Listen to the podcast:
Every week, we talk about important data and analytics topics with data science leaders from around the world on Facebook Live. You can subscribe to the DataTalk podcast on iTunes, Google Play, Stitcher, SoundCloud and Spotify.
This data science video and podcast series is part of Experian’s effort to help people understand how data-powered decisions can help organizations develop innovative solutions and drive more business.
To keep up with upcoming events, join our Data Science Community on Facebook or check out the archive of recent data science videos. To suggest future data science topics or guests, please contact Mike Delgado.
Karen Matthys is the Executive Director, External Partners at Stanford Institute for Computational Mathematical Engineering. Judy Logan serves as the Co-Director of the Women in Data Science Conference at Stanford University.
Mike Delgado: Hello, and welcome to our weekly Data Talk show, featuring some of the smartest people working in data science today. It’s actually a very special broadcast. We have two people joining us. We generally just have one guest, but we have Karen Matthys, who is the Executive Director of External Partners at Stanford Institute for Computational Mathematics and Engineering. We also have Judy Logan. She serves as a Co-Director of the Women in Data Science Conferences at Stanford University. For those who are watching live, feel free to post your questions, your comments. We’ll be posting some of them here on the screen. Karen and Judy, thank you so much for joining us today.
Karen Matthys: Sure. Our pleasure.
Judy Logan: Thanks for having us, Michael.
Mike Delgado: Before we jumped on, you were sharing a little bit about the room you’re in because I do see some … looks like some equations in the background. Where are you located right now?
Karen Matthys: That’s a great question. We’re in the Institute for Computational and Mathematical Engineering at Stanford, which is a mouthful, so we call it ICME. ICME is the institute that launched this Women in Data Science Conference. This is the Gene Golub Room, named after a very famous mathematician who started the predecessor of ICME, our institute, and was the chair of the CS Department here for many years and was beloved by all who knew him.
Mike Delgado: Wow, amazing. Well, it’s wonderful, everything that the organization, that you guys, are doing to help expand women getting involved in data science. For those who are listening to the podcast, I want everyone to check out widsconference.org. That’s where you can learn more about the work Karen and Judy are doing to help bring more women, more girls into data science. There’s a huge gap there — something we definitely discussed in past Data Talks — so it’s exciting to see that. I stumbled upon your website when I was just searching on Google for Women in Data Science. All of a sudden, boom. I saw your conference. I couldn’t believe it.
Judy Logan: I think it’s the top search.
Mike Delgado: Yeah, you guys showed up right away. Unfortunately, the last time I was searching, I just missed it. It was the 2017 conference and I just missed it. I was like, “Oh, my gosh, what an amazing event.” You brought together amazing data science leaders from all over. For those who are listening to the podcast, definitely check out widsconference.org. Amazing things happening there. Before we get started, can you share a little bit about the mission, why the conference started and what you hope you achieve this year?
Karen Matthys: Sure. It’s just been a few years since we launched this conference, so it’s just fantastic to see it grow by leaps and bounds. I think now we’re hitting about 100,000 people worldwide.
Mike Delgado: Wow, really?
Karen Matthys: Yeah. So it’s really cool to see it from its infancy just a few years ago when a few of us were sitting around a table in the local coffee shop, discussing tangible ways we could move the needle and have an impact on this persistent challenge with getting more diversity into fields like data science. We’re discussing some of those challenges with women and gender issues in the field. I’m an engineer, and we were talking with one of our alumni from ICME about this. We thought this would be a great way to do three things, basically, in our mission. One is to inspire the next generation of women who are considering careers related to data science and also to inspire those who are already in the field to go to the next stage in their careers by showing some rock star women already in the field doing amazing things and a real technical conference.
So, inspire, and we wanted to educate this community about the latest technologies and latest uses of data science and machine learning AI across many fields and means. We wanted to connect people because we know that’s really powerful to get people feeling supported and connected. We wanted to provide a forum to do that.
Mike Delgado: That’s wonderful.
Judy Logan: The only thing, Michael, is that this is a truly global conference, so when we started out, we were at Stanford and we ended up having a really long waiting list. So, we added the live-stream, and much to my surprise 6,000 people showed up on live-stream …
Mike Delgado: Really?
Judy Logan: … without any real promotion.
