Diversity in Data Science Working Group

Our Diversity in Data Science working group aims to reduce barriers, and in doing so to increase opportunities, for underrepresented groups in data science and AI. We ask our members to become advocates for others with who might have less privilege and support one another to create a diverse community, which will reduce structural bias in AI training data and optimise our potential to generate innovative solutions to the challenges of our time.

The launch of this group coincides with the 10-year anniversary of DSI – “Decade of Data Science” - as a time to take leadership and action on these issues.

A recent report by the Alan Turing Institute highlighted that only 22% of professionals in data science and AI are women and that women make up only 18% of users across the largest online global data science platforms. Many other groups are likely to be even more strongly underrepresented, and there is likely to be significant intersectionality, yet the data to analyse these trends are extremely limited. We strongly encourage the data science community to learn about these issues (see resources below).

Resources

• Alan Turing Institute “Women in Data Science and AI” Theme
• LGBTQ+ STEM “Improving LGBTQ+ visibility in Science, Technology, Engineering and Mathematics
• 500 Queer Scientists: “We want to: ensure the next STEM generation has LGBTQ+ role models; help the current generation recognize they’re not alone; create opportunities for community connections and greater visibility within STEM.”
• Black History Month + Data Science: collection of articles via the Harvard Data Science Initiative
• Diversity in Artificial Intelligence (divinAI): “The goal of divinAI is to research and develop a set of diversity indicators, related to Artificial Intelligence developments, with special focus on gender balance, geographical representation and presence of academia vs companies."
• Hofstra et al (2020) The Diversity-Innovation Paradox in Science. PNAS 117:9284-9291. “Our analyses show that underrepresented groups produce higher rates of scientific novelty. However, their novel contributions are devalued and discounted.”

This collection is a work in progress. Please contact J.carradus1@lancaster.ac.uk or sally.keith@lancaster.ac.uk if you have any additional resources that we can add, we would really appreciate it.

Download Decade of Data