On Being a Female Data Scientist
How Did I Get Here?
I’m a female mathematician, statistician, and data scientist. How did I get to be all of those things? I wish I could say “I did well in school” or “I always loved computers.” But the reality is, in middle/high school, I was bored stiff. I excelled in art, skipping classes, and bombing exams. At 16, I dropped out of school to begin an illustrious career in office cleaning.
The odds were stacked against me entering the computing industry for many reasons. The biggest obstacle standing in my way? Being female, thanks to a decades-long initiative by the British government to keep women out of computing.
The Good Ol’ Boys Club
The year 1967, the year I was born, should have been a heyday in Britain for women in the blossoming field of computer science. Many female role models emerged during and after World War 2: code breaker and cryptanalysis expert Joan Clarke, who helped Alan Turing break the Enigma Code; American Grace Hopper, who invented COBOL; Jean Bartik, one of the original programmers for the ENIAC computer. In the early 1960s, women were very well represented in the computer sciences and in fact, “…most of the computer operators, programmers, systems administrators, and systems analysts that you were likely to encounter would have been women” (Hicks, 2017).
This may sound like women were in important positions of power, but that was far from the case. If you’ve seen Hidden Figures, you’ll know that prior to the electronic computer, there were human ones: mathematical whizzes who performed complex calculations using advanced calculus, geometry and trigonometry. Yet, the work was seen as dull and repetitive–perfectly suited to women, especially those in the working (blue-collar) class. As mainframe computers became integrated into every corner of business, from tax databases to elections, women rose to the more technical challenges. They migrated from pen and paper to keyboard, and they weren’t just using computers–they were creating code, analyzing data, and maintaining the mainframes. It was a female-dominated industry.
However, women were forbidden from rising up the ranks, and they were paid a lot less than their male colleagues. This low pay/high skill is what Marie Hicks, author of Programmed Inequality: How Britain Discard Women Technologists and Lost its Edge in Computing describes as a “mismatch”. After a series of events, including a strike by women demanding equal pay, “…the [British] government began a decade-long effort to remove women workers from these newly-important technical positions.” They were replaced by men, who were perceived to have better abilities when it came to complex technical work and managing people.
By 1978, the year I entered secondary school, a decade had passed since the government initiative to replace women in the computer industry. Nearly all computer-related jobs were filled by men. As a working class female, I was expected to migrate towards jobs “suited” for a woman of my stature: factory work, bank or office work, typist.
When I left school at age 16 in 1983, I knew how to type (on a manual typewriter), use a slide rule, and efficiently empty a garbage can. At age 18, I was fulfilling societal expectations and cleaning offices for a credit card company.
In 1985, while on a break from cleaning, I saw a notice for a job opening. “Do you like to solve logic problems? Call us to take an aptitude test and become a computer operator!” I must have asked a dozen people what a “computer operator” was (these were the pre-internet days, when you asked people, not Google): no one had a clue. Intrigued, I applied for the job. The hiring manager was impressed by my scores on the aptitude test, and less impressed by my complete lack of qualifications (three CSEs in art, geography, and math). But, as luck would have it, the tide was turning for attitudes towards women in the workplace (I’ll credit Margaret Thatcher, the first female British prime minister for that one) the company was actively seeking to hire women. And I was the only woman who applied for the job.
If I hadn’t seen that intriguing ad, I might still be cleaning offices in England for a living. But that lucky break was the start of a successful career which has, so far, spanned more than 30 years.
As a mainframe computer operator for Access credit card company, I worked shifts (nights, evenings, days) with an all-man crew. I mounted tapes (you know, those old reels that spin back and forward, have flashing lights, and appear in the “lets blow up Skynet” scene in Terminator 2), ran batch jobs, and fed reams of paper into laser printers. During breaks, I was one of the boys: I played pool and drank beer at lunch (yes, this was the 80s), made lewd jokes, and played cards with the boys to a backdrop of porn movies. I said nothing: I wasn’t about to lose the most incredible opportunity of my life by not “fitting in.” I excelled at my job, and earned some significant pay raises. But as much as I enjoyed the work, I was itching for more. In 1990, I started a new position with Ford Motor Company as a network analyst.
My (Attempted)Move Into Management
At Ford, I was the only female network analyst. I worked as hard, if not harder than the guys to prove I could hack it–but years of working in an ‘Ol’ Boys Club’ atmosphere gave me a tough exterior. I was quick to snap a retort, swore like a trooper, and had a temper that blew like a landmine. But I was very good at my job. By this time, I was in my early 20s, and realized I didn’t want to spend the rest of my life as a network analyst–I wanted to get into management. I enrolled in part-time college classes, told my boss I was aiming for management, and then got an unexpected surprise on my yearly appraisal. I will never forget my boss’s words: “…you’re too brusque; You’ll never be a manager.”
The door to advancement slammed shut.