Karen Matthys: It’s a real surprise.
Mike Delgado: Really? 6,000 tuned in with no promotion? That is amazing.
Judy Logan: It really shocked us.
Karen Matthys: It was great.
Judy Logan: We were excited about that, and we realized we had an opportunity to have a bigger impact. That’s why we actually went global with our next conference in 2017. We recruited all the Woods ambassadors from all over the world, and, again, that response was way more than we expected. We ended up having 80 events worldwide in every continent except Antarctica. We ended up … actually between online and regional events, Karen, we reached 75,000 people.
Karen Matthys: Yeah, some of our sponsors are calling it a movement now.
Mike Delgado: I’m sure.
Karen Matthys: I think we’ve struck a chord, and I think it goes back to the fact this community in particular is really hungry for this kind of program. It speaks to a great area of interest by women around the world — an area where they see a lot of opportunity, but they also see some challenges. Hopefully, with this conference and some others, we’re extending it this year, so it’s not just a conference. We’re doing a podcast series and we have a datathon. Hopefully, through all of this, we’re hitting our goals to energize women around the world to be involved in this field.
Mike Delgado: That’s awesome. So, you’re not in Antarctica?
Karen Matthys: Still working on it. So, if you know of somebody there, he is … I think it’s a bandwidth issue. So if someone there you know is willing to take the live-stream, we’re all in favor of it.
Mike Delgado: That’s awesome.
Judy Logan: We are in a lot of great places. We’re going to be, for instance, in both Serbia and Croatia.
Mike Delgado: Oh, really?
Judy Logan: We will also be around the Middle East, and Jordan, and Beirut, and three locations and the Kingdom of Saudi Arabia.
Karen Matthys: One of my favorites is in Peru. We’ve been doing a datathon, and we have so many teams coming out, teams involving a lot of women doing a datathon in parts of Lima.
Mike Delgado: Can you describe what a datathon is? What is happening in Peru?
Karen Matthys: Okay. We decided this year to go beyond just this one-day technical conference, which we think is important, but we want to continue to build momentum and excitement throughout the year. So, leading up to the conference, we decided we would sponsor a predictive analytics challenge. This is our first time doing it. Thanks to some support from Kaggle, which is part of Google Now. The Kaggle team has been extremely helpful. We worked with the Gates Foundation through one of their intermediaries called InterMedia Data. We’ve collected or have their data set around some impoverished areas in developing countries.
The datathon is predicting some elements in that data set. It’s open to women around the world. Actually, we just required that it’s 50 percent or more women on the teams. We’re trying to encourage more women to get involved with datathons, which traditionally have been heavily male-dominated.
Mike Delgado: You mentioned a little bit about the challenges women are having in data science. Can you talk a little bit about the challenges and how this conference is helping them fight through that?
Karen Matthys: Sure. How much time do you have? I think the main thing is that there’s not one magic bullet here from what we’re hearing and seeing. There are challenges starting with young girls, and then there are challenges leading up through women even when they’re middle level, senior level in their career. I’ll just throw out a few examples. The latest data from the national clearinghouse that covers students graduating from undergrad as well as graduate degrees shows this persistent challenge of getting more than around 20 percent women in fields related to CS. So in 2006, the percentage was 20 percent undergrads graduating, and in 2016 it was 19 percent.
Mike Delgado: Really?
Karen Matthys: I mean, there are more women in terms of actual numbers, but percentagewise, it really hasn’t changed. It’s similar in engineering fields. I think it went from 21 percent in 2006 … no, 19 percent in 2006, 21 percent in 2016. A lot of that increase is in the biomedical engineering area. In some of the areas directly related, the computational math and engineering, it really hasn’t changed at all. And so, we’re seeing this persistent problem of really getting to critical math in those fields. The younger women and girls need to be told, “You can do this. It’s cool to be in math. It’s cool to be in computing. These are great fields for women.” Telling them that, encouraging that even if it’s hard to do, you can do it, and then showing role models.
We really think this conference, highlighting these incredible role models, gives girls and women of all ages something to shoot for or something to say, “Oh, I can do that. If they can do that, I definitely see myself there.”
