Women in Data Science: 4 Perspectives

Data science and analytics students are in line for some of the most exciting and rewarding job opportunities in the country. Despite these opportunities and the growing demand for analytical talent, women are currently underrepresented in STEM-related careers. According to Francine Berman, Professor of Computer Science at Rensselaer Polytechnic Institute, “… only 13% of the engineering workforce and 25% of the computer and mathematical sciences workforce are women.”

Recent data suggests that these figures are changing, though. In statistics, “… women are a growing force. More than 40 percent of degrees in statistics go to women, and [women] make up 40 percent of the statistics department faculty poised to move into tenured positions.” In addition, many groups are helping usher in change. For example, the American Statistical Association launched “This is Statistics,” a campaign that encourages young girls and minorities to pursue careers in big data.

With this in mind, we spoke with four data scientists in order to hear more about the current landscape of opportunities for women in big data. Read on to learn about our participants – Claudia Perlich, Lillian Pierson, Jennifer Priestley, and Sarah Aerni – as well as their experiences working in the field, what it takes to be successful, and their advice to students.

Which women in data science (or tech in general) inspire you? Why?

Claudia: Brenda Dietrich was one of my earlier mentors. She led the math department in the TJ Watson IBM research lab when I joined the predictive modeling group there in 2004, and since then she has held various IBM SVP positions. To me, she embodies the success of a deeply skilled technical expert (in optimization and data) in a corporate research environment, where she took on a leadership position not so much (as she confessed to me) because she desired to manage, but rather because the alternatives would have been much worse (my words) for her and all the people who ended up working for her. She is extremely well regarded by both the people working under her and IBM leadership, and she has high external visibility. (Fast Company’s 100 Most Creative People list is one of the recognitions we have in common.) During her very successful career she has also been one of my role models – she has raised four children and was one of the first to congratulate me on the birth of my son months before I even joined IBM research.

I met Kate Crawford two years ago during a two-week stay at the Rockefeller Bellagio center. Kate is inspiring in her brilliance, eloquence and passion for addressing some of the less sexy sides of data science, such as privacy concerns or threats to equality caused by algorithms invading our everyday life. Her career is astonishing in terms of both academic and political influence. She has full-time roles at both MIT and Microsoft Research. I cannot hope to do justice to Kate’s complete curriculum vitae here (Wikipedia might), but on a very personal level, I love Kate for simultaneously challenging me to think a bit harder about how what I do may change the world for the better while being willing to be open and listen to my arguments. She is raising a beautiful baby boy amidst all of her professional travel and is another example of a woman with an unbridled “can-do” attitude.

Lauren Moores is a very good friend and colleague at Dstillery who has taken a very different path in her career – she entered the world of data after completing a Ph.D in economics after having taken some time to raise a family. She calls herself a “digital data geek and strategist” who loves “creating stories from dirty and disparate signals and mentoring others to ‘speak’ and think data.” There is nothing better than seeing Lauren in action connecting worlds and creating a tapestry of stories using the backdrop of data. At this point, we share a passion for speaking externally on the facets of emerging data science and for being involved in the creation of the next generation of “geeks” teaching courses at NYU.

Lillian: I am deeply inspired by Sara-Jayne Terp, Heather Leson, and Justine Mackinnon. They inspire me because they’ve devoted their lives and careers to using data science and/or technology to help people that are in crisis situations in under-privileged nations.

Jennifer: Greta Roberts at Talent Analytics and Radhika Kulkarni at SAS.

Sarah: My manager, Hulya Emir Farinas, is easily one of the most talented and inspiring people in data science (male or female). She is incredibly brilliant, both technically and with her business acumen. She has both inspired and greatly educated me. Beyond that, what I find wonderful is that almost any list of top data scientists will contain several women. I feel this cannot be said for many technical fields, so it does not feel like a boys club!

To what do you attribute your success? What motivates you?

Claudia: I am very fortunate to have entered the field of data science early on under the mentorship of a number of great teachers/friends who shared not only the skills and knowledge about data science but also a passion for discovering the hard-to-discern secrets of data and a desire to be practically relevant. My education combining computer science and business gives me an advantage in solving data problems and modeling challenges within the context of business strategy, leading to more impactful data solutions. Additionally, being involved on the business side allows me to be a better and more comprehensive public speaker by delivering data science concepts in the context of strategy. The combination is what has allowed me to be a spokesperson for both Dstillery and the data science industry in general. I remain committed to participating in both worlds – academia and industry – culminating most recently in my role as General Chair of KDD 2014 in New York City, which had a record of more than 2,500 attendees.

