OnlineEducation.com interviewed three leading women in data science and analytics to better understand the challenges faced by women in this field. We asked each of these experts about their own professional experiences, and about the advice they would offer to women who are considering a career in analytics.
Meta S. BrownMeta S. Brown was finishing a master’s in nuclear engineering when she realized that statistics and statistical analysis was her calling. Although it involved a transition from nuclear engineering to what was a less well defined field at the time, she moved into business analytics and has become a strong proponent for women who now work in what many perceive as a male-dominated field.
Cornelia Lévy-BenchetonCornelia Lévy-Bencheton had a traditional liberal arts education and an affinity for analytical problem solving when she graduated from Stanford. Despite a noticeable paucity of women in senior analytics positions, she entered the world of financial analytics, cultivated technical proficiencies on the job, and has become a leading voice for women in strategic marketing and web analytics.
Claudia PerlichClaudia Perlich found herself in the minority in her first computer science class as an undergraduate. But, her deep interest in machine learning led her to pursue a master’s in computer science, and a PHD in Information Science, which laid the foundation for a career in data science. She was part of the data science team at IBM’s Watson Research Center before taking over as Chief Scientist at the advertising and marketing analytics company Dstllery.
“The first thing that I would want people to know is that a woman in analytics is as common a creature as a man in analytics,” said Meta S. Brown, a Chicago-based business analytics consultant with 20 years of experience in data mining and predictive analytics, in an interview with OnlineEducation.com. Brown’s assessment is in part based on a series of blog posts she began writing in 2014 – Meta’s Binder Fulla Women in Analytics – in response to the perception that it had become difficult to find accomplished women in analytics to speak at data conferences. In Brown’s view, when we define analytics professions too narrowly, we significantly underestimate the number of women who are actually involved in data analytics. However, there are dissenting opinions. Cornelia Lévy-Bencheton wrote a book titled Women in Data and she works in analytics as a strategic marketing and communications consultant. She’s taken note of a potentially alarming trend among women in STEM fields that may carry over to women in fields like business and marketing analytics, data science, and other areas of analytics.
“When I started in this field, women were beginning to come on stream, having careers as well as families,” Lévy-Bencheton explained in a separate interview. “There was excitement and buzz around new challenges and chances to succeed. But, yes, women are underrepresented in the field of analytics and STEM. It’s more than an impression. Women are actively pursuing advanced degrees in higher education and are graduating at a higher rate than males. However, it is important to note that it is not the number or percent of degrees awarded to women that is an issue. Rather, the field of concentration or STEM subject is where there is a lag. According to the US Census Bureau, U.S. women made up 27% of STEM jobs in 2011 and 34% of STEM jobs in 1990. That is not going in the right direction.”
The percentage of women in computing and mathematical occupations dropped from 35% in 1990 to 26% in 2013 (AAUW)
In fact, it is difficult to find good data on the number of women employed in analytics, in part because there is no formal state licensing or standard definition for jobs like business and marketing analysts, data scientists, and other analytics professionals. What we do have are statistics and overall trend lines for STEM jobs and studies like the 2015 American Association of University Women report on Solving the Equation: The Variables for Women’s Success in Engineering and Computing. Among the AAUW’s conclusions is that the percentage of women in computing and mathematical occupations dropped from 35% in 1990 to 26% in 2013, which echoes the shift referenced by Cornelia Lévy-Bencheton. However, the National Science Foundation cautions that such data must be understood within the context of considerable across-the-board growth in computing and mathematics jobs. In their estimation, the number of women in those fields nearly doubled between 1993 and 2010, but those gains were offset by even larger increases in the number of men employed in computing and mathematics.
Beyond the numbers, there are issues of perception and work culture that permeate the discussion about women in data fields. Who do we think of when we envision a data scientist, and if that picture does not include women, what does that mean? Claudia Perlich, a data scientist who worked at IBM’s Watson Research Center for over a decade before taking charge of the machine learning operations at a NYC-based analytics firm, wrote a story for Wired in 2014 titled Women in Data Science Are Invisible. In that piece, she recounts her experiences as the general co-chair for the 2014 KDD Data Science for Social Good conference. The list of keynote and invited speakers she was presented included no women. “Not a single woman was being asked to speak at our biggest industry conference,” she wrote. “Not one.”