Judy Logan: Michael, I think one really important aspect in this conference is that people actually are focusing on the work. It’s not a women’s issues conference. This is about technical work. This is about research and applications happening today by these rock star women. These rock star women are definitely coming to Stanford for the event, but they’re also at the regional events worldwide. There are rock star women all over the world doing amazing work, research and applications in government, nonprofits, industry, academia, all over the place.
Karen Matthys: One thing we found building on that is … I know, certainly, being in engineering, I went to a number of conferences when I was younger that would be 100 percent male speakers. That was not unusual. Even in the past few years, I’ve seen events in other universities where there are 100 percent male speakers on data science–related events. Often when we asked organizers, “Did you consider having some women on that panel or in that speaker lineup?,” they’ll name one woman. They’ll say, “We tried for so and so, and she wasn’t available.” That’s the end of the story. I don’t think these women are getting their fair share.
One thing we’re doing is vetting these speakers and elevating and giving more attention to these women around the world. We’re collecting a database so that, hopefully, we can publish it soon and share with everyone. Here are incredible women in all these different areas related to data science, whether you want biomedicine, financial services, oil and gas, or you name it. They’re out there, and we can share that with everyone, so there should never be the excuse, “Oh, I couldn’t find a woman to be on this panel for this event.”
Mike Delgado: There you go.
Judy Logan: In addition to creating the database of speakers, we’re also really a video repository of not only the Stanford presentations, but also we’re inviting all the Woods ambassadors around the world to videotape their sessions, too, and then share those on a YouTube channel. That’s a worldwide YouTube channel with more work to categorize those videos so they can be used so somebody from Beirut might be able to be featured in something in Buenos Aires for the next meetup if there’s a particular area of interest. For instance, there is this fascinating talk about computational finance that was in Beirut last year. That would actually have applicability all over the world. We just want to highlight these women no matter where they come from, no matter where they’re doing their work, that their work is actually getting a lot … a little bit more attention than they would get otherwise.
Karen Matthys: Yes, and we have the videos from last year and the initial year on the website already and we have a YouTube channel, so anyone can check them out, find interesting speakers. Often, we find that for future events, people say, “I’d like to contact this person and have them speak at my event.” So, we’re seeing a lot of that follow-up activity where these speakers are getting known far and wide outside of their normal areas.
Mike Delgado: For those just joining our live broadcast, I want to encourage everyone to check out widsconference.org. This is the website that Karen and Judy are referring to. This is the Women in Data Science Conference held at Stanford and now all over the world, as you just heard. So powerful to hear about the number of speakers you have and just go see the speaker list. I mean, scroll through the website. You will be absolutely amazed at all these different powerful women doing some aspect of data science who are all thought leaders. They’re going to be speaking at this event, so definitely check it out, widsconference.org, for those just joining.
Yeah, it’s very, very powerful. Karen, you’re talking about just the importance of having someone to look up to, a role model. That is so powerful because like you said back when you were in school, all the speakers were men. How does that make you feel when everybody is a man, and you’re a woman, and you’re like, “Where is my place?”
Karen Matthys: Right. I talked to so many women who struggle with this today or have struggled with this in the past, and it’s great to see those role models. It’s great to have mentors you can reach out to, male or female, mentors and supporters to encourage you as you’re going through different levels of your career. We’re hoping this is also a way to connect people. We have a LinkedIn group. We have other ways to connect the people both online and in person at these local events. I think what’s essential, too, is to give them the connections to build that net of support so they can shine — whether it’s in academia, or government, NGO or business — in what they’re doing.
Mike Delgado: That’s wonderful. What advice do you have for women who want to pursue a career in data science? I read this really interesting stat that said — and I don’t know if you’ve heard this — that 60 percent of undergraduate students are women who are graduating statistics. So of those graduating statistics, 60 percent are women, but yet a lot of them don’t go on to pursue data science careers. That’s a much higher number than men.
Karen Matthys: It is higher, but actually the number … the percentage is a little bit lower than that and has been going down a bit …
Mike Delgado: Oh, really? Okay.
Karen Matthys: … in the last few years in math as well as statistics. We’re tracking that. I can send the link to that report, that snapshot report from the Student Clearinghouse, because we’re quite concerned that that doesn’t go down too much farther, but, yeah, that data is out there.
Judy Logan: See, we got somebody who’s going to be in Torino. Alba, very excited. You’re going to be at Woods, Torino in Italy. Awesome.