There are two things that really motivate me: solving hard challenges and being able to help people. Data science has given me a framework for doing both in good measure. The potential of what can be done with data is tremendous, and having the skills to bring data to bear has allowed me to contribute to addressing a wide array of business opportunities. At the same time, data science is challenging on many levels:

  • With every dataset, I feel like I enter new territory, and I need to spend some time getting familiar with the environment. That means that I have to learn a lot about the new domain and the data collection process.
  • Data always has problems (dirty little secrets). Data review is like a scavenger hunt of letting your intuition guide you when something just does not look right until one either learns something new about the domain or discovers a problem in the data collections that needs to be fixed.
  • The most challenging problem is often trying to figure out what the true problem is that the asker is trying to solve. I tend to need to clarify through many pointed questions and context and ultimately end up in the role of translator between the data and the problem.

Lillian: I attribute my success to determination, hard work, and the gift of a great education. I am motivated by a drive to achieve my goals while helping others in the process.

Jennifer: One word: Students. They are why I come to work everyday. Data science is an explosive and exciting field, and, as we all know, the talent gap is big – and I would argue getting bigger. Helping young students develop the skills necessary to become data scientists and watching them get their first jobs (typically with multiple offers to choose from) is what keeps me in academia. And specifically at our university, many of the undergraduates are first-generation college students, which means that when they complete their education and secure that high-paying data scientist salary, it really makes them the most successful person in their lineage. It’s beautiful to be a small part of that journey.

Sarah: I can say that my curiosity and excitement about data certainly motivate me. When presented with a new dataset, I am extremely greedy in trying to understand what the data is saying. Yes, I validate my findings with experts and ask for guidance, but I’m always so intrigued by what the data confirms about what we already know and finding out what it can tell us that is new. I don’t think I will ever tire of the moment when I build a model that transforms data and crunches numbers, and see that it indeed does have the power to predict something. How amazing is that?

How are the opportunities for women in data science? What do you think the future holds?

Claudia: I see what I would consider a healthy participation of women in the field. They often bring a different intuition to the table and are rising easily into leadership positions. At Dstillery, a digital advertising technology company of 175 people, half of our core data science team are female PhDs from very diverse backgrounds. Similarly, I consistently see nearly half female students taking my MBA course on data mining for business intelligence at NYU. Ultimately, data science is another technical field where women remain statistically a minority, but I do not believe that we need to force the issue and “fight” for a higher female quota. I want to come to work and do what I love and be recognized for what I bring to the table and not waste even one thought on the fact that I am female. Most successful women I know in the field seem to have this attitude and are very comfortable with themselves and their roles.

Lillian: The fields of data science and analytics are absolutely exploding with opportunity. So, for men and women alike, there are plenty of opportunities. The issue for women in this field is getting them trained in STEM areas. We need more women in STEM, period.

Jennifer: I think the current landscape is rich with opportunity – regardless of gender. The trend that excites me the most is that this is really the first career path that is truly interdisciplinary – data science is applicable to biology, chemistry, psychology, marketing, finance, economics, etc. – so courses in applied statistics and applied computer science are filling up with students, again coming from disciplines where we have not previously seen many students, such as psychology, marketing, finance, economics and sociology. This makes for a particularly rich learning environment where students have to solve problems and work in groups with people who think very differently from them. When an engineering student, a chemistry student and a marketing student all have to solve the same analytics problem, they are learning important latent skills related to teaming, communication and project management.

Sarah: I would say that luckily the opportunities for women in data science are the same as those for men. This field benefits from diversity in thinking. There is rarely only one way to approach a problem, and new ideas and input can only lead to more variety in approaches and potentially better outcomes. Not just that, but the consumers of data science exist in every industry, and therefore require us to interact with a wide variety of businesses and individuals. Beyond that, it is a field that values a variety of skills, deep technical abilities, hacking, storytelling, visualization and constant learning. This is something I value, and I think it encourages diverse backgrounds as well.

Are there any disciplines associated with data science that seem to attract more women? Why do you think this is?

