Visit the dedicated Women in data science and AI hub; a place to connect women in DS and AI with resources, news and research, and gather feedback about the needs of the community. The page below contains the background to the Women in Data Science and AI project.
Data scientists and AI professionals are in great demand. Machine learning and data science are now the fastest growing professions in the US, and as our ability to collect and analyse data improves, demand for data scientists will continue to increase. By 2020, more than 2.7 million data scientist job openings are forecast to be advertised in the US alone.
The explosive growth in data science and machine learning roles hides a problematic dynamic: women occupy only a minority of these new positions. In the UK, women represent 47% of the work force, yet they hold less than 17% of all available tech jobs. Although there are no official statistics on the percentage of women in data science or AI roles, there is growing evidence that the gender imbalance that affects the tech sector extends to data science and AI, as well. Many researchers have underlined the benefits of gender equality in the workplace. The presence of women increases a group’s problem-solving abilities (Woolley et al. (2010)) and drives innovation (Sastre (2014)).Gender diversity is also associated with higher sales revenues, larger numbers of customers, as well as greater relative profits (Herring (2009)).
As the national institute for data science and artificial intelligence, The Alan Turing Institute is committed to redressing gender inequality in these fields. We are setting up the women in data science and AI research project as a step towards fulfilling that responsibility. The Turing will contribute towards tackling the gender gap by doing what it does best: using data science and AI to produce research that informs our understanding of the issue. Research, on its own, is rarely sufficient to tackle such a prevalent problem, so we will use the Institute’s convening power to turn our research insights and recommendations into concrete policy measures aimed at increasing the number of women in data science and AI professions. We firmly believe that it is time to address the connection between the lack of diversity in the AI community and bias in the technical products that are produced by the community.
Header image: Margaret Hamilton, American computer scientist, systems engineer and business owner, in 1969, standing next to listings of the software she and her MIT team produced for the Apollo project. Source: Draper Laboratory; restored by Adam Cuerden. [Public domain], via Wikimedia Commons
Explaining the science
The gender imbalance cannot be redressed without understanding the reasons why women are underrepresented in data science and AI as well as what interventions can correct the imbalance. The women in data science and AI research project aims to examine systematically:
- Why so few women enter data science and AI professions
- Why, once they enter these professions, many women leave
- Which interventions work to increase the number of women in data science and AI
- The ways in which the gender deficit shapes both the research agenda and the applications of digital technologies
There are numerous studies and preliminary analyses that shed light on the problem; the project will rigorously examine and interrogate the claims made to map out the scientific evidence.
1 – Why so few women enter data science and AI professions
From a young age, girls feel discouraged when it comes to pursuing subjects such as mathematics or computer science. In 2012, the OECD surveyed the UK’s 15-year old students and found that 41% of the girls agreed with the statement ‘I am just not good at mathematics’, while only 24% of the boys agreed with it. In 2015, the OECD surveyed the country’s 15-year olds again and found that 4.6% of the boys expected to work as ICT professionals at age 30, while only 0.5% of the girls had the same expectations for themselves.
Data science and AI can help us understand how biases in media coverage and public perceptions influence girls’ lack of confidence in their mathematical abilities and their low expectations of pursuing careers in fields such as ICT. Data-driven research can also pinpoint how the educational system discourages girls from engaging with mathematics and the sciences from an early age. For example, there are countries where girls are more confident than boys in their mathematical abilities. The 2012 OECD survey found that in places like the UAE, Qatar, Jordan, Malaysia, Latvia, Bulgaria, and Kazakhstan, a higher percentage of teenage boys than girls agreed with the statement ‘I am just not good at mathematics.’ With the help of data science and AI, we hope to uncover why these countries do a better job at giving girls confidence in maths than countries like the UK.
2 – Why, once they enter these professions, many women leave
Apart from understanding why so few women pursue careers in data science and AI, we also want to identify the reasons behind the large number of women who choose to leave these fields.
Attrition is an enormous problem in tech. Survey evidence published by the Center for Talent Innovation shows that 56% of women in private technology companies leave their organisations at the mid-career point (10-20 years). Of the women who leave, 51% abandon their tech training altogether and move on to other occupations.
Sexism, bullying, and sexual harassment are clear contributors to the high attrition rates of women in data science and AI professions. The gender pay gap, slow career progression for women, male-dominated office culture, lack of access to mentors, and gender bias in hiring are also discouraging women from continuing their careers in data science and AI. Brave, powerful women have been sharing their individual stories of workplace abuse and sexual discrimination for years. Our aim is to use data science and AI to bring these stories together, quantifying the extent of the problem, and motivating policy makers to act.
3 – Which interventions work to increase the number of women in data science and AI
Apart from uncovering the factors that deter women from pursuing studies and careers in data science and AI, we also want to understand what interventions work. Existing research highlights the fact that popular interventions meant to increase diversity, such as diversity trainings or women-only conferences, not only fail at improving the gender balance – they often make it worse (Dobbin and Kalev (2016)). The aim of the research we undertake within the women in data science and AI programme is to identify the interventions that shift the gender imbalance in the right direction.
4 – The ways in which the gender deficit shapes both the research agenda and the applications of digital technologies
In a data-science or AI context, the presence of women brings with it a crucial advantage: it ensures that these technologies and their applications are shaped by the experiences and viewpoints of women. A growing strand of research documents the fact that machine learning algorithms exhibit gender biases (Bolukbasi et al. (2016)), while a separate, but related, strand of the literature argues that questions addressing women-specific issues often get left out of research agendas (McCarthy et al. (2017)). Increasing women’s participation is the only way to ensure that their perspectives and priorities will inform the insights that data scientists will generate, the algorithms that they will build, as well as the research agendas that they will define.
Our aim is to redress the gender imbalance in data science and AI. Digital technologies are changing the way in which we live our lives and it is imperative for women to be equal partners in developing the algorithms, setting the research agendas, and building the applications underpinned by data science and AI.
We want to make data science and AI attractive fields for women and to do that, we need to work alongside policy makers. The research we conduct under the women in data science and AI project has the ultimate goal of identifying the measures that policy makers need to adopt in order to encourage women to pursue careers in data science and AI, as well as to improve the workplace experiences for women already working in these fields.