Around 1992, Ford underwent a series of redundancies (lay-offs), reducing the workforce by several thousand. Voluntary redundancy was offered to almost anyone willing to take it, and I jumped on board, using the money to emigrate to America. I’d spent many weeks backpacking from NY to LA and I’d fallen in love with the place. I was off the the Land of Opportunity.
The American Dream
In the United States, women didn’t fare much better than their British peers. Women struggled to enter the male-dominated workforce in the 1960s (just watch a series of Madmen to find out some of the reasons why), but the the numbers of women in computing rose steadily. In the 1980s, around the time I was cleaning offices, the numbers of women in computing dropped sharply, but not because of a government program to push them out. In an article titled When Women Stopped Coding, NPR reports that one possible answer is that “The share of women in computer science started falling at roughly the same moment when personal computers started showing up in U.S. homes in significant numbers…and these toys were marketed almost entirely to men and boys.”
So I had essentially moved from a country where women were actively pushed out of a male-dominated field, to one where women were welcome if theyhad been lucky enough to have been exposed to computers. Luckily for me, I had been exposed to computers, albeit in my late teens–through that intriguing job ad. The computer rooms I worked in had as many women as men. And, possibly because I was aware of my “brusqueness”, I toned down my act and did my best to fit in.
When I Became a Data Scientist
The single most important step for me to becoming a data scientist was getting my masters degree in mathematics education. That might seem like an odd statement, as I’m sure none of you would define a data scientist as one who holds a degree in something education related. However, what happened at that moment was that people’s attitudes towards me changed to one of respect. That respect didn’t come from all corners.
After graduating with my masters in 2006, I started on my PhD and taught part time at a local university. I started StatisticsHowTo.com to help some of my students who were struggling with statistics. Some of my colleagues noted it’s “lack of professionalism and mathematical rigor”. Other problems surfaced, including the fact that chapters I wrote for a textbook collaboration never made it to the published copy, with no reasons given. It’s possible that I didn’t have the respect of my colleagues because I had a masters degree, and not a PhD. It could be because I thought a little outside the box when it came to teaching. There could be another reason that I’m not aware of. But that is the story of women across these last few decades in any profession: there’s always a “good”, reason why, as a woman, you’re sidelined, disregarded, or passed over for promotion.
I left the world of academia and dropped out of my PhD program, very disillusioned. The next few years were spent solo, building StatisticHowTo.com from scratch. Over the next few years, it would become the most popular statistics education site in the world. How? Mostly thanks to Analytics, and countless hours poring over data in search of trends. But It wasn’t until I started to sideline as a statistical consultant (mostly analyzing corporate data and looking for trends) that I realized my skill set probably fits at least a half dozen definitions of “data scientist.” For example, IBMs Anjul Bhambri thinks that “A data scientist is part digital trendspotter and part storyteller ” while Tableau Software’s Pat Hanrahan“ defines “data scientist” as covering “almost everyone who works with data in an organization. At the most basic level, you are a data scientist if you have the analytical skills and the tools to ‘get’ data, manipulate it and make decisions with it.”
The benefits for being a self-employed female statistician/data scientist are many. Perhaps the most important to me is that, as a parent, I am able to spend a lot of time at home with my children. Most of my consulting work can be performed late in the evening. My “office” is an art studio five blocks from my house (for years, it was the cafe in a local used bookstore which was just as pleasant, if a tad less private).
Women are still very under represented in Data Science. A recent Forbes article states, “Women hold only about 26% of data jobs in the United States. There are a few proposed reasons for the gender gap: a lack of STEM education for women early on in life, lack of mentorship for women in data science, and human resources rules and regulations not catching up to gender balance policies, to name a few.” My advice for any woman considering entering the field is that, if you love logic problems, and want a job where you feel valued, you should absolutely consider data science. It will change your life.
Data scientist, statistician, or dabbler: Whatever the label you want to apply to me, it’s been a long journey but I’m proud to be part of the DSC community. And if I never smell another coffee drenched, left-over lunched, paper filled garbage can again in my life, I will die a happy woman.
Miller, J. (2014). Joan Clarke, woman who cracked Enigma cyphers with Alan Turing. Retrieved February 7, 2019 from: https://www.bbc.com/news/technology-29840653
Hicks, M. (2017). When computer programming was ‘women’s work’. Retrieved February 5, 2019 from: https://www.historyextra.com/period/20th-century/the-changing-role-of-women-in-british-computing/
House of Commons Information Office Women in the House of Commons: House of Commons Information Office Factsheet M4 Appendix B . Retrieved February 7, 2019 from: https://www.parliament.uk/documents/commons-information-office/m04b.pdf
Zhang, V. (2017). Breaking Down The Gender Gap In Data Science.
Retrieved February 8, 2019 from: https://www.forbes.com/sites/womensmedia/2017/08/03/breaking-down-the-gender-gap-in-data-science/#288a338c4287
Link: On Being a Female Data Scientist