Counting the number of women working in the field of analytics today has everything to do with how you define an analytics job and what you call it. This has been a particular concern for Meta S. Brown in her advocacy and mentoring for women in analytics professions. “There are reasons why one job is called a data analyst, and another is called a data scientist, and another is called a statistician, and so on,” she explained. “Those reasons are as political as they are technical. Data science and data scientist are now in vogue. This is coming from the computing world. And the computing world has been pushing women out for the past thirty years. So, we’re letting the term be defined by people who are tech- and programming-driven, rather than people who have an understanding of business, communications skills, and what people tend to call soft skills. It’s just bad terminology. Technical communications skills are as valid a skill as any other technical skill.”
I found 262 women who had written books on analytics. If you looked through to see their job titles, many of them were academics, and if you asked these women what field they worked in, very few would have said, ‘I’m a statistician,’ or ‘I’m a data miner.’ Many of them would have said, ‘I’m a teacher,’ ‘I’m a psychologist,’ and a whole host of other fields… They identify themselves by the application. (Meta S. Brown)
When she was working on her blog posts about women in analytics, Brown discovered many women in the field who didn’t necessarily call themselves data analysts or data scientists. “I was trying to build awareness of highly qualified women who should be prospects for speaking at conferences, just to point out how many there are. I found 262 women who had written books on analytics. If you looked through to see their job titles, many of them were academics, and if you asked these women what field they worked in, very few would have said, ‘I’m a statistician,’ or ‘I’m a data miner.’ Many of them would have said, ‘I’m a teacher,’ ‘I’m a psychologist,’ and a whole host of other fields. Many of the most sophisticated statisticians I’ve worked with have had degrees in psychology or sociology. It’s similar to me studying nuclear engineering. I worked with statistical analysis, but when I finished my degree I would have called myself a nuclear engineer rather than a statistician. A lot of the people doing valuable, applied work in data analytics don’t call themselves data scientists and they may not even call themselves data analysts. They identify themselves by the application.”
Claudia Perlich had similar experience on her path to working as a data scientist at IBM and becoming the Chief Scientist at her current company, the advertising technology firm Dstillery. “The first time I ventured into the field of data was back in 1995,” she recounted in an interview with OnlineEducation.com. “There were plenty of words for data science, but it didn’t have the public recognition it now has as something interesting, important, or sexy. I just happened to stumble upon a course in artificial neural networks with no clue that it was about data. I wasn’t concerned with labeling what it was – I think labeling typically does more harm than good. I just really enjoyed what I was doing, which was learning about the world through data and solving the puzzles that data presented. Throughout that, I stuck to the narrowly, technically true definition of what I was doing, which was supervised modeling and, arguably, the next level above that would be machine learning. But, I didn’t really care as long as I enjoyed what I was doing.”
Again, what’s clear is that there remains a certain lack of clarity about what constitutes a job in analytics. The three analytics professionals we spoke to for this article are examples of very different career paths. For a computer programmer who specializes in machine learning like Perlich, analytics involves the highly technical work of designing algorithms and predictive computer models. Lévy-Bencheton got a lot of her analytics experience on the job after earning a business degree. And Brown applies the statistical training she got as a nuclear engineer to solving complex data mining and business problems. For others that Brown has spotlighted in her writing, data analytics includes a broad range of big data problem-solving activities, some more technical than others, in fields as diverse as finance, psychology, healthcare, and research in science and public policy.
If there is one clear message that the experts we spoke to want to convey to women considering a career in analytics, it’s that there are many paths to success. This is particularly true for education and training in the field. Before there were designated degree programs for data scientists and analytics professionals, a typical path to a career in these fields might begin with a degree in engineering, computer science, applied mathematics, statistics, or business administration. Or not. Claudia Perlich got a master’s in computer science, followed by a PhD in Information Systems from the Stern School of Business at NYU. Meta S. Brown majored in mathematics as an undergraduate at Rutgers, and then got her master’s in nuclear engineering at M.I.T. Cornelia Lévy-Bencheton studied French and Italian before getting her MBA, but points to her Digital Marketing Certificate as a “formative educational experience.”
However, Lévy-Bencheton also emphasizes aspects of her formal undergraduate and graduate work in literature. “Analyzing the ins and outs, themes, sociological, historical, and psychological implications and context of literature, as well as the engineering of novels, plays, poetry et al, is actually quite analytical,” she points out. “Of course, having a start in analytics for a future in the analytics field would appear to be good. I say ‘appear’ because a liberal arts/business background provides a greater context and platform for an analytics degree. I’d vote for a broader background at the bachelor’s level and then a master’s in an analytical discipline for later on.”
Meta S. Brown offers similar advice, to both women and men considering a career in analytics. “Here’s what I would say to anyone: if you have an interest in analytics, that’s not a one-to-one match to a graduate program in data science and a job with a title of data scientist. An interest in data analysis can be a pathway that leads into a variety of other studies and an even wider variety of careers… My view is that you should learn good transferable skills that can be adapted to a wide variety of work.”