Karen Matthys: Yeah, that’s great.
Mike Delgado: What’s going on in Torino?
Judy Logan: In Torino, that’s an event that’s being put on by the ISI Research Institute in Italy. We’ve got a research scientist who’s one of our Woods ambassadors. She’s putting on an event in Turin, Torino, whichever name you like. Daniela is the name of the woman who’s putting it on. She’s doing a fantastic job. She’s going to have some live speakers and combining that with some of the thought leadership. So, we’re thrilled that …
Mike Delgado: That is so cool. I just love how within a short period of time, your message, your movement has gone global.
Judy Logan: Yeah.
Karen Matthys: We try to make it easy for these groups around the world by providing a conference in a box of virtual tools for them to set up whatever kind of event works for them. So, some events, they just simply live-stream what we’re doing, and that’s great. Other events, they want to have their own speakers or intersperse their own speakers with some live-stream. And then other groups, they come up with all sorts of creative hands-on activities and events even before our conference or within a few weeks afterward. We give them the freedom to do whatever they want as long as they feature women doing technical work.
Judy Logan: That’s our one and only requirement is that all speakers, all panelists, all moderators, all emcees, everybody who’s on the stage needs to be female — and they’re in some area related to data science, whether that’s machine learning or artificial intelligence, some area of research applications.
Karen Matthys: Yeah.
Judy Logan: That’s what we’re sticking to. We’re very strict about it.
Karen Matthys: Another message that comes to mind that we’re trying to get out and convey by sharing all these different types of speakers is that there’s not one straight and narrow path into data science. So back to advice that you give to others who are getting into this field, or our daughters, or what have you, we’re trying to feature women who come at data science from very different backgrounds, which I think is great because it’s a nonlinear field to be in. It’s great to see people with background, say in medicine, getting into this area, or someone in education getting involved in data science, or they just come from all over and they’re all ages. We’re trying to share with people that there’s lots of ways to become a data scientist or get involved in the field. You just have to explore what makes the most sense for yourself.
Judy Logan: Even if you look at the speaker panel for Stanford this year, you’re going to see that there’s a particle physicist on the agenda. So, she’s thinking about … if she’s using data science to open up the universe and … to Karen’s point about the nonlinear paths, I mean, you can approach data science from a bunch of different directions in the right way. There’s no one right way. There are a lot of right ways.
Mike Delgado: I like that. There are a lot of right ways.
Karen Matthys: It’s really fun. At the end of the conference, people always get really excited. I mean, they get all these … hear all these talks from these incredible people and the energy and the buzz in the room is palpable at the end. So, it’s an honor to be part of this and then to make it all happen.
Mike Delgado: I love it. I was just at an event recently and they used this metaphor I like. Some people would say to become a data scientist, it’s a marathon. There’s a lot of work that’s involved, but going to an event like this is a sprint because you make these connections, you’re learning a lot. And so, how long this little marathon … I’d say, the Women in Data Science Conference is a sprint, where you’re getting a chance to get a lot of information very, very quickly, making some relationships, finding some mentors to help lead you. So, very, very cool. We just saw a note here from Aaron on Facebook saying that he just checked out your website, looks really, really cool, so awesome.
Again, for those just tuning in, it’s widsconference.org. Just amazing to see the lineup you have there. Again, congratulations to you both for creating a movement so quickly. It’s so exciting to see how many people are getting involved around the world to share the live-stream, to share the videos, to show more role models, to make people realize that, “Hey, you know what, maybe data science is for me.”
Karen Matthys: Yes, that’s the message. Now, with 100,000 people expected who we think will participate in the conference this year and all the surrounding events and the live-stream, the question is how we harness and gather all that energy worldwide and move it to the next step. We’re already planning new things associated with WiDS to catalyze an even greater impact here.
Judy Logan: It’s all about impact.
Mike Delgado: No kidding. Well, you are making an impact, no doubt. I mean, it’s amazing. Now, for those who are in the area of these different events, what level, what … I mean, because some people might feel like, “I really don’t belong there because I don’t have a math background.” Speak to the women, the girls who are listening and who are like, “That sounds intriguing, but I’m looking at the speaker list and the subject matter and I don’t know if that’s for me because it just seems too above my head.”