Claudia: I do not necessarily see a difference within the field of data science, but there seems to be some difference in how people find their way into data science. I see amazing female talent coming from fairly remote areas, such as medicine, biology and the life sciences, in addition to the earlier mentioned business graduates. My experience is that female data scientists often come from diverse backgrounds or specific application domains, whereas many male data scientists have a more direct path into data science having started out in the more technical areas of CS/math or statistics. My personal take is that many women feel very passionate early on about a particular domain and discover their love for data (and its usefulness) in this context. In the end, having data science teams with both trajectories is a true benefit.

Lillian: Well, data science for the environment and social good always tends to attract a proportionally higher number of women. I think that is because, as women, we are raised to care and nurture. I don’t know if that is an evolutionary trait that evolved from bearing and raising offspring, or if it’s a cultural expectation that was thrust upon us, but women tend be motivated by a desire to help causes.

Jennifer: In our own university, psychology majors make up about 30 percent of all students who study applied statistics as a minor field of study. This is likely true because a) if a student wants to pursue an advanced degree in psychology the programs are limited in number and highly competitive, so advanced study in statistics becomes a great point of differentiation, or b) if a student wants to pursue a career in the private sector with a BS in psychology, the combination of a statistics minor makes them far more marketable. The unintended consequence of this is that because psychology (at least at our university) over-indexes females, this is increasing the number of women we are seeing in the applied statistics courses, which for some becomes a “backdoor” entrance into data science.

Sarah: Based on the resumes I have seen, many women enter data science from applied math, statistics, and biomedical domains. I imagine it has to do with the ability for women to see other women in these disciplines, and it’s possible that women are nurtured in these environments. I think it is crucial to avoid feeling singled-out or like an outlier :)


As a result, the number of women who ultimately choose to become data scientists in these fields can be sustained, if not grown. In our own data science team at Pivotal, nearly half of the members are female!

What can be done to encourage more women to pursue data science?

Claudia: Many of the women I meet just lack a bit of confidence and need a little nudge to believe in their abilities. Meeting them in low-key social events, such as meetups or specific career advice sessions, has been very rewarding. Coming from other fields, many junior women doubt whether their coding and stats skills are at par to “qualify” for a data science position. But data science roles come in many flavors, and not all require master’s level CS knowledge. In fact, many senior data scientists in my network recently agree that coding tests are not necessarily a requirement for DS interviews. At Dstillery, we have five female PhDs in data-related positions coming from very diverse fields: information systems, physics, neuroscience, economics and genetics. So on one side, I would encourage junior women to get their feet wet and just start working on projects either related to their work/school or just on their own looking at volunteering opportunities or things like Kaggle competitions. On the other side, I would like to encourage companies to be more risk-taking in their hiring decisions and embrace diverse backgrounds. Additionally, many of the established women happily take on mentorships or speak at events, such as Grace Hopper reaching out and encouraging the next generation.

Lillian: Women need to understand what opportunities are available to them, what those opportunities involve, and what the quality of life looks like for someone in this role. If they had a solid understanding of these, they’d be encouraged.

Jennifer: I think for both genders, the promise of employability is a huge draw. Frequently, when I speak to large groups of students, I like to say that the US has an employment problem… we have too many jobs and not enough people in data science. That usually gets their attention.

Sarah: I think many of the existing programs and meetups out there that focus on bringing women together can help encourage more women to pursue DS. I think the best source of encouragement is from women coming together to support and encourage each other. Having positive role models helps, but certainly knowing that we can elevate each other is also important.

What types of challenges do women (but not men) in data science face?

Claudia: I honestly do not see specific challenges that would not apply to other fields or to men as well. In fact, I think that most of the challenges of building a successful career in DS are the same for both genders. One component that is often overlooked but very important is communication skills with non-DS folks. This includes both the ability to explain DS as well as ask the right questions to deeply understand what the true problem is that needs solving. That might mean that you have to question or even contradict senior management in different business units. That requires political acumen as well as a lot of confidence. I think as women we have a leg up to not appear threatening, but at the same time, we might need to fight a bit harder to be taken seriously. So confidence is key!

Lillian: Women constantly face gender discrimination based on how we look and what we wear. It’s endemic. Employers are even legally allowed to force women to wear makeup, although men cannot be required to do so. We are operating in a system that promotes the underlying message that at least part of our value is in how we look. But then, if a woman is too pretty, people will want her to do extra things to look less attractive. We face a huge control issue, where society (even including other women) is continually judging and controlling us based on our level of attractiveness. It’s pretty lame and irrelevant, IMHO. This is true for data science as it is for any other field, but in data science there is more pressure because there are less women.