Brown also stresses the importance of having a clear career goal. “If you desire a career in analytics, the first question I would ask is what job would you like to do when you graduate. It’s good to have a starting point. I would make an effort to meet and speak with people in that profession, and see if you can confirm that you actually want to work in that area. People’s ideas of what happens in Silicon Valley, for example, and what actually happens in Silicon Valley are not the same. This is true of every workplace and every profession. So, talk to people in the field, find out what it’s really like, and ask them what type of training will help you get that first job. There may be some real preferences, not just in terms of what type of degree you should get, but also what kinds of internships are most useful.”
The trend line for women in STEM fields, including data science and analytics, suggests a decrease in women’s participation in the workforce – According a 2015 report by the American Association of University Women, the percentage of women in computing and mathematical occupations dropped from 35% in 1990 to 26% in 2013.
The actual number of women working in computing and mathematics jobs, like data science and analytics, has been on the rise – The National Science Foundation reports that the number of women in computing and mathematics doubled between 1993 and 2010, and increase that was offset by the even larger number of men who moved into these fields.
Data science and analytics jobs exist across a broad spectrum of industries and professions which may not label those jobs as “data scientist” – “A lot of the people doing valuable, applied work in data analytics don’t call themselves data scientists and they may not even call themselves data analysts,” Meta S. Brown explains. “They identify themselves by the application.”
It is estimated that women hold over 40% of all undergraduate and graduate degrees in statistics, a discipline that is central to analytics – “It seems that statistics has generally had a higher number of women in it than computer science,” says Claudia Perlich. “I think that there is something of this sort going on, and to some extent it is a beneficial side effect of the tremendous demand for data scientists.”
There are a growing number of national and regional professional organizations for women in data science and analytics – Meta S. Brown recently spoke at the first meeting of the Chicago chapter of Women in Machine Learning and Data Science. “They are packing a pretty big room with women who have earned or are in the process of completing graduate degrees and who are intending to have a career in analytics,” she reports.
Networking is as important in data science and analytics as it is in any other growing field – “Find mentors and sponsors and network like crazy where there are role models, examples, advice givers, and influencers who can offer solid help,” advises Cornelia Lévy-Bencheton.
Claudia Perlich teaches a Data Mining for Business Intelligence course at the Stern School of Business and has a somewhat different take on some of the key factors that women should consider regarding a career in analytics. She also has a different perspective on the issue of women in the field. “In all honesty, I don’t think that I have very often even asked myself the ‘as a woman’ question. My own experience has been that since my PhD I have ended up in environments where the team around me was closer to balanced than most places. My PhD advisor had three data scientists graduating under him when I was there, and we were all women. At IBM, half of the predictive modeling group, or maybe 30% were women. And, now at Dstillery, we have at least 30% women in the data science team. I’m not sure why that is, or if that is a representative example.”
Perlich’s advice for women begins on an encouraging note. “There is the question of whether or not you should pursue a career in data science. And I am standing by the belief that there will be an almost unlimited demand for data scientists moving forward. I just can’t see any downside. So, if you have any inclination that this is the kind of work you would enjoy and that you would be good at, you should explore it.”
“What is gender specific about this?,” she asks. “The first advice I would give is, don’t ponder the question too much. At the end of the day, it’s what you’re good at that matters. And the more you ponder the ‘as a woman’ question, the more tension you create for yourself around it. My experience has been that being a woman can be beneficial. It is possible that I was lucky, or just good at selecting environments where nobody seemed to care about me being a woman. I know there are other environments where there is more implicit bias in the culture.”
Perlich goes on to stress the importance of cultivating self-confidence, not just as a defense against implicit bias, but also as an analytics skill. “If you don’t have self-confidence, then you may perceive failure as a personal shortcoming. Whereas failure in data science may just mean that the problem isn’t solvable. Maybe the data you have is just not up to the job. So, there are inherent limits and some things just don’t work, and it is important in data science to realize that that is not a personal failure. Working through what might be perceived as a failure, and appreciating failure as a learning experience or an insight is a valuable skill in data science.”
Regardless of whether or not women are well represented in analytics, there has and continues to be a great deal of energy and discussion about strategies for bringing more women into the field. Annual events such as the Global Women in Data Science Conference and the Women in Statistics and Data Science Conference are a clear manifestation of this; professional organizations like Women in Machine Learning and Data Science and Women in Big Data are another. There is also a general consensus that the entire field of analytics would benefit from the inclusion of more women at all levels of the profession.