Judy Logan: Well, it is a technical conference. So, there is that, but there are different levels of accessibility with different presentations that you’ll find. For instance, in 2015, we had a great presentation from Caitlin Smallwood from Netflix. She gave a great talk, very accessible, but also very rich. She balanced it quite well.
Karen Matthys: Another great one was Diane Bryant at Intel at the time, who gave a great talk about what Intel was doing and talked about implications for health-related issues like Parkinson’s disease. It was powerful, and I think everyone in the room could relate to that regardless of their technical level. So there’s really something for everyone in that. And then what we’re thinking as we hear afterward, too, people get very excited and they say, “How can I go on? What should I do next to get into this field?” So, we want to be a resource for sharing more information about where you would take a course online or where you would go to another conference, or a workshop, or something in your local area so that you can get more involved and go to the next stage. So, I think that’s part of what we’re trying to do here is after we get everyone excited and motivated, really encourage them to take the next steps, however big or small, and get involved in some way.
Mike Delgado: Karen and Judy, parents have a big role to play in how they raise their daughters, and I’m curious about your advice for the parents listening in but have young children and would like to foster a love for data, for mathematics, for computer science, data science. What would be your advice for them to help their children along?
Karen Matthys: We both have daughters who have been friends since they were young.
Mike Delgado: Oh.
Karen Matthys: It’s a great question. I’ll take a stab and you probably, Judy, will have more to add there. I think encouraging them to do whatever they want, to go into whatever field and not give up, to work hard, and encourage them if math is their thing or even if they’re just on the fence to really pursue something around math, computer science. At least try it out, take some courses that they might need a little extra encouragement or try that summer camp. I also think that data science is interdisciplinary. It’s important to have diversity in data science, and it also lends itself well if you have good team skills, team building skills, and communication skills. Leadership skills are important. I would encourage the parents out there to encourage their daughters to work on those skills as well as the hard science skills.
Judy Logan: Yeah. I would also add that I think it is important, in particular with girls, I think society sometimes … the societal pressure to say that you’re not good at math or it’s maybe not cool to be good at math. I think it does take a little bit more encouragement along the way to say, “Yes, you can. Yes, you can do it.”
Karen Matthys: You can do it. Yes.
Judy Logan: And then I find that my own daughter is an accidental data scientist …
Mike Delgado: Really?
Judy Logan: … because she’s an undergraduate and she studies international relations, but she’s been doing policy work. Guess what? She learned how to program last summer and do research work. So, I said, “Well, like it or not, honey, you’re a data scientist.”
Karen Matthys: That’s a great point as well. In fact, I regret that I didn’t take statistics when I was in high school. I would recommend that all students take statistics, because even if you don’t think you’re going into a field related to data science, it’s touching every field and domain these days. It’s worthwhile to get the basics of statistics when you’re in high school, when it’s a little bit, I think, easier than when you’re out in the working world and trying to find time for that.
Mike Delgado: Yes. In fact, I was reading a list … Bill Gates puts out a list of favorite books that he likes every single year, and one of his books that he had was the book … the book has a funny title. It’s called How to Lie with Statistics. It’s a small blue book, and it’s basically meant to teach you what statistics is and how to avoid being fooled by statistics.
Karen Matthys: Right.
Judy Logan: Yeah. That’s a really good point, Michael, which actually tells you why it’s so important to have different voices in data science and to have diversity because data is malleable. The numbers are malleable. A lot of people think there’s … the numbers are cut and dry and that’s just not true. You can actually … the view you take on the data is going to be different based on your life experiences and what your background is in terms of studying studies as well. So, it’s important to have those multiple touch-points, have those multiple people looking in the data and providing those different perspectives.
Karen Matthys: Yes, absolutely. In fact, in our institute, ICME, we are trying to educate our students around data science, machine learning and the potential unintended consequences.
Mike Delgado: Wow.
Karen Matthys: So, how do you think about bias and ethics upfront when you’re working with data and solutions that involve algorithms? These are important issues, and having that diversity of thinking from the start in the design process is critical.
Judy Logan: I think Latanya Sweeney is going to touch on at least those topics.
Karen Matthys: Few of our speakers are involved in that at the conference. So, Latanya Sweeney, who’s the Director of the Harvard Privacy Lab, and then last year Elena Grewal led a breakout session. She’s a Head of Data Science at Airbnb.
