Experian data scientists, Theresa Offwood-le Roux and Berenice Pila Diez dissect the barriers and opportunities for women in data.
Data science has been called the sexiest job of the year as early on as 2012. With FTSE 100 companies and start-ups alike investing time and capital to leverage data for decision making, data scientists are in demand, but is this demand gender-blind?
Data-led businesses make better decisions, build better systems and create sustainable and profitable companies in the long-run. The business case for data analytics is strong, and so is the case for diversity in hiring, particularly in technology.
Women now make up 40 per cent of graduates with degrees in statistics, which is largely regarded as a launch pad for a career in data science. But the talent pipeline for many sectors within technology often lack enough women to enter and grow in these niche fields.
As data scientists and part of Experian’s UK DataLabs team, Theresa Offwood-le Roux and Berenice Pila Diez are no strangers to the trials and tribulations of being a minority in a largely male field. But the joy of working with numbers isn’t lost on Offwood-le Roux.
“There is so much data in the world nowadays, there will be many insights that can be gained from this data and predictions that can be made to help people. To be at the forefront of the new challenges that this brings is very exciting,” she says.
Pila Diez believes data analysis can take on pretty much any modern problem. “Look at what’s going on in the media at the moment around “alternative facts” and “fake news”. Data helps provide clarity and allows you to check opinions and well-crafted arguments against facts, it encourages you to test and explore things rather than take what you are presented with as a ground truth,” she says, outlining what drew her to the field.
A happy accident
Like many women in technology, both Offwood-le Roux and Pila Diez fell into their careers by accident. In her life before data science, Offwood-le Roux did her PhD in financial mathematics in South Africa, which led her into a career in banking as a quantitative analyst. She transferred her skills to data science, intrigued by the relatively new field, and the chance to be part of “the big picture” while still using mathematical and technical skills.
For Pila Diez, her area of expertise was Physics, which she studied in Madrid before moving to the Netherlands to get her PhD in astrophysics.
“It was during that time that I realised I wanted to work in a faster-paced environment than academia and in projects that would have a more tangible impact. However, I was keen on finding a career path in which the scientific method and the data component would be major parts, so data science came as the perfect blend,” she explains. After finishing her PhD, Pila Diez joined one of the big four consultancy firms as a data scientist assistant manager.
The STEM talent pipeline
Despite cross-sector initiatives to drive diversity at senior levels, the vast majority of CIOs are still male, middle-aged and STEM degree-educated. Looking at the talent pipeline, nearly a third more women than men go on to study at degree level in the UK, but in some sectors, like computer science, the ratio of male to female workers has increased dramatically over the past decade. Men now outnumber women almost five to one, a 40 per cent increase from the class of 2005/6, according to figures from the UK Commission for Employment and Skills (UKCES).
“I think there are various reasons for this. It could be that men are naturally more inclined in these fields, like women are more dominant in others. If, however, it is due to girls being influenced into thinking that they cannot do it, then this must be addressed,” says Offwood-le Roux.
For Pila Diez, it’s an issue that goes even deeper. “From my own experience I think it’s mostly a cultural issue in the way we educate children. What’s really interesting for me, having studied in both Spain and the Netherlands before coming to London, is that there are huge national divides. At my university in Spain the ratios were 50:50 between men and women every year, with no explicit incentive or enforcement for female quotas. When I was in Holland I could see a 10 per cent ratio in the bachelor and master levels, and a better ratio at the PhD level, but again due to foreign PhD researchers,” she says.
Research published in the Science journal found that girls as young as six start to believe that specific activities are ‘not for them’ because they think they’re not smart enough, and cultural stereotyping is seen as a big part of that. At the age of five they do not hold this self-deprecating biases. “This is very worrying and an issue I hope is being addressed in the classroom and at home today.”
Getting into data science
The best degrees to study, according to Offwood-le Roux, are computer science, mathematics, statistics or any degrees that teach both indirectly. “You need to be able to code and understand statistical concepts. If you are already in a different scientific career, look at the ‘bootcamps’ out there, they are a great way to get a feel for the industry and to create a good network,” she says.
To get girls into STEM, Pila Diez believes the effort must start early on at home and at school by breaking down the stereotypes that professions and hobbies are gender-specific. “We need to encourage young girls to try and fail – data science is not about getting it right the first time but about discovery and innovation. We need to encourage them to explore, to play with toys different from the usual – computer games, construction toys and logic puzzles – and to choose careers that they like regardless of whether they get the top grade of their class in the subject.”
The way the diversity agenda is portrayed in the media and business world is misleading to Pila Diez. “The real question we should be asking ourselves is what can a diverse workforce offer that a narrow one perhaps can’t, and the answer is easy: different approaches, different solutions, different views.”
A day in the life of a data scientist
Offwood-le Roux’s routine as a data scientists begins at 9AM when she gets to work. “I get right into continuing to solve whatever problem I am currently working on. This can involve data analysis, coding, brainstorming with the team,” she says.
At midday, the team as Experian’s DataLabs has a “standup” where each member takes a moment to explain what they’ve been doing and what they plan to do next, followed by a team lunch. The afternoon is spent as the morning analysing, coding, and so on until they call it a day at 6PM.
“Our role in the DataLab is to explore new avenues to bring the game to the next level: both by developing new products for Experian and by engaging with clients to leverage their data in combination with Experian’s and tackle their own unique challenges,” Pila Diez adds. “We do this by applying big data technologies and machine learning.”
Outside of work, both data scientists have varied interests, but it still involves learning more about the exciting prospects of tech innovation. “I love reading about ways that machine learning is making life easier. From the new Amazon Go stores (I want to know how they control this) to self-driving cars, although it will take time for the general general public to trust these,” says Offwood-le Roux.
“Self-driving cars are definitely very exciting in that they combine sensors, data processing, motion, machine learning and mechanical engineering,” Pila Diez adds. “However, what I would be most excited to see happen in the near future are advancements in energy production and transport engines because they would be linked to conceptual advancements in physics and chemistry.”
“There is a lot of advice I have been given throughout my career,” says Offwood-le Roux. “If I had to choose, don’t be scared to ask questions. Try to learn as much as you can about all parts of the business.”
For Pila Diez, it’s a matter of putting yourself out there, no matter the outcome. “Do not apply just to roles where you meet 90 to 100 per cent of the job specs: they are just a wish list, which is a way to say don’t rule yourself out before the hiring committee does. They are the ones who know the competition.”