (2019), there has been an

increase in gender integration for managerial and professional jobs over the last 40 years. However,

this tendency does not apply to all fields. The tech field counts among fields that are particularly

resistant to gender integration (Alegria, 2019). On the contrary, as previously noted by Henn (2014),

the integration of women has been in decline for the last 40 years. Therefore, women in IT remain

numerical minorities. In alignment with Kanter’s (1993) theory on token women, the numerical

underrepresentation of women in the IT field must result in skewed power/opportunity conditions

between men and women, stereotypical expectations of behavior, and the deprecation of women

by male coworkers. Alegria (2019) agrees with this assumption and states that it can only be

expected that women in tech face challenges as tokens. She argues that the reason for women’s

token status in IT is that most tech work is part of the broader field of engineering, which numerically

(25)

24

and culturally is dominated by men and masculinity. Thus, the soft skills of women will likely make them a better fit for managerial work in the tech field. The exception can be found in organizations where the engineering culture is less dominant, and women fit in more easily (Alegria, 2019), or in other words, in cultures previously defined by Ahuja (2002) as cultures of simulation.

Alegria’s (2019) study shows how women in tech would strive towards managerial work in order to escape the masculine environment of the more technical work. Although some women are accepted into low level managerial positions, provided they exhibit the required level of interpersonal skills, they are unable to reach executive positions (Alegria, 2019). Building on Williams’ 1995 conceptualization of a glass escalator, where men in female dominated workplaces are fast tracked towards leadership roles due to their perceived leadership qualities, Alegria (2019) generates her own metaphorical conceptualization: the concept of a glass step stool. The step stool symbolizes how the interpersonal skills of women can result in promotions, however, women are confined to a single step up (Alegria, 2019).

3. Methodological considerations

This section of the thesis accounts for the methodological considerations of the thesis. These considerations were formed based on my chosen understanding of the philosophical premises that constitute the world, in which research is being done. Additionally, the methodological considerations of the thesis cover the methods used to conduct the research, the thought processes behind the data collection, and the limitations encountered as a result of the chosen methods.

3.1 Research philosophy

The research in this study follows a hermeneutic philosophy. Hermeneutics cover humanistic disciplines that explore humankind as thinking, wanting, and acting individuals (Collin et al., 2012).

In contrast to the paradigm of natural science, where individuals are perceived as objects, and

actions as casual processes, human actions need to be understood and interpreted from the

(26)

25

individual's perception of themselves and the context (Collin et al., 2012). Therefore, this research philosophy was chosen as it allowed me to expand my understanding and gain new insights on the gender gap phenomenon within the field of IT through interpretative and understand-seeking exploration. I aligned the philosophy of research of this study with Gadamer's refinement of Heidegger's hermeneutics, due to his added dimension of history to what it means to understand (Barett et al., 2011). Gadamer argues that people's consciousness is embedded in the particular history and culture that has shaped them. Therefore, historically and culturally influenced pre - understanding or prejudice will affect all interpretation (Mason et al., 2019). This is especially relevant for my study as I am a male researcher seeking to interpret the understandings of women in IT. Historically and culturally, my reality has been shaped differently than theirs, which is important for my interpretation of their experiences. Gadamer’s hermeneutics can essentially be understood as a convergence with human existence through language and action (Barrett et al., 2011). In this convergence, existence is expressed, for which interpretation is used to create a new understanding (Barrett et al., 2011). In other words, all human behavior is based on interpretation.

To understand a phenomenon, one must understand the interpretation hereof. In this study, I sought to understand how I interpreted the problem area. This was done by comparing my pre- understanding of the topic with knowledge acquired throughout the research process. Thus, I attempted to achieve a new understanding of the phenomenon.

In Gadamer’s hermeneutic view, the researcher’s pre-understanding is of great significance because

it creates the foundation for the actual understanding of a phenomenon. One of his more radical

points compared to other hermeneutic philosophers of his time, is that prejudice is inevitable. In

fact, prejudices are conditions for understanding (Barrett et al., 2011). When interpreting

something, it is an attempt to reshape the foreign into the familiar, and because of that, prejudice

is unavoidable. Thus, the researchers pre-understanding becomes significant, as prejudice forms the

true foundation for understanding a phenomenon. In other words, I, as a researcher, am restricted

by my own pre-understanding, which can influence the way I choose to incorporate and interpret

knowledge. Therefore, it is essential for me, as a researcher, to be open towards new

understandings, which is why I chose to conduct semi-structured interviews as a mean for empirical

data collection in this study. In doing so, I assumed my pre-understanding of the phenomenon to

(27)

26

become altered, and that I might gain new understandings on the phenomenon. Essentially, the purpose was not for me to confirm my own pre-understandings and interpretations, but to enter a dialogue with conversation partners, in accordance with Gadamer's thinking.