Jennifer: Street cred. I don’t know why this is, but there is a perception that if you are a female with a manicure and wear jewelry, then you must not know how to program or understand advanced analytics. But a well-groomed guy carrying a cappuccino cup gets immediate respect. I don’t get it. I think that stereotype is slowly dissipating.

Sarah: Again, it is tough for me to generalize. I do not want to assume that we are stereotyped, but I have certainly had my awkward moments. That being said, I imagine my male counterparts have also experienced many such moments for a variety of reasons. Personally, I know that I need to avoid the tendency to fixate on why I may have been singled out, or reading into comments too deeply and assume they are because I am a woman. I think it’s easy to allow yourself to feel that something has happened because you are woman. My approach is certainly not to try to dress differently or act a certain way. That wouldn’t be successful because once I am revealed as an imposter (not as a woman posing as a man, but rather just not true to myself), I will only feel worse. I think it’s important to celebrate your differences, and not become defensive and jump to conclusions. I am happy to say, though, that such occasions are EXTREMELY rare.

How important are mentors for female professionals? Did you have any mentors that had an impact on your success?

Claudia: Mentors are one of the keys to success for anybody. But mentorship does not have to be a formal program and often does not even feel like mentorship. I have had the luck, and maybe good judgment, to work for people who ultimately turned out to be great mentors. I was told that I do not have a huge appreciation for authority. That is not to say that I do not respect authority, but that I perceive my boss primarily as a collaborator and often a friend who has my best interests in mind. I am sure that this is not true in all managing relationships, but it was in mine. As a result, I was encouraged to grow into the best I can be. The list of my mentors includes Dstillery CEO Tom Phillips, my manager at IBM Richard Lawrence, the head of the math department at IBM Brenda Dietrich, my PhD advisor Foster Provost, and my first professor in DS (back then it was called artificial intelligence) Andreas Weigend. Today, I would count all of them as friends.

Lillian: Mentorship can be good. I had some mentors, I suppose, but really, I believe in self-determination. If I want something, I go after it. I believe that if I ever depended on another person to help me get stuff done, then I never would have made it.

Jennifer: Mentors are definitely important. To find someone who will look out for your best interest when you are not in the room, someone who will praise you publicly and then pull you behind closed doors and tell you all the things that you did wrong, and then take you out for coffee … those people are golden. Indispensable. I did have mentors – all of whom were men. So, I don’t think that mentors for young females have to be female – just someone who is willing to provide support and guidance.

Sarah: As with any individual, I think it is critical to find a mentor early on. One of the main challenges is that mentors are most effective when they can see themselves in those they are helping. I actually do not believe that this means women should only be mentored by women. Instead, I think it has to do with the path that you have taken, so mentors can recognize and help with challenges they faced themselves. I have had numerous mentors in my career, both academic and professional. They were not always in my own discipline, and I worked for some and not for others, but they have had such a deep and lasting impact on my life and career. I think they can be found in very unexpected places if you allow yourself to listen to what they have to say.

Any other thoughts?

Claudia: You have to find an environment that values you for who you are, what you bring to the table, and what you are good at. To build a career, you first need to find yourself and what it is that makes you valuable and special. This process takes time and a supportive environment. So in the beginning, go and find a place that feels right and work with people whom you like and from whom you can learn. The big wars can be fought later in your career when you are ready.

Sarah: Of course I believe that individuals are seen for their capabilities, and when we present our work, whether it is to women or men, we are viewed with equal credibility. This does not necessarily mean that our audience is blind to our differences. I think every individual’s differences are something to celebrate, something to make us unique and memorable, and above all, not something to overcome or to be viewed as weaknesses. Instead, I think we must all remain confident in our work and who we are. If we give credibility to the idea that our differences are something that should be hidden, then we are asking the same of others. Instead, our unique qualities might be something that connects us to the person we are working with, presenting to, or trying to help, and makes us that much more successful.

Source: Women in Data Science: 4 Perspectives

Via: Google Alert for Data Science

Pin It on Pinterest

Share This

Join Our Newsletter

Sign up to our mailing list to receive the latest news and updates about homeAI.info and the Informed.AI Network of AI related websites which includes Events.AI, Neurons.AI, Awards.AI, and Vocation.AI

You have Successfully Subscribed!