By Louie Meyer, Junior Researcher


Take a moment to envision your expectations and let the impressions settle.

Were your visions based on circumstances in your own field of work? On previous experience or future aspirations? On the latest science fiction movie or something entirely different?

One of the most well-known and powerful fortune tellers of our time is data science.

When we ask data science to depict the future, the answers are inevitably based on data. Situated, collected, and processed data. But who is in fact fueling this great power of prediction?

Data science is, both historically and currently, known to be represented by a privileged demographic group, predominantly described as straight, white cisgender men (D’Ignazio and Klein, 2020). The lack of diversity in the field often, unintendedly, excludes proper representation and perspectives from minorities, which greatly influence the way we perceive the past, the present and the time to come (D’Ignazio and Klein, 2020).

With my research project in the ETHOS Lab, I aim to explore these power structures focusing on gender gaps within the field of Science, Technology, Engineering, and Mathematics (STEM). For the current iteration of my project, I have decided to focus on data from the World Economic Forum’s Global Gender Gap Report 2021, more specifically from the section ‘Gender Gaps in Jobs of Tomorrow’. With this blogpost, I invite you to follow my findings, reflections, and challenges.


Gender Gaps in Jobs of Tomorrow

During the years 2019 and 2020 the World Economic Forum (WEF) collaborated with the LinkedIn Economic Graph Team to gain insight into emerging jobs, based on real-time labor market data. Through the research, 99 consistently growing roles were defined and divided into eight job clusters namely; Cloud Computing, Engineering, Data and AI, Product Development, Sales, Marketing, People and Culture, and lastly, Content Production (WEF, 2021 p. 59).

Through the research, the WEF (2021, p. 61) found that gender gaps are more likely to appear in clusters that require ‘disruptive technical skills’. These clusters include Data and AI, where women make up 32 percent of the workforce, Engineering in which women represent 20 percent of the workforce, or the field of Cloud Computing for which women make up a staggering 14 percent of the workforce (WEF, 2021 p. 61).

According to the WEF, the growing work clusters can serve as a significant indicator for the prospective gender composition within these fields. As an approaching STEM graduate, I look into my future industry with wonder. What constitutes the significant gender gap in STEM related fields? Who is predicting and portraying this future? Why is gender continuously narrated as a binary distinction in datasets? Are the predictions, in fact, self-fulfilling prophecies, or is there some way to challenge the status quo?


Can We Use Data to Remake the World?

One of the quintessential questions asked by the researchers Lauren Klein and Catherine D’Ignazio (2020) is how we can rewind oppressing power to create a more just present – how can we use data to remake the world? In the book Data Feminism, D’Ignazio and Klein (2020) present seven principles as guidance to utilize intersectional feminism in data science. During my research I have sought inspiration across these principles. In the current phase of my process, I am, however, especially engaged with the principles ‘Elevate Emotion and Embodiment’ and ‘Rethink Binaries and Hierarchies’ as they help me decode the Global Gender Gap (GGG) report even further.

The principle ‘Rethink Binaries and Hierarchies’ highlights the power of counting and classification and encourages inspection of the – seemingly hidden – hierarchies in datasets. D’Ignazio and Klein (2020) have proposed questions that help unveil the binaries and hierarchies, questions such as; Who is represented and who is not? Who is misrepresented? What assumptions have led to the classifications? And so forth.

In addition to this, the principle ‘Elevate Emotion and Embodiment’ challenges the dominant perspective that emotions contradict reason and argues that rebalancing this will allow us to broaden the way we communicate and understand data. In the following sections I unveil some reflections regarding the data in the GGG report through the lens of these two principles.


Which Genders Constitute the Gap?

Based on the GGG report, the future of gender parity is solely evaluated through a binary gender construct. According to the World Health Organization, gender intersects with, yet is different from the three biologically defined sexes, namely intersex, female and male. While the understanding of gender is continuously evolving it is often categorized as female, male or nonbinary (Wamsley, 2021). In addition to this, more than 70 gender identities have been categorized (Allarakha, 2022). And so, it seems that a binary construct neither represents a person’s sex, gender or gender identity. Which brings me back to one of my wonderings – why is gender continuously represented as a binary distinction?

For decades, gender has served as a primary example of a boolean data type for infrastructural systems, government-run systems, marketing algorithms, and educational materials (Broussard, 2021). During my own education at the IT University of Copenhagen this has likewise been the case. From a computer science perspective, the memory space and efficiency required for a boolean data type far surpass that of other data types. As Broussard (2021) states,

“A Boolean for gender, rather than a free text entry field, gives you an incremental gain in efficiency. It also conforms to a certain normative aesthetic known as ‘elegant code’.”

But how come we hold on to the boolean data type if it doesn’t comprehend the reality of gender? According to D’Ignazio and Klein (2020) it inevitably has to do with power, who has it and who doesn’t. According to Broussard (2021), “it’s a matter of will and funding—and in computer science, those can be in short supply when it comes to recognizing the not-so-binary world we live in.”

As opposed to continuing my investigation of ‘why’ the binary gender representation is dominating datasets, I have decided to continue by investigating how I might challenge this practice. This will be one of the main focal points for the continuation of my project in the new year. 


The Future According to Whom?

According to the WEF, the global gender gap in ‘Economic Participation and Opportunity’ is estimated to be closed in 266 years – YIKES!

As a reader I feel a growing sense of urgency from the information presented in the report. Yet, the data is consistently portrayed through conventional, ‘neutral’, histograms and scatter plots – see example below. What do you read from these data visualizations?

I see illustrations of eight growing work clusters focusing on gender composition from different angles. I see female representation as the contrary to male representation. ‘Disruptive technology skills’ as the contrary to female representation – what does that imply for female skills?

In the data visualizations, I do not see the nuances of gender representation, no hint of emotion or embodiment. Despite my own feelings being evoked by reading the report, I experience a subtle distance to my emotions while looking at the ‘neutral’ and ‘objective’ data visualizations. According to D’Ignazio and Klein (2020) the aim to deviate from emotions and instead present data in ‘objective’ and ‘neutral’ manners are rooted in the practices of statistics (amongst others). With reference to the statistician Karl Pearson, they describe the dominant perspective as follows;

The more plain, the more neutral; the more neutral, the more objective; and the more objective, the more true” (D’Ignazio & Klein, 2020 p. 76).

In contrast with this perspective, several feminist researchers have questioned the actual neutrality and objectivity in this practice and argues that “an ideal knowledge situation is one in which neither ethics nor emotions are subordinated to reason.” (D’Ignazio and Klein, 2020, p. 96).

With inspiration from data feminism, I aim to spend the coming months experimenting with emotional end embodied visualizations of data from the GGG report. I hope to be able to invite you to interact with visual and/or technically implemented prototypes come May 2023. Perhaps with this approach, we can envision a different future?



Allarakha, S. (2022). What Are the 72 Other Genders?

Broussard, M. (2021). When Binary Code Won’t Accommodate Nonbinary People.

D’Ignazio, C., & Klein, L. F. (2020). Data feminism. MIT press. 

Wamsley, L. (2021). A Guide To Gender Identity Terms

World Economic Forum. (2021). Global Gender Gap Report 2021.

World Health Organization, Gender and Health.