By being open towards new understandings and allowing my pre-understanding to change, I enabled myself to reach a greater understanding of the phenomenon. Gadamer explains how understanding is a fusion of horizons—a fusion of biased knowledge and new knowledge. The researcher can reach this fusion of horizons by exceeding their pre-understanding (Barrett et al., 2011). Therefore, the fusion of horizons for my study, and a greater understanding, was reached as I concluded my research, allowing me to reflect on my pre-understanding of the phenomenon from the perspective I gained through new understandings. The fusion of horizons can be described through the hermeneutic circle. In the hermeneutic circle, the researcher moves between parts of a ‘whole’ and the ‘whole’ itself in order to establish truth. As explained by Sloan et al. (2013), this is done by discovering and interpreting phenomena: “This circle is the process of understanding a text by reference to the individual parts along with the researcher’s understanding of each individual part, by further reference to the whole document” (p. 1296).

The circular motion is, therefore, created by moving between interpreter and object, or in this case researcher and informant. To understand the ‘whole’, one has to understand the parts and vice versa. In that way, it made sense for this project to perceive the collection of stories told by the informants as a ‘whole’, when it came to understanding the phenomenon, rather than separately fixating on individual stories. However, the data analysis further contributed to g aining an understanding of the whole. By including perspectives from external literature, I was able to not only interpret individual informant cases, but see them as parts of the greater whole. By means of the data collected and the following analysis, I sought to develop a new understanding of the ‘whole’

— that being how Danish women in IT experience factors such as diversity, inclusion, bias, and/or

discrimination in a male-dominated field, and how they these factors may contribute to the

underrepresentation of women in IT — which is why I proceeded to move in a circular motion

between parts of the ‘whole’ and the ‘whole’ itself, in accordance with the hermeneutic principle of

interpretation. Thus, a hermeneutic philosophy allowed me to cultivate and expand my

(28)

27

understanding of the phenomenon as my research process progresses, which is why I deemed it a suitable research philosophy for this particular thesis.

3.1.1 Pre-understanding

As established in the previous section, the researcher’s pre-understanding and prejudice is unavoidable, and the recognition hereof is essential before a new understanding can be reached.

This section of the thesis is a clarification of my pre-understanding, concerning the phenomenon of the gender gap found in the Danish IT sector. In that way, the reader gains insight into the substructure of the interpretation conveyed in the thesis.

To start, a pre-existing knowledge on the subject area formed the basis for my aspiration to achieve greater knowledge on the subject area. My preconceptions on the subject area had been partly derived from one of my previous case studies on the retainment of women at EY Advisory Denmark, partly from insight gained through graduate-level diversity courses, and partly through personal relationships with affected individuals. Furthermore, I work part time in the IT department of the Danish Building Agency. Being embedded in the culture of an IT department, I have gained an understanding of the circumstances that frame IT work such as the skills that are sought after, how people interact with each other, and the general environment of the IT workplace. Considering the insights I had gained so far, along with the general assumptions I had on the subject area - based largely on word-of-mouth narratives and media coverage - my pre-understanding of the phenomenon of a gender gap within the field of IT had been as follows:

First of all, I had believed there to be a generational difference among women working in IT when it

comes to women's interests in IT. I had imagined younger-generation women to be more prone to

either be part of the IT workforce pool, or enter said pool than older-generation women. I had

assumed that the younger generations would be more exposed to technology and could, thus,

be/become more interested in the field, although I had also assumed that this would mostly be the

case with men rather than women. The reason for me to have believed that men from younger

generations were more likely to become interested in IT, through exposure, than women was mainly

(29)

28

due to cultural norms. My impression had been that in western countries, computers are marketed as tools and toys for boys or men. For instance, most video games are directed towards male players, and mainly show male characters. I had believed this, among other things, could be a stepping stone towards gaining an interest in computers and technology.

I had presumed that a lack of female role models in the field of IT could also be a factor, when it comes to the gender gap within the field. Essentially, the limited number of women in the IT pool could have had a deterrent effect on future female candidates. When women become aware of the lack of other women in the field, they could feel discouraged or even pressured into steering away from the field. First of all, I had presumed that it can be disconcerting to enter a field where one would feel misplaced because of cultural norms, but perhaps potential judgement could also have been a factor. Another factor that I had believed could influence women’s decision on entering the IT field is a male dominated work environment. Women could be anxious about being reduced to stereotypes such as “the caring mother”, which could affect their career development opportunities, or they could be anxious about sexual harassment by coworkers.