Meta S. Brown offers a practical argument for more women in analytics. “Personally, I believe that a brain is a brain, and a female brain and a male brain can do the same things. So, theoretically it shouldn’t matter. However, real life has everything to do with culture and experience. If you want to benefit from the information that comes from the full breadth of our culture, you have to involve people who know and have experienced those things. If we only involve a narrow range of people, then we only have their narrow range of problem solving abilities. In practice, I see this. Analysts often make decisions about how to tackle particular problems based on personal experience and observation. There are analytics problems that are so complex that we need people from different life experience and different cultural backgrounds working on them. It’s an argument for diversity, not because it is ethically right, but because it is right for business. Is it also ethically right? Yes, it is. How convenient that it’s both practical and ethical.”
Brown points to realigning public perception regarding the actual work of analytics professionals as one way to bring more women into the field. “The more that the public image of data analysis is based on the computing community’s concept of data science, the worse that perception is for women. Let’s not go too far with that. Women are gravitating to the analytics profession. In fact, I have seen articles that remark on the flood of women coming into analytics. And I’ve seen that informally here in Chicago. A chapter of the Women in Machine Learning and Data Science was recently founded in Chicago, maybe a year ago. I spoke at their first meeting and I’ve been to several other meetings. They are packing a pretty big room with women who have earned or are in the process of completing graduate degrees and who are intending to have a career in analytics. They are in the kinds of educational programs that are perceived as being male dominated, and that probably have been in the past. But that may not last. Women have been attracted to analytics before, they’re attracted to analytics now, and lots of women are coming into even what I would call the most masculine flavored versions of analytics.”
It seems that statistics has generally had a higher number of women in it than computer science. I think that there is something of this sort going on, and to some extent it is a beneficial side effect of the tremendous demand for data scientists. If you are having a hard time finding data scientists to hire, you’re going to hire someone who is qualified. It won’t matter if that person is a woman. (Claudia Perlich)
Cornelia Lévy-Bencheton echoes Brown’s sentiments and sees a potential solution. “This field needs an image makeover. In our culture, analytics and the STEM disciplines are often seen as unfeminine and the purview of males, the proverbial boys’ club where only the boys can thrive. This makes many women turn away before they even start or have a chance to explore. We need to promote STEM career choices to women from a benefits perspective. Making the industry as a whole more attractive to women is partly a marketing and branding challenge, the goal of which should be to help women go down the path of working in a field they love.”
Lévy-Bencheton also encourages women to keep an open mind about what a career in analytics can encompass. “There are many variations and specialties within the data science field and many paths for career development. This is an important consideration for women returning to work after a career break for child rearing or eldercare. Women in STEM can jump back into the workforce, leveraging their skills in new ways for re-entry without the fear of being left behind. Another strategy I would recommend is to find mentors and sponsors and to network like crazy where there are role models, examples, advice givers, and influencers who can offer solid help. STEM fields are highly creative. Problem solving is paramount, and this can be very satisfying and lots of fun. For those who have the curiosity to pull things apart to find out what makes them tick, problem solving, independence, and opportunities to explore are all real benefits.”
And Claudia Perlich ended our interview with her on a distinctly positive note. “Interesting fact,” she notes. “It seems that statistics has generally had a higher number of women in it than computer science. I think that there is something of this sort going on, and to some extent it is a beneficial side effect of the tremendous demand for data scientists. If you are having a hard time finding data scientists to hire, you’re going to hire someone who is qualified. It won’t matter if that person is a woman. By the same token, if you are a women with technical interests, there’s no questions that data science is one the most promising areas to be. This is a field in which you can’t be picky about who has talent. That may push whatever gender biases people have into the background. Talent is rare, and you are glad for any that you can find.
Perlich adds, “I think women tend to be very good communicators. Data science is one of these areas where they talk about the ‘unicorns,’ or people who are supposed to be able to do statistics and math and present or communicate very well. Communication into the business side can be a real challenge in data science. Being able to do that well may not fit the idea of the hardcore data scientist working on math equations. But data communications is a challenge that involves being able grasp the bigger picture, define the problem, and explain that well. That is an aspect of data science that has been welcomed and recognized as an important component.”
"Women in Data Science Are Invisible: We Can Change That," by Claudia Perlich; Wired; October, 2014.
American Association of University Women, "Solving the Equation: The Variables for Women's Success in Engineering and Computing" 2015.
The American Statistical Association, Stanford Institute for Computational and Mathematical Engineering, and Anita Borg Institute’s Grace Hopper Celebration (GHC) of Women in Computing all hold conferences for women in data science and analytics.