Mike Delgado: Oh, wow.
Karen Matthys: She led a breakout session last year that was great.
Mike Delgado: That’s so cool.
Judy Logan: She’s been a real leader in that effort and she needs to be, but she’s really taking to that.
Karen Matthys: And many more of the speakers — Jia Li is speaking from Google and she’s speaking on the democratization of data, which is a really interesting topic and also plays into this. So, if everyone is accessing the data or using it in some way, how do you avoid, like you said, those lies in the data or how do you avoid misconstruing the data and not using it in a proper way?
Judy Logan: Not inserting bias. In fact, we are trying to remove instead of trying to add more. I think that’s very important.
Mike Delgado: Karen and Judy, I know we have to go, but we do have one question from Arron. He says, “As a male undergraduate student here in the UK who’s passionate about data science, getting more diversity into the community, what can I do to get more people, especially young women, interested in and excited by data science? Is it just a case of hosting a talk, or is it more than this?”
Karen Matthys: Oh, great question.
Judy Logan: Great question.
Mike Delgado: I love it, Arron.
Karen Matthys: Of course, we would say that when we have these events associated with WiDS around the world, the person who’s hosting it, we call them an ambassador, can be male or female. So, that’s a great opportunity if you want to host an event in your own community. We welcome everyone to organize an event, whatever makes sense for you at your school, and we provide a lot of tools and support for you to do something successfully.
Judy Logan: We also have a lot of good regional events already happening in the UK. There’s an event that’s happening March 6th in London, and then there’s an event about a week later at The Alan Turing Institute in London. We’re also going to be in Reading at a meet-up organization and then in Buckingham. That’s actually later. That’s not happening until July. So, there are some good opportunities, but you can also invite them just to tune in. This is actually on live-stream on Facebook Live. Encourage all of the women you know, anyone you know, because we want men, women, everybody. We want to have access to these amazing talks. Join us on the live-stream. Encourage your friends to join on the live-stream.
Karen Matthys: They can get the live-stream either from the WiDS website — and they can just go there on the day of the conference and live-stream it that way — or if you like, we can give you the embed codes and you can share that with your own community on your own website. There are a lot of ways to be involved, and if you can help us get out the word, we’d really appreciate that.
Judy Logan: We’re also going to be broadcasting on Facebook Live from our WiDS 2018 Facebook page.
Mike Delgado: Well, I’ll be tuned in. I’ll be sharing, Arron, so you got a call to action.
Karen Matthys: Yay.
Mike Delgado: Get involved. Sounds like a great opportunity to help spread the word. Again, for those listening to the podcast, check out widsconference.org. See what Karen and Judy are up to. If you want to get involved, that’s the place to go. Judy and Karen, any final words?
Karen Matthys: Yeah, thanks for your support.
Judy Logan: Thank you. Aaron, you’re part of the solution and so is everybody else who’s listening, so thank you, Michael, for your time today.
Mike Delgado: Thank you. Thank you so much for sharing your insights and thank you for being our special guests in Data Talk. For those who are new to Data Talk, we host these chats every week, sometimes twice a week. We’re talking about all types of important data science topics, and this is certainly one of them, the gender disparity within the data science community. So, very grateful to have Karen and Judy from Stanford here. Alba just says, “Thank you, thank you, thank you for the talk, Karen and Judy.”
Karen Matthys: Great to talk to you, too. Thanks for the questions.
Mike Delgado: We’ll definitely keep in touch. Here at Experian, we’ll make sure to be retweeting and resharing the things we’re seeing coming out of your conference.
Judy Logan: Sounds good. Thank you. Appreciate it.
Karen Matthys: Excellent. Appreciate it.
Mike Delgado: Great. Take care, and we’ll see you later on live-streams.
Karen Matthys: Bye. Take care.
Judy Logan: Bye-bye.
The Global Women in Data Science (WiDS) Conference aims to inspire and educate data scientists worldwide, regardless of gender, and support women in the field. This annual one-day technical conference provides an opportunity to hear about the latest data science related research and applications in a broad set of domains, All genders are invited to participate in the conference, which features exclusively female speakers.
Check out our upcoming live video big data chats.