With regard to the employment of female It workers, I had presumed that IT companies/departments could also have an effect on women’s employment. Firstly, managers could be under the impression that women put in less hours than men and are less flexible. This had been assumed to be due to the work-life balance stereotype of women spending more time with their children and doing domestic work. Furthermore, men had been assumed not to be required to take the same duration of parental leave as women. Secondly, the way job posts are formulated had been seen as a factor that could discourage women from applying for jobs. This could be in terms of masculine language or overemphasized performance or skill requirements. The same applies for job interviews, where inappropriate questions or inappropriate behavior also had been presumed to be a problem.

Thus, I have had prior knowledge on subjects, which have inspired and presumably influenced the

research for the thesis. The aim of the research process was not necessarily to explore these subjects

in particular to gain deeper insights, as they were not necessarily significant or at all influencing on

(30)

29

the underrepresentation of women. However, as previously mentioned, when interpreting according to Gadamer, one cannot avoid having a pre-understanding of the subject, which means that the knowledge I acquired throughout my research process, to a certain degree, was affected by my pre-understanding, as a researcher of the subject. In the section on data collection, this will be explained more in depth.

3.2 Methodology

This study follows a qualitative research approach in the form of hermeneutic phenomenology.

According to Valentine et al. (2018), the purpose of phenomenology is to understand how phenomena emerge as lived in the world. For phenomenologists, "phenomena are the ways in which we find ourselves being in relation to the world through our day-to-day living" (Vagle, 2014, p.20 as cited by Valentine et al., 2018). The way that phenomena come to be understood is by investigating "the ways in which meanings come to be in relations" (Vagle, 2014, p.12 as cited by Valentine et al., 2018), and Hermeneutic phenomenology assumes one will supply meaning to a phenomenon in order to understand it. The goal of a hermeneutic phenomenological analysis is to interpret how the knowledge of a participant relates to the participants, the phenomenon, and the research as a whole (van Manen, 2014 as cited by Valentine et al., 2018).

A hermeneutic phenomenological approach to methodology means that I focussed on people's perception of the world, or in other words, I studied a phenomenon as experienced by other people (Sloan et al., 2013). In doing so, I applied the hermeneutic circle as a phenomenological method, for which I moved back and forth between my pre-understanding and newfound understandings derived from my interpretation of the experiences of other people. While some phenomenologists remove themselves from the analysis through bracketing, I conformed to the method of bridling, where I restrained but acknowledged my prejudice when understanding the phenomenon, in order to become more reflective and open in my interpretation (Valentine et al., 2018).

Using hermeneutic phenomenology as a methodology, I needed to report on findings through data

analysis. In this case, findings came in the form of a collection of meanings derived from individual

(31)

30

experiences as mentioned by van Manen (2014). Said meanings are typically in the form of attributes ascribed specific experiences related to a phenomenon. Hence, hermeneutic phenomenology “reduces a human subject’s experiences with a phenomenon to a description of its

‘essence’” (Sloan et al., pp. 1293). I, as a qualitative researcher, therefore, proceeded to identify the phenomenon of a gender gap in the field of IT through women's experiences of the phenomenon. I commenced my study by conducting exploratory research with the goal of gaining insight on the topic and obtaining the knowledge necessary to understand the problem areas surrounding it.

3.3 Data collection

Both primary and secondary data were collected for this study. Primary sources include 11 semi - structured interviews supplemented with secondary data in order to get a broader perspective on the phenomenon. Secondary data was mainly in the form of official reports, scientific articles that complemented the primary data collection in making sense of the main perspectives through different scientific- and societal angles.

3.3.1 Interviews

According to hermeneutic phenomenology, as previously mentioned, the researcher’s unavoidable pre-understanding forms the foundation for their perception of a phenomenon (Sloan et al., 2013).

Realizing that my pre-understanding would affect the way I select and interpret knowledge, it was

imperative for me to keep an open mind when it came to different understandings, which was best

achieved through semi-structured interviews. The collected empirical data through interviews, can

aid a researcher in becoming aware of his/her own pre-understandings, while at the same time

creating new understandings on the phenomenon (Sloan et al., 2013). Semi-structured interviews

are ‘non-standardized’ interviews, where the researcher will follow a list of themes and key

questions during the interview. However, that does not mean that questions are fixed and stay the

same between interviews. Questions may vary given the context and interviewee and the order of

questions may vary depending on the flow of the conversation (Saunders et al., 2012). I had

prepared a list of about 22 primarily open-ended questions to keep the conversation on track and

THE GENDER GAP IN THE DANISH IT SECTOR